Properties of the delayed rectifier potassium current in porcine sino-atrial node cells

Abstract

1. Whole-cell currents were recorded in single, spontaneously active cells dissociated from porcine sino-atrial node, and the conductance and gating properties of the delayed rectifier K⁺ current (I_K) were investigated.

2. The isolated cells exhibited spontaneous action potentials at a rate of 80.5 ± 5.4 min⁻¹ (mean ± s.e.m., n = 11). Under Ca²⁺ current block, depolarization from −40 mV to various potentials activated a time-dependent outward current (I_K). The activation curve of I_K showed a half-activation potential (V_½) of 20.5 ± 2.1 mV and a slope factor (S) of 16.4 ± 1.2 mV (n = 8).

3. As the duration of the depolarizing pulse to either +10 or +60 mV was prolonged, the amplitude of the tail current increased in proportion to that of the activated outward current during depolarization.

4. E4031 (2–5 μM), a selective blocker for the rapidly activating component of I_K (I_K,r), hardly affected I_K, but chromanol 293B, a selective blocker for the slowly activating component (I_K,s), inhibited I_K with an IC₅₀ of 8.79 μM.

5. The reversal potential of I_K was −75.2 ± 2.3 mV with 5.4 mM external and 150 mM internal K⁺. The time courses of activation and deactivation of I_K were fitted by the sum of two exponential functions at various potentials. The relationship between the time constants and membrane potential showed a bell-shaped curve with a peak at around −10 mV for both fast and slow components.

6. The results indicate that in porcine sino-atrial node cells I_K is largely derived from I_K,s and that I_K,s plays a functional role in the slow diastolic depolarization. I_K,s may, in part, account for the relatively slower heart rate of pigs than that of rabbit in which I_K,r is a functionally dominant component of I_K.

Slow diastolic depolarization underlies the automaticity of the sino-atrial node (SAN) cells. Until recently, most ionic currents contributing to the slow diastolic depolarization have been identified and analysed extensively in both multicellular and single cell preparations (for review see Irisawa et al. 1993). The currents include the delayed rectifier K⁺ current (I_K), the Na⁺-K⁺ pump current (Sakai et al. 1996), current due to the spontaneous opening of muscarinic K⁺ channels (Ito et al. 1994), L- and T-type Ca²⁺ currents (Hagiwara et al. 1988), hyperpolarization-activated cation current (I_f or I_h; DiFrancesco et al. 1986), background cation current (Hagiwara et al. 1992) and a dihydropyridine-sensitive sustained inward current (Guo et al. 1995). Using the conductance and gating parameters of individual current components, it is possible to reproduce the pacemaker activity in computer simulations (Yanagihara et al. 1980; Noble & Noble, 1984; Wilders et al. 1991). It should be noted, however, that most studies were carried out using rabbit hearts, and that the simulated action potentials only represent automaticity of the rabbit heart, the rate of which is approximately 200 beats min⁻¹. It appears that there are substantial differences in current components among different species. For example, I_K is largely attributed to the rapidly activating component of I_K (I_K,r) in rabbit SAN cells (Ito & Ono, 1995; Verheijck et al. 1995), whereas the slowly activating component of I_K (I_K,s) is a major component in guinea-pig SAN cells (Anumonwo et al. 1992). Since the conductance and gating properties of I_K,r and I_K,s are quite different (Sanguinetti & Jurkiewicz, 1991), their role in forming the spontaneous action potentials may differ depending on the species and heart rate.

In the present study, we isolated single pacemaker cells from porcine hearts. The porcine heart rate, ranging from 60 to 80 beats min⁻¹ in resting conditions, is much slower than that of rabbit. We show that the major component of I_K in porcine heart is I_K,s. The possible contribution of I_K,s to the action potentials of porcine SAN cells will be discussed.

METHODS

Preparation

Porcine heart was obtained from a slaughterhouse near the laboratory. Six-month-old pigs weighing 90–100 kg were killed by electrical shock and exsanguination, and their intra-thoracic and abdominal organs were quickly removed. The heart was also quickly excised, and the right coronary artery was cannulated to enable perfusion with Ca²⁺-free Tyrode solution. Ventricular branches of the coronary artery were clamped in order to facilitate perfusion towards the right atrium including the SAN region. The heart was first perfused with 150 ml Ca²⁺-free Tyrode solution (approximately 30°C) using a disposable syringe and was cooled by subsequent perfusion with 250–300 ml cardioplegic solution (see below) at 4°C. Thereafter, the right atrial region enclosed laterally by the crista-terminalis and the interatrial septum was dissected out, and stored in cardioplegic solution at 4°C for transfer to the laboratory. Approximately 1 h elapsed before the following isolation procedure was commenced in the laboratory.

A piece of SAN tissue measuring approximately 20 mm × 5 mm was cut perpendicular to the crista-terminalis into about 20 strips. In some preparations, the SAN artery of 200–300 μm in diameter could be clearly identified under the stereo-microscope. In such cases, strips were made so as to include the SAN artery at their centre. The pieces of nodal tissue were incubated in Ca²⁺-free Tyrode solution for 5 min and then in the same solution containing 480 U ml⁻¹ collagenase (Wako, Osaka, Japan), 0.1 mg ml⁻¹ elastase (Boehringer, Germany), 10 U ml⁻¹ protease (Type X; Sigma, St Louis, USA) and 3.5 U ml⁻¹ hyaluronidase (Wako) for 45–50 min at 34°C. After the enzyme treatment, strips of SAN tissue were placed into high-K⁺, low-Cl⁻ medium (modified KB medium) and gently teased apart using forceps and a glass pipette in a Petri dish. The cell suspension was stored in modified KB medium at 4°C for later use.

The protocols for isolating SAN cells described in this paper were previously approved by the Animal Committee, Akita University School of Medicine; the ‘Guidelines for Animal Experimentation’ of the University were completely adhered to in all subsequent experiments.

Solutions

Normal Tyrode solution contained (mM): NaCl, 136.9; KCl, 5.4; MgCl₂, 0.53; CaCl₂, 1.8; NaH₂PO₄, 0.33; glucose, 5.5; Hepes, 5.0 (pH = 7.4 with NaOH). Ca²⁺-free Tyrode solution was made by simply omitting CaCl₂ from the above normal Tyrode solution. To block L-type Ca²⁺ current, 0.3 μM nisoldipine was added to the normal Tyrode solution. The cardioplegic solution contained (mM): NaCl, 82; KCl, 20; MgCl₂, 0.5; glucose, 111.1; taurine, 20; Hepes, 10 (pH = 7.4 with NaOH), and was supplemented with 20 units l⁻¹ bovine insulin (Wako). Modified KB medium contained (mM): potassium glutamate, 70; KCl, 20; KH₂PO₄, 10; MgCl₂, 1.0; taurine, 20; glucose, 10; EGTA, 0.3 (pH = 7.2 with KOH).

The pipette solution for conventional whole-cell clamp experiments contained (mM): KOH, 150; HCl, 30; NaCl, 10; CaCl₂, 2; EGTA, 5; Na₂ATP, 5; Na₂GTP, 0.1; MgCl₂, 5; Hepes, 5.0 (pH = 7.2 with aspartic acid). The free concentrations of Ca²⁺ and Mg²⁺ were calculated to be 0.1 μM and 0.55 mM, respectively (Fabiato & Fabiato, 1979). In the perforated patch recording (Fig. 1), the pipette solution contained (mM): KCl, 30; potassium gluconate, 120; NaCl, 10; Hepes, 5.0 (pH = 7.2 with KOH). Nystatin (Sigma) was dissolved in methanol as a 10 mg ml⁻¹ stock solution, and added to the pipette solution to give a final concentration of 200 μg ml⁻¹ (Akaike & Harata, 1994).

Figure 1. Spontaneous action potentials of porcine SAN cells.

A, top: original current trace on the chart recorder. The action potentials were recorded under the nystatin perforated clamp mode. The dashed line indicates the zero current level. For comparison, typical action potentials of rabbit SAN cells are shown in the bottom panel (data from Ono & Ito, 1995). The spontaneous rate was 80.9 min⁻¹ for pig and 202.0 min⁻¹ for rabbit. B, spontaneous action potentials shown on an expanded time scale.

E4031 was kindly donated by Eisai Pharmaceutical Co. and dissolved directly in the external solution. Chromanol 293B, a chromanol derivative, was supplied by Hoechst Co. (Frankfurt, Germany) and dissolved in dimethylsulfoxide as a 10 mM stock solution. Before the experiment, the stock solution was diluted in external solution to obtain the concentrations described in the text.

Electrophysiological experiments

Whole-cell recordings were performed using the technique originally described by Hamill et al. (1981). Patch electrodes were made of glass capillaries of 1.5 mm o.d. and resistance in the range 3–5 MΩ when filled with the internal solution. The currents were recorded with a patch-clamp amplifier (List EPC-7, Darmstadt, Germany). An agar-Tyrode-Ag-AgCl bridge was used as an indifferent electrode. Current and voltage signals were stored on digital audiotape (TEAC RD-101T, Tokyo, Japan) for subsequent computer analysis (NEC PC98Ap, Tokyo, Japan). All experiments were carried out at 35–37°C.

Results are expressed as means ±s.e.m. (n, number of cells). Comparisons between two groups were performed using Student's paired or unpaired t test and values of P < 0.05 were considered statistically significant.

RESULTS

Action potentials of porcine SAN cells

Figure 1 shows a typical record of spontaneous activity recorded using the nystatin perforated patch-clamp technique. The overall electrical activity of porcine SAN cells was qualitatively similar to that recorded in rabbit preparations (Irisawa et al. 1993). Quantitatively, however, the action potential of the porcine SAN cells was characterized by a much slower spontaneous rate, pacemaker depolarization and upstroke, and by a longer duration, compared with that of the rabbit preparation. This is clearly demonstrated in Fig. 1B, where typical action potentials of rabbit SAN cells are shown superimposed on the same time scale. The spontaneous rate, measured in 11 cells, ranged between 47.8 and 118.3 min⁻¹. The action potential parameters are summarized in Table 1. For comparison, the action potential parameters of rabbit SAN cells are also shown (Ono & Ito, 1995).

Table 1.

Action potential parameters of the isolated porcine SAN cells

Parameters	Pig	Rabbit
Cycle length (ms)	781.7 ± 58.7	294.3
Spontaneous rate (min−1)	80.5 ± 5.4	203.9
Amplitude (mV)	75.1 ± 2.4	95.8
Vmax (V s−1)	1.57 ± 0.15	10.3
APD50 (ms)	196.6 ± 19.9	n.d.
MDP (mV)	−58.4 ± 1.9	−70.7
DDR (mV s−1)	34.4 ± 2.4	68.7
Reference	Present study	Ono & Ito (1995)

Values are means ± s.e.m. obtained from 11 cells. V_max, maximum upstroke velocity; APD₅₀, action potential duration measured at 50% repolarization; MDP, maximum diastolic potential; DDR, diastolic depolarization rate. n.d., not determined.

Under conventional whole-cell voltage-clamp conditions using the standard internal solution and Tyrode solution, square pulses of 500 ms duration were applied from a holding potential of −50 mV (Fig. 2). Upon depolarization a transient inward current due to L-type Ca²⁺ current was followed by a slowly activating delayed rectifier K⁺ current (I_K). The activation of I_K is evident as a decaying tail current on repolarization to −50 mV. Hyperpolarization activated a slowly activating inward current, I_f. The current-voltage (I–V) relationships measured at the initial peak (○) and at the end (•) of the pulses are shown in Fig. 2B. The configuration of the I–V curve for the initial current was in rough agreement with that reported in rabbit SAN cells (Irisawa et al. 1993; Ito & Ono, 1995). For the late current, however, there was a difference between the two species: the I–V curve of rabbit SAN cells showed upward curvature between -30 and +20 mV (Irisawa et al. 1993; Ito et al. 1994; Ono & Ito, 1995) whereas the late current in porcine SAN cells increased monotonically as the test potential was increased (Fig. 2B). These findings indicate that the properties of I_K may differ between rabbit and porcine SAN cells.

Figure 2. Whole-cell currents of porcine SAN cells.

A, whole-cell current of a SAN cell in normal Tyrode solution. Traces shown were obtained by applying 500 ms depolarizing (upper traces) or hyperpolarizing (lower traces) pulses from a holding potential of −50 mV in 10 mV increments. Dashed lines indicate the zero current level. B, I-V relationships for the initial current (○) and the current at the end of the pulses (•).

To examine the properties of I_K, long depolarizing pulses were applied during block of the L-type Ca²⁺ current by 0.3 μM nisoldipine (Fig. 3A). In response to a 3 s depolarizing pulse, I_K was activated gradually with little saturation during the pulse. In Fig. 3B, the current densities of the initial (○) and late (•) current measured in eight cells are plotted against the test pulses (V_t). In Fig. 3C, the amplitude of the tail current is plotted against V_t. The relationship was well fitted by the Boltzmann equation:

where V_½ is the half-activation potential and S is a slope factor. The values of V_½and S were determined by a least-squares fit; they were 20.5 ± 2.1 mV and 16.4 ± 1.2 mV, respectively (n = 8).

Figure 3. Delayed rectifier K⁺ current in porcine SAN cells.

A, whole-cell current in external solution containing 0.3 μM nisoldipine. The holding potential (V_h) was −40 mV and 3 s depolarizing pulses were applied in 10 mV increments. V_t, test potential. B, I–V relationships for the initial current (○) and the current at the end of the 3 s pulses (•). Data are means of 8 cells. In this and subsequent figures, the s.e.m. is not shown when the value is smaller than the symbol. C, relationship between the amplitude of the tail current and V_t. The amplitude of the tail current was normalized to the maximum value in each experiment. Data are means ±s.e.m. of 8 cells. The smooth curve is the best fit with the Boltzmann equation: I = I_max/(1 + exp(-(V_t–V_½)/S)), for the averaged data. V_½ and S were 18.4 mV and 14.0 mV, respectively.

Envelope test

The time course of development of the tail current compared with that of the activated outward I_K provides a clue to the relative contribution of I_K,r and I_K,s to net I_K (Sanguinetti & Jurkiewicz, 1990). In the experiment shown in Fig. 4A, the duration of the test pulse at +60 mV (upper traces) and +10 mV (lower traces) was varied. As the duration of the test pulse was prolonged, the amplitude of the tail current increased in proportion to that of the time-dependent outward current activated during depolarization. Data obtained from five cells for +60 mV and six cells for +10 mV are summarized in Fig. 4B, where the tail amplitude (I_tail) is plotted against the activated outward current upon depolarization (I_out). I_out was measured as the time-dependent current during the step depolarization. A linear relationship is evident between I_tail and I_out during the entire course of the pulses in each cell examined. The slope of the linear regression line was 0.50 ± 0.05 for +60 mV and 0.56 ± 0.02 for +10 mV. These findings indicate a single population of I_K in porcine SAN cells.

Figure 4. Envelope test of I_K.

A, tail currents at −40 mV were recorded by changing the duration of the test depolarization to +60 mV (upper traces) and +10 mV (lower traces). B, relationship between the amplitude of the tail current (I_tail) and the activated current (I_out) during the depolarizing pulses. Different symbols indicate different cells.

Pharmacological properties of porcine SAN cell I_K

In rabbit SAN cells, I_K was markedly inhibited by E4031, a selective I_K,r blocker (Ito & Ono, 1995; Ono & Ito, 1995; Verheijck et al. 1995). By contrast, the drug hardly affected the I_K of porcine SAN cells. To monitor the amplitudes of both I_K,r and I_K,s, a pair of 3 s depolarizing pulses to +10 and +60 mV were applied every 60 s. After observation of the control I_K, 2 μM E4031 was applied. Little or no change in the current configuration was observed at both +10 and +60 mV (Fig. 5A). In fact, when the E4031-sensitive current was calculated by subtracting the current in the presence of E4031 from the control current, there was little or no current during the depolarization step or upon repolarization to −40 mV (Fig. 4B). In five cells, the mean amplitude of the activated outward current was 101.0 ± 7.8 pA at +10 mV and 696.2 ± 70.1 pA at +60 mV in the control, and 94.3 ± 7.8 pA at +10 mV and 665.6 ± 70.6 pA at +60 mV in the presence of 2–5 μM E4031. The amplitude of the tail current was 47.5 ± 3.9 pA at +10 mV and 273.6 ± 24.7 pA at +60 mV in the control, and 45.8 ± 4.0 pA at +10 mV and 262.0 ± 24.7 pA at +60 mV in the presence of 2–5 μM E4031. These differences were not statistically significant.

Figure 5. Effects of E4031 on porcine I_K.

A, superimposed currents in response to 3 s test pulses to +60 and +10 mV recorded in the absence and presence of 2 μM E4031. The holding potential was −40 mV and L-type Ca²⁺ current was blocked by 0.3 μM nisoldipine. B, E4031-sensitive current obtained by subtracting the currents in the presence of E4031 from control currents. In A and B, the dashed line indicates the zero current level.

It has been shown that chromanol 293B, a chromanol derivative, selectively inhibits I_K,s in guinea-pig ventricular cells without affecting I_K,r (Busch et al. 1996). In the experiment shown in Fig. 6, whole-cell outward currents were compared before (Fig. 6A) and during (Fig. 6B) application of 30 μM chromanol 293B. In the presence of 30 μM chromanol 293B, no obvious time-dependent current was recorded over the potential range −30 to +60 mV, and I_K tail current was almost completely blocked. These findings may support the view that I_K,r is negligibly small in this preparation. The chromanol 293B-sensitive current was obtained by digitally subtracting the current under the effect of chromanol 293B from the control current at each potential (Fig. 6C). The amplitude became larger as the depolarizing pulse was increased, and rectification was not detected over the voltage range examined.

Figure 6. Block by chromanol 293B of porcine I_K.

Whole-cell current in the control external solution (A) and in the presence of 30 μM chromanol 293B (B). The holding potential was −40 mV and 3 s depolarizing pulses were applied in 10 mV increments. C, chromanol 293B-sensitive current obtained by subtracting the currents in the presence of chromanol 293B from control currents. D, relationship between the amplitude of the tail current and the concentration of chromanol 293B. The amplitude of the tail current was normalized to that recorded in the absence of chromanol 293B. The smooth curve is the best fit with the Hill equation: I = 1/(1+ ([Drug]/IC₅₀)^n_H), where IC₅₀ and n_H are the half-inhibitory concentration and the Hill coefficient, respectively. The values of IC₅₀ and n_H were 8.79 μM and 1.22, respectively. In A-C, the dashed line indicates the zero current level.

The concentration-response relationship was examined in 18 cells and the results are shown in Fig. 6D. The relationship was fitted by the Hill equation:

where IC₅₀ and n_H are the half-inhibitory concentration and the Hill coefficient, respectively. The values of IC₅₀ and n_H were 8.79 μM and 1.22, respectively. The IC₅₀ value was slightly larger than that reported in guinea-pig ventricular cells (IC₅₀ = 2.1 μM; Busch et al. 1996).

Ion selectivity of I_K

The ion selectivity of I_K was examined by measuring the reversal potential (V_rev) of the tail current. In the experiment shown in Fig. 7A, the membrane was first depolarized to +60 mV for 3 s and then repolarized to various test voltages. The initial (○) and late (•) currents recorded during the second pulse were plotted against membrane potential, except at potentials more negative than −70 mV where the currents before I_f activation were plotted as the late current. V_rev, determined as the intersection of these relationships, was −79 mV in this experiment. The mean V_rev, determined in four cells in standard external solution (5.4 mM [K⁺]_o), was −75.2 ± 2.3 mV. In Fig. 7B, summarized results of V_rev measured at variable [K⁺]_o are shown. The linear regression line of the V_rev-log[K⁺]_o relationship had a slope of 52.0 mV per tenfold change in [K⁺]_o. The value was obviously smaller than that for a K⁺-selective electrode expected from the Nernst equation (dotted line in Fig. 7B), indicating that I_K in porcine SAN cells is sensitive to, but not completely selective for, K⁺.

Figure 7. Ion selectivity of porcine SAN cells.

A, left: current traces obtained using the double pulse protocol shown at the top. The dotted line indicates the zero current level. Right: I-V relationships for the initial (○) and late (•) tail current. The late current was measured just before the end of the test pulse, except for data at potentials more negative than −70 mV where the current amplitude was measured before activation of I_f. B, relationship between the reversal potential and [K⁺]_o. The linear regression line had a slope of 52.0 mV per tenfold increase in [K⁺]_o. The dotted line represents a theoretical linear relationship expected for a K⁺-selective electrode calculated from the Nernst equation: V_rev = 60log([K⁺]_o/150).

Time constants

The kinetics of chromanol 293B-sensitive I_K were analysed. The time course of activation in response to depolarizing pulses was fitted with either a single or double exponential function. The data were better fitted by the sum of two exponential functions:

where I_∞ is a free parameter, I₀ is the instantaneous current, and τ_f and τ_s are the fast and slow activation time constants, respectively. It should be noted that, in previous studies, the onset kinetics of the cardiac delayed rectifier K⁺ current exhibited significant sigmoidicity (Matsuura et al. 1987; Balser et al. 1990). However, such sigmoidicity was not obvious in the chromanol 293B-sensitive current of porcine SAN cells, as illustrated in Fig. 6C.

The deactivating tail current was obtained using the double pulse protocol (see Fig. 7). The tail current was also fitted with the sum of two exponential functions, except at potentials more negative than −60 mV where a single exponential function was sufficient. Data obtained from five cells are summarized in Fig. 8. Both the slow and fast time constants were clearly dependent on the membrane voltage and, when plotted against membrane potential, showed a bell-shaped curve with a maximum around −10 mV. The slope was much steeper at negative potentials whereas both time constants showed only weak voltage dependency at potentials more positive than +10 mV.

Figure 8. Time constants of activation and deactivation of I_K.

A, time constants for slow (filled symbols) and fast (open symbols) components plotted against membrane potential. Activation phases (circles) were recorded during depolarization for 3 s from −40 mV to various potentials, while deactivation phases (squares) were recorded by repolarization from +60 mV. Each phase was fitted with the sum of two exponentials. Data are means ±s.e.m. of 5 experiments. Inset, original current traces. The dashed line indicates the zero current level. B, relative amplitude of the fast component (F/(F + S)) plotted against membrane potential. S, slow component. Data obtained from activation (○) and deactivation (•) phases are shown.

Effect of chromanol 293B on the spontaneous action potentials

Complete block of I_K,s by chromanol 293B abolished the spontaneous activity of porcine SAN cells. In the experiment shown in Fig. 9, the action potential was recorded using the nystatin-perforated patch-clamp method. After confirming regular action potentials, 10 and 30 μM chromanol 293B was cumulatively applied, causing complete cessation of the action potential in a reversible manner. Essentially similar results were obtained in three other cells.

Figure 9. Effects of chromanol 293B on the spontaneous action potentials of porcine SAN cells.

A, after obtaining a stable firing of action potentials in a nystatin-perforated patch, 10 and 30 μM chromanol 293B was superfused into the bath. The bar indicates the period of drug application. B, traces at the times indicated in A shown on an expanded time scale.

DISCUSSION

I_K,s is characterized as a slowly activating I_K; the activation threshold is around −20 mV and V_½ is about +10 mV. The time course of activation and deactivation requires several seconds and the fully activated I–V relationship shows little rectification. By contrast, I_K,r is activated at more negative potentials (V_½ is around −25 mV) and shows strong inward rectification (Sanguinetti & Jurkiewicz, 1990; Ono & Ito, 1995). These currents can also be distinguished pharmacologically. I_K,r is specifically blocked by class III antiarrhythmic agents (Sanguinetti & Jurkiewicz, 1990), whereas I_K,s is inhibited by chromanol 293B (Busch et al. 1996). In the present study, we have demonstrated that I_K of porcine SAN cells possesses electrophysiological and pharmacological properties quite similar to those of I_K,s. The reversal potential of −75 mV, measured with 5.4 mM external and 150 mM internal K⁺, is ∼10 mV positive to the equilibrium potential expected for a K⁺-selective electrode from the Nernst equation. This is consistent with reports that I_K,s is sensitive to, but not completely selective for, K⁺ (Matsuura et al. 1987; Sanguinetti & Jurkiewicz, 1990; Anumonwo et al. 1992). Furthermore, block of I_K,s inhibited the spontaneous action potentials of porcine SAN cells. All these findings indicate that not only is I_K,s the dominant outward K⁺ current, but also that it plays a functional role in the pacemaker activity of porcine SAN cells.

The isolated porcine SAN cells exhibited spontaneous activity of which the spontaneous rate was similar to the native heart rate of pigs, indicating that the mechanism underlying different heart rates among various species can be attributed to intrinsic properties of isolated SAN cells in individual species. The action potentials of porcine SAN cells showed a relatively slower upstroke, longer duration and slower diastolic depolarization compared with those of rabbit SAN cells (Table 1). The slow upstroke indicates that the density of the Ca²⁺ current might be low, or that the property of the Ca²⁺ current might be different from that of rabbit SAN cells which usually show the maximal upstroke, around 10 V s⁻¹ (see Table 1; Wilders et al. 1991; Ono & Ito, 1995). On the other hand, the relatively longer duration of the action potential and slower diastolic depolarization of porcine SAN cells than those of rabbits may be brought about, at least in part, by the difference in I_K kinetics between the two species. The functional role of time-dependent outward current, I_K, in the pacemaker potential has long been suggested (Irisawa et al. 1993). Namely, the activation of I_K during the preceding action potential drives the membrane towards repolarization, and I_K deactivation triggers the slow diastolic depolarization after the maximum diastolic potential (Irisawa et al. 1993). In the case of rabbit SAN cells, I_K,r has been shown to play this active role (Ono & Ito, 1995; Verheijck et al. 1995). Application of chromanol 293B to the rabbit SAN cells did not affect spontaneous activity (Lei et al. 1997). This finding indicates that I_K,s, which has been reported to exist in rabbit SAN cells (Ono & Ito, 1995; Lei & Brown, 1996), cannot participate in regulating the spontaneous action potentials of rabbit SAN cells, probably due to its slower gating and relatively higher threshold compared with I_K,r. On the other hand, the present study has demonstrated that chromanol 293B inhibited the spontaneous action potentials in porcine SAN cells in a reversible manner. This effect is not likely to be due to possible block of L-type Ca²⁺ current by the drug, since chromanol 293B selectively inhibited I_K,s in human and guinea-pig ventricular cells without affecting L-type Ca²⁺ current and Na⁺ current (Bosch et al. 1998). Also, I_f of porcine SAN cells was not affected by chromanol 293B (K. Ono & T. Iijima, unpublished data). Thus our results indicate that I_K,s is required for generating pacemaker activity in porcine SAN cells. Furthermore, I_K,s is the only time-dependent outward current that can drive the membrane towards the maximum diastolic depolarization after the peak of the action potential. The longer action potential duration of pigs may enable I_K,s to replace the role of rabbit I_K,r in porcine SAN cells, thereby causing a slower heart rate compared with that of rabbits.

Although we failed to record I_K,r in porcine SAN cells which showed little striation and were spindle shaped, I_K,r did exist in large, rod-shaped quiescent cells in the same batches of preparation (authors’ unpublished data). I_K,r and I_K,s may be distributed heterogeneously in the SAN region depending on the cell shape. Heterogeneous distribution of I_K,r or I_K,s has been suggested in various species. Recent molecular studies have revealed that the minK channel (also termed IsK), which is considered to form I_K,s channels along with another K⁺ channel subunit, KvLQT1 (Barhanin et al. 1996; Sanguinetti et al. 1996a), is expressed in the SAN region of the guinea-pig (Freeman & Kass, 1993a), ferret (Brahmajothi et al. 1996) and mouse (Kupershmidt et al. 1999). On the other hand, mRNA of ERG, which forms the I_K,r channel (Sanguinetti et al. 1996b) together with another subunit, MiRP1 (Abbott et al. 1999), is substantially expressed in the SAN region of the rabbit heart (Wymore et al. 1997) and in the crista-terminalis of ferret heart including the SAN region (Brahmajothi et al. 1997). Thus, the distribution of I_K,s and I_K,r channels varies depending on the species, and could be potentially related to the different spontaneous rate of SAN cells in various species. In this respect, it would be interesting to determine the expression and distribution of various K⁺ channels including ERG, minK, MiRP1 and KvLQT1 in the SAN region of porcine hearts.

Apart from the molecular analysis described above, direct measurement of membrane currents from mammalian SAN cells has been performed in rabbits (for review see Irisawa et al. 1993), guinea-pigs (Anumonwo et al. 1992; Guo et al. 1997) and pigs (present study). The different kinetics of I_K,s and I_K,r may, in part, account for the difference between the heart rate of rabbits (approximately 200 beats min⁻¹) and pigs (∼80 beats min⁻¹). It is interesting to note that I_K of guinea-pig SAN cells is also derived from I_K,s, not I_K,r, despite a marked difference in the spontaneous rate compared with that of porcine SAN cells (Anumonwo et al. 1992; Guo et al. 1997). It should be noted that the properties of I_K,s appear to be different, depending on the mammalian species. The half-activation voltage was +32 mV in guinea-pig (Anumonwo et al. 1992) and +20 mV in porcine SAN cells (Fig. 3). The relationship between fast and slow time constants and the membrane potential showed a clear bell-shaped curve peaking near 0 to +20 mV in guinea-pig SAN cells. By contrast, the voltage dependence of porcine I_K,s was not obvious at potentials more positive to +10 mV (Fig. 8A). Furthermore, the values for fast and slow time constants are generally slower for porcine I_K,s (present study) than for I_K,s of guinea-pig SAN cells (Anumonwo et al. 1992). These properties might be beneficial for forming a longer action potential duration in porcine SAN cells, compared with that in guinea-pig. In addition to the different gating properties of I_K,s, the current density may also provide an important variable for determining the contribution of I_K,s to the action potential in various species. The current density of I_K,s is only 8 pA pF⁻¹ at +60 mV in porcine SAN (Fig. 3), whereas Anumonwo et al. (1992) reported a current size of approximately 1 nA at +80 mV in guinea-pig SAN cells which have cell capacitance of less than 10 pF. Such a large current density might be required for I_K,s to play a functional role in the spontaneous activity of guinea-pig SAN cells.

It should also be noted that the properties of I_K,s do not only depend on the species but also on the cell type in the same heart. Anumonwo et al. (1992) reported that the time constants for activation and deactivation of I_K,s were faster in guinea-pig SAN cells than those in ventricles. Freeman & Kass (1993b) reported that regulation of I_K,s in guinea-pig SAN cells is not temperature dependent, in contrast to ventricular I_K,s in which regulation by kinases A and C was diminished by lowering the temperature (Walsh & Kass, 1988). Non-stationary fluctuation analysis predicted a single channel conductance of 6 pS in the SAN I_K,s channel, which is larger than that estimated for ventricular I_K,s channels of guinea-pigs (Freeman & Kass, 1993b).

Possible contributing factors that could explain the differences in I_K,s kinetics among various species and cell types described so far remain to be elucidated. One possibility is that primary structure (KvLQT1 and/or minK) and the resulting kinetic properties of the I_K,s channel may differ depending on the species. A second possibility may be related to previous observations that coexpression of minK with KvLQT1 slowed activation kinetics without much alteration in the deactivation process, decreased the single channel amplitude of KvLQT1 (Romey et al. 1997), and slowed the blocking rate of KvLQT1 by chromanol 293B (Loussouarn et al. 1997). Romey et al. (1997) proposed that a tetramer of KvLQT1 can bind one to four minK subunits depending on the concentration of minK in the membrane, thereby producing the complex kinetic behaviour of I_K,s channels. It might be speculated that the stoichiometry of minK in the I_K,s channel complex is different in various regions of the heart. Finally, there may be KvLQT1 isoforms that have different gating properties or serve various functions, like that reported in human heart (Jiang et al. 1997).

Not only I_K but also various other ionic channels are important in controlling the rate of diastolic depolarization of SAN cells. In rabbit SAN cells, the currents involved include T- and L-type Ca²⁺ current (Hagiwara et al. 1988), background cation current (Hagiwara et al. 1992), I_f (DiFrancesco et al. 1986), sustained inward current (Guo et al. 1995), Na⁺-K⁺ pump current (Sakai et al. 1996) and ionic current due to spontaneous opening of muscarinic K⁺ channels (Ito et al. 1994). Shibata et al. (1999) have demonstrated recently that the activation kinetics of I_f in porcine SAN cells are quite similar to those of rabbit. As for other current systems, however, little is known of whether gating property and/or current density of these currents are different between rabbits and pigs. Further studies are clearly necessary to elucidate the mechanism underlying the pacemaker activity in porcine SAN cells.

In conclusion, we have recorded for the first time the spontaneous action potentials of porcine SAN cells, and demonstrated that I_K,s plays a major role in the membrane repolarization of these cells. The difference in heart rate between various species, which might be caused by differences in current components generating the pacemaker potential and/or the current density of various ionic channels in SAN cells, should be given attention for understanding the pacemaker mechanism of the mammalian heart, including that of humans.