INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE CALCIUM-INDEPENDENT transient outward potassium current (I_to) is an important repolarizing current in rat ventricular myocytes. The expression of this current is not uniform throughout the rat heart, and regional variations in the density, biophysical properties, and regulation of I_to exist (3, 8, 22). For example, the density of I_to in the left ventricle is higher in the epicardium compared with the endocardium whereas endocardial currents show a greater degree of rate dependence, and endocardial and epicardial currents are differentially regulated by thyroid hormone (3, 22). Consistent with heterogeneity of transient outward currents in rat heart, three different K⁺-channel genes that encode I_to-like currents (i.e., Kv1.4, Kv4.2, and Kv4.3) are known to be expressed at the mRNA level in the rat ventricle (5, 6, 16). In the left ventricle, the expression of Kv4.2 mRNA correlates with the density of I_to across the left ventricular wall (6). On the other hand, Kv4.3 and Kv1.4 mRNAs show uniform transmural expression in the left ventricle (5, 6). It is possible, therefore, that the molecular composition of I_to may vary across the left ventricular wall and that differences in the relative contribution of Kv4.2, Kv4.3, and Kv1.4 to epi- and endocardial I_to may underlie differences in the density, biophysical, and regulatory properties of this current.

In addition to variations in I_to across the left ventricular wall, differences in the density and regulatory properties of I_to may also exist between other regions of the rat heart. The density of I_to is lower in septum compared with left ventricular free wall, for example, and I_to in the free wall but not septum are reduced during the development of myocardial hypertrophy in the rat (8). The molecular basis of these differences, however, has not been addressed. In the present study, we combined electrophysiology and molecular biology to characterize I_to in the right ventricular free wall and the interventricular septum of the adult rat ventricle and to evaluate the relative contributions of Kv4.2, Kv4.3, and Kv1.4 to I_to in these regions. In addition, because it has been reported that the mRNA levels of K⁺ channels are not always predictive of protein levels (1, 28), we also used Western blot analysis to evaluate the potential contribution of K⁺-channel gene products to I_to in the right ventricular free wall and septum.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of adult rat ventricular myocytes. Adult rat ventricular myocytes were isolated as previously described (27) with minor modifications. Briefly, rats (250 g, Sprague-Dawley, Charles River) were heparinized and killed (under 75 mg/kg pentobarbital sodium) by cervical dislocation. The heart was removed and retrogradely perfused for 3 min with Tyrode solution of the following composition (mM): 132 NaCl, 5.4 KCl, 1 CaCl₂, 1.2 MgSO₄, 10 HEPES, and 10 D-glucose, pH 7.4. The hearts were then perfused with nominally calcium-free Tyrode for a further 5 min, followed by perfusion with Tyrode solution containing collagenase (type II, 0.6 mg/ml, Boehringer-Mannheim), protease (type XIV, 0.05 mg/ml, Sigma Chemical, St. Louis, MO), and CaCl₂ (25 µM). Once digestion was complete (typically 7-9 min), hearts were perfused for an additional 5 min with enzyme-free high-K⁺ solution of the following composition (mM): 120 potassium glutamate, 20 KCl, 20 HEPES, 1 MgCl₂, 10 D-glucose, and 0.5 K-EGTA. The entire right ventricular free wall (from 5 hearts) and interventricular septum (from 4 hearts) were removed under a dissecting microscope and minced in high-K⁺ solution. Cells were liberated with gentle mechanical agitation, filtered through nylon mesh, and stored in high-K⁺ solution until required. Only calcium-tolerant, quiescent, rod-shaped myocytes with clear cross striations were selected for electrophysiological recordings. All experimental protocols conform with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH) [DHHS Pub. no. (NIH) 85-23, revised 1996]. The protocols were approved by the Animal Care Committee of the Research Institute at the Toronto Hospital.

tsa-201 cell culture and transfection. tsa-201 cells were maintained in MEM supplemented with 10% fetal bovine serum and gentamicin (50 µg/ml) in an incubator at 37°C with a humidified atmosphere of 5% CO₂. Medium was replaced every 48-72 h. Twenty-four hours before transfection, tsa-201 cells were harvested by brief trypsinization (0.5 mg/ml in phosphate-buffered saline) and replated at a density of 3 × 10⁵ cells per 35-mm culture dish. Cells were transfected using Lipofectamine reagent (GIBCO BRL) according to the manufacturer's instructions. Cells were incubated for a period of 5 h with a mixture of 10 µl Lipofectamine, 1 µg green fluorescent protein (GFP), and 0.5-1 µg pRcCMVKv4.2, 1 µg pcDNA3Kv4.3, 0.03-1 µg pGW1HKv1.4, or a similar amount of vector alone in OPTI-MEM serum-free media. Twenty-four to forty-eight hours posttransfection, cells were prepared for electrophysiological evaluation. Cells were removed from the culture dish by brief trypsinization (as described in Isolation of adult rat ventricular myocytes), collected by centrifugation (1,000 rpm, 5 min), and replated in growth medium at low density. All reagents for cell culture were purchased from GIBCO BRL.

Electrophysiological recording. Adult right ventricular myocytes were placed in a bath and perfused (~1 ml/min) with extracellular Tyrode solution of the following composition (mM): 140 NaCl, 2 CaCl₂, 4 KCl, 1 MgCl₂ · 6H₂O, 10 glucose, 10 HEPES, and 0.5 CdCl₂ (for myocyte recordings only), pH 7.4 with NaOH. Electrophysiological recordings from tsa-201 cells were made with the cells adhered to the 35-mm culture dishes in which they were plated. Culture medium was replaced with extracellular solution immediately before recordings. Pipette tips were heat polished to a resistance of 1-3 M when filled with an intracellular solution of the following composition (mM): 140 KCl, 1 MgCl₂ · 6H₂O, 10 EGTA, 10 HEPES, and 5 MgATP, pH 7.2-7.3 with KOH. In some experiments with cultured cells a KF-based pipette solution was used to improve seal stability. The composition (mM) of the modified Tyrode (KF) solution was 100 KF, 40 KCl, 5 NaCl, 2 MgCl₂, 10 HEPES, 5 EGTA, and 5 glucose, pH 7.2-7.3 with KOH. Recovery kinetics were similar with KCl- and KF-based intracellular solutions. All recordings were made at room temperature (22-24°C) within 12 h of cell isolation (myocytes) or replating (tsa-201 cells). Successfully transfected tsa-201 cells were identified by their green fluorescence under appropriate conditions.

Preparation of total RNA. Hearts were removed rapidly, and the right ventricle and septum were isolated, rinsed briefly in 0.9% NaCl (wt/vol), and snap frozen in liquid nitrogen. Ventricular tissue was homogenized in Trizol reagent (GIBCO) and RNA precipitated with isopropyl alcohol. The integrity of RNA samples was confirmed by the presence of sharp bands after brief electrophoresis through a 1% agarose gel. The concentration of RNA was measured spectrophotometrically and confirmed by agarose gel electrophoresis. RNA was isolated independently from four adult rat hearts.

RNase protection assays. RNase protection assays were performed using an RPAII Ribonuclease Protection Assay Kit (Ambion, Austin, TX) as previously described (27). The Kv4.2 and Kv4.3 probes were kindly provided by Dr. David McKinnon (State University of New York at Stony Brook) and have been described previously (5). A 429-bp fragment of rat Kv1.4 (kindly provided by Dr. David McKinnon) was subcloned into pGEM11 (Hind III-Nsi I) to make a Kv1.4 probe capable of protecting a 331-bp fragment of Kv1.4 mRNA. The cyclophilin probe was purchased from Ambion. Abundance of mRNA transcripts was quantified by densitometry (Bio-Rad GS670 Imaging densitometer). Signals were normalized to a cyclophilin internal standard to ensure that findings were not influenced by minor variations in loading. Absolute cyclophilin levels (densitometric units) were not significantly different between right ventricle and septum in the present study (right ventricular cyclophilin levels were 129 ± 20% of septal levels; n = 24; P > 0.05, two-tailed, paired t-test), indicating that this gene was expressed uniformly between these regions. Right ventricular wall mRNA levels for each rat were normalized to mRNA levels in the septum of the same animal.

Isolation of protein from right ventricular wall and septum. Hearts were removed rapidly, and the right ventricle and septum were isolated as described in Isolation of adult rat ventricular myocytes, rinsed briefly, and homogenized in buffer A [0.32 M sucrose, 5 mM Tris (pH 7.4) containing protease inhibitors phenylmethylsulfonyl fluoride (100 µM), o-phenanthroline (1 mM), iodoacetamide (1 mM), and benzamidine (1 mM)]. Homogenate was centrifuged to remove particulate matter. Membranes were collected from the supernatant by centrifugation at 27,000 g for 45 min and resuspended in buffer C (0.32 M sucrose, 5 mM HEPES solution containing protease inhibitors as described above). Membrane protein was isolated independently from three to five adult rat hearts. Protein concentration was determined by the Lowry method with minor modifications (11), and the membranes were aliquoted and stored at 70°C until further use. Rat brain membranes were prepared as described previously (9). After protein determination, rat brain membranes were aliquoted and frozen in liquid nitrogen.

Western blot analysis. For Western blot analysis, 10- to 50-µg (heart) or 3.5- to 10-µg (brain) aliquots of freshly thawed membrane protein were tritrated in 2× SDS sample buffer containing -mercaptoethanol, heated for 5 min at 95°C, and centrifuged to pellet any insoluble debris. Supernatant proteins were then resolved by electrophoresis on 7.5% SDS-polyacrylamide gels and transferred electrophoretically to nitrocellulose (0.45 µm Kv1.4) or polyvinylidene difluoride (Kv4.2). The blots were then blocked in Tris-buffered saline (TBS) containing 5% nonfat dried milk, washed with TBS, and probed with anti-Kv antibody raised in rabbit (Dr. O. T. Jones, Department of Pharmacology, University of Toronto) at a 1:1,000 dilution in blocking solution. After consecutive washes with TBS containing 0.05% Tween-20 and TBS alone, the blots were incubated for 2 h with secondary antibody [horseradish peroxidase-conjugated donkey anti-rabbit (Amersham), 1:4,000 in blocking solution], rewashed, and developed by enhanced chemiluminescence (ECL, Amersham). Gel loading was checked by staining total proteins with Ponceau S, and molecular masses were determined using prestained markers (Kaleidoscope, Bio-Rad). Densitometric analysis of films was done with a Bio-Rad model GS-670 imaging densitometer.

Plasmid constructs. The long splice variant of human Kv4.3 (10) was amplified from a hippocampus cDNA library (Clontech) using overlap extension. First, three overlapping pieces covering the entire Kv4.3 coding sequence were amplified using three primer pairs: 5'-TCTCAAGCTTCCACCATGGCGGCCGGAGTTGCGGCCTGGCT-3' and 5'-CGAGGGCATCGATTCCTGGTTGTTCTCCGAGTCGTTG-3'; 5'-CACCAGGAATCGATGCCCTCGCTCAGCTTCCGCCAGAC-3' and 5'-GCAGATGGAGCCGAAGATCTTCCCTGC-3'; and 5'-GCAGGGAAGATCTTCGGCTCCATCTGC-3' and 5'-TAGCTCTAGATTACAAGACAGAGAGACCTTGACAACATTGC-3'. Overlap extension was performed using 20 ng of each amplified fragment, using the first and last primers. The overlap product was subcloned into pcDNA3 (Invitrogen), and the insert was confirmed by sequencing.

Statistics. All data are expressed as means ± SE. Statistical significance was determined using a suitable (hetero- or homoscedastic) unpaired two-tailed t-test. For RNase protection assays and Western blot analysis, the null hypothesis (i.e, that mRNA/protein levels were not different in septum and right wall) was tested using paired two-tailed t-tests. A P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

I_to in right ventricular free wall and septum. I_to density (at 60 mV) was 29.5 ± 3.0 (n = 20 cells from 5 hearts) and 15.9 ± 3.0 (n = 19 cells from 4 hearts) pA/pF in myocytes isolated from the right ventricular free wall (cell capacitance 139.3 ± 8.0 pF) and the interventricular septum (cell capacitance 167.0 ± 7.2 pF), respectively. When comparing the biophysical properties of these currents, we focused exclusively on the kinetics of recovery from inactivation because the rate of recovery from inactivation has previously been shown to vary between different regions of the left ventricle and at different developmental stages (22, 27). Furthermore, dramatically different rates of recovery from inactivation have been observed between Kv1.4 and the Shal-related K⁺ channels Kv4.2 and Kv4.3 in Xenopus oocytes (6, 14, 19, 20, 25), thereby potentially providing an electrophysiological fingerprint for the different channel types. Recovery from inactivation was measured using the double-pulse protocol shown in the inset to Fig. 1C and described in METHODS. In these studies we used a very low stimulation frequency (0.05 Hz) to ensure sufficient time for all currents to completely recover (see below). In both the right ventricular free wall and interventricular septum the recovery from inactivation of I_to was dominated by a rapid component. Complete recovery from inactivation, however, frequently required very long interpulse intervals, resulting in a biphasic time course of recovery. The slowly recovering component of I_to was observed in less than one-half of the cells isolated from the right ventricular free wall (7/20) and in virtually all cells isolated from the septum (17/19). The biphasic nature of recovery from inactivation for I_to is clearly visible from the voltage-clamp recordings displayed in Fig. 1 for myocytes from the right wall (Fig. 1A) and septum (Fig. 1B). For both cells I_to recovered rapidly to ~80% of the control level, but complete recovery required prolonged interpulse intervals. The plots of normalized current against interpulse interval for the same cells are shown in Fig. 1, C (right wall) and D (septum). These figures show that the rapid phase of recovery was complete within 200 ms but that full recovery required an interpulse interval of 20 s. Consistent with the presence of a slowly recovering component of I_to, the amplitude of I_to was decreased in some cells during repetitive, high-frequency stimulation. This is shown in Fig. 2, C and D, in which voltage-clamp recordings from a septal cell show that the amplitude of I_to was reduced by 20% during repetitive stimulation with 500-ms depolarizing pulses at a rate of 1 Hz.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Biphasic recovery from inactivation of transient outward potassium current (Ito) in myocytes isolated from adult rat heart. Recovery from inactivation was measured using double-pulse protocol shown in inset in C. Typical voltage-clamp recordings showing recovery from inactivation of Ito in a right ventricular free cell (A) and a septal cell (B), with increasing interpulse intervals (10, 20, 40, 100, 200, and 600 ms, 1 , 2, and 20 s), are shown. Note that although recovery is dominated by a rapid component, complete recovery required prolonged interpulse intervals in both cells. Ito (peak current  current remaining at end of 500-ms pulse) elicited by second pulse was normalized to that elicited by first pulse and plotted against interpulse interval in C (right wall) and D (septum). Recoveries were best fit with a biexponential function, with fast components accounting for 80.2% of the recovery in right wall and 85.4% in septum [time constant for recovery of fast component (fast) = 26.7 and 49 ms for right wall and septal cells, respectively], with a slow component [time constant for recovery of slow component (slow) = 3,478 and 12,772 ms for right wall and septal cells, respectively] accounting for the remainder.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Reduction of Ito amplitude during 1-Hz stimulation in a myocyte isolated from interventricular septum. Superimposed voltage-clamp recordings shown in A are first 5 sweeps recorded from a septal cell stimulated at a frequency of 1 Hz after a rest period of 30 s (sweep 1 is top trace). Ito amplitude (peak  sustained) was normalized to Ito amplitude on sweep 1 and plotted against sweep number in B. This figure shows that Ito amplitude was abruptly reduced between sweep 1 and sweep 2.

Recovery from inactivation of Kv4.2-, Kv4.3-, and Kv1.4-based currents in a mammalian cell line. To understand the molecular basis of the rapidly and slowly recovering components of I_to in the right wall and septum, we measured the recovery kinetics of candidate K⁺-channel gene products (Kv4.2, Kv4.3, and Kv1.4) after transient transfection of mammalian tsa-201 cells. Cells transfected with Kv4.2 (Fig. 3B), Kv4.3 (Fig. 3C), or Kv1.4 (Fig. 3D) expressed robust, transient outward-like currents. Similar currents were never recorded from nontransfected cells or cells transfected with vector DNA only (although these cells do express a small endogenous delayed rectifier-like current; Fig. 3A). Figure 3, E and G, shows typical voltage-clamp recordings of recovery from inactivation of Kv4.3- and Kv4.2-based currents, respectively. As shown, recovery from inactivation for both these gene products was relatively rapid, with the normalized current versus interpulse interval being fit with monoexponential functions with time constants of 84 (Kv4.3, Fig. 3F) and 51.4 (Kv4.2, Fig. 3H) ms. Similar observations were made in a total of six cells per group. On average, recovery from inactivation was slower for Kv4.3 (mean ± SE  value was 141 ± 23 ms) compared with Kv4.2 ( value was 73.1 ± 9.5 ms; P = 0.02). The recovery kinetics of Kv4.3 and Kv4.2 based currents expressed in tsa-201 cells resembled the rapid component of I_to measured in myocytes. These findings suggest that Kv4.3 and/or Kv4.2 may underlie the rapidly recovering component of I_to in these regions.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Recovery from inactivation of Kv4.3, Kv4.2, and Kv1.4 expressed in tsa-201 cells. Current recordings showing families of outward currents recorded from tsa-201 cells transfected with green fluorescent protein alone, Kv4.2, Kv4.3, or Kv1.4 are shown in A-D, respectively (for clarity only currents elicited after steps to 40, 20, 0, 20, 40, and 60 mV are shown). Recovery from inactivation was measured using double-pulse protocol shown in insets in F, H, and J. Typical voltage-clamp recordings showing recovery of a Kv4.3- or Kv4.2-based current with increasing interpulse interval (20-ms steps starting at 20 ms) are shown in E and G, respectively. Kv4.2/Kv4.3 currents (defined as peak current  current remaining at end of 500-ms pulse) elicited by second pulse were normalized to those elicited by first pulse and plotted against interpulse interval (Kv4.2, F; Kv4.3, H). In both cases, recovery was well fit with a single exponential function with  of 84 (Kv4.3) and 51.4 (Kv4.2) ms. Typical voltage-clamp recordings showing recovery of a Kv1.4-based current with increasing interpulse interval (from 200 ms to 5 s) are shown in C. Recovery of this Kv1.4-based current was well fit with a single exponential with a  of 1,784 ms (D).

Kv4.2, Kv4.3, and Kv1.4 mRNA levels in right ventricular free wall and interventricular septum. To gain further insight into the molecular basis of I_to in the right ventricular free wall and septum, we measured mRNA levels of Kv4.2, Kv4.3, and Kv1.4 in these regions. Figure 4, A, C, and E, shows representative gels from RNase protection assays showing Kv4.2, Kv4.3, and Kv1.4 mRNA levels in the right ventricular free wall, the interventricular septum, and brain. Robust transcriptional expression of Kv4.2, Kv4.3, and Kv1.4 was observed in both the right ventricular free wall and septum. Transcript levels (normalized to the levels of cyclophilin mRNA and to respective mRNA levels in the septum) are plotted in Fig. 4, B, D, and F. The mRNA levels of Kv4.2, Kv4.3, and Kv1.4 tended to be higher in the right ventricular free wall compared with the septum (right ventricular mRNA levels were 248 ± 46, 113 ± 9, and 133 ± 26% of those in the septum for Kv4.2, Kv4.3, and Kv1.4, respectively; n = 4/group). However, only the Kv4.2 levels were significantly different between right ventricle and septum (P = 0.049). The higher level of expression of Kv4.2 mRNA in the right wall compared with the septum correlated well with the twofold higher density of the rapidly recovering component of I_to in the right wall compared with the septum. Furthermore, the observation that Kv1.4 mRNA was readily detected supports the suggestion that the product of this gene may underlie the slowly recovering component of I_to.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Kv4.2, Kv4.3, and Kv1.4 mRNA levels in right ventricular free wall and septum measured using an RNase protection assay. Gels show Kv4.2 (A)-, Kv4.3 (C)-, and Kv1.4 (E)-protected mRNA fragments, along with control cyclophilin mRNA levels for samples containing 10 µg of total RNA isolated from rat brain (Br), 4 right ventricular free walls (RV1-4), and 4 septums (S1-4). mRNA levels were quantified by densitometry and normalized to cyclophilin mRNA levels in same sample. mRNA level in right ventricle was further normalized to septum level in same heart, and mean values are plotted in bar graphs for Kv4.2 (B), Kv4.3 (D), and Kv1.4 (F) in right ventricular wall (solid bars) and septum (open bars). Bars represent means of 4 determinations, and vertical lines represent SE.

Kv1.4 and Kv4.2 protein levels in right ventricular free wall and interventricular septum. Because mRNA levels may not always be predictive of expression at the protein level (28), we also conducted Western blot analyses using specific anti-Kv4.2 and anti-Kv1.4 antibodies. Figure 5A shows that Kv1.4 was detectable in samples isolated from the right ventricular free wall and septum, albeit at considerably lower levels than in brain. In the present study, the anti-Kv1.4 antibody labeled two distinct bands, in both brain (seen with reduced exposures; Fig. 5A, inset) and myocyte protein. The molecular masses of these two bands were ~91.5 and 100.5 kDa, which are very similar to those previously identified in protein extracted from cos-1 cells transfected with Kv1.4 (18) and cultured neonatal rat ventricular myocytes (27). The basis for the presence of two bands is unclear, but studies have established that Shaker, Kv1.3, and Kv1.1 channels have glycosylated and nonglycosylated fractions leading to two distinct bands (4, 18, 23). Both bands could be eliminated by preincubation of the antibody with the peptide against which the antibody was raised (Kv1.4N 12 µg/µl; Fig. 5B). Taken together, these results suggest that both bands likely represent Kv1.4-related proteins. Total Kv1.4 immunoreactivity in the right ventricular free wall, although somewhat greater, was not statistically (P = 0.06) different from the septum in five hearts studied.

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Kv1.4 and Kv4.2 protein levels in right ventricular free wall and septum. Representative gels from Western blots showing Kv1.4 (A) and Kv4.2 (D) levels in protein samples (see METHODS) obtained from adult rat brain (Br, 10 µg), adult rat right ventricular wall (R1-R3, 50 µg), and adult rat interventricular septum (S1-S3, 50 µg). Anti-Kv1.4 antibody labeled 2 distinct bands in myocyte protein with molecular masses of ~91.5 and 100.5 kDa. As shown in inset in A, 2 bands were also seen in rat brain samples. Because of higher expression of Kv1.4 in brain samples, it was difficult to resolve these 2 bands under conditions required to visualize cardiac proteins (right lane of inset). However, two bands were evident with reduced exposure of film (left lane of inset). Blots of protein from rat heart probed with anti-Kv1.4 or anti-Kv4.2 antibodies after preincubation with peptide are shown in B and E, respectively. Absence of bands in these blots, despite prolonged exposures, indicates that bands seen in A and D were specific. Protein levels were quantified by densitometry and normalized to levels in septum of same heart. Relative expression levels are plotted as bar charts for Kv1.4 (C) and Kv4.2 (F). Solid bars represent relative expression in right ventricular free wall; open bars show relative expression in septum. Each bar is mean ± SE from 3-5 hearts.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Differences in the density and regulatory properties of I_to exist between anatomically distinct regions of the rat heart (3, 8, 22). The purpose of the present study was to characterize and compare I_to in the right ventricular free wall and the interventricular septum of the adult rat ventricle and to evaluate the possible relative contributions of Kv4.2, Kv4.3, and Kv1.4 to I_to in these regions. In the present study, we found that the density of I_to was significantly greater in the right ventricular free wall compared with the septum. Recovery from inactivation (using slow stimulation frequencies) revealed two kinetically distinct components to I_to in both regions. Recovery from inactivation of I_to was dominated by a rapidly recovering component in both the right ventricular free wall and the interventricular septum. The kinetics of this rapid component were similar to the kinetics of recovery of Kv4.2 expressed in mammalian cells (Refs. 7 and 29 and present study) and to Kv4.3 expressed in mammalian cells (present study) and in Xenopus oocytes (6, 20), suggesting that these Shal-related K⁺-channel genes, either as homo- or heterotetramers, contribute importantly to the rapidly recovering component of I_to in the right wall and septum. Consistent with this, we found that Kv4.2 and Kv4.3 mRNAs and Kv4.2 protein were robustly expressed in both the right ventricle and the septum. The finding that Kv4.2 mRNA and protein levels were significantly higher in the right wall compared with the septum suggests that differences in Kv4.2 expression account for the difference in the density of the rapidly recovering component of I_to between these regions. The observation that Kv4.3 mRNA levels were similar in these regions suggests that Kv4.3 may be proportionately more important in the septum compared with the right wall. Because we found that recovery from inactivation for Kv4.3 is marginally slower than that for Kv4.2, it is possible that differences in the quantities of Kv4.2 and Kv4.3 in different regions of the adult rat heart could explain the finding that the recovery kinetics of the fast component differed between the right wall and the septum, as suggested previously for differences observed between the endo- and epicardium (6).

In addition to the dominant, rapidly recovering component of I_to, a smaller, slowly recovering component of I_to was also evident in a portion of right wall cells and the septum in the present study. This current appeared proportionately more important in the septum, in which on average it accounted for ~20% of I_to. The kinetics of recovery of this current was remarkably similar to that recorded from Kv1.4 expressed in the mammalian cell line. Moreover, the absolute current density of the slow component was similar between regions, which coincided with the absence of significant differences in the Kv1.4 protein and mRNA levels. Although it remains possible that the slowly recovering component of I_to may represent some other (i.e., non-Kv1.4) channel and that the source of the Kv1.4-like protein may be nerve or vascular smooth muscle (i.e., noncardiac), the simplest interpretation of our findings is that Kv1.4 underlies the slowly recovering component of I_to in the right wall and septum of the adult rat heart, as has been previously suggested for the slowly recovering I_to recorded in rabbit ventricle (26). These findings are also consistent with previous studies in human and rat myocardium showing that a slowly recovering I_to is expressed predominantly in cells from the endocardial layer of the left ventricular free wall (12, 22) and that Kv1.4 protein is preferentially expressed in the endocardial layer of the ferret left ventricle (2). In addition, the slowly recovering component of I_to was only detected in a portion of right wall cells. It is possible that the right wall cells displaying the slow component originate from the endocardium portion of the right wall. Further studies are required to address this possibility. Our findings differed from previous studies that failed to detect Kv1.4 protein in adult rat heart (1, 28). It is possible that this discrepancy may be explained by the use of different strains of rats, differences in the methods of protein isolation, and/or by the use of different anti-Kv1.4 antibodies.

Although Kv1.4-based currents seem to contribute to I_to under the conditions of the present study (i.e., long interpulse intervals), the functional contribution of such slowly recovering channels under more physiological conditions is unclear. Indeed, it has been suggested that at normal heart rates these currents would be permanently inactivated (15). In fact, however, the behavior of these currents under physiological conditions is very difficult to predict. Temperature, action potential amplitude and/or duration, extracellular K⁺, redox, and posttranslational modification could all impact on the degree of inactivation and/or recovery from inactivation and potentially render these currents available at physiological heart rates. Interestingly in this regard, one recent report has shown that Ca/calmodulin-dependent protein kinase II-mediated phosphorylation of the NH₂-terminus of Kv1.4 slows the rate of inactivation and accelerates the recovery kinetics of Kv1.4-based currents (17).

In the present study we have sought to correlate gene expression (mRNA and protein) with function (electrophysiology) to identify the molecular correlates of I_to in the right wall and the septum of the rat heart. The RNase protection assay is a powerful technique for the identification of rare messages such as ion channels. However, this technique can only provide information on the relative expression of a given transcript between regions and does not allow for the measurement and comparison of absolute copy number of multiple molecular species within a region. Further studies using alternative molecular techniques (such as quantitative PCR) would be helpful in this regard. Correlations between gene expression and function, although persuasive, also require support using alternative techniques. We are currently examining the possibility of using recombinant adenoviruses to deliver dominant negative constructs into cardiac muscle before cell isolation as an alternative strategy for determining the molecular nature of native currents in the rat heart.

In summary, the results of the present study show that I_to is composed of both rapidly and slowly recovering components in the right wall and septum of the rat ventricle. Both Kv4.2 and Kv4.3 probably contribute to the rapid component, and Kv1.4 appears to underlie the slow component of I_to. Kv4.2 expression predominates in the right wall, whereas Kv4.3 and Kv1.4 may be proportionately more important in the septum. Such regional differences in the contribution of Kv4.2, Kv4.3, and Kv1.4 may account for regional differences in the properties and regulation of I_to.