Modulation of the decay of Ca²⁺-activated Cl⁻ currents in rabbit portal vein smooth muscle cells by external anions

Abstract

1. The effects of external anions on the decay kinetics of Ca²⁺-activated Cl⁻ currents (I_Cl(Ca)) were studied in smooth muscle cells isolated from rabbit portal vein using the perforated patch whole-cell voltage clamp technique.

2. In normal NaCl-containing external solution the decay of spontaneous Ca²⁺-activated Cl⁻ currents (STICs) and Ca²⁺-activated Cl⁻‘tail’ currents (I_tail) was described by a single exponential with a time constant (τ) that was prolonged by external anions which are more permeable than Cl⁻ (Br⁻, I⁻ and SCN⁻) and accelerated by less permeant anions. However, intracellular I⁻ did not affect the τ of STICs and I_tail.

3. There was a positive correlation between the ability of an external anion to affect the decay τ of I_Cl(Ca) and its permeability relative to Cl⁻.

4. The voltage dependence of STIC and I_tail decay was not affected by external or internal anions.

5. External permeating anions were not obligatory for activation of I_Cl(Ca) and STIC τ was not altered in Cl⁻-free external solution.

6. Modulation of τ by mole fractions of SCN⁻ and Cl⁻ ions was fitted by a logistic curve, suggesting competition between SCN⁻ and Cl⁻ ions for a binding site.

7. In conclusion, external anions affect the decay of I_Cl(Ca) by a mechanism compatible with an interaction with a binding site which modulates Cl⁻ channel kinetics.

Ca²⁺-activated chloride currents (I_Cl(Ca)) have been recorded in many types of smooth muscle and it has been proposed that this conductance may be involved in agonist-induced contraction and some forms of spontaneous activity in smooth muscle (see Large & Wang, 1996). Apart from its Ca²⁺ dependence, little is known about the gating of I_Cl(Ca) because the unitary conductance is small and in isolated patches single channel activity runs down rapidly, which precludes systematic investigation of the gating of I_Cl(Ca) at the single channel level (Klöckner, 1993; van Renterghem & Lazdunski, 1993).

Some insight into the gating of I_Cl(Ca) has been obtained by studying macroscopic currents, and in particular spontaneous transient inward current (STICs). These are randomly occurring Ca²⁺-activated chloride currents which are triggered by the sporadic release of Ca²⁺ ions from the sarcoplasmic reticulum (Wang et al. 1992). In rabbit portal vein STICs have a rise time of about 40 ms and decay exponentially with a time constant (τ) of 80-100 ms. We have argued previously that the STIC time course is longer than that of the intracellular calcium signal and therefore the decay of STICs reflects kinetic properties of Ca²⁺-activated chloride channels in rabbit portal vein smooth muscle cells (see Hogg et al. 1993 and Large & Wang, 1996 for a full discussion). Subsequently in vascular smooth muscle localized calcium signals (‘sparks’) were recorded just beneath the sarcolemmal membrane (Nelson et al. 1995) and it is likely that these signals activate STICs. The rise time and half-decay time of calcium ‘sparks’ were about 20 ms and 40 ms, respectively, i.e. briefer than the time course of STICs and therefore these results support the proposal that STICs may be used to study the gating of I_Cl(Ca). Consequently the voltage-dependent mono-exponential decay of STICs can be explained in terms of a minimal model where Ca²⁺-activated chloride channels exist in two states, open and closed. This hypothesis was supported by a later study in which I_Cl(Ca) was manifest as tail currents evoked by the influx of Ca²⁺ ions through voltage-dependent Ca²⁺ channels in response to brief depolarizing pulses (Greenwood & Large, 1996). It was found that Ca²⁺-activated chloride tail currents (I_tail) also decayed exponentially and the voltage-dependent decay time constant was similar to that of STICs (Greenwood & Large, 1996).

In a previous investigation where we were studying the ionic selectivity of STICs and I_tail, it was apparent that substitution of external Cl⁻ ions by other anions not only altered the reversal potential but also had a profound effect on the STIC and I_tail decay time constant. In this report we describe these effects in detail and demonstrate that anions modulate the time course of STICs and I_tail by binding to a site accessible from the external surface of the membrane. Moreover it appears that the alteration of the STIC time course is related to the relative permeability of the anions and a link between ion permeation and gating of I_Cl(Ca) is discussed.

METHODS

Experiments were carried out on smooth muscle cells freshly dispersed from rabbit portal vein. Female New Zealand White rabbits were killed by overdose of sodium pentobarbitone injected into the ear vein and portal veins were excised, dissected free of fat and connective tissue and then incubated in physiological salt solution (PSS) containing 50 μM CaCl₂ at 37°C for 10 min. Tissues were then incubated in PSS containing protease type 8 (0·3 mg ml⁻¹; Sigma, UK) for 5 min followed by collagenase type 11 or type 1A (1 mg ml⁻¹; Sigma, UK) for 10 min. Cells were isolated by mechanical agitation with a wide-bore pipette. The dissociation PSS had the following composition (mM): NaCl, 126; KCl, 6 mM; MgCl₂, 1·2; CaCl₂, 0·05; Hepes, 10; and glucose, 11; and was adjusted to pH 7·2 with NaOH.

Whole-cell currents were measured at room temperature (21-25°C) with the perforated patch method using a patch clamp amplifier (List EPC 7; List-Electronic, Darmstadt, Germany). The perforated patch technique was used to minimize perturbation of physiological Ca²⁺ buffers and regulatory mechanisms that are affected by conventional whole-cell recording. To obtain a perforated patch the pipette solution contained amphotericin B dissolved in dimethyl sulphoxide (DMSO) to a final concentration of 100-200 μg ml⁻¹ which was prepared every 3 h.

Experiments were performed in K⁺-free conditions to remove contaminating K⁺ currents so that I_Cl(Ca) was recorded in isolation. The K⁺-free extracellular solution had the following composition (mM): NaCl, 126; MgCl₂, 1·2; CaCl₂, 2; Hepes, 10; and glucose, 11; and was adjusted to pH 7·2 with NaOH. The effect of different anions on the kinetics of STICs and I_tail were investigated using a modified external solution in which NaCl was replaced by an equimolar amount of NaI, NaBr, NaSCN, sodium glutamate or sodium isethionate. In all experiments a K⁺-free pipette solution was used which contained (mM): CsCl or NaCl, 126; MgCl₂, 1·2; Hepes, 10; glucose, 11; and EGTA, 0·1; and the pH was adjusted to 7·2 with CsOH. In all experiments, changes in liquid junction potential were minimized by the use of a 150 mM KCl-agar bridge connecting the main recording bath and a side bath containing the normal intracellular solution. Junction potentials of the different external solutions were measured using the technique of Neher (1992) and were not greater than 5 mV. In a series of experiments the effect of internal iodide on the kinetics of I_Cl(Ca) was investigated using a modified pipette solution that had the following composition (mM): NaI, 126; MgSO₄, 1·2; Hepes, 10; glucose, 11; and EGTA, 0·1; and the pH was adjusted to 7·2 with NaOH. In experiments where spontaneous transient outward currents (STOCs) were recorded, the extracellular and intracellular solutions were similar to that used in the K⁺-free experiments except the PSS contained 6 mM KCl and the pipette solution contained 126 mM KCl instead of NaCl. In a final series of experiments the effect of total removal of external Cl⁻ on the activation of I_Cl(Ca) was investigated. Cells were bathed initially in a normal control solution except that CaCl₂ and MgCl₂ were replaced by calcium gluconate (5 mM) and MgSO₄ (1·2 mM), respectively. When sufficient STICs had been recorded for analysis the external solution was replaced by one which was completely Cl⁻ free and NaCl was replaced by 250 mM mannitol to maintain tonicity.

All voltage protocols were generated and membrane currents were recorded using Cambridge Electronic Design (CED) hardware and voltage clamp software. For analysis of voltage-evoked currents signals were low-pass filtered at 1 kHz prior to digitization at 2·5 kHz. Spontaneous transient inward currents (STICs) were recorded over a 3-5 min period and low-pass filtered at 1 kHz. Between 10 and 30 individual currents were sampled at 2·5 kHz and averaged using the SIGAVG signal-averaging programme via a CED 1401 interface (both systems Cambridge Electronic Design, Cambridge, UK) and the mean amplitude was calculated. Inward ‘tail’ currents (I_tail) were evoked by stepping the cell from a holding potential of -50 mV to +20 mV for 100-200 ms to generate calcium currents (see Greenwood & Large, 1996 for full description); I_tail was recorded upon repolarization to -50 mV. The voltage dependence of the decay of I_tail was determined by stepping the cell to various potentials between -110 and +40 mV for 500 ms following the activation of I_Ca. The amplitude of I_tail was measured after the capacitative transient had settled completely which was within 20 ms. In some cells the amplitude of I_tail recorded in an extracellular solution containing 2 mM CaCl₂ was relatively small and so the bathing solution was replaced by one containing 10 mM CaCl₂ to increase the amplitude of I_Ca. This manipulation enhanced the amplitude of I_Ca but did not alter the rise time of STICs or the decay kinetics of STICs and I_tail. The voltage dependence of the decay of STICs was determined by manually changing the holding potential to different test potentials between -50 to +50 mV. Cells were then held at the new potential for at least 2 min to capture sufficient individual currents for analysis. The decay of both I_tail and averaged STICs was fitted by an exponential described by the equation I_t=I₀exp-(t/τ), using a non-linear least-squares fitting routine, where I_t and I₀ are the amplitudes of the current at times t and 0 ms. The relative permeability of external anions with respect to Cl⁻ (P_A/P_Cl) was determined using the Goldman-Hodgkin-Katz potential equation:

where E_rev is the reversal potential, the subscripts ‘o’ and ‘i’ denote external and internal ion species and R, T, z and F have their usual meaning. [Cl⁻] and [A⁻] are the activities of chloride and the test anion, respectively, which were determined by multiplying the ion concentration by an activity coefficient of 0·75 (Bormann et al. 1987).

Statistics

All data are the means ±s.e.m. of n individual cells.

RESULTS

Using conventional Cl⁻-containing pipette and external solutions, STICs had a rise time of about 40 ms at -50 mV (mean of 7 cells was 38 ± 2 ms) and then declined in a characteristic exponential manner (Fig. 1A and B) with a time constant (τ) at -50 mV of between 70 and 115 ms (mean value was 85 ± 7 ms; n= 9). This value was similar to the τ of STICs recorded in earlier studies in the same cell type (Hogg et al. 1993; Greenwood & Large, 1996). Initial experiments were performed to investigate whether more permeable anions such as iodide (I⁻) and thiocyanate (SCN⁻) (Amédée et al. 1990) affected the decay of STICs. At a holding potential of -50 mV replacement of the normal bathing solution with one that contained 126 mM NaI reduced the STIC amplitude (Fig. 1A). This was due to a change in the driving force as the STIC reversal potential (E_rev) is shifted to more negative potentials in a NaI external solution (E_rev was -34 ± 1 mV in external NaI and -2 ± 0·5 mV in external NaCl; n= 5). STICs in 126 mM NaI-containing external solution had a similar rise time to STICs in normal extracellular solution (42 ± 4 ms) and had a decay that could be described by a single exponential. However, the τ value was significantly longer in NaI-based external solution (Fig. 1A) and the mean τ at -50 mV was 134 ± 4 ms (n= 7) compared with a control τ of 85 ms (see above). Replacement of chloride by SCN⁻, which is more permeable than I⁻ through Ca²⁺-activated Cl⁻ channels (Amédée et al. 1990), caused STICs to become outward at -50 mV due to the change in reversal potential (E_rev was -60 ± 4 mV; n= 6) and markedly prolonged the STIC decay (Fig. 1B). Thus, at -50 mV the STIC decay in 126 mM NaSCN was 252 ± 9 ms (n= 5). Bromide (Br⁻), which is also considered to be a more permeable anion than Cl⁻ through Ca²⁺-activated Cl⁻ channels but is less permeable than I⁻ and SCN⁻ (Amédée et al. 1990), caused a small prolongation of the STIC decay with a mean τ value in NaBr of 96 ± 7 ms (n= 5).

Figure 1. Effect of substitution of external NaCl with NaI or NaSCN on the decay of STICs and I_tail at -50 mV.

A and B show averaged STICs recorded at -50 mV in external solutions containing NaCl (Ai and Bi) or NaI (Aii) and NaSCN (Bii). In each case the currents decayed exponentially and the fit is shown superimposed on the trace with the time constant of decay (τ) shown next to the current. C and D show I_tail evoked by a step depolarization from -50 mV to +10 mV for 200 ms. The initial downward deflection is I_Ca evoked by the step depolarization which is followed on repolarization to -50 mV by an inward current that is labelled I_tail. I_tail was recorded in normal extracellular solution (Ci and Di) and in NaI-containing (Cii) and NaSCN-containing (Dii) solutions. The dotted horizontal line shows the resting current level at -50 mV.

As reported previously (Greenwood & Large, 1996), I_tail evoked as a consequence of Ca²⁺ influx following brief membrane depolarizations also decayed exponentially with a τ value at -50 mV similar to those reported for STICs (Greenwood & Large, 1996). In the present study, under control conditions (NaCl-based external solution) I_tail evoked by a 200 ms depolarization from -50 mV to +10 mV decayed with a mean τ of 97 ± 7 ms (n= 9) at -50 mV (e.g. Fig. 1C). Replacement of the normal external solution with one containing 126 mM NaI reduced the amplitude of I_tail consistent with a change in E_rev and prolonged the decay of I_tail at -50 mV (Fig. 1C) with a mean τ of 157 ± 37 ms (n= 5). Replacement of the normal external solution with one containing 126 mM NaSCN caused I_tail to become outward at -50 mV (Fig. 1D) and prolonged the decay of I_tail. Thus, the mean τ of I_tail at -50 mV in external NaSCN was 278 ± 19 ms (n= 10). Thus, replacement of external Cl⁻ with more permeant anions prolongs the decay of STICs and I_tail by a similar degree.

Other experiments were undertaken to investigate whether relatively impermeant anions such as glutamate and isethionate affected the decay of STICs and I_tail. Replacement of the normal PSS with one that contained sodium glutamate or sodium isethionate reduced significantly the amplitude of I_tail at -50 mV (Fig. 2A). Figure 2B shows the mean data from five cells where the current-voltage relationship of I_tail has been determined in NaCl- and sodium isethionate-based extracellular solutions. It can be seen that sodium isethionate shifts the reversal potential of I_tail to about +18 mV (mean of 5 cells was +25 ± 5 mV) which is equivalent to a relative permeability to Cl⁻ of 0·11. Sodium isethionate also reduced the amplitude of I_tail at all potentials studied (Fig. 2B) which was associated with a reduction of the instantaneous chord conductance at -50 mV from 2·7 ± 0·7 nS in control conditions to 0·85 ± 0·4 nS in sodium isethionate (n= 5). Replacement of the external NaCl with sodium isethionate also increased the decay rate of I_tail (Fig. 2A), and τ at -50 mV decreased from 90 ± 5 ms in 126 mM NaCl to 67 ± 5 ms in 126 mM sodium isethionate (n= 9). STICs were also recorded in smooth muscle cells bathed in normal (NaCl) and sodium glutamate-based extracellular solution and the mean τ values at -50 mV were 83 ± 11 and 63 ± 5 ms, respectively (n= 5). A similar increase in the rate of STIC decay was observed in two cells when the extracellular NaCl in the bathing solution was replaced by sodium isethionate.

Figure 2. Effect of impermeant anions on the decay of I_tail.

A shows I_tail recorded at -50 mV in NaCl-containing and sodium isethionate-containing PSS. I_tail was recorded following a 200 ms step depolarization from -50 mV to +10 mV. The dotted horizontal line shows the holding current at -50 mV. Exponential fits are shown superimposed on the traces. B shows the mean data from 5 cells for the amplitude of I_tail recorded at different potentials between -110 and +40 mV in NaCl-containing (•) and sodium isethionate-containing (^) PSS. The horizontal axis is the test potential (in mV) and the vertical axis is the peak current amplitude (in pA).

It was apparent that anions more permeant than Cl⁻ decreased the decay rate of STICs and I_tail whereas less permeant anions increased the decay rate of these currents. Consequently, we investigated whether the effect on the time course of I_Cl(Ca) was related to the relative permeability of the anions. Figure 3 shows the permeability of the anion relative to Cl⁻, calculated from E_rev measurements with the Goldman-Hodgkin-Katz equation, plotted against τ measured at -50 mV. It can be seen that there is a strong correlation between the relative permeability of the anion and the effect on τ and therefore it seems that the ability of an anion to affect the gating of I_Cl(Ca) is related to the physico-chemical factors which determine the ability of the anion to permeate the channel pore.

Figure 3. Relationship between anion permeability and τ at -50 mV.

Data shows the pooled data from STICs and I_tail recorded at -50 mV in different extracellular solutions as indicated by each point. The data represent the mean of between 7 and 15 cells with error bars showing s.e.m. The vertical axis is the τ at -50 mV and the horizontal axis is the permeability of the external anion relative to chloride calculated from changes in reversal potential according to the Goldman-Hodgkin-Katz equation. The line represents a linear fit of the data with r= 0·996.

Effect of anions on the voltage dependence of the decay of STICs and I_Cl(Ca)

It has been shown previously that the exponential decay of STICs in rabbit portal vein smooth muscle cells becomes slower as the membrane potential is depolarized, with τ increasing e-fold for an approximately 110 mV change in voltage (Hogg et al. 1993). To determine if the voltage dependence of STICs was affected by the presence of various external anions we recorded individual currents at different potentials between -70 and +50 mV in the presence of external NaI and NaBr. Figure 4A shows the data from one cell in which STICs recorded in normal external solution decayed with a τ value of 82 ms at -30 mV which increased to 136 ms at +30 mV. Replacement of the normal extracellular solution by one containing 126 mM NaI prolonged the τ values of the STIC to 136 and 301 ms at -30 and +30 mV, respectively. Figure 4B shows the mean data from cells bathed in normal PSS with τ increasing e-fold with a 120 mV change in voltage. Replacement of external NaCl with NaI or NaBr prolonged the STIC decay at all potentials studied without an effect on the voltage dependence of STIC decay (Fig. 4B). Thus, τ values changed e-fold with 115 and 118 mV changes in voltage in NaI and NaBr, respectively.

Figure 4. Effect of anions on the voltage dependence of STIC decay.

A shows averaged STICs recorded at -30 and +30 mV in extracellular solution containing NaCl (left) and NaI (right). All currents decayed exponentially and the time constant of decay (τ) is shown next to the trace. B shows data from 4-8 cells for STICs recorded in external NaCl (▪), NaBr (□) and NaI (^). Each point is the mean of 4-8 cells with error bars representing the s.e.m. The horizontal axis is the test potential (mV) and the vertical axis is τ (ms) plotted on a logarithmic scale.

The decay of I_tail has a similar voltage dependence to that reported for STICs (Greenwood & Large, 1996). Figure 5A shows the data from one cell where in a normal NaCl-based external solution the τ of I_tail increased from 38 ms at -110 mV to 120 ms at -20 mV. Figure 5B shows I_tail in the same cell bathed in 126 mM NaSCN and it can be seen that the decay of I_tail remained exponential and was prolonged at all test voltages. Mean data on the voltage dependence of τ is shown in Fig. 5C; NaI and NaSCN had no significant effect on the voltage dependence of the decay of I_tail. These data show that anions more permeable than Cl⁻ such as Br⁻, I⁻ and SCN⁻ prolong the decay of STICs and I_tail without altering the intrinsic voltage dependence of the current decay. The permeable anions I⁻ and SCN⁻ also had no effect on the instantaneous chord conductance (n= 5). Conductances were recorded at -110 and +40 mV to allow for changes in reversal potential under the various ionic conditions. Thus, at -110 mV the chord conductances in NaCl, NaI and NaSCN were 2·2 ± 0·6, 2·7 ± 0·6 and 2·2 ± 0·34 nS (n= 5) and at +40 mV the conductances were 2·5 ± 0·5, 2·4 ± 0·9 and 2·9 ± 0·5 nS in NaCl, NaI and NaSCN, respectively. These data suggest that the effects of the permeable anions on I_Cl(Ca) were not due to a change in the intrinsic voltage dependence or single channel conductance.

Figure 5. Effect of anions on the voltage dependence of I_tail decay.

A and B show representative families of ‘tail’ currents evoked by a step depolarization from -50 mV to +10 mV for 100 ms in cells bathed in NaCl-containing (A) and NaSCN-containing (B) extracellular solutions. I_tail was recorded at different potentials between -110 and -20 mV. In each case I_tail decayed exponentially and τ is shown next to the current. C shows the mean data for I_tail recorded at different potentials in extracellular solutions containing NaCl (▪), NaI (□) and NaSCN (^). The horizontal axis is the test potential (mV) and the vertical axis represents the τ value (ms) plotted on a logarithmic scale. Each point is the mean of 5-10 cells with error bars showing the s.e.m.

Effect of external Cl⁻ concentration on the decay of I_Cl(Ca)

Some Cl⁻ channels are gated by extracellular Cl⁻ ions (Pusch et al. 1995; Chen & Miller, 1996; Rychkov et al. 1996) and therefore it was of interest to investigate whether simply reducing the external Cl⁻ concentration affects the decay of I_Cl(Ca). Figure 6 shows data from an experiment where a family of tail currents were recorded at different potentials from cells bathed in an extracellular solutions containing 126 mM NaCl (normal) or one where the NaCl was reduced to 25 mM. In the latter case the osmolarity was maintained by including 212 mM sucrose in the bathing solution. Under these conditions I_tail reversed at -2 ± 1 mV in 126 mM NaCl and +21 ± 4 mV in 25 mM NaCl, and these reversal potentials were similar to the calculated chloride equilibrium potential (E_Cl) under these conditions (-3 and +25 mV, respectively). In 126 mM NaCl the τ value of I_tail at -50 mV was 129 ± 11 ms (n= 4) which was not significantly different from the τ value of I_tail recorded in a bathing solution containing 25 mM NaCl (131 ± 8 ms). Furthermore, Fig. 6C shows that the decay of I_tail had a similar voltage dependence in both bathing solutions. These data suggest that neither a reduction in extracellular Cl⁻ nor a change in driving force affects the decay of I_tail and that it is the identity of the external anion which alters the decay of I_Cl(Ca).

Figure 6. Effect of reducing extracellular Cl⁻ concentration on I_tail.

A and B show families of I_tail recorded at different potentials after activation of I_Ca by step depolarization from -50 mV to +10 mV for 100 ms in an extracellular solution containing 126 mM NaCl (A) or in 25 mM NaCl (B). The test potential at which I_tail was recorded was changed in 30 mV increments from -110 to +40 mV. C shows the τ of I_tail recorded at different potentials in normal PSS (•) and PSS containing 25 mM NaCl (^). The horizontal axis is the test potential (mV) and the vertical axis represents the τ value (ms) plotted on a logarithmic scale. Each point is the mean of 3 cells with error bars showing the s.e.m.

We extended these experiments to investigate whether there was an obligatory requirement for external Cl⁻ to activate I_Cl(Ca). A test external solution was used which contained no NaCl and the osmolarity of the solution was maintained by the addition of 250 mM mannitol (the osmolarities of the normal and NaCl-free solutions were 274 and 263 mosmol l⁻¹, respectively). The amplitude of I_Ca was markedly inhibited by the mannitol-containing solution and consequently I_tail was difficult to record. However, in seven cells STICs could be recorded in the NaCl- and mannitol-containing, Cl⁻-free external solutions (Fig. 7A). In normal extracellular solution containing 126 mM NaCl, STICs had a mean amplitude and decay τ of 19 ± 3 pA and 104 ± 9 ms, respectively. It has been shown previously that removal of external Na⁺ has no effect on the amplitude and E_rev of STICs in rabbit portal vein myocytes (Wang et al. 1992) and in the present study the mean amplitude and decay of STICs recorded in a NaCl-free external solution containing 250 mM mannitol was not significantly different from control values (21 ± 4 pA and 98 ± 6 ms, respectively). However, the frequency of STIC discharge was increased by removal of extracellular NaCl (Fig. 7), which may be due to an increase in intracellular Ca²⁺ concentration by reverse mode Na⁺-Ca²⁺ exchange activity which has been shown to occur in this cell type (Leblanc & Leung, 1995). Therefore STICs could be recorded with a standard Cl⁻-containing pipette solution from cells bathed in a Cl⁻-free extracellular solution and, moreover, the decay of STICs in Cl⁻-free external solution was not significantly different from STICs recorded under normal conditions. Consequently, I_Cl(Ca) does not have an obligatory requirement for external Cl⁻ for activation. However, Fig. 7A and B shows that the amplitude of the STICs in this cell was far less than would be expected from the increased driving force due to the shift of E_Cl to very positive values. It is obvious from Fig. 7B that the current-voltage relationship of STICs recorded in Cl⁻-free external solutions showed marked rectification. This may reflect a change in ion permeation under the disproportional ionic conditions employed due to reduction in Cl⁻ conductance produced by Cl⁻ removal from the external solution.

Figure 7. Effect of the total removal of extracellular Cl⁻ on the activation of STICs.

A shows a long-term trace from a cell held at -50 mV in which STICs were recorded in control solution (126 mM NaCl) and then in a NaCl-free extracellular solution containing 250 mM mannitol (zero NaCl). This caused a reduction in the holding current and increased the discharge of STICs. Insets show the average of 12-15 STICs recorded in normal PSS and NaCl-free PSS. Exponential fits of the decay of the currents are shown as superimposed smooth lines. B shows the current-voltage relationship of STICs from the cell shown in A recorded in NaCl-free PSS (•); ^ shows mean value at -50 mV in 126 mM NaCl solution. Each point is the mean of 5-12 individual STICs with error bars showing the s.e.m.

Investigation into the mechanism of anion-dependent modulation of I_Cl(Ca) kinetics

Experiments were performed to determine if the effect of SCN⁻ on the decay of I_tail was due to a mechanism compatible with an interaction with a binding site which exhibited saturable characteristics or whether the effects on τ increased monotonically with an increase in SCN⁻ concentration (i.e. was the modulation of current decay independent of any ion : ion competition?). I_tail was recorded in normal extracellular solution containing 126 mM NaCl and in the presence of progressively increasing concentrations of NaSCN. In all cases the extracellular Cl⁻ concentration was reduced concomitant with an increase in SCN⁻ so that the total anion concentration was maintained at 148 mM. Figure 8 shows that as the SCN⁻ mole fraction was increased the τ of I_tail was prolonged in a sigmoidal manner at all potentials investigated and the data can be fitted by a logistic curve with an apparent half-maximal effect in 69, 70 and 57 mM SCN⁻ at -80, -50 and -20 mV, respectively. It is therefore evident that there was not a linear relationship between τ and SCN⁻ concentration and these data suggest that Cl⁻ and SCN⁻ ions compete for a binding site which, when occupied, modifies the decay of I_tail.

Figure 8. Relationship between the decay of I_tail and the SCN⁻:Cl⁻ mole fraction.

Data show τ recorded at -20 mV (▪), -50 mV (•) and -80 mV (□) at different mole fractions of SCN⁻ and Cl⁻. The data were fitted by a logistic equation, shown by the continuous lines, of the form 1/(1 + (X/X₅₀)ⁿ) where X denotes the SCN⁻ mole fraction, X₅₀ denotes the mole fraction that produces a half-maximal increase in τ and n represents the slope of the curve. Points are the mean of 5 cells with error bars denoting the s.e.m. There is no data point at -50 mV for a mole fration of 0·8 because the reversal potential of I_tail was close to -50 mV under these ionic conditions. Vertical axis is τ (ms) and the horizontal axis is the SCN mole fraction where 0 = 126 mM NaCl, 0·5 = 63 mM NaCl and 63 mM NaSCN and 1·0 = 126 mM NaSCN.

Effect of internal I⁻ on time course of I_Cl(Ca)

Experiments were performed to investigate whether changing the identity of the intracellular anion affected the decay of STICs and I_tail. We used two different pipette solutions which contained either 126 mM NaCl or 126 mM NaI and in both cases MgSO₄ replaced MgCl₂ so that there was no Cl⁻ in the NaI pipette solution. Figure 9Ai shows an averaged STIC recorded from a cell using a NaI-based pipette solution and an extracellular solution that contained 126 mM NaCl, which had a τ value of 89 ms. The mean τ for STICs recorded at -50 mV with a NaI-based pipette solution in external NaCl was 102 ± 8 ms (n= 7), which was similar to the mean τ value of STICs recorded with a NaCl-containing pipette solution in external NaCl (104 ± 9 ms; n= 7). Figure 9Aii shows that when the conventional extracellular solution (126 mM NaCl) was replaced by one containing 126 mM NaI the decay τ of the STIC at -50 mV increased from 89 to 180 ms and the mean τ under these conditions was 201 ± 25 ms (n= 4). Similarly at -50 mV I_tail recorded with an NaI-based pipette solution from cells bathed in NaCl-containing PSS had a similar decay τ (108 ± 13 ms; n= 12) to I_tail recorded using a Cl⁻-based pipette solution, but τ was increased by the replacement of the extracellular solution for one that contained 126 mM NaI (197 ± 22 ms; n= 7). Figure 9B shows that the voltage dependence of STICs and I_tail recorded using a NaI-based pipette was not different from that recorded with a NaCl-containing pipette solution.

Figure 9. STICs recorded using a NaI-based pipette solution.

A shows averaged STICs recorded at -50 mV using a pipette solution containing NaI (NaI_i) in normal (126 mM NaCl; NaCl_o) extracellular solution (trace i) and 126 mM NaI-based extracellular solution (NaI_o) (trace ii). Each current decayed exponentially with the fit, and the τ value is illustrated next to the current. B shows the voltage dependence of τ of STICs and I_tail recorded using a NaI-based pipette solution in cells bathed in either external NaCl (•) or NaI (^). The horizontal axis is the test potential (mV) and the vertical axis represents the τ value (ms) plotted on a logarithmic scale. Each point is the mean of 7 cells with error bars showing the s.e.m.C shows the current-voltage relationship of STICs recorded in one cell using a NaI-based pipette solution and extracellular solutions containing NaCl (•) and NaI (^).

To test whether the cell was dialysed effectively with I⁻ from the NaI-containing pipette solution the reversal potential of I_tail and STICs was measured with an iodide-based pipette solution from cells bathed in either NaCl- or NaI-containing extracellular solution. Figure 9C shows data from the same cell as Fig. 9A where STICs were recorded at various potentials under these conditions. When the cell was bathed in an external solution containing 126 mM NaCl the STICs reversed at +42 mV and this changed to +2 mV when the external NaCl was replaced by NaI. The mean reversal potential for STICs and I_tail with NaI in the pipette solution was +27 ± 5 mV with external NaCl (n= 7; Fig. 9B) and +4 ± 3 mV (n= 7) with external NaI. These data suggest that changing the intracellular anion does not affect the decay of STICs and I_tail.

The data obtained with a pipette solution containing NaI also suggest that the effect of the various external anions on the decay of STICs and I_tail was not due to modulation of the time course of the intracellular Ca²⁺ signal that activates I_Cl(Ca). This was further supported by the observation that the decay of spontaneous transient outward currents (STOCs), which are Ca²⁺-activated K⁺ currents evoked by Ca²⁺‘sparks’ released from the sarcoplasmic reticulum (Nelson et al. 1995), was not affected by substitution of normal external solution for one that contained 126 mM NaSCN. As the decay of STOCs is proportional to the amplitude of the current, we measured the half-decay time of currents less than 50 pA and currents greater than 100 pA. At -10 mV with a NaCl-containing extracellular solution, STOCs less than 50 pA and greater than 100 pA had half-decay times of 21 ± 2 and 41 ± 3 ms, respectively (n= 3). When NaSCN was substituted for NaCl the half-decay times for currents less than 50 pA and greater than 100 pA were 19 ± 2 and 42 ± 5 ms, respectively (n= 3). Consequently, substitution of external NaCl for NaSCN had no effect on the time course of STOCs.

DISCUSSION

The main observation from the present work is that external anions have a marked effect on the time course of I_Cl(Ca) when manifest as STICs and I_tail. There are two pieces of experimental evidence that rule out an effect of external anions on the time course of the intracellular Ca²⁺ signal that activates the Cl⁻ channels or on a biochemical process such as phosphorylation which might modulate the channel. First, external SCN⁻, which more than doubled the STIC τ, did not alter the time course of STOCs which are a sensitive indicator of subsarcolemmal Ca²⁺ and which are often activated simultaneously with STICs (Large & Wang, 1996). Secondly, intracellular dialysis with I⁻ from the pipette solution did not change the time course of I_Cl(Ca) thus eliminating an intracellular site of action for I⁻ either on the Ca²⁺ signal or on the Cl⁻ channel. Therefore it can be concluded that anions alter the gating of Ca²⁺-activated Cl⁻ currents by modulating Cl⁻ channel kinetics directly and not by modulating the activating Ca²⁺ signal or possible biochemical processes.

In comparision to studies on ClC-0 isolated from Torpedo electroplax and skeletal muscle ClC-1, where external Cl⁻ is required for full activation and voltage dependence of the channel (Pusch et al. 1995; Chen & Miller, 1996; Rychkov et al. 1996), replacement of NaCl by sucrose or mannitol did not change the time course of STICs and I_tail. Indeed with zero Cl⁻ and 10 mM gluconate as the only anion (impermeant) in the external solution, STICs were readily recorded, which shows that external permeant anions are not obligatory for activation of I_Cl(Ca). However, anions more permeable than Cl⁻ increased the τ of STICs and I_tail and less permeable anions reduced τ. Permeable anions also slowed the rate of inactivation of swell-induced Cl⁻ currents observed at positive test potentials in BC₃H1 myoblasts (Voets et al. 1997) and HeLa cells (Stutzin et al. 1998). In terms of the proposed two-state model for Ca²⁺-activated Cl⁻ channels where the channels exist either in the open or closed configuration a simple interpretation is that the more permeable anions favour the open state, for example by increasing the mean open time. This is analagous to the ‘foot in the door’ situation observed with various ion channels where permeable ions stabilize the open conformation (Yellen, 1997), although single channel studies are required to confirm this hypothesis. In the present study there was a strong correlation between P_A/P_Cl and the effect on τ, which suggests that gating is linked to permeability. A similar link between permeation and gating has been proposed for ClC-0 channels (Pusch et al. 1995; Chen & Miller, 1996). However, the inability of internal I⁻ to influence the decay kinetics of I_tail and STICs in the present study suggests that it is not the permeating ion per se which determines channel kinetics. Therefore the physico-chemical factors which determine the selectivity of the channel to different anions appear also to modulate the kinetics of the Cl⁻ channel.

The ability of external anions to modify the time course of calcium-activated Cl⁻ currents followed the Hofmeister or lyotropic series (SCN⁻ > I⁻ > Br⁻ > Cl⁻), which reflects the ability of these anions to adsorb to lipid bilayers or proteins. Such an interaction may alter the surface potential in the vicinity of the channel by neutralizing positively charged amino acid residues. Thus, it has been shown in frog skeletal muscle fibres that anions shifted the activation and steady-state inactivation curves of Na⁺ currents by about 10-20 mV in the negative direction (Dani et al. 1983). In the present study external I⁻ and SCN⁻ increased the τ of Cl⁻ currents two- to threefold. The voltage dependence of τ is such that a similar change in τ would require an alteration of the membrane potential by greater than 100 mV in the positive direction. Thus, it is unlikely that the marked effects of external anions on τ are due to a lyotropic modulation of surface potential, although a direct lyotropic effect on channel conformation (i.e. of a gating particle) is not excluded.

It is clear from the present study that anions modulate the gating of I_Cl(Ca) by access from the external surface of the membrane. In an external solution containing NaCl and with pipettes containing 126 mM NaI the τ of STICs was the same as values obtained from cells dialysed intracellularly with NaCl. Subsequent replacement of external NaCl by NaI produced the characteristic lengthening of STIC time course. It seemed that adequate dialysis of the cell interior by I⁻ was achieved because with 126 mM NaI in the pipette and bathing solutions the STIC reversal potential was close to 0 mV, the calculated anion equilibrium potential. Similarly, Stutzin et al. (1998) showed that inactivation of swell-induced currents in HeLa cells was not modulated by intracellular I⁻ but was prolonged by external I⁻. The data of the present study are in contrast to those of Evans & Marty (1986) who suggested that I⁻ altered the gating of I_Cl(Ca) in rat lacrimal glands by binding to a site located on the intracellular side of the membrane. In that work, relaxations of I_Cl(Ca) in response to voltage steps were studied with whole-cell recording with intracellular Ca²⁺ concentration buffered to fixed values. It was found that the exponential relaxation produced by voltage steps to negative potentials was slowed but relaxations at positive potentials were speeded up by external I⁻ and similar effects were produced by the inclusion of NaI in the pipette solution. In the present study the time course of I_Cl(Ca) was lengthened by external I⁻ at all potentials and these discrepancies between the present work and that of Evans & Marty (1986) might be due to the use of different experimental protocols. However, in a few experiments we recorded voltage-dependent relaxations following activation of I_Cl(Ca) by low concentrations of caffeine that were lengthened by I⁻ at all potentials, similar to the effects of I⁻ on STICs and I_tail (authors' unpublished data). Consequently the contrasting results between rat lacrimal glands and rabbit portal vein smooth muscle cells might indicate a difference between tissues in the gating of I_Cl(Ca).

In experiments where the ratio of Cl⁻:SCN⁻ was varied there was not a simple monotonic relationship between the mole fraction and the effect on I_Cl(Ca) gating. Instead the data were better explained by the interaction of the anions in a competitive manner with a saturable binding site. Since the ability of the anion to alter the τ of I_tail and STICs was related to its permeability relative to Cl⁻, it is possible that the anions bind to a site which represents the selectivity filter and also modulates gating of I_Cl(Ca). An anionic binding site which depends on accessibility from the external bulk solution has also been proposed to modulate the activation and gating of ClC-0 and ClC-1 (Pusch et al. 1995; Chen & Miller, 1996; Rychkov et al. 1996). It has been proposed that in these voltage-dependent channels it is the binding of Cl⁻ to the binding site which confers the marked voltage dependence on channel activation (see Chen & Miller, 1996 and Rychkov et al. 1996 for fuller discussion). In comparison, the Ca²⁺-activated Cl⁻ channel studied in the present work has a relatively weak voltage dependence of τ which was not altered by any of the anions tested or by a reduction in the extracellular Cl⁻ concentration. This suggests that the part of the channel structure responsible for voltage dependence may be distinct from the proposed site involved in gating. Furthermore, STIC decay recorded after the complete removal of external Cl⁻ was not different from the decay of STICs under control conditions which suggests that the Cl⁻ channel has an intrinsic rate of closure which is independent of anion binding and not influenced by the presence of Cl⁻ ions. However, binding of more permeable anions slows the rate of channel closure, whereas impermeant anions increase the rate of deactivation.

In conclusion, the present data suggest that permeating anions may modulate the kinetics of Ca²⁺-activated Cl⁻ channels in smooth muscle cells by binding to a site located on the external surface of the channel which may be part of the channel representing the selectivity filter.