Molecular diversity of the repolarizing voltage-gated K⁺ currents in mouse atrial cells

Abstract

1. Voltage-clamp studies on atrial myocytes isolated from adult and postnatal day 15 (P15) C57BL6 mice demonstrate the presence of three kinetically distinct Ca²⁺-independent, depolarization-activated outward K⁺ currents: a fast, transient outward current (I_to,f), a rapidly activating, slowly inactivating current (I_K,s) and a non-inactivating, steady-state current (I_ss). The time- and voltage-dependent properties of I_to,f, I_K,s and I_ss in adult and P15 atrial cells are indistinguishable.

2. Pharmacological experiments reveal the presence of two components of I_K,s: one that is blocked selectively by 50 μM 4-aminopyridine (4-AP), and a 4-AP-insensitive component that is blocked by 25 mM TEA; I_ss is also partially attenuated by 25 mM TEA. There are also two components of I_K,s recovery from steady-state inactivation.

3. To explore the molecular correlates of mouse atrial I_K,s and I_ss, whole-cell voltage-clamp recordings were obtained from P15 and adult atrial cells isolated from transgenic mice expressing a mutant Kv2.1 α subunit (Kv2.1N216Flag) that functions as a dominant negative, and from P15 atrial myocytes exposed to (1 μM) antisense oligodeoxynucleotides (AsODNs) targeted against Kv1.5 or Kv2.1.

4. Peak outward K⁺ current densities are attenuated significantly in atrial myocytes isolated from P15 and adult Kv2.1N216Flag-expressing animals and in P15 cells exposed to AsODNs targeted against either Kv1.5 or Kv2.1.

5. Analysis of the decay phases of the outward currents evoked during long (5 s) depolarizing voltage steps revealed that I_K,s is selectively attenuated in cells exposed to the Kv1.5 AsODN, whereas both I_K,s and I_ss are attenuated in the presence of the Kv2.1 AsODN.

6. In P15 and adult Kv2.1N216Flag-expressing atrial cells, mean ± s.e.m. I_K,s and I_ss densities are also significantly lower than in non-transgenic atrial cells. In addition, pharmacological experiments reveal that the TEA-sensitive component I_K,s is selectively eliminated in P15 and adult Kv2.1N216Flag-expressing atrial cells.

7. Taken together, the results presented here reveal that both Kv1.5 and Kv2.1 contribute to mouse atrial I_K,s, consistent with the presence of two molecularly distinct components of I_K,s. In addition, Kv2.1 contributes to mouse atrial I_ss.

Voltage-gated potassium (K⁺) channel currents are important in controlling the amplitudes and durations of action potentials in cardiac cells, as well as refractoriness and automaticity. These channels are also important targets for hormones, neurotransmitters and anti-arrhythmic drugs (Anumonwo et al. 1991; Gadsby, 1995). Electrophysiological studies have documented the expression of multiple types of voltage-gated K⁺ channels in cells isolated from different species and from different regions of the heart in the same species (Barry & Nerbonne, 1996; Roden & George, 1997). These currents can be classified into two broad categories: (1) rapidly activating and inactivating transient outward K⁺ currents, I_to; and (2) slowly or non-inactivating K⁺ currents, typically referred to as I_K. Molecular cloning techniques have revealed considerable potential for generating functional K⁺ current diversity (Barry & Nerbonne, 1996; Deal et al. 1996; Nerbonne, 1998), and have facilitated efforts focused on defining the molecular correlates of the various voltage-gated K⁺ channels evident in myocardial cells.

In adult mouse atrial myocytes, three kinetically distinct components of the total whole-cell Ca²⁺-independent depolarization-activated K⁺ currents have recently been distinguished: (1) a fast, transient outward current, I_to,f, (2) a rapidly activating, slowly inactivating current, I_K,s, and (3) a non-inactivating, steady-state current, I_ss (Xu et al. 1999c). Although it has been demonstrated that Kv4 α subunits underlie mouse atrial I_to,f (Xu et al. 1999c), the molecular correlates of I_K,s and I_ss in mouse atrium have not been defined. Recent reports, however, do suggest roles for α subunits of the Kv1, probably Kv1.5 (London et al. 1998), and the Kv2 (Xu et al. 1999a) subfamilies in the formation of I_K,s in mouse ventricle. The present study was initiated to investigate the molecular correlates of I_K,s and I_ss in mouse atrial myocytes, and to test the specific hypothesis that both Kv1.5 and Kv2.1 also play roles in the genesis of I_K,s and/or I_ss in these cells. Experiments were performed on adult and postnatal day 15 (P15) atrial myocytes isolated from transgenic mice expressing a truncated, epitope-tagged Kv2.1 α subunit, Kv2.1N216Flag, which functions as a dominant negative (Xu et al. 1999b), and on P15 atrial myocytes exposed to antisense oligodeoxynucleotides (AsODNs) targeted against the translation start site of either Kv2.1 or Kv1.5 (Bou-Abboud & Nerbonne, 1999). The results presented reveal that both Kv1.5 and Kv2.1 contribute to the generation of I_K,s in mouse atrial myocytes, and that Kv2.1 also contributes to mouse atrial I_ss.

METHODS

All experiments reported here were conducted in compliance with the guidelines of the Washington University Medical School Animal Use Committee. All of the protocols employed in the generation of transgenic mice, in the harvesting of mouse tissues and in the isolation of mouse atrial myocytes have been approved by the Animal Use Committee.

Kv α subunit expression in mouse atria

The methods used in designing and characterizing the epitope tagged, truncated Kv2.1, Kv2.1N216Flag, construct and for the generation of transgenic mice expressing this construct in the myocardium have recently been described (Xu et al. 1999a). Briefly, for generation of the mice, Kv2.1N216Flag was subcloned downstream of the α myosin heavy chain (α-MHC) promoter in an α-MHC expression vector obtained from Dr Jeffrey Robbins (University of Cinncinnati). This construct/vector has been used extensively to allow cardiac specific expression of transgenes (Lyons et al. 1990; Ng et al. 1991; Barry et al. 1998; Xu et al. 1999a, c). As reported previously, on gross examination, there were no differences evident between Kv2.1N216Flag-expressing transgenic animals and their non-transgenic littermates (Xu et al. 1999a). Small, but statistically significant differences were seen, however, in the properties of ventricular myocytes isolated from the transgenic, compared with the non-transgenic (wild-type), animals. Mean ±s.e.m. whole-cell membrane capacitances, for example, were higher, and input resistances were lower, in Kv2.1N216Flag-expressing ventricular cells compared with ventricular myocytes isolated from non-transgenic animals (Xu et al. 1999a). No differences in the properties, input resistances or whole-cell membrane capacitances of atrial myocytes isolated from the Kv2.1N216Flag-expressing transgenics, however, were observed in the present study (see Results).

For reverse transcription polymerase chain reaction (RT-PCR) analysis of wild-type Kv2.1, Kv1.5 and Kv2.1N216Flag expression, mRNA was prepared from the atria and ventricles of adult Kv2.1N216Flag-expressing transgenic and non-transgenic mice using the Macro-FasTrack mRNA isolation kit (Invitrogen, Carlsbad, CA, USA). Atrial and ventricular cDNA were synthesized in 20 μl reaction mixtures containing 50 mM Tris-HCl (pH 8.3), 5 mM MgCl₂, 75 mM KCl, 4 mM sodium pyrophosphate, 5 mM deoxynucleotides triphosphates (dNTPs), 10 mM dithiothreitol (DTT), 0.5 mg of oligo (dT)_12–18, 0.2 mg of mRNA, 10 units of RNAse inhibitor and 5 units of avian myeloblastosis virus reverse transcriptase (Invitrogen). After 1 h incubations at 42°C, reactions were terminated by heating to 95°C for 2 min; approximately 2 μl of each mixture was used for PCR amplification. RT-PCR was carried out in 25 μl reaction mixtures, each containing 10 mM Tris-HCl, pH 9.0, 50 mM KCl, 2 mM MgCl₂, 1 mM DTT, 0.5 mM of each primer, 0.2 mM dNTPs, and 1 unit of Taq DNA polymerase (Sigma). Each reaction proceeded for 30 cycles as follows: 94°C for 30 s; 58°C for 45 s; and 68°C for 90 s. The forward and reverse primers used for RT-PCR to probe for actin, Kv1.5, wild-type Kv2.1 and Kv2.1N216Flag are given in Fig. 3. For analysis, 20 μl of each of the amplified PCR mixtures was analysed on 1 % agarose gels, stained with ethidium bromide and photographed.

Figure 3. Expression of Kv α subunits in mouse atria and ventricles.

A, RT-PCR analysis reveals that Kv1.5 and Kv2.1 are readily detected in wild-type (wt) mouse atria and ventricles, whereas Kv2.1N216Flag is only detected in the hearts of transgenic (Kv2.1N216Flag-expressing) mice. For comparisons among samples, probes against muscle-specific α actin were employed. B, forward and reverse primers used in the PCR analysis.

Antisense oligodeoxynucleotides and electrophysiological recordings

Atrial myocytes were dispersed from atria isolated from adult and postnatal day 15 (P15) C57BL6 wild-type (WT) and Kv2.1N216Flag-expressing transgenic mice. For this purpose, each animal was anaesthesized (5 % halothane) and, once deep anaesthesia was confirmed, the chest cavity was opened and the heart was excised rapidly. Atrial appendages were removed and dissociated using enzymatic and mechanical methods similar to those previously described (Feng et al. 1997; Bou-Abboud & Nerbonne, 1999). Isolated cells were resuspended in Medium 199 (Irvine, Santa Ana, CA, USA), plated onto laminin-coated coverslips and maintained in a 95 % air and 5 % CO₂ incubator at 37°C (Bou-Abboud & Nerbonne, 1999; Xu et al. 1999a, b, c). To examine the effects of antisense oligodeoxynucleotide (AsODN) generated against the translation start site of either mouse Kv1.5 (nucleotides 4–18; AsODN sequence CAC CAG GGA GAT CTC) or mouse Kv2.1 (nucleotides 16–30; AsODN sequence CGA GCC ATG CTT CGT), P15 atrial cells were placed in media containing 1 μM AsODN approximately 2–4 h after plating. Incubations were performed at 37°C in the presence of 4–8 ng ml⁻¹ of lipofectamine (Sigma) to facilitate uptake of the AsODNs (Bou-Abboud & Nerbonne, 1999). In control experiments, cells were exposed to lipofectamine alone (see Results). The specificity of the AsODNs has been demonstrated previously in experiments completed on heterologously expressed Kv α subunits (see Results and Bou-Abboud & Nerbonne, 1999). Additional control experiments examined the effects of scrambled AsODNs containing the same deoxynucleotides as in the Kv1.5 AsODN or the Kv2.1 AsODN but in the following (scrambled) sequences: ATC CAA GCG GCG TAC (for Kv1.5) and CGC GTA CAG TCC GTT (for Kv2.1). All AsODNs were obtained from Integrated DNA Technologies (Coralville, IA, USA) and all were 5′ tagged with fluorescein to allow visualization of uptake into isolated myocytes.

Whole-cell patch-clamp recordings (Hamill et al. 1981) from isolated adult and P15 C57BL6 mouse atrial myocytes were obtained at room temperature (22–24°C) within the first 24 h after isolation. Experiments were performed using an Axopatch 1D amplifier (Axon Instruments) interfaced to a Gateway 400 MHz Pentium computer with a Digidata1200 analog/digital interface (Axon) and the pCLAMP 7 software package (Axon). The extracellular bath solution contained (mM): 136 NaCl, 4 KCl, 2 MgCl₂, 1 CaCl₂, 10 glucose and 10 Hepes; pH 7.4 (NaOH); tetrodotoxin (TTX; 20 μM) and CdCl₂ (200 μM) were also added to suppress voltage-gated Na⁺ and Ca²⁺ currents, respectively. Recording pipettes routinely contained (mM): 135 KCl, 4 NaCl, 10 EGTA, 10 Hepes, 5 glucose, 3 Mg-ATP and 0.5 Na₃-GTP; pH 7.2 (KOH). 4-Aminopyridine (4-AP) stock solutions were prepared in water, and diluted in bath solution immediately prior to use. Tetraethylammonium (TEA)-containing bath solutions were prepared by equimolar substitution of TEACl for NaCl in bath solution. 4-AP or TEA was applied to isolated myocytes during recordings using narrow bore (∼200 μm) capillary tubes place within ∼200 μm of the cell. Recording pipettes, fabricated from borosilicate glass (Sutter Instruments, Novato, CA, USA) using a Sutter Model P-87 puller, had resistances at 1.5–3 MΩ when filled with the recording solution.

After establishing the whole-cell configuration, ± 5 mV (25 ms) voltage steps were applied from a holding potential (V_h) of −60 mV to allow measurement of whole-cell membrane capacitances (C_m). Series resistance (R_s) was estimated from the decay of the capacitative transients, and the mean ±s.e.m. value of R_s was 7.7 ± 0.7 MΩ; (n = 26). In each cell, R_s was compensated electronically by 80–90 %. Because peak outward K⁺ current amplitudes at potentials < +50 mV in mouse atrial myocytes were less than 2 nA, voltage errors resulting from the uncompensated series resistance were always < 4 mV, and were not corrected. Tip potentials were zeroed before membrane pipette seals were formed. Only data obtained from cells with seal resistances > 2 GΩ were analysed. Outward K⁺ currents were evoked by depolarizing voltage steps to potentials between −50 and +50 mV from a V_h of −60 mV. Inwardly rectifying K⁺ currents (I_K1) were recorded in response to hyperpolarizing voltage steps to potentials between −70 and −120 mV from the same V_h (−60 mV). Currents were low-pass filtered at 2–5 kHz, digitized at 1–10 kHz, and stored for subsequent off-line analysis.

Data analysis

Whole-cell membrane capacitances (C_m) were determined by integrating the capacitative current transients evoked during brief (25 ms) ± 5 mV voltage steps from −60 mV. Input resistances (R_in) were determined from the currents remaining at the end of 25 ms ± 10 mV (hyperpolarizing or depolarizing) voltage steps from −60 mV. Mean ±s.e.m. (n = 28) C_m and R_in values for adult wild-type mouse atrial myocytes were 46.8 ± 4.0 pF and 2.7 ± 0.3 GΩ, respectively. Peak outward K⁺ currents at each test potential (in each cell) were measured (relative to the zero current level) as the maximal amplitude of the currents recorded during the first 100 ms of the depolarizing voltage steps. I_K1 was measured as the amplitude of the current at the end of 100 ms hyperpolarizing voltage steps to −120 mV. Peak outward K⁺ current densities were obtained by dividing the measured currents by the whole-cell membrane capacitance, C_m (determined in the same cell). The decay phases of the outward currents recorded in response to long (5 s) depolarizing voltage steps were fitted using the equation:

where t is the time and τ_fast and τ_slow are the time constants of inactivation of I_to,f and I_K,s, respectively; I_ss is the steady-state, non-inactivating component of the current (Xu et al. 1999a).

The time course of recovery from steady-state inactivation of the outward K⁺currents in adult (and P15) atrial myocytes was examined using a double pulse protocol (P₀= P₁= 5 s) with interpulse intervals ranging from 0.01 to 30 s (see protocol in Fig. 1). The amplitudes of the individual current components (I_to,f, I_K,s and I_ss) were determined from double exponential fits to the decay phases of the total outward K⁺ currents recorded during the test depolarization (P₁) to +30 mV, and these values were subsequently normalized to the current (I_to,f, I_K,s and I_ss) amplitudes determined in P₀. The recovery data for I_K,s were well described by a biexponential function of the form:

where τ_rec,fast and τ_rec,slow are the time constants of the fast and slow components of recovery, respectively. All data are presented as means ±s.e.m. A two-tailed Student's t test was used to assess statistical significance between the different experimental groups; P values are given in the text and figures.

Figure 1. Two components of adult mouse atrial I_K,s recovery from steady-state inactivation.

After inactivating the currents during 5 s prepulses to +30 mV, cells were hyperpolarized to −70 mV for times ranging from 10 ms to 30 s before a second test depolarization to +30 mV (to assess the extent of recovery) was presented; the protocol is illustrated in the lower left panel. A, representative (total outward K⁺) current waveforms recorded during the conditioning step and after varying recovery times are presented; currents recorded after brief recovery times are displayed in the inset. The amplitudes of I_to,f, I_K,s and I_ss at each recovery time (in each cell) were determined from double exponential fits to the decay phases of the total outward K⁺ currents. These values were then normalized to the amplitudes of the currents (I_to,f, I_K,s and I_ss) evoked after the 40 s recovery period (in the same cell). B, mean ±s.e.m. (n = 7) normalized I_K,s amplitudes are plotted as a function of the interpulse interval (IPI); the recovery data at early times are shown on an expanded time scale in the inset. As is evident, the mean normalized recovery data for I_K,s are well fitted (continuous line) by the sum of two exponentials with time constants of ≈0.5 and ≈10 s (see text).

RESULTS

Two components of I_K,s in mouse atrial myocytes

The waveforms of the depolarization-activated outward K⁺ currents in adult mouse atrial myocytes are similar to those in adult mouse ventricular cells (Xu et al. 1999b) although, as reported previously, peak outward K⁺ current densities are significantly lower in atrial (than in ventricular) cells (Xu et al. 1999c). Also similar to the currents in the vast majority of adult mouse ventricular cells (Xu et al. 1999b), three kinetically distinct outward K⁺ current components have been distinguished in adult mouse atrial myocytes from analysis of the decay phases of the outward K⁺ currents evoked during prolonged (5 s) depolarizing voltage steps (Xu et al. 1999c). Adopting the terminology used to describe the depolarization-activated outward K⁺currents in mouse ventricular myocytes (Xu et al. 1999b), these current components are referred to as I_to,f, I_K,s and I_ss (Xu et al. 1999c). Although previous studies have demonstrated that, as in ventricular cells, α subunits of the Kv4 subfamily underlie I_to,f in mouse atrial myocytes (Xu et al. 1999c), the molecular correlates of mouse atrial I_K,s and I_ss have not been defined. The aim of the study here, therefore, was to explore the molecular correlates of I_K,s and I_ss in mouse atrial cells.

In preliminary experiments aimed at more detailed characterization of the outward K⁺ currents in adult mouse atrial myocytes than was completed in previous studies (Xu et al. 1999c), evidence for the presence of two components of I_K,s was obtained. In experiments exploiting a standard two-pulse protocol (P₁= P₂= 0.5 s; see protocol in Fig. 1), for example, the time course of recovery from steady-state inactivation of the atrial K⁺ currents was examined. Following 5 s depolarizations to +30 mV (from a V_h of −70 mV) to inactivate the currents, cells were repolarized to −70 mV for varying times, ranging from 0.01 to 30 s, prior to the test depolarizations to +30 mV to assess the extent of recovery (Fig. 1A). The amplitudes of the individual current components, I_to,f, I_K,s and I_ss, at each recovery time were determined from double exponential fits to the decay phases of the currents recorded during the test depolarizations and normalized to the corresponding (I_to,f, I_K,s and I_ss) control current amplitudes determined during the prepulse (to +30 mV). The normalized recovery data for I_K,s (Fig. 1B) follow a biexponential time course characterized by a fast recovery time constant (τ_rec,fast) of ∼0.5 s and a much slower recovery time constant (τ_rec,slow) of ∼10 s); the recovery data at short interpulse intervals are displayed in the inset to Fig. 1B. The mean ±s.e.m. (n = 7) τ_rec,fast and τ_rec,slow determined from these experiments were 370 ± 70 ms and 7.6 ± 2.6 s, respectively. The relative amplitudes are also revealed in these analyses and, on average, the fast component of recovery contributes 38 ± 8 % and the slowly recovering component contributes 62 ± 8 % to the total I_K,s (Fig. 1B).

Pharmacological experiments also revealed the presence of two components of I_K,s in adult mouse atrial myocytes (Fig. 2). As illustrated in Fig. 2B, outward K⁺ current amplitudes in adult mouse atrial myocytes are attenuated in the presence of low (50 μM) concentrations of 4-AP. Analysis of the 50 μM 4-AP-sensitive currents (Fig. 2C) revealed that I_K,s amplitude is reduced by ∼50 % and that I_to,f is attenuated by ∼20 % by 50 μM 4-AP. In the presence of 5 mM 4-AP, peak outward K⁺ currents are further reduced (Fig. 2D). Analysis of the 5 mM 4-AP-sensitive K⁺ currents (Fig. 2E) revealed that ∼50 % of I_K,s remains, whereas I_to,f is almost completely (∼90 %) eliminated (in the presence of 5 mM 4-AP). The residual, 5 mM 4-AP-insensitive component of I_K,s (Fig. 2D), however, is blocked effectively by 25 mM TEA (Fig. 2F). Similar results were obtained in five cells, and on average, 45 ± 8 % of the control I_K,s remains in the presence of 5 mM 4-AP and this 5 mM 4-AP-insensitive current is blocked almost completely by 25 mM TEA. Similar to the recovery data presented above, the findings suggest the presence of two distinct components of I_K,s.

Figure 2. 4-AP-sensitive and insensitive components of I_K,s in adult mouse atrial myocytes.

Outward K⁺ currents were recorded during 5 s depolarizations to test potentials between 0 and +50 mV from a V_h of −60 mV; all of the records displayed were obtained from the same cell. Control currents (A) and currents in the presence of 50 μM 4-AP (B) or 5 mM 4-AP (D) were recorded. The waveforms of the 50 μM 4-AP-sensitive (C) and the 5 mM 4-AP-sensitive (E) currents were obtained by off-line digital subtraction of the records in the presence of 50 μM (B) or 5 mM (D) 4-AP from the controls (A). To determine the effect of TEA on the currents remaining in the presence of 5 mM 4-AP (D), cells were exposed to 25 mM TEA (F); as is evident, most of residual current was blocked in the presence of TEA (F). The component of the 4-AP-insensitive currents (D) that is blocked by TEA (G) was obtained by off-line digital subtraction of the records in the presence of 25 mM TEA (F) from those in the presence of 5 mM 4-AP (D). Similar results were obtained in five cells.

Expression of Kv2.1, Kv1.5 and Kv2.1N216Flag in mouse atria

To explore the role of Kv α subunits of the Kv2 subfamily (and specifically Kv2.1) in the generation of functional voltage-gated K⁺ channels in mouse atrial myocytes, experiments focused on determining the effects of exposure to an antisense oligodeoxynucleotide (AsODN) targeted against Kv2.1, and of expression of an epitope-tagged, truncated form of Kv2.1, Kv2.1N216Flag, which functions as dominant negative (Xu et al. 1999a), were completed. Importantly, Kv2.1N216Flag, as well as wild-type, full length Kv2.1 (and Kv1.5), is readily detected in the atria, as well as in the ventricles, of Kv2.1N216Flag-expressing transgenic animals (Fig. 3), suggesting that experiments focused on examining the functional consequences of Kv2.1N216Flag expression in adult mouse atrial myocytes could be completed. Similar to previous findings in rat atrial myocytes (Bou-Abboud & Nerbonne, 1999), however, preliminary experiments also revealed that the uptake of the fluorescein-tagged AsODNs into isolated mouse atrial myocytes required the presence of lipofectamine and that uptake is age dependent. No uptake of the labelled AsODNs was evident, for example, in cells isolated from animals older than postnatal day 17 (P17). At P15, uptake could be visualized in ∼20 % of the cells, suggesting that the antisense experiments could be completed on cells isolated from P15 animals. Prior to these studies, however, experiments aimed at comparing the densities and the properties of the depolarization-activated K⁺ currents in atrial myocytes isolated from wild-type P15 animals with those of the currents recorded in adult mouse atrial cells were completed.

Outward K⁺ currents in P15 and adult mouse atrial myocytes are similar

In response to brief (∼100 ms) depolarizations, outward K⁺ currents in mouse atrial myocytes activate rapidly and decay only slightly, and the waveforms of the evoked currents in adult (Fig. 4A) and P15 (Fig. 4B) atrial cells are similar. The peak amplitudes of the depolarization-activated outward K⁺ currents in P15 and adult mouse atrial cells are not significantly different (not illustrated). The mean ±s.e.m. peak outward K⁺ current amplitudes measured in depolarization to + 50 mV in P15 and adult cells, for example, were 1025 ± 194 pA (n = 30) and 1398 ± 147 pA (n = 28), respectively. Cells isolated from P15 animals, however, are significantly (P < 0.001) smaller than adult cells; the mean ±s.e.m. (n = 30) C_m for P15 cells was 18.2 ± 2.2 pA pF⁻¹, compared with a mean ±s.e.m. (n = 28) C_m of 46.8 ± 4.0 pF in adult cells. Normalization of current amplitudes for differences in cell sizes (C_m), therefore, revealed that peak outward K⁺ current densities are actually lower in adult (Fig. 4B and D), than in P15 (Fig. 4A and C) mouse atrial cells. Similar results were obtained in many experiments and, as illustrated in Fig. 4E, mean ±s.e.m. peak outward K⁺ current densities are significantly (P < 0.01) higher in P15, than in adult, cells at all test potentials (Fig. 4E). In contrast to the differences in peak outward K⁺ current densities, however, there were no statistically significant differences in the densities of the inwardly rectifying K⁺ currents (I_K1) in adult and P15 mouse atrial cells; mean ±s.e.m.I_K1 densities (at −120 mV) were 15.0 ± 1.5 (n = 28) and 18.7 ± 2.4 pA pF⁻¹ (n = 30) in adult and P15 atrial myocytes, respectively. The finding that peak outward K⁺ current amplitudes are the same (although peak current densities are different) in adult and P15 mouse atrial myocytes suggests that the number of functional voltage-gated K⁺ channels on the surface may not vary measurably between P15 and the adult (see Discussion). It is certainly possible, however, that the properties of the underlying K⁺ channels change in complex ways during development, thereby resulting in macroscopic total outward K⁺current waveforms that are similar in P15 and adult cells (Fig. 4). Subsequent experiments, therefore, were aimed at further characterizing the properties of the currents in P15 mouse atrial myocytes.

Figure 4. Comparison of outward K⁺ currents in adult and postnatal day 15 (P15) mouse atrial myocytes.

Depolarization-activated outward currents, evoked during 100 ms (A and B) and 5 s (C and D) voltage steps to varying test potentials (−50 to +50 mV) from a holding potential of −60 mV, were measured and subsequently normalized to the whole-cell membrane capacitance (C_m) in the same cell to correct for differences in cell size and determine current densities. The decay phases of the outward K⁺ currents evoked during 5 s voltage steps in both P15 (C) and adult (D) cells were well fitted by the sum of two exponentials; the fits (continuous lines) are superimposed on the experimental data. E, normalized mean ±s.e.m. peak outward K⁺ current densities in P15 and atrial cells are plotted as a function of test potential.

As reported previously (Xu et al. 1999c), exponential fits to the decay phases of the currents evoked during prolonged (5 s) voltage steps revealed that the depolarization-activated outward K⁺ currents in adult mouse atrial cells (Fig. 4D) are well described by the sum of two exponentials, reflecting I_to,f and I_K,s, and that there is a steady-state non-inactivating current, I_ss (Xu et al. 1999c). Neither inactivating time constant displays any appreciable voltage dependence, and the mean ±s.e.m. (n = 11) decay time constants (τ_decay) derived from the (double exponential) fits to the decay phases of the outward K⁺ currents evoked during 5 s depolarizing voltage steps to 0 mV in adult cells (in records such as those in Fig. 4D) were 91 ± 9 and 1333 ± 96 ms for I_to,f and I_K,s, respectively. These values are very similar to those reported previously (Xu et al. 1999c). The decay phases of the outward K⁺ currents in P15 cells (Fig. 4C) were also well fitted by the sum of two exponentials with mean ±s.e.m. (n = 12) τ_decay of 114 ± 15 and 1536 ± 111 ms. These values are not significantly different from those determined in adult cells.

Analysis of the decay phases of the outward K⁺ currents in P15 and adult mouse atrial myocytes also provided the amplitudes of I_to,f, I_K,s and I_ss and revealed that, similar to the peak outward K⁺ currents, the amplitudes of I_to,f, I_K,s and I_ss in P15 cells are not measurably different in P15 and adult cells. The densities of I_to,f, I_K,s and I_ss, however, are significantly (P < 0.01) higher in P15 than in adult cells. Mean ±s.e.m. (at +50 mV) I_to,f densities, for example, were 39 ± 9.6 and 19 ± 3.9 pA pF⁻¹ in P15 (n = 12) and adult (n = 18) atrial myocytes, respectively. Mean ±s.e.m.I_K,s densities (at +50 mV) in P15 and adult cells were 14.4 ± 2.1 and 9.0 ± 1.6 pA pF⁻¹, respectively, whereas mean ±s.e.m.I_ss densities (at +50 mV) in P15 and adult cells were 11.7 ± 1.9 and 5.8 ± 0.5 pA pF⁻¹, respectively.

The densities of I_to,f, I_K,s and I_ss at P15 are approximately double those in adult cells. The relative contributions of I_to,f (∼55 %), I_K,s (∼25 %) and I_ss (∼20 %) to the peak outward K⁺ currents in adult and P15 mouse atrial myocytes, however, are similar. In addition, the voltage dependences of activation and inactivation of I_to,f, I_K,s and I_ss in P15 and adult mouse atrial myocytes are indistinguishable. The finding that the densities of the currents (I_to,f, I_K,s and I_ss) are all lower in adult (than P15) cells is attributed to the marked increase in whole-cell membrane capacitances in mouse atrial myocytes between P15 and the adult. The fact that the time- and voltage-dependent properties of I_to,f, I_K,s and I_ss in adult and P15 cells are indistinguishable suggests that the molecular correlates of the underlying K⁺ channels are the same. Further support for this hypothesis is the finding that the outward K⁺ currents in adult and P15 mouse atrial myocytes have similar pharmacological properties (see below). Nevertheless, it must be noted that it certainly remains a possibility that distinct molecular entities with very similar (indistinguishable) properties underlie the individual current components at different developmental stages (see Discussion).

Peak K⁺ currents are reduced in Kv2.1N216Flag-expressing atrial cells

As reported previously, whole-cell voltage-clamp recordings from ventricular myocytes isolated from adult Kv2.1N216Flag-expressing transgenic animals revealed that whole-cell membrane capacitances were increased and input resistances were reduced relative to those in the cells isolated from non-transgenic littermates. These observations suggest the presence of some (cellular) ventricular hypertrophy in Kv2.1N216Flag-expressing hearts (Xu et al. 1999a), perhaps owing to the overexpression of the truncated Kv2.1 protein. In contrast, no significant differences in whole-cell membrane capacitances or input resistances were evident when the properties of the Kv2.1N216Flag-expressing atrial myocytes were compared with cells isolated from non-transgenic (wild-type) littermates. The mean ±s.e.m. (n = 22) R_in and C_m in adult Kv2.1N216Flag-expressing atrial myocytes were 2.9 ± 0.3 GΩ and 54.3 ± 4.5 pF, values indistinguishable from the mean ±s.e.m. (n = 28) R_in and C_m of 2.7 ± 0.3 GΩ and 46.8 ± 4.0 pF determined in wild-type adult mouse atrial cells. The properties of P15 atrial myocytes isolated from Kv2.1N216Flag-expressing and wild-type animals are also indistinguishable.

Whole-cell voltage-clamp recordings from atrial cells isolated from Kv2.1N216Flag-expressing animals revealed that outward K⁺ currents are attenuated compared with the currents in wild-type cells. In cells isolated from P15 Kv2.1N216Flag-expressing animals (Fig. 5B), for example, peak outward K⁺ current densities are attenuated relative to the currents in wild-type P15 cells (Fig. 5A). As illustrated in Fig. 5E, mean ±s.e.m. peak outward K⁺ current densities at all test potentials are significantly (P < 0.01) lower in P15 Kv2.1N216Flag-expressing cells than in P15 wild-type cells. In contrast, mean ±s.e.m.I_K1 densities in wild-type and Kv2.1N216Flag-expressing P15 mouse atrial cells (16.3 ± 2.4 pA pF⁻¹, n = 16) are not significantly different. Similar results were obtained when cells isolated from adult wild-type and Kv2.1N216Flag-expressing animals were compared. These results reveal an important contribution of Kv α subunits of the Kv2 subfamily in the generation of the repolarizing K⁺ currents in (adult and P15) mouse atrial myocytes. Because Kv2.1 (as well as Kv1.5) is abundant in mouse atria (Fig. 3), subsequent experiments were undertaken to assess directly the role of Kv2.1 (and Kv1.5) in the formation of functional voltage-gated K⁺ channels in these cells.

Figure 5. Peak outward K⁺ currents are attenuated in atrial myocytes isolated from P15 Kv2.1N216Flag-expressing transgenics and in P15 atrial cells exposed to AsODNs targeted against either Kv1.5 or Kv2.1.

Whole-cell outward K⁺ currents were recorded as described in the legend to Fig. 4 from P15 cells isolated from wild-type (WT) (A) or Kv2.1N216Flag-expressing (B) mice, and from P15 cells exposed to either the Kv2.1 AsODN (C) or the Kv1.5 AsODN (D). Representative outward K⁺ current waveforms (and densities) are presented in A-D. Compared with wild-type P15 cells, mean ±s.e.m. peak outward K⁺ current densities (E) are significantly (P < 0.01) lower in P15 cells exposed to either the Kv1.5 (n = 15) or the Kv2.1 (n = 15) AsODN. Mean ±s.e.m. peak outward currents are also significantly (P < 0.001) lower in Kv2.1N216Flag-expressing (n = 18), than in WT (n = 30) cells (E).

Outward K⁺ currents are reduced in atrial myocytes exposed to Kv2.1 or Kv1.5 AsODN

To explore directly the role of Kv2.1 in the generation of voltage-gated mouse atrial K⁺ currents, the effects of an AsODN targeted against Kv2.1 were examined. Previous experiments demonstrated that the Kv2.1 AsODN specifically and selectively attenuates Kv2.1-induced currents in transiently transfected HEK 293 cells (Bou-Abboud & Nerbonne, 1999). The Kv2.1 AsODN only affects the currents recorded from cells expressing Kv2.1; no effects of the Kv2.1 AsODN were evident in cells expressing Kv1 or Kv4 α subunits (Bou-Abboud & Nerbonne, 1999). Similar to previous studies on rat atrial myocytes (Bou-Abboud & Nerbonne, 1999), however, preliminary experiments revealed that uptake of the AsODNs is age dependent. No uptake, for example, was detected in mouse atrial cells isolated from > P17 animals; at P15, uptake was visualized in ∼ 20 % of the cells. In addition, uptake is only detected in the presence of lipofectamine. Experiments aimed at examining the effects of the Kv2.1 (and the Kv1.5) AsODN on mouse atrial K⁺ currents were, therefore, by necessity, completed in cells isolated from P15 animals and were conducted in the presence of lipofectamine (see Discussion).

Whole-cell voltage-clamp recordings from (P15) mouse atrial myocytes exposed to the Kv2.1 AsODN (in the presence of lipofectamine) revealed that peak outward K⁺ currents are markedly reduced (Fig. 5C) compared with the currents recorded in wild-type P15 cells exposed to lipofectamine alone (Fig. 5A). As in the Kv2.1N216Flag-expressing cells, mean ±s.e.m. peak outward K⁺ current densities at all test potentials (> −20 mV) are significantly lower in P15 atrial myocytes exposed to the Kv2.1 AsODN than in wild-type P15 cells (Fig. 5E). In contrast, outward K⁺ currents in cells exposed to a scrambled [Kv2.1] oligodeoxynucleotide (ODN; see Methods) were indistinguishable from those in wild-type cells. Mean ±s.e.m. peak outward current densities at +50 mV in wild-type P15 atrial myocytes and in P15 cells exposed to the scrambled [Kv2.1] ODN were 57.1 ± 6.0 (n = 28) and 54.8 ± 7.5 pA pF⁻¹ (n = 9), respectively.

Experiments were also completed on cells exposed to an AsODN targeted against the translation start site of Kv1.5. Similar to the Kv2.1 AsODN, previous studies have demonstrated that the effects of this AsODN are specific for Kv1.5 (Bou-Abboud & Nerbonne, 1999). Similar to the findings in cells exposed to the Kv2.1 AsODN, marked reductions in peak outward K⁺ current densities were evident in P15 mouse atrial cells incubated in the Kv1.5 AsODN (Fig. 5D and E). In addition, mean ±s.e.m. peak outward K⁺ current densities in P15 cells exposed to the Kv1.5 AsODN are significantly lower (than in wild-type P15 cells) at all test potentials positive to −20 mV (Fig. 5E). Outward K⁺ currents in cells exposed to a scrambled [Kv1.5] ODN (see Methods), in contrast, were indistinguishable from those in wild-type cells. The mean ±s.e.m. peak outward current density at +50 mV in P15 cells exposed to the scrambled [Kv1.5] ODN, for example, was 54.8 ± 8.1 pA pF⁻¹ (n = 9), a value that is not significantly different from the mean ±s.e.m. (n = 30) peak current density of 57.1 ± 6.0 pA pF⁻¹ determined in wild-type P15 atrial cells.

In contrast to the effects on outward K⁺ currents, the densities of the inwardly rectifying K⁺ currents (I_K1) were not affected in cells exposed to the Kv1.5 AsODN or the Kv2.1 AsODN (Fig. 5). Mean ±s.e.m.I_K1 densities at −120 mV in Kv1.5 AsODN (16.6 ± 3.4 pA pF⁻¹; n = 10) and in Kv2.1 AsODN (13.2 ± 4.2 pA pF⁻¹; n = 11) treated cells were not significantly different from each other or from the mean ±s.e.m.I_K1 density of 18.7 ± 2.4 pA pF⁻¹ (n = 28) determined in P15 atrial cells exposed to lipofectamine alone. In addition, I_K1 densities were unaffected following exposure of P15 atrial cells to the scrambled [Kv1.5] and [Kv2.1] ODNs.

Selective attenuation of I_K,s and I_ss in Kv2.1N216Flag-expressing atrial myocytes

As noted above, analysis of the decay phases of the outward currents in P15 and adult mouse atrial cells reveals the presence of three kinetically distinct components: I_to,f, I_K,s and I_ss (Fig. 4). Subsequent experiments, therefore, were focused on identifying the component(s) of the peak outward K⁺ currents affected by the Kv1.5 AsODN, the Kv2.1 AsODN and/or the expression of Kv2.1N216Flag. The decay phases of the currents evoked during 5 s voltage steps in cells exposed to the AsODNs or isolated from the Kv2.1N216Flag-expressing transgenic animals were analysed (see Methods) and compared with the currents in wild-type cells. Representative outward K⁺ currents recorded at 0 mV in a wild-type P15 atrial cell (Fig. 6A), a Kv2.1N216Flag-expressing P15 atrial cell (Fig. 6B), and in P15 atrial cells exposed to the Kv1.5 AsODN or the Kv2.1 AsODN (Fig. 6C and D) are presented in Fig. 6. Analysis of the decay phases of the currents in records such as those presented in Fig. 6 revealed that in cells exposed to the Kv1.5 AsODN (Fig. 6D), I_K,s is selectively attenuated (Fig. 6E), whereas both I_K,s and I_ss current densities are reduced in cells exposed to the Kv2.1 AsODN (Fig. 6C and E). Analysis of the outward K⁺ currents in Kv2.1N216Flag-expressing P15 cells (Fig. 6B) also revealed that the densities of I_K,s and I_ss are selectively reduced in these cells (Fig. 6E). In spite of the marked reductions in I_K,s amplitudes (densities), there were no changes in the time constants of I_K,s (or I_to,f) inactivation (Fig. 6F) in any of these experiments. Similar results were obtained in experiments completed on atrial myocytes (n = 15) isolated from Kv2.1N216Flag-expressing adult animals, i.e. I_K,s and I_ss densities are reduced relative to the densities in wild-type adult cells, whereas I_to,f densities and the time constants of I_K,s (and I_to,f) decay are unaffected.

Figure 6. Kv1.5 contributes to I_K,s, whereas Kv2.1 contributes to both I_K,s and I_ss.

Outward K⁺ currents were recorded in response to 5 s depolarizing voltage steps to 0 mV from −60 mV in P15 wild-type (WT) atrial myocytes (A), cells exposed to either the Kv1.5 AsODN (D) or the Kv2.1 AsODN (C), and in Kv2.1N216Flag-expressing cells (B). The decay phases of the currents were fitted to the sum of two exponentials to provide the amplitudes of I_to,f, I_K,s and I_ss. E, in cells exposed to the Kv1.5 AsODN (n = 8) or the Kv2.1 AsODN (n = 7) and in cells isolated from Kv2.1N216Flag-expressing transgenics (n = 11), I_K,s (density) is significantly (**P < 0.01) lower than I_K,s density in WT (n = 12) cells. F, the time constants of I_K,s inactivation, however, were not affected by the (Kv1.5 or Kv2.1) AsODNs or by the expression of the (Kv2.1N216Flag) transgene. Mean ±s.e.m.I_ss densities are also attenuated significantly (*P < 0.05, **P < 0.01) in P15 atrial myocytes exposed to the Kv2.1 AsODN and in Kv2.1N216Flag-expressing cells (E).

The results of the experiments presented above reveal that there are two components of mouse atrial I_K,s, consistent with the results of the pharmacological experiments completed in wild-type cells, suggesting the presence of two (pharmacologically distinct) components of I_K,s, one that is sensitive to micromolar concentrations of 4-AP and another that is 4-AP insensitive, but is blocked by millimolar concentrations of TEA. Subsequent experiments, therefore, examined the pharmacological sensitivity of the component of I_K,s remaining in the Kv2.1N216Flag-expressing cells and compared the results with those obtained in wild-type atrial cells. As illustrated in Fig. 7, bath application of 50 μM 4-AP attenuates the amplitudes of both I_to,f and I_K,s in Kv2.1N216Flag-expressing adult atrial myocytes (Fig. 7B), and analysis of 50 μM 4-AP-sensitive currents in records such as those in Fig. 7C revealed that, on average, I_to,f is attenuated by approximately 30 % and I_K,s by approximately 50 % in Kv2.1N216Flag-expressing cells (n = 6). Exposure to 5 mM 4-AP completely blocks both I_to,f and I_K,s in these cells, and only I_ss remains (Fig. 7D). In contrast to wild-type atrial cells (Fig. 2D), therefore, there does not appear to be a (5 mM) 4-AP-insensitive component of I_K,s in Kv21N216Flag-expressing (P15 and adult) atrial myocytes. Consistent with this interpretation, experiments aimed at examining the effects of TEA on the outward K⁺ currents in Kv21N216Flag-expressing cells revealed that 25 mM TEA has little effect on the K⁺ currents in these cells. As illustrated in Fig. 8, bath application of 25 mM TEA has no detectable effect on I_K,s (or I_ss) in Kv2.1N216Flag-expressing atrial myocytes. Subtraction of the records in the presence of 25 mM TEA (Fig. 8E) from the controls (Fig. 8D) provided the 25 mM TEA-sensitive currents (Fig. 8F). Analysis of the 25 mM TEA-sensitive currents (Fig. 8F) revealed only the presence of I_to,f (n = 5). The TEA-sensitive components of I_K,s and I_ss that are evident in wild-type mouse atrial cells (Fig. 2G), therefore, are not detected in Kv2.1N216Flag-expressing cells.

Figure 7. The 4-AP-sensitive component of I_K,s is evident in Kv2.1N216Flag-expressing mouse atrial myocytes.

Outward K⁺ currents were recorded as described in the legend to Fig. 2; all of the records displayed were obtained from the same cell. Control currents (A) and currents in the presence of 50 μM 4-AP (B) or 5 mM 4-AP (D) were recorded. The waveforms of the 50 μM 4-AP-sensitive (C) and the 5 mM 4-AP-sensitive (E) currents were obtained by off-line digital subtraction of the records in the presence of 50 μM (B) or 5 mM (D) 4-AP from the controls (A). In contrast to the findings in wild-type cells (Fig. 2), there does not appear to be a 4-AP-insensitive component of I_K,s in Kv2.1N216Flag-expressing atrial cells. Similar results were obtained in five cells.

Figure 8. The TEA-sensitive component of I_K,s, evident in wild-type mouse atrial myocytes is not detected in cells isolated from the Kv2.1N216Flag-expressing transgenic animals.

Outward K⁺ currents were recorded as described in the legend to Fig. 2; the records displayed in A, B and C were obtained from the same cell, as were the records in D, E and F. Control currents (A and D) and currents in the presence of 25 mM TEA (B and E) were recorded in wild-type (WT) (A and B) and Kv2.1N216Flag-expressing cells (D and E). The waveforms of the 25 mM TEA-sensitive (C and F) were obtained by off-line digital subtraction of the records in the presence of 25 mM TEA (B and E) from the controls (A and D). Similar results were obtained in five WT and five Kv2.1N216Flag-expressing cells.

DISCUSSION

Two components of mouse atrial I_K,s

In adult and P15 wild-type (C57BL6) mouse atrial cells, three kinetically distinct current components, I_to,f, I_K,s and I_ss, are revealed on analysis of the decay phases of the currents evoked during long (5 s) depolarizing voltage steps. Peak outward K⁺ current densities and the densities of the three K⁺ current components (I_to,f, I_K,s and I_ss), however, are higher in P15 than in adult, cells. Although the differences in current densities are small, they are, nevertheless, statistically significant. In contrast, peak current amplitudes are not significantly different in cells isolated from P15 and adult animals. Mean ±s.e.m. peak outward K⁺ current amplitudes at +50 mV, for example, were 1025 ± 194 pA in P15 (n = 30), and 1398 ± 147 pA (n = 28) in adult cells. Cell size, as reflected in whole-cell membrane capacitance, however, increases markedly between P15 and adult: mean ±s.e.m. whole-cell membrane capacitances were 18.2 ± 2.2 (n = 30) and 46.8 ± 4 pF (n = 28) in P15 and adult mouse atrial myocytes, respectively. The observed decreases in peak outward K⁺ current densities, as well as in the densities of I_to,f, I_K,s and I_ss, between P15 and adult, therefore, are attributed to the large increases in whole-cell membrane capacitances that occur during this period of development. In contrast to the differences in current densities, the time- and voltage-dependent properties and the pharmacological sensitivities of the currents in P15 and adult cells are indistinguishable, suggesting that the molecular correlates of the underlying K⁺ channels are also the same during development from P15 to adult.

Unexpectedly, analysis of the kinetics of the recovery of mouse atrial I_K,s from steady-state inactivation revealed the presence of two components of I_K,s characterized by recovery time constants differing by approximately an order of magnitude. The relative amplitudes of the two components of recovery are similar (Fig. 1). In addition, although mouse atrial I_K,s is sensitive to low (50 μM) concentrations of 4-AP, the current is not blocked completely by 5 mM 4-AP and the residual 4-AP-insensitive I_K,s is blocked effectively by 25 mM TEA (Fig. 2). Similar to previous findings in ventricular cells (Xu et al. 1999a), therefore, these observations suggest that there are two distinct components of I_K,s in mouse atrial cells. The molecular analysis of the currents described here also supports this conclusion.

Molecular correlates of mouse atrial I_K,s and I_ss

The experiments completed here demonstrate that both I_K,s and I_ss are attenuated in atrial myocytes isolated from (P15 and adult) transgenic mice expressing the dominant negative Kv2.1 α subunit, Kv2.1N216Flag (Xu et al. 1999a). The densities of I_K,s and I_ss are also reduced in P15 mouse atrial myocytes exposed to an AsODN targeted against the translation start site of Kv2.1. In addition, pharmacological experiments revealed that the 25 mM TEA-sensitive component of I_K,s is eliminated in Kv2.1N216Flag-expressing atrial myocytes, whereas the 50 μM 4-AP-sensitive component of I_K,s remains. Taken together, these results suggest that Kv2.1 contributes to both I_K,s and I_ss in (P15 and adult) mouse atrial myocytes and further that Kv2.1 underlies the 25 mM TEA-sensitive components of I_K,s and I_ss in these cells. The components of I_K,s and I_ss encoded by Kv2.1 could reflect two distinct gating pathways of the same population of depolarization-activated K⁺ channels. It is certainly possible, however, that the two currents reflect functionally distinct channels rather than different kinetic/gating states of the same channel, perhaps reflecting differences in post-translational processing and/or to interactions with accessory subunits. Alternative experimental approaches will be needed to explore these and other possible mechanistic interpretations directly.

The results presented here also reveal that, although I_K,s and I_ss are attenuated in (P15 and adult) Kv2.1N216Flag-expressing atrial myocytes and in wild-type P15 atrial cells exposed to the Kv2.1 AsODN, neither current is eliminated. Rather, both currents are reduced to approximately 50 % of the control I_K,s and I_ss densities (Fig. 6). These observations suggest that there are two molecularly distinct components of both I_K,s and I_ss in these cells, i.e. one component that reflects expression of Kv2.1 and another that reflects the functioning of another Kv α subunit or subunits. Consistent with this hypothesis, the Kv1.5 AsODN also attenuates I_K,s in P15 atrial myocytes and, interestingly, current densities are also reduced by ∼50 %. Analysis of the decay phases of the currents in Kv2.1N216Flag-expressing atrial cells and in wild-type cells exposed to the Kv1.5 AsODN or the Kv2.1 AsODN reveal no significant differences in inactivation kinetics. These results suggest that the rates of inactivation of the two components of mouse atrial I_K,s are indistinguishable, and further that the functional roles of these two current components in shaping action potentials will be the same. There are, however, two components of recovery of I_K,s from steady-sate inactivation and, although both are slow, one component recovers very slowly (τ_rec,slow∼10 s) and may not play a physiological role at the normal heart rate (650–700 beats min⁻¹) in the mouse. Importantly, both components of I_K,s recovery remain in the Kv2.1N216Flag-expressing atrial myocytes (n = 3) and the mean ±s.e.m. recovery time constants (252 ± 35 and 3753 ± 854 ms) are similar to those (370 ± 69 and 7566 ± 2610 ms; n = 7) determined in wild-type atrial cells (Fig. 1). These results reveal the complex gating of Kv1.5 encoded I_K,s channels.

In contrast to I_K,s, the molecular identity of the component of I_ss that remains in the Kv2.1N216Flag-expressing transgenics and in atrial myocytes exposed to the Kv2.1 AsODN is not known. Because this component is not affected in atrial myocytes isolated from transgenic mice expressing Kv4.2W362F (Xu et al. 1999c) or Kv2.1N216Flag (present study), it seems reasonable to suggest that a voltage-gated K⁺ channel α subunit (or subunits) that is not a member of the Kv4 or the Kv2 subfamily underlies this (25 mM TEA-sensitive) component of I_ss.

Relationship to previous studies

It has been reported that I_K,s in mouse ventricular myocytes is reduced or eliminated in cells isolated from transgenic animals expressing a truncated Kv1α subunit, Kv1.1N206Tag (London et al. 1998; Zhou et al. 1998). The experimental observations that mouse ventricular I_K,s is sensitive to micromolar concentrations of 4-AP (Fiset et al. 1997; London et al. 1998) and that Kv1.5 protein levels are reduced in Kv1.1N206Tag-expressing mouse ventricles (London et al. 1998) have been interpreted as suggesting a role for Kv1.5 in the generation of mouse ventricular I_K,s. The results presented here are consistent with a role for Kv1.5 in the formation of I_K,s in atrial cells. Nevertheless, it will be of interest to examine the functional consequence of Kv1.1N206 expression (London et al. 1998) on the K⁺ currents in mouse atrial myocytes.

The finding here that Kv2.1 contributes to I_K,s in mouse atrial myocytes is consistent with recent studies demonstrating that a component of I_K,s in mouse ventricular cells is also eliminated in Kv2.1N216Flag-expressing transgenic animals (Xu et al. 1999a). In contrast to the results in atrial cells presented here, however, the rate of inactivation of the component of I_K,s remaining in Kv2.1N216Flag-expressing ventricular myocytes is accelerated compared with I_K,s in wild-type cells (Xu et al. 1999a). In addition, although wild-type mouse ventricular I_K,s is blocked by 25 mM TEA (Xu et al. 1999b), I_K,s in ventricular cells isolated from Kv2.1N216Flag-expressing mice is insensitive to 25 mM TEA (Xu et al. 1999a). As noted above, a component of mouse atrial I_K,s is TEA sensitive (Fig. 2), and experiments completed here have revealed that this component of I_K,s is selectively eliminated in Kv2.1N216Flag-expressing mouse atrial myocytes (Fig. 8). In contrast, the 4-AP-sensitive currents in the Kv2.1N216-expressing cells are indistinguishable from those in wild-type cells.