Elimination of the transient outward current and action potential prolongation in mouse atrial myocytes expressing a dominant negative Kv4 α subunit

Abstract

1. Analyses of whole-cell voltage-clamp recordings from isolated adult (C57BL6) mouse atrial myocytes reveal the presence of two prominent Ca²⁺-independent depolarization-activated K⁺ currents: a rapidly activating and inactivating, transient outward K⁺ current, I_to,f; and a non-inactivating, steady-state, K⁺ current, I_ss.

2. The properties of I_to,f and I_ss in adult mouse atrial myocytes are similar to those of the analogous currents recently described in detail in adult mouse ventricular cells. A slowly inactivating K⁺ current, which is similar to I_K,slow in ventricular cells, is detected in ≈40 % of adult mouse atrial myocytes, and when expressed, the density of this current component is substantially lower than the density of I_to,f or I_ss.

3. The similarity between atrial and ventricular I_to,f and the finding that both the Kv4 subfamily α subunits, Kv4.2 and Kv4.3, are expressed in wild-type mouse atria prompted us to determine if atrial I_to,f is affected in transgenic mice expressing a mutant Kv4.2 α subunit, Kv4.2W362F, that functions as a dominant negative.

4. Similar to findings in ventricular cells, electrophysiological recordings reveal that I_to,f is selectively eliminated in atrial myocytes isolated from transgenic mice expressing Kv4.2W362F, thereby demonstrating directly that Kv4 subfamily members also underlie mouse atrial I_to,f.

5. Neither the steady-state, non-inactivating K⁺ current I_ss, nor the inwardly rectifying K⁺ current I_K1, in atrial myocytes is affected by the expression of Kv4.2W362F.

6. In contrast to previous findings in Kv4.2W362F-expressing mouse ventricular myocytes, there is no evidence that electrical remodelling occurs in atrial cells when I_to,f is functionally eliminated.

7. The elimination of I_to,f is accompanied by marked increases in atrial action potential durations, although no electrocardiographic abnormalities attributable to, or suggestive of, altered atrial functioning are evident in Kv4.2W362F-expressing animals.

In the mammalian heart, two types of voltage-gated K⁺ channels that function to control action potential heights and durations have been distinguished based on differences in time- and voltage-dependent properties as well as pharmacological sensitivities: (1) rapidly activating and inactivating, i.e. transient outward, K⁺ currents, I_to; and (2) delayed, usually slowly activating, outwardly rectifying K⁺ currents, I_K (for review, see Campbell et al. 1995; Barry & Nerbonne, 1996; Nerbonne, 1998). These are broad classifications, however, and the detailed properties of I_to and I_K in myocardial cells isolated from different species, as well as from different regions of the heart in the same species are somewhat heterogeneous (Nerbonne, 1998). Voltage-gated K⁺ channels are important targets for the actions of endogenous neurotransmitters and hormones, as well as exogenous drugs, and it is clear that the properties and densities of these channels are affected in the damaged and/or diseased myocardium (Potreau et al. 1995; Gidh-Jain et al. 1997; Van Wagoner et al. 1997). Consequently, there is considerable interest in understanding the molecular basis of functional voltage-gated K⁺ channel diversity in the heart and in delineating the mechanisms involved in the regulation, modulation and functional expression of these channels.

Several recent studies have provided important insights into the voltage-gated K⁺ channel (Kv) pore forming α subunits that underlie myocardial I_to and I_K. Two K⁺ channel genes, HERG and KvLQT1, identified as loci of mutations leading to long QT syndromes 1 and 2 (Curran et al. 1995; Wang et al. 1996), for example, have been shown to encode I_Kr (I_K,rapid) and I_Ks (I_K,slow), respectively, in cardiac cells (Sanguinetti et al. 1995, 1996; Trudeau et al. 1995; Barhanin et al. 1996). For other voltage-gated K⁺ channels, a variety of in vitro (Feng et al. 1997; Fiset et al. 1997; Johns et al. 1997; Bou-Abboud & Nerbonne, 1999) and in vivo (Barry et al. 1998; London et al. 1998a, b) experimental strategies are being exploited to define the molecular correlates of functional cardiac K⁺ channels. In ventricular myocytes isolated from transgenic mice expressing a mutant Kv1.1 subunit (Kv1.1N206Tag) driven by the α-myosin heavy chain promoter (α-MHC) to direct cardiac specific expression (Lyons et al. 1990; Ng et al. 1991; Palermo et al. 1996), for example, it has been demonstrated that the slowly inactivating outward K⁺ current, I_K,slow, is selectively attenuated (London et al. 1998a). Because Kv1.1N206Tag functions as a dominant negative in vitro (Folco et al. 1997), these observations have been interpreted as suggesting that Kv1 subfamily members contribute to I_K,slow (London et al. 1998a). In ventricular myocytes isolated from transgenic mice expressing a mutant Kv4.2 subunit Kv4.2W362F that also functions as a dominant negative, in contrast, I_to is selectively eliminated (Barry et al. 1998). These observations are consistent with previous in vitro studies, using antisense oligonucleotides or adenoviral constructs encoding a truncated Kv4.2 subunit, on isolated rat ventricular myocytes that have suggested a role for Kv4 α subunits in the generation of ventricular I_to (Fiset et al. 1997; Johns et al. 1997).

The experiments here were undertaken to examine the properties of I_to in adult mouse atrial myocytes and to test directly the hypothesis that members of the Kv4 subfamily also underlie I_to in mammalian atrial myocytes. Experimental results are presented demonstrating that Kv4.2W362F (driven by the α-MHC promoter; Barry et al. 1998) is readily detected in the atria, as well as the ventricles, of Kv4.2W362F-expressing transgenics, whereas Kv4.2W362F is not detectable in brain. Electrophysiological recordings reveal that the properties of I_to in adult mouse atrial cells are similar to those of the ‘fast’ transient outward K⁺ current I_to,fast (I_to,f) recently characterized in detail in mouse ventricular cells (Xu et al. 1999), and that I_to,f is selectively eliminated in all atrial myocytes isolated from Kv4.2W362F-expressing mice. As with ventricular I_to,f, therefore, we conclude that the Kv4 subfamily of Kv α subunits underlies atrial I_to,f. The elimination of I_to,f leads to marked increases in atrial action potential durations, although no electrocardiographic abnormalities indicative of altered atrial functioning are evident in the Kv4.2W362F-expressing transgenic animals.

METHODS

All experiments here involving animals were carried out according to the guidelines established by the Washington University Animal Use and Care Committee, and all procedures and protocols employed here have been approved by this committee.

Expression of Kv4.2W362F-FLAG in atria and ventricles of transgenic mice

As described previously, the Kv4.2W362F-FLAG coding sequence was subcloned downstream from the α-myosin heavy chain (α-MHC) promoter to direct cardiac-specific expression of the transgene (Lyons et al. 1990; Ng et al. 1991). A (7 kb) fragment that included the α-MHC promoter, the first three non-coding exons of the α-MHC gene, the Kv4.2W362F-FLAG coding sequence and the human growth hormone (HGH) polyadenylation signal sequence was isolated, purified and injected (at 1 ng μl⁻¹) into fertilized C57BL6 mouse oocytes (Barry et al. 1998). Following transfer to pseudopregnant C57CBA adult mice, 57 offspring were obtained, three of which were positive for the transgene by Southern blot analysis of (tail) genomic DNA using a probe directed against the HGH polyadenylation sequence (Barry et al. 1998). All three (male) founders were crossed with wild-type C57BL6 adults and three Kv4.2W362F-expressing transgenic lines were established.

Expression of Kv4 α subunits in mouse ventricles and atria

For RT-PCR analysis, mRNA was prepared from the atria, ventricles, and brains of adult Kv4.2W362F-expressing transgenic and non-transgenic (control) littermates using the Micro-FasTrack mRNA isolation kit (Invitrogen, Carlsbad, CA, USA). To harvest tissues, animals were deeply anaesthetized with 5 % halothane and the hearts and brains were rapidly excised and frozen. cDNA was synthesized in a 20 μl reaction mixture containing 50 mM Tris-HCl (pH 8.3), 5 mM MgCl₂, 75 mM KCl, 4 mM sodium pyrophosphate, 5 mM deoxynucleotide triphosphates (dNTP), 10 mM dithiothreitol (DTT), 0.5 μg of oligo(dT)_12–18, 0.2 μg of mRNA, 10 units of RNase inhibitor and 5 units of avian myeloblastosis virus reverse transcriptase (Invitrogen). After incubation (1 h) at 42°C, the reaction was terminated by heating at 95°C for 2 min. Approximately 2 μl of the resulting reaction mixture was used for polymerase chain reaction (PCR) amplification. PCR was carried out in a 25 μl reaction mixture containing 10 mM Tris-HCl, pH 9.0, 50 mM KCl, 2 mM MgCl₂, 1 mM DTT, 0.5 μM of each primer (see Table 1), 0.2 mM dNTP, and 1 unit of Taq DNA polymerase (Sigma). The reaction proceeded for 30 cycles as follows: 94°C for 30 s, 58°C for 45 s, followed by 68°C for 90 s. The forward and reverse primers used for RT-PCR to probe for actin, Kv4.2W362F-FLAG Kv4.2, Kv4.3 and Kv2.1 were: 5′-GTGTTACGTCGCCCTTGATT-3′ and 5′-GCTGGAAGGTGGACAGAGAG-3′ (actin); 5′-GAAACCGGTAAATGGCAGCCGGTGTTGCAGCAT-3′ and 5′-CTTGCATCGTCGTCCTTGTAGTC-3′ (Kv4.2 W362F-FLAG); 5′-GAAACCGGTAAATGGCAGCCGGTGTTGCAGCAT-3′ and 5′-TGGTCATGGTGACAATGGTATAC-3′ (Kv4.2); 5′-GCTGGGTAGCACAGAGAAGG-3′ and 5′-GTGTCCAGGCAGAAGAAAGC-3′ (Kv4.3); and 5′-CCGAGACCAGCTCCAGTAAG-3′ and 5′-CTCCACGAAGAAACCAGAGC-3′ (Kv2.1). The amplified PCR products were analysed on 1 % agarose gels, stained with ethidium bromide.

Table 1.

Outward K⁺ currents in mouse atrial and ventricular cells

Cell type		Ipeak	Ito,f	IK,slow	Iss
Atrial
	τact (ms)	2.8 ± 0.2	—	—	—
	τdecay (ms)	—	83 ± 5	720 ± 137	—
	Density* (pA pF−1)	19.9 ± 13	13.8 ± 1.1	4.9 ± 0.6	4.4 ± 0.4
	n	22	22	8	22
	Mean % of Ipeak (n = 14)	—	74	—	26
	Mean % of Ipeak (n = 8)	—	62	22	20
Ventricular
	τact (ms)	2.3 ± 0.1	—	—	—
	τdecay (ms)	—	79 ± 3	1154 ± 54	—
	Density* (pA pF−1)	48.3 ± 3.4	26.4 ± 2.4	15.6 ± 2.4	6.3 ± 0.4
	Mean % of Ipeak	—	55	32	13
	n	30	30	30	30

^*Current densities were determined from analyses of records obtained on depolarization to +40 mV from a V_h of −70 mV; values are given as means ± s.e.m.

Electrophysiological recordings

Atrial and ventricular myocytes were isolated from wild-type and Kv4.2W362F-expressing adult (6–8 weeks) C57BL6 mice. Briefly, animals were anaesthetized with 3 % halothane (97 % oxygen) and, once deep anaesthesia was confirmed, the hearts were rapidly removed, cannulated and perfused (Barry et al. 1998), and whole-cell voltage and current-clamp recordings were obtained at room temperature within 48 h of disaggregation. For voltage-clamp recordings, the bath solution routinely contained (mM): 136 NaCl; 4 KCl; 1 CaCl₂; 2 MgCl₂; 5 CoCl₂; 10 Hepes; 10 glucose and 0.02 tetrodotoxin (TTX); pH at 7.4; 295–300 mosmol l⁻¹. The TTX and Co²⁺ were eliminated when action potentials were recorded. For both current- and voltage-clamp experiments, the recording pipette solution contained (mM): 135 KCl; 1 MgCl₂; 10 EGTA; 10 Hepes and 5 glucose; pH 7.2; 295–310 mosmol l⁻¹. Currents, evoked during 500 ms or 4.5 s depolarizing voltage steps to test potentials between −40 and +60 mV from a holding potential (V_h) of −70 mV, were routinely recorded; prior to each voltage step, a 20 ms prepulse to −20 mV was presented to activate and inactivate the TTX-insensitive voltage-gated Na⁺ current (I_Na), thereby minimizing contamination of the outward K⁺ current waveforms by I_Na. Importantly, the use of the prepulse (20 ms to −20 mV) did not compromise the ability to record the voltage-gated K⁺ currents in these cells (Xu et al. 1999). Inwardly rectifying K⁺ currents (I_K1) were evoked during hyperpolarizing voltage steps to test potentials between −90 and −120 mV. Action potentials were evoked by brief (5 ms) depolarizing current injections, presented at a frequency of 2 Hz.

Experiments were performed using an Axopatch 1D patch-clamp amplifier interfaced either to an IBM-compatible 486 computer with a Tecmar Labmaster-1 (TL-1) analog/digital interface (Axon Instruments) or a Gateway 350 MHz Pentium with a Digidata 2000 analog/digital interface, and either the pCLAMP6 or the pCLAMP7 software package (Axon). Data were filtered at 5 kHz prior to storage. Electrodes were fabricated from soda lime glass, coated with Sylgard (Dow Corning) and fire polished; tip resistances were 1.5–2.0 MΩ. Series resistances were in the range of 3–4 MΩ and were compensated electronically by 80–90 %; voltage errors resulting from the uncompensated series resistance were ≤ 6 mV and were not corrected.

Surface electrocardiograms were recorded from adult wild-type and Kv4.2W362F-expressing mice as previously described (Barry et al. 1998). Briefly, mice were lightly anaesthetized with 3 % halothane (97 % O₂), and needle electrodes were placed on the limbs under the skin. Standard ECG recordings were first obtained from leads I, II and III simultaneously at a frequency response of 0.05 to 500 Hz using a Gould Model 13–4615 ECG amplifier (Gould Inc., Santa Clara, CA). The electrode normally on the left front leg was placed on the left shoulder, and the right front leg electrode was placed at the base of the sternum. This configuration facilitated resolution of T waves that are difficult to resolve in mice (London et al. 1998a; Barry et al. 1998). Signals were digitized at 2 kHz and recorded on a 486 personal computer. Subsequently the mice were killed.

Data analysis

Voltage- and current-clamp data were compiled and analysed using Clampfit (Axon and Excel (Microsoft)). For each cell, the spatial control of the membrane voltage was assessed by analysing the decays of the capacitative transients evoked during ±10 mV voltage steps from the holding potential; only cells with capacitative transients well described by single exponentials were analysed further. Whole-cell membrane capacitances were determined by integration of the capacitative transients evoked during brief (25 ms) subthreshold (±10 mV) voltage steps from the holding potential. The whole-cell membrane capacitances of adult wild-type and Kv4.2W362F-expressing atrial myocytes are indistinguishable, with mean ± s.e.m. values of 47 ± 4 pF (n = 22) and 48 ± 3 pF (n = 17) in wild-type and Kv4.2W362F-expressing cells, respectively. The input resistances of adult mouse wild-type (n = 22) and Kv4.2W362F-expressing atrial myocytes (n = 17) were also indistinguishable, with mean ± s.e.m. values of 1.8 ± 0.3 GΩ and 1.9 ± 0.2 GΩ, respectively. Leak currents were always less than 100 pA, and were not corrected. The plateau outward K⁺ current was defined as the current remaining 4.5 s after the onset of the depolarizing voltage steps, and the peak outward current was defined as the maximum value of the outward K⁺ current during 200 ms voltage steps. Current amplitudes, measured in individual cells, were normalized for differences in cell size (whole cell membrane capacitance), and current densities (in pA pF⁻¹) are reported.

The time constants of outward current activation in wild-type and Kv4.2W362F-expressing myocytes were determined from single exponential fits to the rising phases of the currents after the initial delay. The time constants of inactivation of the depolarization-activated outward K⁺ currents in wild-type and Kv4.2W362F-expressing atrial and ventricular myocytes were determined from single or double exponential fits to the decay phases of the current, using one of the two following equations:

or

where A_t is the amplitude of the current at time, t; A₁ and τ₁ and A₂ and τ₂ represent the amplitudes (A) and the time constants (τ) of the fast and slow components of current decay; and A_ss is the amplitude of the non-inactivating current. Correlation coefficients, determined to assess the quality of the fits, were ≥ 0.980. For analysis of electrocardiograms, the onsets and offsets of the P, Q, R, S and T waves were determined by measuring the earliest (onset) and the latest (offset) times from the three leads. Observed differences among data sets were analysed using ANOVA and the Student's t test; results are quoted as means ± s.e.m.P values are presented in the text.

RESULTS

Depolarization-activated currents in adult mouse atrial myocytes

In initial experiments, whole-cell depolarization-activated outward K⁺ currents in adult C57BL6 mouse atrial myocytes were recorded, analysed and compared with the currents in ventricular cells. With voltage-gated Ca²⁺ and Na⁺ currents blocked, outward currents were routinely recorded during 500 ms and 4.5 s depolarizing voltage steps to potentials between −40 mV and +60 mV from a holding potential of −70 mV. Outward K⁺ current waveforms recorded in a typical adult mouse atrial myocyte are illustrated in Fig. 1A; representative ventricular outward K⁺ currents are presented in Fig. 1B, for comparison. In both cell types, the rates of rise and the amplitudes of the currents increase with increasing depolarization; the largest and most rapidly activating current in each panel 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 recorded and analysed here, therefore, are assumed to reflect only the activation of Ca²⁺-independent, depolarization-activated K⁺ channels.

Figure 1. Depolarization-activated outward K⁺ current waveforms in adult mouse atrial myocytes are similar to those recorded in ventricular cells.

Whole-cell outward K⁺ currents were evoked in isolated adult mouse atrial (A) and ventricular (B) myocytes during 500 ms (left panels) and 4.5 s (right panels) depolarizing voltage steps to potentials between −50 and +60 mV from a holding potential of −70 mV. Note that the K⁺ currents are considerably lower in amplitude in atrial (A), compared with ventricular (B) cells (see text). C, K⁺ current-voltage relations in mouse atrial and ventricular myocytes are indistinguishable (means ± s.e.m.). The amplitudes of the rapidly inactivating component of the outward currents (i.e. I_to,f; see text) and of the plateau outward K⁺ currents (i.e. I_ss; see text) in atrial (n = 22) and ventricular (n = 30) cells were determined (see text) and normalized to the currents evoked on depolarization to +30 mV (in the same cell), and mean ± s.e.m. normalized I_to,f (circles) and I_ss (triangles) amplitudes in atrial (open symbols) and ventricular (filled symbols) are plotted (C) as a function of test potential.

As is evident in Fig. 1, the waveforms of the depolarization-activated outward K⁺ currents in adult mouse atrial and ventricular myocytes are similar in that the currents at all test potentials activate fast, and subsequently inactivate rapidly to a steady-state level (Figs 1A and B). The normalized peak and plateau current-voltage relations in the two cell types are indistinguishable and the time constants of peak outward current activation, determined from single exponential fits to the rising phases of the currents after the initial delay, are similar (Table 1). Comparison of the records in Fig. 1, however, suggests that peak outward K⁺ current amplitudes are substantially lower in atrial, than in ventricular, cells. Analyses of currents recorded from many cells revealed that this was a consistent finding, i.e. peak outward K⁺ currents (at all voltages) are significantly (P < 0.001) lower in atrial, than in ventricular cells: peak outward K⁺ current amplitudes at +40 mV, for example, were 962 ± 121 pA (n = 22) and 6576 ± 481 pA (n = 30), respectively. Atrial cells, with whole-cell membrane capacitances (C_m) of 47 ± 4 pF (n = 22), are considerably smaller than ventricular cells, with mean C_m = 140 ± 6 pF (n = 30). This difference in cell size, however, does not account for the lower current amplitudes in atrial cells: peak outward K⁺ current densities are significantly (P < 0.001) lower in adult mouse atrial, compared with ventricular, myocytes. Peak outward K⁺ current densities at +40 mV in atrial and ventricular myocytes, for example, were 19.9 ± 1.3 pA pF⁻¹ (n = 22) and 49.3 ± 3.4 pA pF⁻¹ (n = 30) (Table 1). The input resistances of adult mouse atrial and ventricular cells, in contrast, are similar with values of 1.8 ± 0.3 GΩ (n = 22) and 1.5 ± 0. 3 GΩ (n = 30), respectively.

In all wild-type adult mouse atrial cells, there is a rapid component of outward K⁺ current inactivation (Fig. 1), and analyses of the decay phases of the currents evoked during long depolarizations revealed that, in the majority (14/22) of cells, current decay is well described by a single exponential characterized by a decay time constant (τ_decay) of 81 ± 6 ms (n = 14); the decay rates do not vary appreciably with voltage. This time constant is similar to the mean τ_decay of 78 ± 3 ms (n = 30) of the rapidly decaying transient outward K⁺ current in mouse ventricular cells (Wang & Duff, 1997; Barry et al. 1998; Xu et al. 1999). This has recently been referred to as I_to,fast, or I_to,f, (rather than ‘I_to’) to distinguish it from another, more slowly inactivating, transient outward K⁺ current (I_to,slow or I_to,s) evident in a subset of mouse ventricular myocytes (Xu et al. 1999). The current-voltage relations of the rapidly inactivating current (I_to,f) in atrial and ventricular cells are indistinguishable (Fig. 1C). Similar to I_to,f in ventricular cells, the rapidly inactivating K⁺ current in atrial myocytes is sensitive to milli-, but not micromolar, concentrations of 4-aminopyridine (not shown).

A distinguishing property of mouse ventricular I_to,f is a very rapid recovery (τ_recovery ∼30 ms) from steady-state inactivation (Xu et al. 1999). In subsequent experiments, therefore, the time course of recovery from steady-state inactivation of the rapidly inactivating transient current in mouse atrial was examined and compared with the rate of recovery of I_to,f in ventricular cells. In these experiments, cells were first depolarized to +50 mV for 4.5 s to inactivate the currents (longer depolarizations did not lead to further inactivation), subsequently hyperpolarized to −70 mV for varying times ranging from 6 to 9064 ms, and finally, stepped to +50 mV to activate the currents and assess the extent of recovery. Typical records obtained in mouse atrial (Fig. 2A) and ventricular (Fig. 2B) cells are presented in Fig. 2. Current amplitudes at each recovery time were measured and normalized to the current recorded following the 9064 ms recovery time. Normalized current amplitudes are plotted as a function of recovery time in Fig. 2; the data are well described by a single exponential with a time constant of 38 ms (continuous line), a value indistinguishable from the recovery time constant of 27 ms determined in adult mouse ventricular cells (Fig. 2). The properties of the rapidly inactivating transient outward K⁺ current in adult mouse atrial myocytes, therefore, appear indistinguishable from mouse ventricular I_to,f and for this reason, we refer to this current as mouse atrial I_to,f (see Discussion).

Figure 2. The rates of recovery from steady-state inactivation of the rapidly inactivating outward current in mouse atrial ventricular myocytes I_to,f are similar.

After inactivating the currents during long prepulses to +50 mV, cells were hyperpolarized to −70 mV for times ranging from 6 to 9064 ms prior to a second (test) depolarizations to +50 mV (to assess the extent of recovery); the experimental protocol is illustrated between the records. Typical current waveforms recorded in cells isolated from adult mouse atria (A) and ventricles (B) during the +50 mV conditioning step and the +50 mV test depolarization following varying recovery times are displayed. The amplitudes of the transient currents (I_to,f) evoked at +50 mV following each recovery period were determined from double exponential fits to the decay phases of the total outward currents (see text), and normalized to the current amplitudes evoked following the 9064 ms recovery period (in the same cell). Mean ± s.e.m. normalized recovery data for the transient current in atrial (^) and ventricular (•) myocytes are plotted in C; the initial phase of recovery of the currents is shown on an expanded time scale in the inset. As is evident, recovery of transient currents in the two cells types follows a similar time course (see text).

In the other eight wild-type atrial myocytes studied, two exponentials with τ_decay values of 87 ± 9 ms and 720 ± 137 ms were required to fit the decay phases of the currents; neither time constant displays any appreciable voltage dependence. The faster time constant (τ_decay 87 ± 9 ms) is indistinguishable from that of the rapidly decaying current (I_to,f) in the majority of atrial cells (above), suggesting that I_to,f is expressed in all mouse atrial myocytes; the τ_decay for I_to,f is 83 ± 5 ms (n = 22; Table 1). The density of I_to,f in atrial cells varies over the range of 3.9 to 23.7 pA pF⁻¹ at +40 mV, with a mean density (at +40 mV) of 13.8 ± 1.1 pA pF⁻¹ (n = 22; Table 1). Similar to the comparison of peak outward K⁺ current density, therefore, the density of I_to,f in adult mouse atrial myocytes is significantly (P < 0.001) lower than the I_to,f density (26.4 ± 2.4 pA pF⁻¹ at +40 mV; n = 30) in ventricular cells (Table 1). The τ_decay of the slower component of current inactivation that is evident in this subset (8/22) of atrial myocytes is similar to that of I_K,slow in ventricular cells (see below and Discussion). The density of this slowly decaying current component, however, is quite low compared with I_to,f in adult mouse atrial cells, as well as I_K,slow (and/or I_to,f) densities in ventricular cells (Table 2). In addition, in all adult mouse atrial cells, there is also a non-inactivating K⁺ current, referred to as I_ss (Xu et al. 1999), remaining at the end of 4.5 s depolarizing voltage steps. The density of I_ss is also variable among cells with a density (at +40 mV) of 4.4 ± 0.4 pA pF⁻¹ (n = 22).

Table 2.

Action potential prolongation in Kv4.2W362F-expressing myocytes

Cell type	APD25 (ms)	APD50 (ms)	APD75 (ms)	APD90 (ms)	RMP (mV)	APA (mV)
Atrial
Wild-type (n = 8)	2.8 ± 0.6	5.0 ± 0.9	9.0 ± 1	27 ± 8	−64 ± 1	125 ± 8
Kv4.2W362F (n = 4)	7.3 ± 2.2*	11.1 ± 2.1**	25 ± 7**	132 ± 34**	−68 ± 3	130 ± 4
Ventricular
Wild-type (n = 5)	2.9 ± 0.2	7.5 ± 1.3	10.0 ± 2	36 ± 10	−68 ± 1	141 ± 6
Kv4.2W362F (n = 5)	5.7 ± 1.3	13.2 ± 2.2	37 ± 9*	117 ± 14**	−68 ± 2	137 ± 8

All values are means ± s.e.m.

^*P < 0.05

^**P < 0.01.

Expression of Kv4 α subunits in wild-type and transgenic mouse atria

The similarity between the properties of adult mouse atrial and ventricular I_to,f revealed in the experiments described above suggested the interesting possibility that the molecular correlates of mouse atrial and ventricular I_to,f are also similar (or the same). As noted in Introduction, several recent studies have clearly indicated a role for Kv α subunits of the Kv4 subfamily in the generation of cardiac I_to (Fiset et al. 1997; Johns et al. 1997; Sanguinetti et al. 1997). In addition, it was recently reported that mouse ventricular I_to,f is selectively eliminated in transgenic mice expressing a mutant Kv4.2 α subunit (Kv4.2W362F) that functions as a dominant negative (Barry et al. 1998), clearly demonstrating the functional role of Kv4 α subunits in the generation of I_to,f in adult mouse ventricles. Although not previously demonstrated, we do find that both Kv4.2 and Kv4.3 are readily detected in adult mouse atria, as well as ventricles (Fig. 3), consistent with a role for these subunits in the generation of functional voltage-gated K⁺ channels.

Figure 3. Expression of Kv4 α subunits in adult mouse atria.

Using subunit-specific probes (see text), RT-PCR analysis (see Methods) reveals that both Kv4.2 and Kv4.3 are readily detected in wild-type adult mouse atria, as well as in adult mouse ventricles and brain. To allow comparisons to be made between samples, probes against muscle-specific actin were employed in screening the atrial and ventricular samples and probes against Kv2.1 were employed for comparing brain samples. In contrast to the widespread distribution of the wild-type Kv4 α subunits, Kv4.2W362F-FLAG expression (see text) is only detected in mouse heart, consistent with the α-MHC promoter driving cardiac-specific expression of the (Kv4.2W362F) transgene. wt, wild-type; tg, transgenic.

To generate Kv4.2W362F-expressing transgenic mice, the mutant construct was epitope tagged at the 3′ end, and introduced into an α-myosin heavy chain vector (Lyons et al. 1990; Ng et al. 1991; Palermo et al. 1996) behind the α-MHC promoter to drive cardiac specific expression of the transgene (Barry et al. 1998). As expected (Lyons et al. 1990; Ng et al. 1991), expression of Kv4.2W362F is robust in the atria, as well as the ventricles of these animals, although no expression of the transgene is detected in the brain (Fig. 3). These observations allowed us to test directly the hypothesis that the molecular correlates of atrial and ventricular I_to,f are the same, and subsequent experiments examined the functional consequences of atrial expression of Kv4.2W362F.

Outward K⁺ currents are altered in Kv4.2W362F-expressing atrial myocytes

Whole-cell voltage-clamp recordings revealed marked differences in the waveforms of the depolarization-activated outward K⁺ currents in atrial myocytes isolated from non-transgenic and transgenic (Kv4.2W362F-expressing) littermates (Fig. 4). As is evident in the records presented in Fig. 4, peak outward K⁺ current amplitudes at all test potentials are substantially lower in cells isolated from transgenic animals (Fig. 4B) compared with the currents typically recorded in myocytes isolated from wild-type non-transgenic littermates (Fig. 4A). Similar results were obtained in 17 myocytes from Kv4.2W362F-FLAG transgenic animals, and the mean peak outward densities at +40 mV were 19.9 ± 1.3 (n = 22) and 9.9 ± 1.1 pA pF⁻¹ (n = 17) in wild-type and Kv4.2W362F-expressing atrial myocytes (Fig. 4C), respectively; these values are significantly different (at the P ≤ 0.001 level). In contrast to the marked reductions in peak outward K⁺ current densities in atrial myocytes isolated from the Kv4.2W362F-expressing transgenic animals, no measurable effects on the densities of either the plateau outward K⁺ currents, determined as the currents remaining at the end of 4.5 s voltage steps, or of the hyperpolarization-activated, inwardly rectifying K⁺ current, I_K1, were observed. Mean plateau outward current densities at +40 mV, for example, were 4.5 ± 0.4 (n = 22) and 4.9 ± 0.8 pA pF⁻¹, (n = 17) in wild-type and Kv4.2W362F-expressing myocytes, respectively, (Fig. 4C). Peak I_K1 densities evoked at −120 mV from a holding potential of −70 mV in wild-type and Kv4.2W362F-expressing myocytes were 9.2 ± 1.0 (n = 18) and 7.9 ± 1.5 pA pF⁻¹ (n = 14), respectively, (Fig. 4C).

Figure 4. The rapidly activating and inactivating transient outward current, I_to,f (see text) is eliminated in atrial myocytes isolated from Kv4.2W362F-expressing transgenic animals.

Whole-cell outward K⁺ currents, recorded from atrial myocytes isolated from adult non-transgenic (A) and transgenic (B) littermates, were evoked during 500 ms (left panels) or 4.5 s (right panels) depolarizing voltage steps to potentials between −40 and +60 mV from a holding potential of −70 mV. As is evident, the amplitudes and the rates of decay of the peak outward K⁺ currents recorded from atrial myocytes isolated from Kv4.2W362F-expressing animals (B) are reduced compared with those recorded from non-transgenic (A) cells. C, means ± s.e.m. K⁺ current densities in non-transgenic and Kv4.2W362F-expressing atrial are illustrated. Peak and plateau outward K⁺ currents, evoked at +40 mV from a holding potential of −70 mV were measured in individual cells and normalized to the whole-cell membrane capacitance (determined in the same cell). Mean ± s.e.m. peak and plateau densities determined in Kv4.2W362F-expressing (n = 18) and non-transgenic (n = 22) atrial myocytes are plotted. The amplitudes of inwardly rectifying K⁺ currents (I_K1), evoked on membrane hyperpolarizations to −120 mV from a holding potential of −70 mV, were measured in individual cells and normalized to the whole-cell membrane capacitance (determined in the same cell). I_K1 densities in wild-type (n = 18) and Kv4.2W362F-expressing (n = 14) atrial myocytes are plotted. Note that only peak outward K⁺ current density is attenuated in Kv4.2W362F-expressing atrial cells (see text).

Action potentials are prolonged in Kv4.2W362F-expressing animals

Whole-cell current-clamp recordings revealed that action potential waveforms recorded in isolated adult atrial myocytes (Fig. 5A) are similar to those recorded in adult mouse ventricular cells (Fig. 5B). In both cell types, action potentials are very brief, lasting less than 50 ms (Fig. 5, Table 2). Also, similar to the findings in ventricular cells (Barry et al. 1998), action potentials recorded from Kv4.2W362F-expressing atrial myocytes are substantially broader than action potentials recorded from atrial cells isolated from non-transgenic littermates (Fig. 5A), consistent with the elimination of I_to,f. Analysis of action potential durations at 25, 50, 75 and 90 % repolarization (APD₂₅, APD₅₀, APD₇₅ and APD₉₀) revealed that mean action potential durations at all repolarization times are prolonged significantly compared with action potential durations in wild-type atrial cells (Table 2). In contrast, action potential amplitudes and resting membrane potentials of the Kv4.2W362F-expressing myocytes are not significantly different from those measured in wild-type cells (Table 2).

Figure 5. Action potentials are prolonged in Kv4.2W362F-expressing cells.

Whole-cell current-clamp recordings were obtained from isolated adult mouse atrial (A) and ventricular (B) myocytes; recordings were obtained from cells isolated from Kv4.2W362F-expressing transgenic (right panels) and non-transgenic (left panels) littermates. Note the difference in the time base in the right hand panel of A (see text and Table 2).

As reported previously, the functional consequences of expression of Kv4.2W362F in mouse heart were assessed by examining surface electrocardiograms (ECGs) recorded from Kv4.2W362F-expressing transgenic and non-transgenic littermates (Barry et al. 1998). Analysis of the ECGs revealed a significant (P < 0.001) prolongation of the QT interval in the transgenic animals, consistent with an underlying defect in ventricular repolarization; mean corrected QT intervals (London et al. 1998a) in wild-type and transgenic animals, for example, were 57 ± 3 (n = 5) and 79 ± 3 ms (n = 12), respectively, (Barry et al. 1998). Other electrocardiographic parameters, such as heart rates (RR intervals), QRS durations, PQ intervals as well as P, QRS and T wave morphologies, in contrast, were unchanged in Kv4.2W362F-expressing transgenics suggesting that atrial functioning is not affected measurably in these animals in spite of the loss of I_to,f and the marked increase in action potential durations (see Discussion).

DISCUSSION

Depolarization-activated K⁺ currents in adult mouse atrial myocytes

The results presented here demonstrate the presence of three kinetically distinct Ca²⁺-independent, voltage-gated K⁺ currents in isolated adult mouse atrial myocytes that are very similar to the corresponding currents recently described in detail in adult mouse ventricular cells (Xu et al. 1999), and are referred to as: I_to,f, I_K,slow and I_ss. The densities of all three current components (and particularly I_to,f and I_K,slow), however, are significantly lower in atrial cells than in ventricular cells. Functionally, I_to,f and I_K,slow underlie the peak outward currents in mouse atrial cells (Fig. 1) and I_ss determines current amplitude at times late after the onset of membrane depolarization(s). Although I_to,f and I_ss were found in all (n = 22) wild-type atrial myocytes studied, I_K,slow was detectable only in about 40 % (8/22) of the cells. Even when detected, the density of I_K,slow in atrial cells is low, particularly compared with I_to,f in the same cells (Table 1). In addition, the density of I_K,slow in atrial cells is significantly (P < 0.001) lower than in ventricular cells. It seemed possible, therefore, that I_K,slow was not detected in all cells simply because the density of the current (particularly relative to I_to,f) was too low to be resolved reliably. Consistent with this hypothesis, I_K,slow is evident in all Kv4.2W362F-expressing atrial myocytes, and the density (4.3 ± 1.3 pA pF⁻¹ at +40 mV; n = 8) is not significantly different from I_K,slow density in wild-type cells (density at +40 mV = 4.9 ± 0.6 pA pF⁻¹; n = 8) in which this current was resolved.

Although not characterized in detail here, the properties of atrial and ventricular I_K,slow (like I_to,f) also appear to be quite similar. Clearly, studies focussed on determining the molecular correlate(s) of functional I_K,slow channels will resolve this question. Recently, London and coworkers (1998a) reported the generation and characterization of Long QT mice expressing a truncated Kv1.1 subunit (Kv1.1N206Tag) that functions as a dominant negative (Folco et al. 1997). Expression of the Kv1.1N206Tag selectively attenuates I_K,slow in mouse ventricular myocytes, consistent with the hypothesis that a member of the Kv1 subfamily, probably Kv1.5, underlies this current (London et al. 1998a; Fiset et al. 1998; Zhou et al. 1998). Clearly, it will be of interest to examine the functional consequences of Kv1.1N206Tag expression in mouse atrial myocytes to test directly the hypothesis that the molecular correlates of mouse atrial and ventricular I_K,slow are (like I_to,f) also the same.

Functional consequences of atrial expression of Kv4.2W362F

The results presented here reveal that I_to,f is functionally eliminated in atrial myocytes isolated from Kv4.2W362F-expressing animals, demonstrating directly that members of the Kv4 α subunit subfamily underlie atrial (as well as ventricular) I_to,f in mouse. The functional ‘knockout’ of I_to,f results in increased action potential durations in single atrial myocytes. Indeed, all phases of action potential repolarization are increased significantly in Kv4.2W362F-expressing atrial cells and the prolongation evident is substantially greater than observed in Kv4.2W362F-expressing ventricular cells. Electrocardiographic recordings however, reveal that although QT intervals are prolonged significantly in Kv4.2W362F-expressing animals, consistent with a defect in ventricular repolarization, no effects on P or QRS wave morphologies or durations, PR intervals or heart rates were seen in these animals. In spite of the elimination of atrial I_to,f and the marked increases in atrial action potential durations, therefore, no electrocardiographic abnormalities attributable to, or suggestive of, altered atrial functioning are evident in Kv4.2W362F-expressing animals.

Different effects of Kv4.2W362F expression in mouse atria and ventricles

In ventricular myocytes isolated from Kv4.2W362F-expressing transgenics, the presence of a ‘novel’ rapidly activating and slowly inactivating K⁺ current (that is distinct from both I_to,f and I_K,slow in wild-type cells) was reported (Barry et al. 1998). More recent experiments, however, have revealed that a slow, transient outward K⁺ current is expressed in a subset of wild-type mouse ventricular myocytes (Xu et al. 1999). Specifically, this slow transient current, which was referred to as I_to,slow or I_to,s, was evident in cells isolated from the left ventricular septum (Xu et al. 1999). Interestingly, the properties of I_to,s in these cells (Xu et al. 1999) are very similar to the ‘novel’ current detected in (all) Kv4.2W362F-expressing ventricular myocytes (Barry et al. 1998), suggesting that the ‘novel’ current in Kv4.2W362F-expressing ventricular myocytes could reflect the upregulation of I_to,s, rather than the expression of a conductance pathway (and the underlying K⁺ channel subunits) not normally expressed in the mouse heart. Detailed analysis of the depolarization-activated K⁺ currents in wild-type and Kv4.2W362F-expressing transgenic atrial myocytes however, reveals that only I_ss and I_K,slow are present. There is no evidence for ‘upregulation’ of wild-type currents or for the presence of ‘novel’ voltage-gated K⁺ currents, i.e. currents not seen in wild-type cells, in Kv4.2W362F-expressing atrial myocytes. These observations suggest that, in contrast to findings in ventricular cells, electrical remodelling does not occur in mouse atria when I_to,f is eliminated. Clearly, studies focussed on delineating the molecular mechanisms underlying the tissue-specific remodelling that occurs in Kv4.2W362F-expressing animals will be of interest.

The absence of electrocardiographic abnormalities suggestive of altered atrial functioning in the Kv4.2W362F transgenics was initially surprising because of the profound effect of Kv4.2W362F expression on action potential repolarization in atrial myocytes. Indeed, the increases in atrial action potential durations are significantly larger than those in ventricular cells (Table 2). Although the functional consequences of Kv1.1N206Tag expression in mouse atria have not been described, it has been reported that ventricular action potentials and QT intervals are also prolonged significantly in Kv1.1N206Tag-expressing transgenics (London et al. 1998a). In addition, the Kv1.1N206Tag animals display premature ventricular beats and spontaneous ventricular arrhythmias (London et al. 1998a). Interestingly, the increases in ventricular action potential durations and the QT prolongation in the Kv4.2W362F-expressing animals are both larger than those seen in the Kv1.1N206Tag-expressing transgenics (London et al. 1998a). Nevertheless, neither premature beats nor spontaneous ventricular arrhythmias are seen in the Kv4.2W362F transgenics (Barry et al. 1998). The large differences in the functional consequences of atrial and ventricular expression of Kv4.2W362F, as well as the marked differences between the Kv4.2W362F and the Kv1.1N206Tag transgenics, suggest that additional factors, i.e. other than prolonged repolarization, must also play an important role in determining the propensity to develop and to sustain arrhythmias. Clearly, studies focussed on exploring the molecular mechanisms underlying the profound tissue specific (i.e. atrial versus ventricular) functional effects of cardiac-specific expression of both Kv4.2W362F and Kv1.1N206Tag are warranted.