Sodium-potassium pump current in smooth muscle cells from mesenteric resistance arteries of the guinea-pig

Abstract

1. The Na⁺-K⁺ pump current was studied in smooth muscle cells from mesenteric resistance arteries of guinea-pigs by the use of the perforated patch-clamp technique in the presence of blockers for various ion channels and exchangers.

2. When the Na⁺ concentration in the pipette solution ([Na⁺]_i) was 50 mM, an increase in the extracellular K⁺ concentration ([K⁺]_o) from 0 to 10 mM caused an outward current. Both the removal of K⁺ from the bath solution and the application of 10 μM ouabain abolished this current. Thus, this K⁺-induced and ouabain-sensitive current was considered to be the Na⁺-K⁺ pump current.

3. The amplitude of the Na⁺-K⁺ pump current increased as the membrane potential was made more positive until around 0 mV, while the amplitude saturated at more positive potentials than 0 mV.

4. An increase in [K⁺]_o or [Na⁺]_i amplified the Na⁺-K⁺ pump current. For [K⁺]_o, the binding constant (K_d) was 1.6 ± 0.3 mM and the Hill coefficient (n_H) was 1.1 ± 0.2 (n = 6). For [Na⁺]_i, K_d was 22 ± 5 mM and n_H was 1.7 ± 0.5 (n = 4–19).

5. The presence of various monovalent cations other than Na⁺ in the bath solution also evoked the Na⁺-K⁺ pump current. The order of potency was K⁺ ≥ Rb⁺ > Cs⁺ ≫ Li⁺.

6. Ouabain inhibited the Na⁺-K⁺ pump current in a dose-dependent manner with a K_d of 0.35 ± 0.03 μM and an n_H of 1.2 ± 0.1 (n = 6–8).

7. The Na⁺-K⁺ pump current increased as temperature increased. The temperature coefficient (Q₁₀; 26–36 °C) was 1.87 (n = 9).

8. In summary the present study characterized for the first time the Na⁺-K⁺ pump current in vascular smooth muscle cells by the use of the voltage-clamp method. The use of this method should provide essential information for Na⁺,K⁺-ATPase-mediated changes in the cell functions of vascular smooth muscle cells.

The maintenance of electrochemical gradients for extracellular Na⁺ and intracellular K⁺ ions is essential for almost all cells and is mainly achieved by the action of Na⁺,K⁺-adenosine triphosphatase (ATPase). The electrogenic nature of the Na⁺-K⁺ pump has been described: one forward cycle provides the outward current resulting in the expulsion of three Na⁺ ions and the uptake of two K⁺ ions for each hydrolysed ATP. To assess the function of Na⁺,K⁺-ATPase, including its electrogenic characteristics, the Na⁺-K⁺ pump currents have been measured by voltage-clamp techniques in several cell types, such as cardiac myocytes (Gadsby & Nakao, 1989; Nakao & Gadsby, 1989; Gao et al. 1995; Sakai et al. 1996) and vascular endothelial cells (Oike et al. 1993).

In smooth muscle cells, the physiological roles of Na⁺,K⁺-ATPase in cell function have been examined mainly by measuring ion flux, isometric tension and membrane potential with blockers for Na⁺,K⁺-ATPase such as ouabain, low temperature or incubation with K⁺-free solution (Casteels et al. 1977; Widdicombe, 1981; Casteels, 1981; Stekiel et al. 1986; Magliola et al. 1986; Fujii et al. 1997). Since the inhibition of the Na⁺-K⁺ pump by such procedures would dynamically change the intracellular Na⁺ concentration ([Na⁺]_i), which in turn affects the activity of Na⁺,K⁺-ATPase, a precise evaluation of Na⁺,K⁺-ATPase is difficult. In addition, some of these procedures for inhibiting the Na⁺-K⁺ pump might also affect other currents (such as background K⁺ conductances) non-specifically (Hirst & van Helden, 1982). The recording of the Na⁺-K⁺ pump current by the patch-clamp technique would be of advantage in evaluating the function of Na⁺,K⁺-ATPase, since intra- and extracellular ionic conditions could be controlled and other ionic currents could be eliminated by various blockers during the experiment. However, no study has evaluated the Na⁺-K⁺ pump current in smooth muscle cells, although one study has shown ouabain-sensitive background currents in intestinal smooth muscle cells (Burke & Sanders, 1990). In the present study, we examined the properties of the Na⁺-K⁺ pump current in smooth muscle cells from mesenteric artery by use of the perforated patch-clamp technique.

METHODS

Cell isolation procedure

Female guinea-pigs (body weight, 200–250 g) were anaesthetized with ether and then decapitated. Single smooth muscle cells were obtained by collagenase treatment from mesenteric resistance arteries (diameter < 300 μm), as described previously (Ohya et al. 1993; Setoguchi et al. 1997). The cells were stored in modified KB solution (Isenberg & Klöckner, 1988) until use (for composition, see below). The study protocol was approved by the ethical committee for animal experimentation at Kyushu University.

Perforated patch-clamp method

The perforated patch voltage-clamp technique was performed with a patch pipette through a voltage-clamp amplifier (Axopatch-1D, Axon Instruments Inc.). The conditions and procedures were basically the same as those described previously (Horn & Marty, 1988; Nakamura et al. 1998). Recording electrodes with a tip resistance of 5–7 MΩ were made from Pyrex glass capillary tubing. The bath chamber was placed on the stage of an inverted microscope with continuous superfusion of the bath solution at 0.5 ml min⁻¹.

From a holding potential of −40 mV, a ramp pulse of amplitude from −80 to 40 mV over 100 ms was applied every 60 s. Membrane currents were low-pass filtered at 2 kHz, digitized with a sampling frequency of 5–10 kHz, and stored in a personal computer system for subsequent analysis. Traces are finally shown after being low-pass filtered at 1 kHz. The capacitive component of the current was minimized by the cancellation network in the voltage-clamp amplifier. Electrical recordings were performed at a temperature of 36 ± 0.5°C, unless otherwise stated.

Solutions and chemicals

The modified KB solution contained the following (mM): KCl, 85; MgSO₄, 5; Na₂ATP, 5; EGTA, 0.2; sodium pyruvate, 5; succinate, 5; creatine, 5; glucose, 20; taurine, 20; and K₂HPO₄, 30; pH 7.30 (adjusted with KOH).

For the current recording, the pipette solution contained (mM): caesium aspartate, 110; sodium aspartate, 30; NaCl, 20; MgCl₂, 2; Hepes, 10; pH 7.3 (adjusted by CsOH); and 30 μg ml⁻¹ nystatin (Sigma). When the low-Na⁺ pipette solution was prepared, sodium aspartate and NaCl were replaced with an equimolar amount of caesium aspartate and CsCl, respectively.

The K⁺-free bath solution contained (mM): NaCl, 145; glucose, 10.8; Hepes, 5; MgCl₂, 2; NiCl₂, 1; BaCl₂, 1; TEA-Cl, 10; 4-aminopyridine (4-AP; to inhibit K⁺ currents), 1; and CaCl₂, 1; pH 7.3 (adjusted with NaOH). The K⁺ concentration in the bath solution ([K⁺]_o) was varied from 0 to 10 mM by addition of KCl. When the external cation dependence of the Na⁺-K⁺ pump current was examined, KCl was replaced with equimolar RbCl, CsCl or LiCl.

Nystatin was first dissolved in 100 % dimethyl sulphoxide (30 mg ml⁻¹), frozen in aliquots, and then diluted within 1 h before use. Ouabain (Sigma) was dissolved in distilled water (10 or 30 mM stock solution) and diluted at least 1000 × when used.

Data presentation

Data are expressed as means ± s.e.m. Fitting of the data to each equation was performed by the non-linear least-squares method.

RESULTS

Recording of extracellular K⁺-induced current and ouabain-sensitive current

Membrane currents were recorded at a holding potential of −40 mV in the presence of blockers for K⁺ and Ca²⁺ channels and Na⁺-Ca²⁺ exchangers. These were Cs⁺ in the pipette solution with TEA, 4-AP and Ba²⁺ in the bath solution to inhibit K⁺ channels and with Ni²⁺ in the bath solution to inhibit Ca²⁺ channels and Na⁺-Ca²⁺ exchangers. In the presence of 50 mM [Na⁺]_i, an increase in [K⁺]_o from 0 to 10 mM caused an outward current (Fig. 1). This current remained almost constant for more than 20 min (20 min after the recording, the amplitude was 93 ± 3 % of the initial amplitude, n = 5). The outward current disappeared when extracellular K⁺ was removed and reappeared with an addition of extracellular K⁺. The application of 10 μM ouabain abolished this outward current.

Figure 1. Extracellular K⁺-induced and ouabain-sensitive currents in vascular smooth muscle cells.

A chart recording of membrane current recorded at −40 mV in the absence and presence of 10 mM K⁺ in the bath solution. The change in K⁺ concentration in the bath solution ([K⁺]_o) is shown at the top of the figure. Ouabain (10 μM) was added to the bath solution during the period indicated by the horizontal filled bar. The dashed line indicates the current level at 0 mM [K⁺]_o. The zero current level, the net 0 pA, is indicated by a horizontal bar at the left of the current trace. Cell capacitance was 18 pF. Na⁺ concentration in the pipette solution ([Na⁺]_i) was 50 mM.

The K⁺-induced and ouabain-sensitive currents were compared (Fig. 2). Figure 2A shows the K⁺-induced change in the current at −40 mV before and after the application of 10 μM ouabain. Figure 2B shows the current traces obtained with a ramp pulse from −80 to 40 mV, before and after the application of K⁺, and subsequently after the application of ouabain. The K⁺-induced current was isolated by subtraction of the current recorded at 0 mM [K⁺]_o from that recorded at 10 mM [K⁺]_o (Fig. 2C). A ouabain-sensitive component was isolated by subtraction of the current recorded with 10 μM ouabain from that recorded without ouabain (Fig. 2C). The K⁺-induced current and the ouabain-sensitive current almost overlapped at potentials between −80 and 40 mV (Fig. 2C). Thus, the K⁺-induced current was almost identical to the ouabain-sensitive current, and was thus considered to be the Na⁺-K⁺ pump current. The current density of the Na⁺-K⁺ pump current at 40 mV was 1.2 ± 0.2 pA pF⁻¹ (n = 19) with 50 mM [Na⁺]_i and 10 mM [K⁺]_o.

Figure 2. Presence of Na⁺-K⁺ pump current.

A, chart recording of membrane current recorded at −40 mV. An increase in [K⁺]_o from 0 to 10 mM evoked the outward current. Ouabain (10 μM) inhibited the outward current. Vertical excursions on the current trace indicate currents evoked by a ramp pulse of −80 to 40 mV over 100 ms, which correspond to i, ii and iii in B. The meanings of the dashed line, the bar at the left of the trace and the filled bar are the same as in Fig. 1. Cell capacitance was 20 pF and [Na⁺]_i was 50 mM. B, current traces recorded with a ramp pulse at 0 mM [K⁺]_o (i), 10 mM [K⁺]_o (ii) and 10 mM [K⁺]_o after application of 10 μM ouabain (iii). C, current traces of the isolated external K⁺-induced current obtained by subtraction (ii - i) and the isolated ouabain-sensitive current obtained by subtraction (ii - iii).

The current-voltage (I-V) relationship of the Na⁺-K⁺ pump current was approximately linear between −80 and −30 mV, while the amplitude was saturated at potentials more positive than 0 mV.

Effect of varying [K⁺]_o on the Na⁺-K⁺ pump current

In order to ascertain the [K⁺]_o dependence of the Na⁺-K⁺ pump current, [K⁺]_o was cumulatively increased from 0 to 10 mM in the presence of 50 mM [Na⁺]_i. The amplitude of the Na⁺-K⁺ pump current increased dose dependently with the increase in [K⁺]_o (Fig. 3A and B). Figure 3C summarizes the relationship between the current density of the Na⁺-K⁺ pump current (the amplitude normalized by the cell capacitance) and [K⁺]_o (n = 6 cells). When the data were fitted to the Hill equation, the current densitywas I_max/(1 + (K_d/[K⁺]_o)^n_H), where the maximal current density (I_max) was 1.3 ± 0.1 pA pF⁻¹, the binding constant (K_d) was 1.6 ± 0.3 mM and the Hill coefficient (n_H) was 1.1 ± 0.2.

Figure 3. Effect of varying [K⁺]_o on Na⁺-K⁺ pump current.

A, chart recording of the current with a cumulative increase in [K⁺]_o from 0 to 10 mM. The change in [K⁺]_o is shown at the top of the figure. Membrane potential was −40 mV and [Na⁺]_i was 50 mM. Cell capacitance was 20 pF. Vertical excursions on the current trace indicate currents evoked by a ramp pulse of −80 to 40 mV and the artifact due to the change in the solutions. Some excursions are cut off due to the frequency limitation of the chart recorder. The dashed line, the bar at the left of the trace and the filled bar are the same as in Fig. 1. The transient decrease in the Na⁺-K⁺ pump current in this figure was an artifact which resulted from the procedure to change the bath solution. B, current traces of Na⁺-K⁺ pump current evoked by a ramp pulse at various [K⁺]_o. Traces were obtained by subtraction of the currents recorded at 0 mM [K⁺]_o from those at various [K⁺]_o. C, relationship between amplitude of Na⁺-K⁺ pump currents and [K⁺]_o. The current amplitude was normalized by the cell capacitance and is shown as the current density. The current densities at 40 mV with various [K⁺]_o were plotted against [K⁺]_o. Values are means ± s.e.m. (n = 6 cells). The data were fitted to the Hill equation; the maximal current density (I_max) was 1.3 ± 0.1 pA pF⁻¹, the binding constant (K_d) was 1.6 ± 0.3 mM and the Hill coefficient (n_H) was 1.1 ± 0.2. D, I-V relationship of Na⁺-K⁺ pump current with various [K⁺]_o. Current densities were measured every 10 mV and then plotted against membrane potential. Data points are means ± s.e.m. (n = 4–6 cells). ^, 10 mM [K⁺]_o; •, 5 mM [K⁺]_o; ▴, 1 mM [K⁺]_o. E, normalized I-V relationship of Na⁺-K⁺ pump current at 1 and 10 mM [K⁺]_o. The current density at 40 mV was normalized as 1.0. The data are the same as in D.

Figure 3D shows the I-V relationship of the Na⁺-K⁺ pump current at 1, 5 and 10 mM [K⁺]_o. Current densities at every 10 mV point were obtained from six cells, and the average values were plotted against the membrane potentials. The normalized I-V relationship clearly shows that the I-V curve was not altered by the change in [K⁺]_o (Fig. 3E).

Effect of varying [Na⁺]_i on Na⁺-K⁺ pump current

The Na⁺-K⁺ pump current was recorded in the presence of various [Na⁺]_i values (Fig. 4). Figure 4A shows the Na⁺-K⁺ pump currents at 10 and 50 mM [Na⁺]_i recorded from two different cells. The amplitude was larger at the higher [Na⁺]_i. Figure 4B summarizes the relationship between the current densities of Na⁺-K⁺ pump currents and [Na⁺]_i (n = 4–19). The values of K_d and n_H were obtained as 22 ± 5 mM and 1.7 ± 0.5, respectively, from the dose-response curve.

Figure 4. Effect of varying [Na⁺]_i on Na⁺-K⁺ pump current.

A, Na⁺-K⁺ pump currents evoked by 10 mM [K⁺]_o recorded at 10 and 50 mM [Na⁺]_i. A ramp pulse from −80 to 40 mV was applied. The K⁺-induced components were isolated by subtraction and are shown. Two different cells were used for 10 and 50 mM [Na⁺]_i (cell capacitances were 20 and 18 pF, respectively). B, relationship between amplitude of Na⁺-K⁺ pump currents and [Na⁺]_i. Current densities at 40 mV with various [Na⁺]_i are plotted against [Na⁺]_i. Values are means ± s.e.m. (n = 4–19). The curve was drawn by fitting the data to the Hill equation; I_max was 1.3 ± 0.4 pA pF⁻¹, K_d was 22 ± 5 mM and n_H was 1.7 ± 0.5. C, I-V relationship of Na⁺-K⁺ pump current at 10 mM [Na⁺]_i and 50 mM [Na⁺]_i. ^, 50 mM [Na⁺]_i (n = 19); ▴, 10 mM [Na⁺]_i (n = 4). Data are shown as means ± s.e.m.D, normalized I-V relationship of Na⁺-K⁺ pump current at 10 mM [Na⁺]_i and 50 mM [Na⁺]_i. The current density at 40 mV was normalized as 1.0. Data are the same as in C.

The I-V relationships of the Na⁺-K⁺ pump current at 10 and 50 mM [Na⁺]_i are shown in Fig. 4C and D. The I-V relationship at 10 mM [Na⁺]_i was shifted in a more positive direction compared with that at 50 mM [Na⁺]_i.

Effect of ouabain on Na⁺-K⁺ pump current

Ouabain inhibited the Na⁺-K⁺ pump current in a dose-dependent manner (Fig. 5A). The dose-response relationships of the ouabain action on the Na⁺-K⁺ pump current are shown in Fig. 5B. When the dose-response relationship was fitted to the Hill equation, a K_d of 0.35 ± 0.03 μM and an n_H of 1.2 ± 0.1 were obtained (n = 6–8 cells) (Fig. 5C). Ouabain did not affect the I-V relationship (Fig. 5D and E).

Figure 5. Effect of ouabain on Na⁺-K⁺ pump current.

A, chart recording of Na⁺-K⁺ pump current evoked with 10 mM [K⁺]_o before and after application of ouabain. Ouabain concentration increased cumulatively, as indicated by the horizontal filled bar at the top of the figure. The holding potential was −40 mV and [Na⁺]_i was 50 mM. The vertical excursions, the dashed line and the horizontal bar at the left of the current have the same meanings as in Fig. 1. B, current traces of Na⁺-K⁺ pump current evoked by the ramp pulse from −80 to 40 mV before (Control) and after application of 1 μM ouabain. Cell capacitance was 18 pF. C, dose-response relationships of ouabain action on Na⁺-K⁺ pump currents. The relative amplitudes of the current at 40 mV with various ouabain concentrations were plotted against the ouabain concentration. The amplitude before application of ouabain was normalized to 1.0. Data are shown as means ± s.e.m. The curve was drawn by fitting the data to the Hill equation, where K_d was 0.35 ± 0.03 μM and the Hill coefficient was 1.2 ± 0.1 (n = 6–8). D, I-V relationship of Na⁺-K⁺ pump current in the absence and presence of 1 μM ouabain. Data are means ± s.e.m. (n = 6 cells). ^, without ouabain; ▴, 1 μM ouabain. E, normalized I-V relationship of Na⁺-K⁺ pump current in the absence and presence of 1 μM ouabain. The current density at 40 mV was normalized as 1.0. The data are the same as in D.

Temperature dependence of Na⁺-K⁺ pump current

Figure 6A demonstrates the changes in the Na⁺-K⁺ pump current when the temperature increased. The amplitude of the Na⁺-K⁺ pump current increased as the temperature increased from 26 to 36°C. The relationship between the amplitude of the Na⁺-K⁺ pump current and temperature is shown in Fig. 6B. The temperature coefficient, Q₁₀, was calculated as 1.87 (n = 9).

Figure 6. Temperature dependence of Na⁺-K⁺ pump current.

A, change in Na⁺-K⁺ pump current during an increase in temperature from 26 to 36 °C. Change in [K⁺]_o (upper line), bath temperature (middle line) and current traces recorded at −40 mV (lower trace). The lower and upper dashed lines indicate the current level at 26 °C without and with 10 mM K⁺, respectively. Cell capacitance was 18 pF. B, relationship between amplitude of Na⁺-K⁺ pump current and temperature. The relative current amplitudes at 26, 31 and 36 °C are shown; the amplitude at 36 °C is registered as 1.0. The straight line was obtained from fitting by linear regression. The temperature coefficient Q₁₀ was calculated as 1.87. Values are means ± s.e.m. (n = 9 cells).

Activation of Na⁺-K⁺ pump current by monovalent cations

To clarify whether various monovalent cations could activate the Na⁺-K⁺ pump, the Na⁺-K⁺ pump current was recorded in the presence of K⁺, Cs⁺, Rb⁺ and Li⁺. Figure 7A represents the Na⁺-K⁺ pump current evoked in the presence of 10 mM K⁺, Cs⁺, Rb⁺ or Li⁺. The currents recorded with Cs⁺, Rb⁺ and Li⁺ were also due to the activation of the Na⁺-K⁺ pump, since all currents were abolished by 10 μM ouabain (data not shown). The amplitude of current evoked with Cs⁺, Rb⁺ or Li⁺ was expressed as a relative value to that with K⁺ in Fig. 7B. The relative strength of the Na⁺-K⁺ pump activation was as follows: K⁺ (normalized as 1.0) ≥ Rb⁺ (0.84 ± 0.05, n = 4) > Cs⁺ (0.26 ± 0.04, n = 5) ≫ Li⁺ (0.09 ± 0.01, n = 4).

Figure 7. Cation dependence of extracellular binding site of Na⁺-K⁺ pump current.

A, current traces of Na⁺-K⁺ pump current evoked in the presence of various monovalent cations: 10 mM K⁺, 10 mM Rb⁺, 10 mM Cs⁺ and 10 mM Li⁺. Current was recorded with a ramp pulse from −80 to 40 mV. The cation-induced component was isolated by subtraction and is shown. B, relative amplitude of Na⁺-K⁺ pump current evoked with various monovalent cations. The current was first recorded with K⁺ and then with Rb⁺, Cs⁺ or Li⁺. The current amplitude with K⁺ was normalized to 1.0. The bars indicate means ± s.e.m. (n = 4–6 cells).

DISCUSSION

The present study characterized, for the first time, the Na⁺-K⁺ pump current in vascular smooth muscle cells. The basic characteristics of the Na⁺-K⁺ pump current in vascular smooth muscle cells are similar to those reported in other cells such as cardiac myocytes (Gadsby & Nakao, 1989; Nakao & Gadsby, 1989; Gao et al. 1995; Sakai et al. 1996) and vascular endothelial cells (Oike et al. 1993).

Recording of the Na⁺-K⁺ pump current

For the isolation of the Na⁺-K⁺ pump current, other voltage-dependent and time-dependent currents were inhibited by the following conditions: (i) a high Cs⁺ concentration in the pipette solution and TEA, 4-AP and Ba²⁺ in the bath solution to inhibit K⁺ channels such as delayed K⁺ channels, Ca²⁺-activated channels and inward rectifying K⁺ channels or (ii) Ni²⁺ added to the bath solution to inhibit Ca²⁺ channels (L- and T-type) and Na⁺-Ca²⁺ exchangers. Accordingly, the background current was time and voltage independent in the absence of extracellular monovalent cations other than Na⁺. Since we did not use Cl⁻ channel blockers, a minute contamination by Cl⁻ current could not be eliminated. However, most of the K⁺-induced current would be due to activation of the Na⁺-K⁺ pump, since the K⁺-induced current was abolished by ouabain.

The perforated patch-clamp method was used to minimize the possible washout of the intracellular component, which usually occurs in the conventional patch-clamp method. Nakao & Gadsby (1989) reported that the Na⁺-K⁺ pump current of guinea-pig ventricular myocytes runs down by 50 % in about 20 min with the use of a wide-tip recording pipette. In human endothelial cells, Oike et al. (1993) reported that the amplitude of the Na⁺-K⁺ pump current was small when recorded with the conventional patch-clamp method compared with that recorded with the perforated patch-clamp method. In the present study, the Na⁺-K⁺ pump current did not apparently run down for more than 20 min with the perforated patch-clamp method.

Dependence of the Na⁺-K⁺ pump current on extracellular K⁺ and intracellular Na⁺

The Na⁺-K⁺ pump current depended on extracellular K⁺. The K_d value of [K⁺]_o was 1.6 mM and the n_H was 1.1 in the present study. Previous studies using multicellular tissue of intestinal and uterine smooth muscle showed that the K_d for [K⁺]_o-induced change in the ouabain-sensitive K⁺ flux or Na⁺ flux ranged from 4.7 to 8 mM and the n_H value was more than 2 (Widdicombe, 1981). The K_d and n_H values, which were higher than those in the present study, might be due to the problems of accumulation and depletion of ions in the extracellular spaces of multicellular tissues. On the other hand, studies employing the voltage-clamp method for single cardiac muscle cells and single endothelial cells showed that K_d and n_H values were similar to those calculated in the present study; rabbit sino-atrial node cells (K_d = 1.4 mM, n_H = 1.2; Sakai et al. 1996), guinea-pig ventricular cells (K_d = 1.5 mM, n_H = 0.95; Nakao & Gadsby, 1989) and human endothelial cells (K_d = 2.4 mM; Oike et al. 1993).

The intracellular Na⁺ dependence of the Na⁺-K⁺ pump current in smooth muscle cells was not known previously. The present study showed that the K_d was 22 mM and the n_H was 1.7 for [Na⁺]_i. The K_d and n_H values were comparable to those in the voltage-clamp studies for rabbit sino-atrial node cells (K_d = 14 mM, n_H = 1.3; Sakai et al. 1996), guinea-pig ventricular cells (K_d = 11 mM, n_H = 1.4; Nakao & Gadsby, 1989) and sheep Purkinje fibres (K_d = 20 mM; Sejersted et al. 1988).

Studies using Na⁺-sensitive dye showed that the resting [Na⁺]_i in smooth muscle cells ranges from 10 to 20 mM (Ye et al. 1996; Lamont et al. 1998). Various conditions such as the presence of agonists and a pH change increased [Na⁺]_i by about 10 mM. Since the resting [Na⁺]_i was almost the same as the K_d value, a change in [Na⁺]_i under both physiological and pathological conditions would substantially alter the activity of the Na⁺-K⁺ pump.

Voltage dependence of the Na⁺-K⁺ pump current

The I-V curve of the Na⁺-K⁺ pump current was steep at −80 to 0 mV, while it was almost flat at more positive potentials. This voltage dependence of the Na⁺-K⁺ pump current was also observed in cardiac muscle cells (Nakao & Gadsby, 1989; Stimers et al. 1993; Sakai et al. 1996), but was not evident in endothelial cells (Oike et al. 1993). This would have a significant influence on the membrane potential, since depolarization would enhance the Na⁺-K⁺ pump current, which in turn would cause repolarization.

The change in [Na⁺]_i shifted the I-V relationship, but the change in [K⁺]_o did not. This result is basically the same as that observed in guinea-pig ventricular cells and rabbit sino-atrial node cells (Nakao & Gadsby, 1989; Sakai et al. 1996). However, such a shift by the change in [Na⁺]_i was not observed in rat cardiac muscle cells (Stimers et al. 1993). The mechanism for the formation of voltage dependence of the Na⁺-K⁺ pump current has not been clarified in the present study.

Na⁺-K⁺ pump activation by external monovalent cations

External monovalent cations such as Rb⁺, Cs⁺ and Li⁺, as well as K⁺, activated the Na⁺-K⁺ pump in the present study. The potency for evoking the Na⁺-K⁺ pump current was of the following order: K⁺ ≥ Rb⁺ > Cs⁺≫ Li⁺. The flux studies of multicellular smooth muscle tissues showed a similar sequence (K⁺ = Rb⁺ > Cs⁺ ≫ Li⁺) (Widdicombe, 1981). This result is also the same as that found in other cells such as guinea-pig ventricular myocytes (Eisner & Lederer, 1980), rabbit sino-atrial node cells (Kurachi et al. 1981) and rabbit Purkinje cells (Bielen et al. 1991), as determined with the patch-clamp method.

Effect of temperature on Na⁺-K⁺ pump current

The present study showed that the amplitude of the Na⁺-K⁺ pump current depended on the temperature. This result is in good accord with the temperature-dependent change in membrane potentials: lowering the temperature depolarizes the membrane potential of smooth muscle cells (Casteels, 1981; Stekiel et al. 1986).

In the present study, the Q₁₀ value for the Na⁺-K⁺ pump current was 1.87. This value was similar to that observed in rabbit sino-atrial node cells (Q₁₀ = 2.1; Sakai et al. 1996) and sheep Purkinje fibres (Q₁₀ = 1.67; Glitsch & Pusch, 1984), assessed by use of the voltage-clamp method.

Effect of ouabain on Na⁺-K⁺ pump current

Ouabain inhibited the Na⁺-K⁺ pump current dose dependently with a K_d of 0.35 μM in the present study. This ouabain sensitivity is in good accord with the previous studies, which examined the action of ouabain on membrane potential in vascular smooth muscle from rabbit and guinea-pig (Hendrickx & Casteels, 1974; Casteels et al. 1977; Harder & Sperelakis, 1978), while vascular smooth muscle from rats showed a lower sensitivity to ouabain (Magliola et al. 1986; Stekiel et al. 1986; Fujii et al. 1997). When compared with other tissues, the K_d value in the present study was almost the same as that in chick cardiac cells (0.5–10 μM; Stimers et al. 1991) and canine Purkinje cells (3.27 μM; Cohen et al. 1987), but was lower than that in human vascular endothelial cells (21 μM; Oike et al. 1993). In contrast to these cells, two K_d values were obtained in canine and guinea-pig ventricular cells (a high-affinity binding site with a K_d of 0.05 or 0.75 μM, and a low-affinity binding site with a K_d of 65 or 72 μM) (Mogul et al. 1989; Gao et al. 1995). The differences in the K_d values as well as the presence of two K_d values in different tissues and species are probably due to the presence of different α-subunits of Na⁺,K⁺-ATPase.

Several lines of evidence suggest that extracellular K⁺ competes with ouabain binding in the α-subunit (Eisner & Smith, 1991). The hypothesis is that glycosides bind preferentially to the phosphorylated α-subunit (known as E₂P) and stabilize this conformation. Extracellular K⁺ promotes the dephosphorylation of the α-subunit as an initial step in translocating the cation into the cytosol, thus decreasing the affinity of the α-subunit to ouabain by a type of kinetic competition. It was shown in chick cardiac cells that the K_d for ouabain increased from 0.56 to 10 μM when [K⁺]_o increased from 0.3 to 10.8 mM (Stimers et al. 1991). Since the ouabain action was examined at 10 mM [K⁺]_o in the present study, ouabain sensitivity might be slightly higher in the presence of physiological [K⁺]_o.

Contribution of Na⁺-K⁺ pump current to membrane potential

The current density of the Na⁺-K⁺ pump current in mesenteric artery cells was 1.2 ± 0.2 pA pF⁻¹ at 40 mV with 50 mM [Na⁺]_i and 10 mM [K⁺]_o in the present study. This value is similar to that observed in cardiac cells (1.1 ± 0.1 pA pF⁻¹ at 40 mV, Gadsby & Nakao, 1989; 1.5 ± 0.2 pA pF⁻¹ at 0 mV, Sakai et al. 1996), and in endothelial cells (approximately 1.4 pA pF⁻¹; Oike et al. 1993) under similar conditions. It is thus likely that the Na⁺-K⁺ pump makes an important contribution to cell function including the maintenance of membrane potential, as in these tissues. Indeed, ouabain application depolarized the membrane potential by 5–10 mV in tissue preparations from rabbit ear artery (Hendrickx & Casteels, 1974), rabbit main pulmonary artery (Casteels et al. 1977), guinea-pig mesenteric artery (Harder & Sperelakis, 1978), rat caudal artery (Hermsmeyer & Harder, 1986) and rat mesenteric artery (Stekiel et al. 1986; Fujii et al. 1997). The Na⁺-K⁺ pump current was 2–5 pA at 80 mV with 50 mM [Na⁺]_i and 5 mM [K⁺]_o (Fig. 3). If the input resistance of the cell was assumed to be 3 GΩ, the contribution of Na⁺-K⁺ pump current to the resting membrane potential would be 6–15 mV. We also observed that 10 μM ouabain depolarized single cells by 5 ± 2 mV (n = 5) in the current-clamp configuration under physiological conditions (20 mM [Na⁺]_i and 5 mM [K⁺]_o) (Y. Nakamura & Y. Ohya, unpublished observations).

In conclusion, the present study characterized for the first time the properties of Na⁺-K⁺ pump currents in vascular smooth muscle cells. Due to its electrogenic characteristics, voltage dependence and ion dependence, Na⁺,K⁺-ATPase may play an important role not only in the maintenance of the electrochemical gradients of Na⁺ and K⁺, but also in the control of membrane excitability, thus contributing to the maintenance of vascular tone under normal and pathological conditions.