Four Kinetically Distinct Depolarization-activated K⁺ Currents in Adult Mouse Ventricular Myocytes

Depolarization-activated K⁺ currents play key roles in determining the amplitudes and durations of action potentials in cardiac cells, and several distinct types of voltage-gated K⁺ channels that subserve these functions have been identified (for reviews, see Anumonwo et al., 1991; Campbell et al., 1995; Kass, 1995; Barry and Nerbonne, 1996; Giles et al., 1996). This diversity has a physiological significance in the heart in that the various K⁺ currents underlie distinct phases of action potential repolarization (Anumonwo et al., 1991; Campbell et al., 1995; Kass, 1995; Barry and Nerbonne, 1996; Giles et al., 1996), and cell type–specific differences in K⁺ channel expression contribute to regional variations in action potential waveforms (Antzelevitch et al., 1995; Barry and Nerbonne, 1996). Voltage-gated K⁺ channels are primary targets for the actions of a variety of endogenous neurotransmitters and neurohormones, as well as exogenous drugs that modulate cardiac functioning (Anumonwo et al., 1991; Gadsby, 1995). In addition, changes in the densities and/or the properties of K⁺ currents occur in conjunction with myocardial damage or disease and these changes can have profound physiological consequences, including leading to the generation of life threatening arrhythmias (Ten Eick et al., 1989; Näbauer et al., 1993; Boyden and Jeck, 1995; Wilde and Veldkamp, 1997; VanWagoner et al., 1997; Näbauer and Kääb, 1998). For all of these reasons, there is considerable interest in defining the molecular correlates of functional cardiac K⁺ channels (Barry and Nerbonne, 1996; Deal et al., 1996), and in understanding the mechanisms controlling the regulation, modulation, and functional expression of these channels.

A number of voltage-gated K⁺ channel pore-forming (α) and accessory (β) subunits have now been cloned from heart cDNA libraries, and a variety of experimental approaches are being exploited to probe the molecular basis of functional K⁺ channel diversity in mammalian cardiac cells (for reviews, see Barry and Nerbonne, 1996; Deal et al., 1996; Nerbonne, 1998). Of these, transgenic and knockout strategies seem particularly promising because the functional consequences of manipulating K⁺ channel expression in vivo can also be explored directly. Indeed, mice with targeted Kv α subunit deletions, including Kv3.1, Kv1.1, and Kv1.4, have been described recently (Ho et al., 1997; London et al., 1998b; Smart et al., 1998). None of these animals appears to display a cardiac phenotype, however, leading to suggestions that Kv1.1, Kv3.1, and Kv1.4 likely do not play roles in the generation of functional cardiac K⁺ channels, at least in the mouse (Ho et al., 1997; London et al., 1998b; Smart et al., 1998). London et al. (1998a), however, recently reported the generation of transgenic mice expressing a truncated Kv1.1 (Kv1.1N206Tag), driven by the α-myosin heavy chain promoter to direct cardiac-specific expression of the transgene (Lyons et al., 1990; Ng et al., 1991; Palermo et al., 1996). In vitro, Kv1.1N206Tag functions as a dominant negative, reducing or eliminating heterologously expressed Kv1.4 and Kv1.5 currents (Folco et al., 1997). In ventricular myocytes isolated from Kv1.1N206Tag-expressing mice, a 4-aminopyridine–sensitive, slowly decaying outward K⁺ current, I_K,slow, was found to be selectively attenuated (London et al., 1998a). In addition, action potentials and QT intervals are prolonged in the Kv1.1N206Tag-expressing transgenics, and electrocardiographic recordings revealed increased frequency of premature ventricular beats and spontaneous ventricular tachycardia in these animals (London et al., 1998a).

A dominant negative strategy has also been exploited by Barry et al. (1998) in studies focussed on identifying the molecular correlate of the cardiac transient outward K⁺ current, I_to. Using an approach previously used to make Kv1 α subunits that gate on membrane depolarization but do not conduct (Perozo et al., 1993; Ribera, 1996), mutations were introduced into the coding sequence of the pore region of Kv4.2 to convert the tryptophan (W) in position 362 to phenylalanine (F) to produce Kv4.2W362F, and, in in vitro experiments, Kv4.2W362F was shown to function as a dominant negative against Kv4.2 and Kv4.3 (Barry et al., 1998). In ventricular myocytes isolated from Kv4.2W362F-expressing animals, I_to (I_to,f) is eliminated, and action potential durations are increased significantly (Barry et al., 1998). Although QT intervals are also markedly prolonged in Kv4.2W362F-expressing transgenics, these animals do not develop spontaneous arrhythmias (Barry et al., 1998). However, a “novel” rapidly activating, slowly inactivating K⁺ current, not clearly evident in wild-type cells, was identified in ventricular myocytes isolated from Kv4.2W362F-expressing animals, an observation interpreted as suggesting that electrical remodeling occurs in the myocardium when the expression of endogenous K⁺ channels is altered (Barry et al., 1998).

In spite of the growing interest in using transgenic and knockout mice, the electrophysiological properties of adult mouse cardiac myocytes have not been characterized in detail to date. This fact and the finding of the novel K⁺ current in Kv4.2W362F-expressing ventricular myocytes prompted us to undertake studies focussed on examining the time- and voltage-dependent properties and the pharmacological sensitivities of the voltage-gated K⁺ currents in adult mouse ventricular myocytes. The results of these experiments reveal the presence of four kinetically and pharmacologically distinct voltage-gated K⁺ currents in these cells: (a) a rapidly activating and inactivating, “fast transient” K⁺ current, I_to,f; (b) a rapidly activating, slowly inactivating, “slow transient” current, I_to,s; (c) a rapidly activating, very slowly inactivating current, I_K,slow; and (d) a slowly activating, noninactivating K⁺ current, I_ss. Interestingly, the results presented here reveal that I_K,slow and I_ss are expressed in all adult mouse ventricular cells, whereas I_to,f and I_to,s are differentially distributed.

Ventricular myocytes were isolated from adult (after postnatal day 45) C57BL6 mice using a procedure previously developed and used to isolate rat cardiomyocytes (Xu et al., 1996). In brief, hearts were excised from anesthetized (5% halothane/95% O₂) adult animals, mounted on a Langendorf apparatus, and perfused retrogradely through the aorta with 40 ml of a Ca²⁺-free HEPES-buffered Earles balanced salt solution (Gibco Laboratories) supplemented with 6 mM glucose, amino acids, and vitamins (Buffer A). Hearts were then perfused with 50 ml of Buffer A containing 0.8 mg/ml collagenase B (Boehringer Mannheim Biochemicals) and 10 μM CaCl₂, and the temperature of the tissue and the perfusate were maintained at 34–35°C. The enzyme solution was filtered (at 5 μm) and recirculated through the heart for ∼15–20 min.

In initial experiments, the lower two thirds of the left and right ventricles were removed after the perfusion, and, after mincing in enzyme-containing Buffer A, were transferred to fresh (enzyme-free) Buffer A supplemented with 1.25 mg/ml taurine, 5 mg/ml bovine serum albumin (Sigma Chemical Co.), and 150 μM CaCl₂ (Buffer B). In experiments focussed on determining whether there are regional differences in K⁺ current expression, the heart was cut open after perfusion, and the ventricular septum and the top ∼0.3 mm of tissue at the apex of the left ventricle were removed. The tissue pieces from the apex and septum were placed separately in fresh Buffer B. After mechanical dispersion (by gentle trituration), cell suspensions were filtered to remove large undissociated tissue fragments, and cells were collected by sedimentation. Isolated myocytes were resuspended in fresh Buffer B, plated on laminin-coated coverslips, and placed in a 95% air/5% CO₂ incubator at 37°C. Approximately 30 min after plating, serum-free medium-199 (M-199; Irvine Scientific), supplemented with antibiotics (penicillin/streptomycin), was added. Ca²⁺-tolerant ventricular myocytes adhered to the laminin, and damaged cells were removed by replacing the medium with fresh M-199 ≈ 1 h after plating. Cells were examined electrophysiologically within 48 h of isolation.

The conventional whole-cell gigaohm seal recording technique (Hamill et al., 1981) was employed to record Ca²⁺-independent, depolarization-activated K⁺ currents from isolated adult mouse ventricular myocytes. Electrophysiological recordings were only obtained from Ca²⁺-tolerant, rod-shaped ventricular cells, and all experiments were conducted at room temperature (22–24°C). The bath solution contained (mM): 136 NaCl, 4 KCl, 1 CaCl₂, 2 MgCl₂, 5 CoCl₂, 10 HEPES, 0.02 tetrodotoxin, and 10 glucose, pH 7.35, and 295–305 mOsm. Recording pipettes contained (mM): 135 KCl, 1 MgCl₂, 10 EGTA, 10 HEPES, 5 glucose, pH 7.2, and 300–310 mOsm. α-dendrotoxin (α-DTX; Alamone Labs),¹ Heteropoda toxin-3 (HpTx-3; NPS Pharmaceuticals), and 4-aminopyridine (4-AP; Sigma Chemical Co.) stock solutions were prepared in distilled water, and diluted to the appropriate concentration in bath solution immediately before use. Tetraethylammonium (TEA; Sigma Chemical Co.)-containing bath solutions were prepared by equimolar substitution of TEACl for NaCl in the standard bath solution. α-DTX, HpTx-3, 4-AP, or TEA was applied to isolated myocytes during recordings using narrow-bore capillary tubes (300 μm i.d.) placed within ≈ 200 μm of the cell.

Experiments were conducted using an Axopatch-1D (β = 1) (Axon Instruments) or a Dagan 3900A (Dagan Corp.) amplifier. Recording pipettes, fabricated from soda lime glass, had tip diameters of 1–2 μm and resistances of 2–3 MΩ when filled with recording solution. Tip potentials were zeroed before membrane pipette seals were formed; seal resistances were ≥5 GΩ. After establishing the whole-cell configuration, ±10-mV steps were applied to allow measurement of whole cell membrane capacitances and input resistances. Series resistances were estimated by dividing the time constant of the decay of the capacitative transient by the membrane capacitance. Whole-cell membrane capacitances and series resistance were routinely compensated (≥85%) electronically; voltage errors resulting from the uncompensated series resistance were ≤8 mV and were not corrected. Only data obtained from cells with input resistance ≥0.7 GΩ were analyzed. Voltage-gated outward K⁺ currents were routinely evoked during 500-ms or 4.5-s depolarizing voltage steps to potentials between −40 and +60 mV from a holding potential of −70 mV; voltage steps were presented in 10-mV increments at 15-s intervals. Experiments were controlled and data were collected using a Gateway 300 MHz microcomputer equipped with a Digidata 1200 (Axon Instruments) analogue/digital interface and pClamp 6.1 (Axon Instruments). Data were acquired at variable sampling frequencies (ranging from 0.1 to 50 kHz), and current signals were filtered on-line at 5 kHz before digitization and storage.

Analyses of digitized data were completed using pClamp 6.1. Whole-cell membrane capacitances were determined by integrating the capacitative transients evoked during ±10-mV voltage steps from a holding potential of −70 mV (before series resistance and capacity compensation). Peak currents at each test potential were measured as the difference between the maximal outward current amplitudes and the zero current level. The waveforms of the 4-AP–, TEA-, and α-DTX–sensitive currents were determined by subtraction of the currents recorded in the presence of 4-AP, TEA, or α-DTX from the control currents (in the same cell).

Activation time constants were determined from single exponential fits to the rising phases of the outward K⁺ currents evoked during depolarizing voltage steps to test potentials between 0 and +60 mV from a holding potential of −70 mV; the fits were constrained to data points acquired 300 μs after the onset of the voltage step (thereby ignoring the delay) to the peak of the outward current. The decay phases of the currents evoked during long (4.5 s) depolarizing voltage steps to test potentials between +10 and +60 mV from a holding potential of −70 mV were fitted by the sum of two or three exponentials using one of the following expressions: y(t) = A₁ * exp(−t/τ₁) + A₂ * exp(−t/ τ₂) + A_ss or y(t) = A₁ * exp(−t/τ₁) + A₂ * exp(−t/τ₂) + A₃ * exp(−t/τ₃) + A_ss, where t is time, τ₁, τ₂, and τ₃ are the time constants of decay of the inactivating K⁺ currents, A₁, A₂, and A₃ are the amplitudes of the inactivating current components, and A_ss is the amplitude of the steady state, noninactivating component of the total outward K⁺ current. For all fits, time zero was set at the peak of the outward current. For all analyses, correlation coefficients (R) were determined to assess the quality of fits, R values for the fits reported here were ≥0.98. All averaged and normalized data are presented as means ± SEM. The statistical significance of observed differences between groups of cells or between different parameters describing the properties of the currents were evaluated using a one way analysis of variance or a two-tailed Student's t test; P values are presented in the text, and statistical significance was set at P < 0.05.

In the initial experiments here, whole-cell depolarization-activated outward K⁺ currents in myocytes isolated from the (lower two thirds of the left and right) ventricles of adult C57BL6 mice were recorded and analyzed (see materials and methods). With voltage-gated Ca²⁺ and Na⁺ currents blocked, outward currents were routinely recorded during 500-ms (Fig. 1, A and C) and 4.5-s (Fig. 1, B and D) depolarizing voltage steps to potentials between −40 and +60 mV from a −70-mV holding potential (n = 72). The rates of rise and the amplitudes of the currents increase with increasing depolarization; the largest and most rapidly activating current in Fig. 1 was evoked at +60 mV. No outward K⁺ currents were recorded during depolarizing voltage steps when the K⁺ in the pipettes was replaced by Cs⁺ (n = 6). The currents characterized here, therefore, reflect only the activation of Ca²⁺-independent, depolarization-activated K⁺ channels.

Although the outward K⁺ currents in all cells activated rapidly, the absolute current amplitudes (densities) and the decay phases of the currents were somewhat variable among cells (Fig. 1). In the majority of cells (65 of 72, ≈90%), outward current waveforms similar to those in Fig. 1, A and B, were recorded. There is a rapid component of current decay in these cells, consistent with the presence of a transient outward K⁺ current, previously described in these cells (Benndorf and Nilius, 1988; Wang and Duff, 1997), as well as in cardiac myocytes in other species (Campbell et al., 1995; Barry and Nerbonne, 1996; Giles et al., 1996). We refer to this current as I_{to, fast}, or I_to,f, to distinguish it from another, more slowly inactivating, transient outward K⁺ current, I_to,slow, or I_to,s, seen in some cells (see below). As noted in the introduction, it was recently demonstrated that I_to,f is eliminated in adult mouse ventricular myocytes isolated from animals expressing a mutant Kv4.2 α subunit (Kv4.2W362F) that functions as a dominant negative (Barry et al., 1998), an observation consistent with previous suggestions that members of the Kv4 α subunit subfamily underlie cardiac I_to,f (Dixon and McKinnon, 1994; Barry et al., 1995; Dixon et al., 1996).

In the other (7 of 72, ≈10%) cells studied in these initial experiments, I_to,f was not evident (Fig. 1, C and D). Peak outward K⁺ current densities in this subset of cells were significantly (P < 0.001) lower than in the cells with I_to,f; the mean ± SEM peak outward K⁺ current densities at +40 mV in cells with and without I_to,f, for example, were 47.0 ± 2.5 (n = 65) and 31.5 ± 4.1 (n = 7) pA/pF, respectively (Table I). In other respects, however, the properties of the cells with and without I_to,f were indistinguishable; for example, the mean ± SEM whole cell membrane capacitances and input resistances were 142 ± 4 pF and 1.00 ± 0.09 GΩ for the cells with I_to,f (n = 65) and 151 ± 14 pF and 0.95 ± 0.15 GΩ for the cells lacking I_to,f (n = 7). Subsequent experiments were focussed, therefore, on characterizing the depolarization-activated K⁺ current components in these cells in further detail.

Analysis of the decay phases of the outward K⁺ currents evoked during long depolarizations in the majority of cells (Fig. 1 B) revealed that current decay is well described by the sum of two exponentials, with decay time constants (τ_decay) that differ by an order of magnitude, and a noninactivating (i.e., steady state) current. The mean ± SEM τ_decay (n = 65) for the fast and slow components derived from these fits were 85 ± 2 and 1,162 ± 29 ms (Table I); neither time constant displays any appreciable voltage dependence (Fig. 2 A). Two components of outward K⁺ current decay (with τ_decay values of 80 ms and 1 s) in adult mouse ventricular myocytes were previously described, although both were attributed to inactivation of I_to (Wang and Duff, 1997). In another recent study, however, the two components of inactivation were considered distinct current components: the rapidly inactivating current was referred to as I_to and the slowly inactivating current was termed I_K,slow (London et al., 1998a; Zhou et al., 1998). Consistent with this distinction, I_K,slow was selectively attenuated in ventricular myocytes isolated from transgenic mice expressing a truncated Kv1.1 construct, Kv1.1N206Tag (London et al., 1998a) that functions as a dominant negative (Folco et al., 1997). These observations were interpreted as suggesting that Kv1 α subunits, likely Kv1.5, underlie mouse ventricular I_K,slow (London et al., 1998a). The slowly decaying current component will also be referred to here as I_K,slow and as noted above the rapidly inactivating current is referred to as I_to,f to distinguish it from I_to,s.

The densities of I_to,f and I_K,slow vary considerably among cells. For I_to,f, for example, the peak current density at +40 mV ranged from 7.9 to 62.4 pA/pF, with a mean ± SEM of 26.2 ± 1.6 pA/pF (n = 65; Table I). I_K,slow density at +40 mV ranged from 2.8 to 39.9 pA/pF with a mean ± SEM of 14.9 ± 0.9 pA/pF (n = 65; Table I). The noninactivating K⁺ current component that remains at the end of 4.5-s voltage steps (Fig. 1 B) is referred to here as I_ss (steady state). It is important to note that I_ss is not the same as the “sustained” K⁺ current recently described by Fiset et al. (1998) that was measured at the end of 500-ms voltage steps, and almost certainly reflects the sum of two K⁺ currents, I_K,slow and I_ss (see discussion). As with I_to,f and I_K,slow, I_ss densities varied measurably among cells: at +40 mV, for example, I_ss density ranged from 1.7 to 10.5 pA/pF, with a mean ± SEM of 5.5 ± 0.3 pA/pF (n = 65; Table I).

Analysis of the decay phases of the outward currents evoked during 4.5-s depolarizations in cells lacking I_to,f (Fig. 1 D) also revealed two inactivating components with mean ± SEM (n = 7) decay time constants (τ_decay) of 196 ± 7 and 1,368 ± 101 ms (Table I); neither time constant displays any appreciable voltage dependence (Fig. 2 B). The τ_decay (196 ± 7 ms) for the faster component of current decay in these cells is significantly (P < 0.001) larger than the τ_decay (85 ± 2 ms) for I_to,f. In addition, when all of the fast τ_decay values are compared (Fig. 2 C), the cells with a mean ± SEM τ_decay of 196 ± 7 ms fall well outside of the otherwise normal distribution of decay time constants. Taken together, these observations suggest a distinct subpopulation of cells (see below), and this current is referred to here as I_to,s for transient outward, slow (see below). I_to,s density in these cells ranged (at +40 mV) from 4.8 to 16.8 pA/pF, with a mean ± SEM (n = 7) density of 11.4 ± 1.9 pA/pF (Table I). In contrast to the differences in the rates of inactivation of the rapid components (i.e., I_to,f and I_to,s) of current decay, the mean ± SEM τ_decay (1,368 ± 101 ms) for the slowly inactivating current in the cells lacking I_to,f is not significantly different from the mean ± SEM τ_decay for I_K,slow (1,162 ± 29 ms) in cells with I_to,f (Table I), consistent with the presence of I_K,slow in both populations of cells. In addition, the mean ± SEM (n = 7) I_K,slow density of 15.6 ± 2.1 pA/pF in cells lacking I_to,f (Table I) is not significantly different from the mean ± SEM I_K,slow density (of 14.9 ± 0.9 pA/pF) in cells with I_to,f (Table I). In cells lacking I_to,f, however, I_K,slow contributes, on average, 50% to the peak outward currents; i.e., substantially more than the average contribution (32%) of this current to the peak in cells with I_to,f (Table I). I_ss is also evident in adult mouse ventricular myocytes expressing I_to,s (Fig. 1 D); I_ss density at +40 mV in these cells ranged from 2.5 to 6.2 pA/pF, with a mean ± SEM (n = 7) of 4.5 ± 0.5 pA/pF, a value that is not significantly different from the mean ± SEM I_ss density of 5.5 ± 0.3 pA/pF (n = 65) in cells with I_to,f (Table I).

The finding of the slowly inactivating, transient outward K⁺ current, I_to,s, in a subset of adult mouse ventricular myocytes prompted us to consider the possibility that this current might also be present in cells with I_to,f, but simply not be readily detected because the density is too low (particularly relative to I_to,f) to be resolved. To test this hypothesis, the decay phases of the currents were fitted assuming three exponential components (see materials and methods) with fixed inactivation time constants of τ₁ = 80 ms (for I_to,f), τ₂ = 1,200 ms (for I_K,slow), and τ₃ = 200 ms (for I_to,s). These analyses revealed that the decay phases of the total outward K⁺ currents in some (40 of 65) cells could indeed be fitted by the sum of three exponentials; in the other 25 cells, the fits did not converge or the amplitude of the component with the τ_decay = 200 ms was negative. Importantly, even when current records could be fitted to the sum of three exponentials, the quality of the fits was not improved significantly by the inclusion of the third component. Based on these analyses, therefore, it was not possible to determine whether some mouse ventricular cells do indeed express both I_to,f and I_to,s.

The finding of cells expressing I_to,s and lacking I_to,f prompted us to explore the possibility that there might be regional differences in the expression/densities of these functionally distinct K⁺ conductance pathways in mouse ventricle. To test this hypothesis directly, tissue pieces were dissected from the ventricular septum and from the apex of the left ventricle and dispersed (see materials and methods). Electrophysiological recordings revealed that the waveforms of the outward K⁺ currents in left ventricular myocytes isolated from these two regions are indeed distinct (Fig. 3). Peak outward K⁺ current densities in cells from the apex (mean ± SEM = 57.2 ± 3.5 pA/pF, n = 35) are significantly (P < 0.001) greater than in cells isolated from the septum (mean ± SEM = 28.5 ± 1.5 pA/pF, n = 28; Table II). In addition, for all cells isolated from the apex, analysis of the waveforms of the outward currents evoked during long (4.5 s) depolarizations revealed the presence of I_to,f, I_K,slow, and I_ss (Table II). The mean ± SEM τ_decay for I_to,f in these cells was 59 ± 2 ms (n = 35; see discussion); I_to,f density (at +40 mV) in these cells ranged from 13.8 to 79.5 pA/pF, with a mean ± SEM (n = 35) density of 34.6 ± 2.6 pA/pF and, on average, I_to,f contributes 60% to the peak outward K⁺ currents in apex cells (Table II). Mean ± SEM (n = 35) I_K,slow and I_ss densities at +40 mV in these cells were 17.2 ± 1.1 and 5.5 ± 0.4 pA/pF (Table II).

(Fig. 3) The waveforms of the outward K⁺ currents in cells isolated from the septum appear different and are more variable than those seen in cells from the apex. Two distinct outward K⁺ current waveforms were recorded in cells isolated from the septum (Fig. 3, B and C). Some cells, for example, clearly lack I_to,f (Fig. 3 C), and analyses of the decay phases of the currents in these cells revealed two components with τ_decay of 194 ± 11 and 1,143 ± 64 ms (Table II). These values are indistinguishable for those determined above for I_to,s and I_K,slow (Table I). I_to,s density (at +40 mV) ranged from 2.8 to 8.5 pA/pF, with a mean ± SEM (n = 6) density of 5.4 ± 0.9 pA/pF (Table II) and, on average, I_to,s contributes 26% to the peak outward K⁺ currents in these cells (Table II). The mean ± SEM I_K,slow density (10.4 ± 1.5 pA/pF) in septum cells lacking I_to,f (Table II) is similar to the density of this current component in cells with I_to,f (Tables I and II). In cells lacking I_to,f, however, I_K,slow contributes, on average, ≈50% to the peak outward currents; i.e., substantially more than the average contribution (of 30–40%) seen in cells with I_to,f (Tables I and II). In the other subset of cells from the septum (Fig. 3 B), three exponentials with mean ± SEM (n = 22) time constants of 53 ± 2, 258 ± 15, and 1,180 ± 45 ms were required to fit the decay phases of the currents; the decay time constants for these three components suggest the coexpression of I_to,f, I_to,s, and I_K,slow (as well as I_ss; Table II). The mean ± SEM I_to,f density in these cells (6.8 ± 0.5 pA/pF at +40 mV, n = 22) in these cells is significantly lower (P < 0.001) than the mean ± SEM I_to,f density (34.6 ± 2.6 pA/pF at +40 mV, n = 35) in cells from the left ventricular apex, and I_to,f only contributes ∼20% to the peak outward currents in these cells, compared with 60% in apex cells (Table II). In addition, the densities of I_K,slow and I_ss in cells isolated from the septum are slightly lower than in apex cells. I_ss density at +40 mV in these cells ranged from 2.5 to 6.2 pA/pF, with a mean ± SEM (n = 7) of 4.5 ± 0.5 pA/pF. Also, in septum cells, I_ss contributes significantly more to the total outward current than does I_ss in apex cells (Table II).

Subsequent experiments were focussed on examining the pharmacological sensitivities of the outward K⁺ currents in adult mouse ventricular myocytes and on determining if the kinetically distinct K⁺ current components could also be distinguished pharmacologically. In initial experiments, the effects of varying concentrations (10 μM to 5 mM) of 4-AP on whole-cell K⁺ currents were examined. Control K⁺ currents, evoked during 500-ms and 4.5-s voltage steps, were recorded before superfusion of 4-AP–containing bath solutions and, when the effect of 4-AP had reached a steady state, outward currents were again recorded. To obtain the current(s) blocked by 4-AP, records obtained in the presence of each concentration of 4-AP were digitally subtracted from the controls. Examples of typical experiments and the currents blocked by varying concentrations of 4-AP are presented in Fig. 4. The currents blocked by 10 μM 4-AP activate rapidly and inactivate slowly (Fig. 4 C). Analysis of the decay phases of the 10 μM 4-AP–sensitive currents suggests that I_K,slow is selectively attenuated (by ≈35%), consistent with previous reports of the 4-AP sensitivity of this conductance pathway (Fiset et al., 1998; London et al., 1998a). In the presence of 50 μM 4-AP, I_K,slow is further reduced to <50% of control (n = 7), whereas I_to,f is slightly (≈16%) reduced, and I_ss is unaffected (Table III).

Exposure of adult mouse ventricular myocytes to concentrations of 4-AP > 100 μM also blocks I_to,f (Table III). After application of 0.5 mM 4-AP, for example, peak outward currents are attenuated markedly (Fig. 4 E), and comparison of the control currents (Fig. 4 D) and the currents blocked by 0.5 mM 4-AP (Fig. 4 F) revealed that I_K,slow is blocked by ≈80% and I_to,f is reduced by ≈50% at 0.5 mM 4-AP; I_ss, in contrast, is unaffected by 0.5 mM 4-AP (Table III). When the 4-AP concentration is increased to 5 mM, outward currents are markedly reduced (Fig. 4 H), and analysis of the control (Fig. 4 G) and the 5 mM 4-AP–sensitive (Fig. 4 I) currents revealed that I_to,f (as well as I_K,slow) is blocked completely at this (5 mM) 4-AP concentration. The currents remaining in 5 mM 4-AP (Fig. 4 H) are interpreted as reflecting only I_ss, although I_ss is attenuated by ≈35% at high concentrations of 4-AP (Table III). Experiments conducted on mouse ventricular myocytes isolated from the septum reveal that I_to,s is also blocked by 4-AP (Fig. 5; Table III). Application of 10 μM 4-AP to septum cells selectively attenuates I_K,slow (Fig. 5, B and C), and higher concentrations of 4-AP are required to also affect I_to,s (Fig. 5, D and E). In the presence of 0.5 mM 4-AP, I_to,s is attenuated by ≈70% and I_K,slow is blocked completely (Table III; Fig. 5).

To test the validity of the separation of the currents based on differential sensitivities to 4-AP (Figs. 4 and 5), the effects of varying concentrations of other K⁺ channel blockers, TEA, α-DTX, and HpTx-3 were also examined. As in the experiments with 4-AP, outward K⁺ currents were recorded before and after bath applications of these blockers, and the drug-sensitive currents were obtained by off-line digital subtraction of these records. (Fig. 6 B) Exposure to 25 mM TEA resulted in marked attenuation of the peak currents and the currents remaining at the end of long depolarizing voltage steps. Analysis of the 25 mM TEA-sensitive currents (Fig. 6 C) reveals that these currents activate rapidly and inactivate slowly to a steady state level. The decay phases of the 25 mM TEA-sensitive currents were well described by single exponential with a mean ± SEM τ_decay of 1,234 ± 197 ms (n = 4), a value that is similar to the τ_decay of I_K,slow (Tables I and II). In addition, comparison of the current waveforms in Fig. 6 B reveals that both I_ss and I_K,slow are reduced by ≈60% at this TEA concentration, whereas I_to,f is unaffected (Table III). In similar experiments completed on cells isolated from the septum, I_to,s is also found to be unaffected by 25 mM TEA (Table III). When the TEA concentration was increased to 135 mM, the currents remaining are rapidly inactivating (Fig. 6 E). Both I_ss and I_K,slow are blocked completely and I_to,f is blocked by ≈40% by isotonic TEA (Table III).

In experiments completed with α-DTX at concentrations up to 100 nM (n = 4), no measurable effects on the outward K⁺ currents in adult mouse ventricular myocytes were observed (not shown). In contrast, local applications of 100–300 nM HpTx-3 resulted in the selective attenuation of I_to,f (Fig. 7). In cells isolated from the apex of the left ventricle, for example, 100 nM HpTx-3 reduced I_to,f by 30–40% (Fig. 7, left). In cells isolated from the septum that express I_to,f, this current is selectively attenuated by HpTx-3 (Fig. 7, middle), whereas I_to,s in cells with (Fig. 7, middle) and without (Fig. 7, right) I_to,f is unaffected by HpTx-3 at concentrations up to 300 nM.

Having identified four distinct depolarization-activated K⁺ currents, I_to,f, I_to,s, I_K,slow, and I_ss, in adult mouse ventricular myocytes, subsequent experiments were focussed on characterizing the time- and voltage-dependent properties of these currents. For most of these analyses, the amplitudes of the currents were determined from exponential fits to the decay phases of the currents evoked during long depolarizing voltage steps. In some cases, the individual current components, separated based on differential sensitivities to 4-AP and TEA, were also analyzed. In the pharmacological analyses, I_K,slow was defined as the 10-μM 4-AP–sensitive current (Fig. 4 C); the current remaining in the presence of 5 mM 4-AP (Fig. 4 H) was analyzed as I_ss, and the current remaining in 135 mM TEA (Fig. 6 E) was analyzed as I_to,f. Most of the initial experiments were completed on randomly dispersed adult mouse ventricular myocytes; detailed analysis of I_to,s was completed on cells isolated from the septum. In some cases, the properties of I_to,f, I_K,slow, and I_ss in apex and septum cells were determined and compared. To examine the voltage dependences of activation, the amplitudes of the individual current components at each test potential (in each cell) were measured and normalized to the current amplitude determined (in the same cell) at +30 mV; mean normalized currents are plotted as a function of test potential in Fig. 8 A. Although the voltage dependences of activation of I_to,f, I_to,s, I_K,slow, and I_ss are similar in that the currents begin to activate at less than −30 mV, the foot of the current–voltage plot for I_to,f is less steep than for I_to,s, I_K,slow, or I_ss (Fig. 8 A).

Time constants of activation for I_to,f, I_K,slow, and I_ss were determined from single exponential fits to the rising phases of the currents separated using 4-AP and TEA. For these analyses, I_K,slow was defined as the 10 μM 4-AP–sensitive current (Fig. 4 C), and I_to,f and I_ss were defined as the currents remaining in the presence of 135 mM TEA (Fig. 6 E) or 5 mM 4-AP (Fig. 4 H), respectively. For all current components, the rising phases of the currents at each test potential were well described by single exponentials. Mean (± SEM) activation time constants for I_to,f, I_K,slow, and I_ss are plotted as a function of test potential in Fig. 8 B. As is evident, I_to,f and I_K,slow activate rapidly, and with similar activation time constants. I_ss, in contrast, activates much more slowly than either I_to,f or I_K,slow at all test potentials (Fig. 8 B). In addition, analyses of the rising phases of the currents in records such as those in Figs. 4 C and 6 C revealed that the time constants of activation of the 25-mM TEA-sensitive currents and 10-μM 4-AP–sensitive currents are indistinguishable: mean ± SEM activation time constants at +40 mV for the 25-mM TEA-sensitive currents (Fig. 6 C) and the 10-μM 4-AP–sensitive currents (Fig. 4 C), for example, are 1.9 ± 0.5 ms (n = 4) and 2.1 ± 0.2 ms (n = 4), respectively. I_K,slow, therefore, is blocked selectively by both 10 μM 4-AP and 25 mM TEA.

The activation time constants (tau, τ) are voltage dependent, decreasing with increasing depolarization, and the variations with voltage are well described by single exponential functions of the form: tau(V) = a + b [exp(−V_m/c)], where V_m is the test potential and c is a constant that defines the steepness of the voltage dependence. The best fits (Fig. 8 B, continuous lines) to the data yielded: c = 27.1 (a, 0.54; b, 11.2) for I_to,f; c = 18.1 (a, 1.1; b, 10.1) for I_K,slow; and c = 13.0 (a, 12.5; b, 3.9) for I_ss. Comparison of the c values derived from these fits indicates that I_to,f and I_K,slow activation rates vary similarly as a function of voltage, whereas the voltage dependence of I_ss activation is much steeper (Fig. 8 B).

The voltage dependences of steady state inactivation of I_to,f, I_to,s, and I_K,slow were examined during voltage steps to +50 mV presented after 5-s conditioning prepulses to potentials between −100 and −10 mV; the protocol is shown below the current records in Fig. 9. Experiments were completed on randomly selected myocytes, as well as on cells isolated from the left ventricular apex (Fig. 9 A) and septum (B). In the first two cases, the decay phases of the currents evoked at +50 mV from each prepulse potential were fitted to the sum of two exponentials to provide I_to,f and I_K,slow. For cells isolated from the septum, the amplitudes of I_to,s, I_to,f (when present) and I_K,slow were determined from (double or triple) exponential fits to the decay phases of the currents evoked at +50 mV from each prepulse potential. The amplitudes of I_to,f, I_to,s, and I_K,slow evoked from each conditioning potential were then normalized to their respective maximal current amplitudes (in the same cell) evoked from −100 mV. Mean (± SEM) normalized I_to,f, I_to,s, and I_K,slow amplitudes are plotted as a function of conditioning potential in Fig. 9 C; the continuous lines represent the best Boltzmann fits to the averaged data.

The steady state inactivation data for I_to are well described by single Boltzmann with a V_1/2 of −24 mV (k = 3.8 mV; Fig. 9 C). For I_to,s and I_K,slow, in contrast, the variations in current amplitudes with conditioning voltage were not well described by a single Boltzmann, and two Boltzmanns were required to fit the data (Fig. 9 C). For I_K,slow, the V_1/2 values derived from these fits were −73 mV (k = 6.7 mV) and −19 mV (k = 6.0 mV), and for I_to,s, the V_1/2 values were −64 mV (k = 9.0 mV) and −19 mV (k = 6.3 mV). The voltage dependences of steady state inactivation of I_to,s and I_K,slow are very similar (Fig. 9 C), both in terms of the steepness (k values) of the curves and the relative amplitudes (≈35%, V_1/2 = −73 mV, and ≈65%, V_1/2 = −19 mV) of the two components. The finding of two components of steady state inactivation suggest the presence of two populations of channels contributing to I_K,slow (and I_to,s) or, alternatively, the complex gating of a single population of I_K,slow (I_to,s) channels (see discussion).

To examine the time dependences of recovery from steady state inactivation of I_to,f, I_to,s, and I_K,slow, cells were first depolarized to +50 mV for 9.5 s to inactivate the currents (longer depolarizations did not lead to further inactivation), subsequently hyperpolarized to −70 mV for varying times ranging from 10 to 9,064 ms, and finally stepped to +50 mV (to activate the currents and assess the extent of recovery); the protocol is illustrated in Fig. 10. Experiments were completed on randomly selected myocytes, as well as on cells isolated from the left ventricular apex (Fig. 10 A) and septum (B). Typical records obtained from cells isolated from the apex and septum are illustrated in Fig. 10, A and B. For the randomly selected and the apex cells, the decay phases of the currents evoked at +50 mV after each recovery period were fitted to the sum of two exponentials to provide I_to,f and I_K,slow. For cells isolated from the septum, the amplitudes of I_to,s, I_to,f (when present), and I_K,slow were determined from (double or triple) exponential fits to the decay phases of the currents evoked at +50 mV after each recovery period; note that for the cell illustrated in Fig. 10 B, I_to,f (as well as I_to,s) is present. The amplitudes of I_to,f, I_to,s, and I_K,slow after each recovery period were then normalized to their respective maximal current amplitudes (in the same cell) evoked after the 9.5-s recovery time. Mean (± SEM) normalized I_to,f, I_to,s, and I_K,slow amplitudes are plotted as a function of recovery time in Fig. 10 C; the continuous lines represent the best single exponential fits to the averaged data.

The mean normalized recovery data for I_to,f (in left ventricular apex and septum cells) are well described by a single exponential characterized by a time constant of 27 ms (Fig. 10 C, solid line); note that the recovery data for cells isolated from the septum and apex are indistinguishable. The mean normalized recovery data for I_to,s and I_K,slow also follow a monoexponential time course (Fig. 10 C), and the recovery data are well described by single exponentials with time constants of 1,298 (I_to,s) and 1,079 (I_K,slow) ms. Thus, in addition to differences in inactivation rates, I_to,f and I_to,s also recover from steady state inactivation at markedly different rates (Fig. 10 C).

The results presented here demonstrate the presence of four kinetically and pharmacologically distinct Ca²⁺-independent, voltage-gated K⁺ currents in isolated adult mouse ventricular myocytes: I_to,f, I_to,s, I_K,slow, and I_ss. Although I_K,slow and I_ss were found in all (n = 132) adult mouse ventricular myocytes studied, I_to,f and I_to,s were not. In randomly selected mouse ventricular cells, I_to,f was evident in the majority (65 of 72, ≈90%) of the cells, whereas I_to,s was identified in only 7 of 72 (≈10%) cells; i.e., the 7 cells lacking I_to,f. Importantly, the densities and the properties of I_K,slow and I_ss in I_to,s-expressing cells were indistinguishable from the corresponding currents in cells with I_to,f. Subsequent experiments revealed regional differences in I_to,f and I_to,s expression in mouse left ventricles: I_to,f was identified in all cells (n = 35) isolated from the apex and I_to,s was not detected in these cells (n = 35); in the septum, by contrast, all cells expressed I_to,s (n = 28), and in the majority (22 of 28, 80%) of cells, I_to,f was also present. The density of I_to,f (mean ± SEM at +40 mV = 6.8 ± 0.5 pA/pF, n = 22) in septum cells, however, is significantly (P < 0.001) lower than I_to,f density in cells from the apex (mean ± SEM at +40 mV = 34.6 ± 2.6 pA/pF, n = 35). In addition to differences in inactivation kinetics, I_to,f, I_to,s, and I_K,slow were also distinguished here by marked differences in the rates of recovery (from inactivation), as well as differential sensitivities to 4-AP, TEA, and HpTx-3; none of the currents in mouse ventricular myocytes was found to be sensitive to the dendrotoxins.

The absolute amplitudes of the individual current components varied among cells and, in all randomly selected cells with I_to,f (n = 65) and in cells from the apex (n = 35), the density of I_ss was substantially less than the density of either I_to,f or I_K,slow (Tables I and II). On average, the ratio of current densities in these (randomly selected and apex) cells was 5–6:3:1 for I_to,f, I_K,slow, and I_ss, respectively. In the I_to,s-expressing cells from the septum, peak outward K⁺ current densities are significantly (P < 0.001) lower (in cells with and without I_to,f) than the peak outward current density in cells from the apex (Table II), although no differences in cell sizes or input resistances were evident when these two groups of cells were compared. The mean ± SEM I_K,slow density is also significantly (P < 0.001) lower in septum than in apex cells (Table II). There were no significant differences, however, in mean ± SEM I_ss densities among the various populations of adult mouse ventricular (i.e., randomly selected) cells from the apex or cells from the septum (both the cells with and without I_to,f; Tables I and II). On average, the ratio of current densities in I_to,s-expressing cells was 1:2:1 for I_to,s, I_K,slow, and I_ss, respectively, in cells lacking I_to,f and 1:1:2:0.7 for I_to,f, I_to,s, I_K,slow, and I_ss, respectively, for the cells with I_to,f (Table II).

Functionally, I_to,f, I_to,s, and I_K,slow underlie the peak outward currents in all mouse ventricular cells (Figs. 1 and 3); since activation is slow, however, I_ss does not contribute appreciably to the peak. Rather, I_ss, together with I_K,slow, determines current amplitude at times late after the onset of depolarization(s), whereas I_to,f and I_to,s, which inactivate rapidly (τ_decay ≈ 60 and 200 ms, respectively) also do not contribute to the late currents. The fact that peak current densities are significantly lower in I_to,s- than I_to,f-expressing cells and that I_K,slow densities are similar suggests that action potentials likely are broader in I_to,s-expressing adult mouse ventricular myocytes than in cells lacking I_to,s and expressing I_to,f. The finding that I_to,f and I_to,s are differentially distributed also suggests that there will be differences in action potential waveforms in cells isolated from different regions of the ventricles. In addition, the marked differences in the rates of inactivation and recovery from inactivation for I_to,f and I_to,s suggest regional differences in rate-dependent variations in action potential waveforms. Further experiments will be necessary to test these hypotheses directly.

Several previous studies have examined outward K⁺ currents in adult mouse ventricular cells, and have been focussed primarily on I_to,f (Benndorf et al., 1987; Benndorf and Nilius, 1988; Wang and Duff, 1997). The time- and voltage-dependent properties of I_to,f described in these studies are quite similar to those reported here, in that current activation and inactivation are rapid. The results presented here also show that mouse ventricular I_to,f recovers from steady state inactivation very rapidly (τ = 27 ms), and the currents resemble those of I_to,f in a variety of other species (Campbell et al., 1995; Barry and Nerbonne, 1996; Giles et al., 1996). The slowly inactivating outward K⁺ current, I_K,slow, has also been identified previously in adult mouse ventricular myocytes (London et al., 1998a; Zhou et al., 1998). The mean I_K,slow inactivation time constant reported in these studies was 595 ± 29.3 ms (n = 7), a value that is considerably smaller than the time constant (mean ≈1,200 ms, n = 132) determined here for I_K,slow. The reason for the discrepancy between the absolute values of these time constants is unclear. Similar to previous findings (London et al., 1998a), however, I_K,slow is highly sensitive to 4-AP. A current sensitive to micromolar concentrations of 4-AP and referred to as I_sus (for sustained), has also been described by Fiset et al. (1998). Because I_sus was measured at the end of 500-ms voltage steps, however, it almost certainly reflects the sum of I_ss and I_K,slow.

In contrast to I_to,f and I_K,slow, neither I_to,s nor I_ss has been characterized previously in adult mouse ventricular myocytes. Interestingly, however, I_to,s is similar to the novel current recently described in ventricular myocytes isolated from adult Kv4.2W362F-expressing transgenic mice, in which I_to,f is eliminated (Barry et al., 1998). The time constants of inactivation of the novel current and I_to,s are similar, and preliminary experiments suggest that the pharmacological properties of the two currents are also similar (H. Xu and J.M. Nerbonne, unpublished observations). Thus, it seems reasonable to suggest that I_to,s is the novel current seen in the Kv4.2W362F transgenics and, further, that the functional expression of this conductance pathway is influenced, and perhaps regulated, by I_to,f expression. Further experiments will be necessary to test this hypothesis directly.

Several recent studies focused on identifying the molecular correlates of cardiac transient outward K⁺ currents have provided considerable evidence supporting the hypothesis (Dixon and McKinnon, 1994) that α subunits of the Kv4 subfamily underlie I_to,f (Fiset et al., 1997; Johns et al., 1997; Barry et al., 1998; Xu, H., H. Li, and J.M. Nerbonne, manuscript submitted for publication). In mouse heart, for example, it has been demonstrated that I_to,f is selectively eliminated in both ventricular and atrial myocytes isolated from transgenic animals expressing the dominant negative construct, Kv4.2W362F (Barry et al., 1998; Xu, H., H. Li, and J.M. Nerbonne, manuscript submitted for publication). The finding that the novel transient outward current (i.e., I_to,s) is expressed in all transgenic cells lacking I_to,f, in contrast, suggests that Kv4 α subunits do not contribute to I_to,s. The slow kinetics of inactivation and recovery from inactivation of I_to,s, however, are reminiscent of the properties of another Kv α subunit, Kv1.4 (Tseng-Crank et al., 1990), that generates slowly decaying transient K⁺ currents when expressed in heterologous systems (Comer et al., 1994; Petersen and Nerbonne, 1999). In addition, it was reported recently that regional differences in the properties of the transient outward K⁺ currents in ferret left ventricular epicardial and endocardial myocytes are correlated with differences in the expression of Kv4.2/Kv4.3 (epicardium, fast I_to) and Kv1.4 (endocardium, slow I_to) (Brahmajothi et al., 1998). Although these observations make it tempting to speculate that Kv1.4 underlies mouse ventricular I_to,s, it has recently been reported that targeted deletion of the Kv1.4 gene does not affect the membrane properties of adult mouse ventricular myocytes (London et al., 1998b). In these experiments, however, mouse ventricular myocytes were randomly dispersed and a rather small number of cells was studied. It is certainly possible, therefore, that the small subset of cells expected to be affected by elimination of Kv1.4 (if indeed this subunit underlies I_to,s) might not have been identified. Additional experiments aimed at examining regional differences in the K⁺ currents expressed in the Kv1.4 knockout mice and the molecular correlate(s) of I_to,s will certainly be of interest. In addition, two components of steady state inactivation of I_to,s were identified in the experiments here. These could reflect two (molecularly) distinct populations of I_to,s channels or, alternatively, the complex gating of a single population of I_to,s channels. Further experiments focused on resolving this question will be of interest.

A dominant negative strategy was also exploited recently in the generation of transgenic mice expressing a truncated Kv1.1, Kv1.1N206Tag (London et al., 1998a). Electrophysiological experiments revealed that the 4-AP–sensitive, slowly decaying outward K⁺ current I_K,slow was selectively attenuated in ventricular myocytes isolated from Kv1.1N206Tag-expressing mice (London et al., 1998a). These observations, together with the sensitivity of I_K,slow to 4-AP, suggest that a member of the Kv1 subfamily, likely Kv1.5, underlies I_K,slow (London et al., 1998a). Interestingly, the experiments here have revealed the presence of two components of steady state inactivation of I_K,slow. The finding of two components of steady state inactivation could reflect the fact that there are two (molecularly) distinct populations of K⁺ channels contributing to I_K,slow or, alternatively, the complex gating of a single population of I_K,slow channels. It will be of interest to pursue further studies on ventricular myocytes isolated from Kv1.1N206Tag-expressing animals to determine whether one (or both) component of steady state inactivation is reduced by Kv1.1N206Tag expression. Further experiments will also be necessary to define the molecular correlate of the slowly activating and noninactivating component of the mouse ventricular K⁺ currents, I_ss.

The authors thank Drs. Dianne Barry and Elias Bou-Abboud for many helpful discussions during the course of this work. Also, we thank Ms. Janice Delmar and NPS Pharmaceuticals (Salt Lake City, UT) for providing us with Heteropoda toxin-3.

This study was supported by the National Heart, Lung and Blood Institute of the National Institutes of Health and the Monsanto/Searle/Washington University Biomedical Agreement.

4-AP

4-aminopyridine

DTX

dendrotoxin

HpTx-3

Heteropoda toxin 3

TEA

tetraethylammonium