Feedback Remodeling of Cardiac Potassium Current Expression
A Novel Potential Mechanism for Control of Repolarization Reserve
Background— Inhibition of individual K+ currents causes functionally based compensatory increases in other K+ currents that minimize changes in action potential duration, a phenomenon known as repolarization reserve. The possibility that sustained K+ channel inhibition may induce remodeling of ion current expression has not been tested. Accordingly, we assessed the effects of sustained inhibition of one K+ current on various other cardiac ionic currents.
Methods and Results— Adult canine left ventricular cardiomyocytes were incubated in primary culture and paced at a physiological rate (1 Hz) for 24 hours in the presence or absence of the highly selective rapid delayed-rectifier K+ current (IKr) blocker dofetilide (5 nmol/L). Sustained dofetilide exposure led to shortened action potential duration and increased repolarization reserve (manifested as a reduced action potential duration–prolonging response to IKr blockade). These repolarization changes were accompanied by increased slow delayed-rectifier (IKs) density, whereas IKr, transient-outward (Ito), inward-rectifier (IK1), L-type Ca2+ (ICaL), and late Na+ current remained unchanged. The mRNA expression corresponding to KvLQT1 and minK (real-time polymerase chain reaction) was unchanged, but their protein expression (Western blot) was increased, suggesting posttranscriptional regulation. To analyze possible mechanisms, we quantified the muscle-specific microRNA subtypes miR-133a and miR-133b, which can posttranscriptionally regulate and repress KvLQT1 protein expression without affecting mRNA expression. The expression levels of miR-133a and miR-133b were significantly decreased in cells cultured in dofetilide compared with control, possibly accounting for KvLQT1 protein upregulation.
Conclusions— Sustained reductions in IKr may lead to compensatory upregulation of IKs through posttranscriptional upregulation of underlying subunits, likely mediated (at least partly) by microRNA changes. These results suggest that feedback control of ion channel expression may influence repolarization reserve.
Received December 5, 2007; accepted June 11, 2008.
Cardiac repolarization is a key cellular function. Disruption of cardiac repolarization leads to potentially lethal ventricular tachyarrhythmias.1,2 The repolarization process is governed by the interplay of multiple ion channels. An important recently developed concept is the notion that the complex of multiple repolarizing ion channels provides repolarization reserve in that dysfunction or inhibition of a K+ current causes functionally based compensatory increases in other K+ currents that minimize changes in action potential (AP) duration (APD).1–3 Impairments in repolarization reserve predispose the heart to lethal ventricular arrhythmias.1,2 The basic notion of repolarization reserve is that a delay in the repolarization process caused by impaired function of one K+ channel causes greater current-carrying capacity of other K+ channels, generally by enhancing voltage and time-dependent activation. This functionally based repolarization reserve is most typically seen for rapid and slow delayed-rectifier K+ currents (IKr and IKs), with IKs preventing excessive APD prolongation in response to IKr reduction.4–6
Editorial p 981
Clinical Perspective p 992
To the best of our knowledge, the possibility that APD prolongation by sustained inhibition of a K+ current can elicit increased repolarization reserve by affecting cardiac ion channel expression has not been addressed. Cardiac ion channel expression is a regulated function, although little is known about the potential feedback mechanisms and their functional consequences.7 In transgenic mice engineered to lack the rapid transient-outward K+ current (Itof), evidence exists for compensatory upregulation of the slow transient-outward component (Itos) carried by Kv1.4 subunits.8,9 Similarly, in mice lacking IK,slow1 carried by Kv1 subunits, an apparent compensatory regulatory response of Kv2.1 subunits to increase IK,slow2 can be observed.10 The present study was designed to assess the effects of APD prolongation by highly selective IKr inhibition with dofetilide on cardiac repolarization and repolarization reserve and on the expression of other cardiac ionic currents. To allow for well-controlled conditions, we used an in vitro model of paced canine ventricular cardiomyocytes.
Additional details are available in the Materials and Methods section in the online-only Data Supplement.
In Vitro Cellular Pacing Model
All animal care and handling procedures were approved by the animal research ethics committee of the Montreal Heart Institute. Male adult mongrel dogs (21 to 35 kg) were anesthetized with pentobarbital (30 mg/kg IV) under artificial ventilation. Hearts were excised via left lateral thoracotomy and immersed in oxygenated Tyrode’s solution at room temperature. The transmural free wall (≈30×50 mm) of the anterior left ventricle was dissected, and the perfusing coronary artery was cannulated. Cell isolation was performed as previously described.11 Epicardial cells were removed and kept in culture medium (for contents, see the online-only Data Supplement) for further studies.
Cell Culture and Pacing
Cell culture and pacing were performed under aseptic conditions as previously described.12 The cell suspension was centrifuged at 500 rpm (1 minute at 4°C), and cell pellets were resuspended in culture medium. Cells were plated at ≈1×104 cells/cm2 on rectangular glass coverslips or 4-well rectangular Petri dishes precoated with laminin (15 μg/mL). Cardiomyocytes were maintained at 37°C in a humidified, 5% CO2–enriched environment. After 4 hours, any dead or unattached myocytes were washed off to leave a homogeneous layer of adherent rod-shaped cells. Cells were divided into 2 groups: control (medium without dofetilide) and dofetilide (culture media containing 5 nmol/L dofetilide). Both control and dofetilide cells were continuously paced (pulse duration, 5 milliseconds; pulse voltage, 43 V) at 1 Hz for 24 hours in the incubator. Capture efficiency, verified by eye and by video-recording of cell shortening, was ≈100%. After 24 hours, cells on coverslips were kept in high-K+ storage solution at 4°C for electrophysiological studies. Cells were scraped off culture plates and centrifuged at 1000 rpm for 5 minutes at 4°C. Cell pellets were fast-frozen in liquid N2 and kept at −80°C for subsequent biochemical analyses.
Whole-cell patch clamp was applied for ionic current and AP recording at 36±0.5°C. Ionic currents were recorded with tight-seal patch clamp in voltage clamp mode, and APs were recorded in current clamp mode with perforated patch techniques. Borosilicate-glass electrodes had tip resistances between 2 and 5 MΩ. Cell capacitance and series resistance were compensated by ≈80% to 90%. Junction potentials (≈10 mV) were corrected for AP recordings only. Cell capacitance averaged 160.6±5.7 pF in control (n=37) and 169.1±6.0 pF in dofetilide (n=44; P=NS) cells. Currents are expressed as current density (normalized to cell capacitance).
Standard Tyrode’s solution contained (mmol/L) NaCl 136, KCl 5.4, MgCl2 1, CaCl2 1, NaH2PO4 0.33, HEPES 5, and dextrose 10 (pH 7.35 with NaOH). The high-K+ storage solution contained (mmol/L) KCl 20, KH2PO4 10, dextrose 10, mannitol 40, L-glutamic acid 70, β-OH-butyric acid 10, taurine 20, and EGTA 10 and 0.1% BSA (pH 7.3, KOH). Standard pipette solution contained (mmol/L) K-aspartate 110, KCl 20, MgCl2 1, MgATP 5, GTP 0.1, HEPES 10, Na-phosphocreatine 5, and EGTA 5 (for current recording) or 0.025 (for AP recording) (pH 7.3, KOH).
For AP recording, nystatin (60 μg/mL) was back-filled into the pipette tip, and external solutions contained 2 mmol/L CaCl2. For all K+ current recordings, atropine (1 μmol/L) and either CdCl2 (200 μmol/L) or nimodipine (5 μmol/L for IKr and IKs) were added to external solutions to eliminate muscarinic K+ currents and block Ca2+ currents. For K+ currents other than transient-outward current (Ito), 1 mmol/L 4-aminopyridine was added. Inward-rectifier current (IK1) was studied as 1 mmol/L Ba2+-sensitive current. HMR1566 (1 μmol/L) was added to inhibit IKs for IKr recording, and E4031 (5 μmol/L) was used to inhibit IKr for IKs recording.
For L-type calcium current (ICa,L) studies, external solutions contained (mmol/L) tetraethylammonium chloride 136, CsCl 5.4, CaCl2 2, MgCl2 0.8, HEPES 10, and dextrose 10 (pH 7.4, CsOH). Niflumic acid (50 μmol/L) was added to inhibit ICl,Ca. The pipette solution contained (mmol/L) CsCl 20, Cs-aspartate 110, MgCl2 1, MgATP 5, GTP 0.1, Na2-phosphocreatine 5, EGTA 10, and HEPES 10 (pH 7.2, CsOH).
Late Na+ (late INa) was recorded as described by Maltsev et al.13 The pipette solution contained (mmol/L) NaCl 5, CsCl 133, MgATP 2, tetraethylammonium-chloride 20, EGTA 10, and HEPES 5 (pH 7.3, CsOH). The bath solution contained (mmol/L) NaCl 140, CsCl 5.4, CaCl2 1.8, MgCl2 2, nifedipine 0.002, and HEPES 5 (pH 7.3, NaOH). Late INa (current 200 to 220 ms after depolarization onset) was recorded 5 minutes after cell rupture at room temperature, with 2-second voltage steps (0.2 Hz) from −120 mV to test potentials between −70 and 40 mV.
Real-Time Polymerase Chain Reaction
Cell samples were homogenized in TRIzol (Invitrogen, Carlsbad, Calif) and RNA extracted with chloroform and isopropanol precipitation. Genomic DNA was eliminated with DNase I incubation (0.1 U/μL, 37°C, 30 minutes), followed by phenol-chloroform acid extraction and gel verification. RNA was quantified spectrophotometrically at 260 nm. RNA samples were stored in diethyl pyrocarbonate H2O at −80°C. First-strand cDNA was synthesized by reverse transcription with 2-μg RNA samples, random primers, and Moloney murine leukemia virus reverse transcriptase. Real-time polymerase chain reaction (PCR) was conducted with TaqMan (KCNH2 and KCNE2 transcripts) or SYBR green (KCNQ1 and KCNE1 transcripts). Canine β2-microglobulin (TaqMan) and 18S rRNA (SYBR green) were used as internal controls. Primers and probes for real-time PCR are listed in Table I of the online-only Data Supplement. Each sample was run in duplicate, and PCR products were verified with gel electrophoresis or dissociation curves (SYBR green). KCNH2 and KCNE2 results were normalized to β2-microglobulin; KCNQ1 and KCNE1 were normalized to 18S rRNA. Results were obtained from the same samples at the same time.
Protein Extraction and Western Blot
Membrane proteins were isolated with extraction buffer (for contents, see the online-only Data Supplement), followed by homogenization. After initial centrifugation (1000 rpm, 4°C, 5 minutes), supernatants containing cell membranes were centrifuged (100 000g for 1 hour). Pellets were resuspended in extraction buffer and 1% Triton and stored at −80°C. Protein concentration was determined by Bradford assay. Membrane protein samples (40 μg) were denatured with Laemmli buffer, fractionated on 8% or 12% SDS-PAGE, and then transferred electrophoretically to Immobilon-P polyvinylidene fluoride membranes in 25 mmol/L Tris base, 192 mmol/L glycine, and 20% ethanol (0.3 A for 1 hour). Membranes were blocked in Tris-buffered saline with 0.1% Tween 20 and 5% nonfat dry milk and incubated with primary antibodies (anti-KvLQT1, 1:500, Alomone Labs, Jerusalem, Israel; anti-minK, 1:2000, from Dr Jacques Barhanin) overnight at 4°C. After washing and reblocking, membranes were incubated with horseradish peroxidase–conjugated goat anti-rabbit IgG secondary antibody (1:10 000, Jackson Immunolabs, West Grove, Pa). Antibody was detected with Western Lightening Chemiluminescence Reagent Plus. Membranes were also probed with anti-GAPDH for protein loading.
Rapid Amplification of cDNA Ends
The Ambion RNA ligase-mediated 3′ rapid amplification of cDNA ends (RACE) kit was used to determine 3′ untranslated region (UTR) sequences of canine KCNE1 and KCNQ1. Gene-specific primers (for sequences, see the online-only Data Supplement) were designed on the basis of canine KCNE1 (GenBank No. XM_544868) and KCNQ1 (GenBank No. XM_540790) cDNAs.
The TaqMan MicroRNA Assay (Applied Biosystems, Foster City, Calif) was used in conjunction with real-time PCR with TaqMan probes for quantification of miR-133 (miR-133a and miR-133b) and miR-1 transcripts. RNA samples were isolated from cultured cells with mirVana miRNA isolation kits. Reactions contained TaqMan MicroRNA Assay sets specific for miR-133a, miR-133b, or RNU24 as a positive endogenous control. Quantitative reverse-transcriptase PCR was performed on a GeneAmp 5700 thermocycler (40 cycles) after first determining the appropriate cycle threshold with the baseline determination feature. Fold variations in expression between samples were calculated after normalization to RNU24.
Data Acquisition and Analysis
Voltage protocols were applied at 0.1 Hz unless stated otherwise. Clampfit 6.0 (Axon Instruments, Union City, Calif), GraphPad Prism 3.0, and Quantity One (Bio-Rad Laboratories, Hercules, Calif) were used for data analysis. Nonlinear least-square curve-fitting algorithms were performed for curve fitting. Paired or unpaired Student’s t tests were applied for single comparisons between groups. Comparisons involving repeated-measures analyses were performed by ANOVA with Bonferroni-adjusted t tests in the case of statistically significant intergroup differences. Two-tailed values of P<0.05 were taken to indicate statistical significance. Group data are expressed as mean±SEM.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Influence of Dofetilide Exposure During Culture on APD
We selected a dofetilide concentration for cardiomyocyte incubation (5 nmol/L) on the order of the EC50 of the drug14,15 to ensure significant IKr inhibition and APD prolongation while facilitating dofetilide washout at the end of the culture period. To confirm effective APD prolongation by this dofetilide concentration, we studied its effects on fresh cells and noted an 18±4% prolongation at 1 Hz (online-only Data Supplement Figure I).
Original AP recordings obtained from control and dofetilide cells after 24 hours in culture are shown in Figure 1A. Resting membrane potential averaged −75.4±1.0 mV in control (n=18) and −77.0±0.9 mV in dofetilide (n=14; P=NS) cells. Figure 1B shows mean APD at 90% repolarization (APD90) at different frequencies (0.1, 0.5, 1, and 2 Hz). Cells exposed to dofetilide long term showed significantly shorter APD90 compared with parallel controls (Figure 1B). For example, APD90 at 1 Hz averaged 300±21 ms in control cells versus 237±17 ms in dofetilide-incubated cells (P<0.05; n=18 and 14, respectively). These results suggest that sustained dofetilide exposure results in reduced intrinsic APD. We then evaluated the effects of sustained dofetilide exposure on repolarization reserve by measuring APD in control and dofetilide cells before and after superfusion with 5 nmol/L dofetilide. AP recordings from a control and dofetilide cell before and after short-term exposure to dofetilide are shown in Figure 1C. Cells cultured in the presence of dofetilide had reduced APD and a reduced APD response on reexposure to the compound, indicating enhanced repolarization reserve. Mean data in Figure 1D confirm the statistically significantly reduced APD response of dofetilide cells to IKr blockade.
Effects of Sustained Exposure to Dofetilide on Ionic Currents
Figure 2A shows examples of IKs recorded on stepping to 40 mV in control and dofetilide cells. IKs step and tail-current densities were significantly increased after sustained dofetilide exposure (Figure 2B). Mean current densities at 30 mV were 0.83±0.08 (step) and 0.31±0.03 pA/pF (tail) for control cells compared with 1.17±0.09 (step) and 0.44±0.03 pA/pF (tail) for dofetilide-incubated cells. The voltage dependence of IKs activation (Figure 2C; based on Boltzmann fits to tail currents) was not significantly altered; mean half-activation voltages averaged 31.1±7.5 mV for control and 28.1±2.6 mV for dofetilide cells. Similarly, IKs kinetics were unchanged (online-only Data Supplement Figure IIA).
Representative IKr recordings from control and dofetilide-incubated cells are shown in Figure 3A. Sustained dofetilide exposure did not significantly affect IKr properties after cell isolation and washing. IKr tail-current density showed no statistically significant alterations (Figure 3B). For example, mean IKr tail density at 10 mV averaged 0.15±0.01 and 0.16±0.02 pA/pF in control and dofetilide cells, respectively. The IKr half-activation voltage (Figure 3C; based on Boltzmann fits to tail currents) averaged −22.4±4.3 mV for control cells and −24.1±2.7 mV for dofetilide cells (n=8 per group; P=NS). IKr kinetics also were unchanged (online-only Data Supplement Figure IIB).
Other currents that can contribute to determining ventricular repolarization also were studied. Figure 4A and 4B shows examples of Ito recorded from a control and a dofetilide cardiomyocyte. No differences were found in mean Ito densities between control and dofetilide cells (Figure 4C). For example, Ito densities at 30 mV averaged 15.9±3.0 pA/pF in control and 14.8±1.4 pA/pF in dofetilide (P=NS versus control) cells. Neither activation (Figure 4D) nor inactivation (Figure 4E) kinetics were altered. Figure 5 shows IK1 recordings from a control (Figure 5A) and a dofetilide (Figure 5B) cardiomyocyte, along with mean current density-voltage relations (Figure 5C). IK1 density was not changed by long-term dofetilide exposure; eg, IK1 density at −120 mV averaged −23.3±1.5 pA/pF in control and −26.2±2.3 pA/pF in dofetilide (P=NS versus control) cells. ICa,L was similar in control (Figure 6A) and dofetilide (Figure 6B) cells. Mean ICa,L density was unchanged (Figure 6C), eg, averaging (at 10 mV) −3.2±0.6 pA/pF in control and −3.3±0.4 pA/pF in dofetilide cells. ICa,L inactivation kinetics were similarly unaffected (Figure 6D). Late INa recordings are shown in online-only Data Supplement Figure IIIA, with corresponding mean data at −30 mV in Figure IIIB. Sustained dofetilide exposure had no significant effect on late INa density.
The ionic current recording data suggest that sustained dofetilide exposure increases repolarization reserve by upregulating IKs. If this is the case, the response to IKs inhibition should differ in control and dofetilide cells. We tested this possibility by studying the APD response to 100 nmol/L HMR1566, as illustrated in online-only Data Supplement Figure IV. Whereas IKs inhibition had a small, nonsignificant effect on control cells, HMR1566 significantly delayed repolarization of dofetilide cells, consistent with IKs upregulation.
Expression of Delayed-Rectifier K+ Channel Subunits
We then addressed the potential molecular basis of IKs upregulation in response to dofetilide exposure by assaying mRNA and protein expression of the principal subunits believed to participate in forming IKr and IKs. Figure 7A shows mean mRNA expression values of the IKr-related subunits KCNH2 and KCNE2 (P=NS), and Figure 7B shows mRNA expression values for the IKs-related subunits KCNQ1 and KCNE1 (P=NS) in control and dofetilide cells. Incubation in the presence of dofetilide did not affect the mRNA expression of subunits underlying IKr and IKs. We then analyzed the protein expression of the IKs-related subunits. The top panels in Figure 7C and 7D show examples of KvLQT1 and minK bands, detected at ≈78 kDa for KvLQT1 and ≈20 kDa for minK, with corresponding GAPDH bands from the same lanes (≈34 kDa). The bottom panels show mean expression levels for KvLQT1 and minK protein normalized to GAPDH. The protein expression values for both KvLQT1 and minK were significantly increased, by 21±5% and 26±5%, respectively, in dofetilide-exposed cells (P<0.05 for KvLQT1; P<0.01 for minK).
MicroRNA Expression After Sustained Exposure to Dofetilide
Sustained dofetilide exposure increased IKs subunit protein expression but not corresponding mRNA levels, suggesting changes in posttranscriptional regulation of the subunits underlying IKs. Recently, it has been found that microRNAs can regulate ion channel protein expression and transmural distribution and may be important in posttranscriptional regulation of ion channel expression in the heart.16–18 MiR-1 and miR-133 are muscle-specific microRNAs (miRNAs) that are strongly and specifically expressed in adult cardiac tissues.19 Evidence has been presented to show that miR-133 can repress KvLQT1 protein expression without altering mRNA expression, whereas miR-1 represses minK protein but not mRNA expression.17 To assess whether changes in microRNA regulation are a candidate to explain IKs subunit protein expression alterations caused by sustained dofetilide exposure, miR-133 and miR-1 expression levels were quantified in control and dofetilide cells. Because we were unable to find documented 3′UTR sequences of canine KCNQ1 (encoding KvLQT1) and KCNE1 (encoding minK) genes, we first performed 3′-RACE experiments on dog cDNAs to obtain the 3′UTR sequences of canine KCNQ1 and KCNE1 genes. The results are shown in online-only Data Supplement Figures V and VI. The 3′UTR of canine KCNQ1 that we sequenced is 919 bp (GenBank accession No. EU162137), whereas the corresponding length for KCNE1 is 303 bp (GenBank accession No. EU162136). Complementary binding sites for miR-133 and miR-1 on canine KCNQ1 and KCNE1 3′UTRs were computationally analyzed and are shown in Figure 8 (left). We found 3 miR-133 potential binding sites but no miR-1 binding sites on the 3′UTR sequence of KCNQ1. We found 1 potential miR-1 binding site on the 3′UTR of the canine KCNE1 gene, on which no miR-133 binding sites were observed. Both miR-133a and miR-133b are mature forms of the miR-133 family, having the same core miR-133 sequence (in italics below) that is the main functional sequence for the effect of miR-133 on target mRNAs (MiR-133a 3′-5′, UGUCGACCAACUUCCCCUGGUU; miR-133b 3′-5′, AUCGACCAACUUCCCCUGGUU). The expression levels of miR-133a, miR-133b, and miR-1 were quantified by real-time PCR. As shown in Figure 8B, the expression levels of miR-133a and miR-133b were significantly decreased in dofetilide cells compared with control, a change opposite of the alterations in expression of KvLQT1 protein and consistent with the inhibitory effect of miR-133 on KvLQT1 protein expression.17 In contrast, the expression of miR-1 was not different in dofetilide cells compared with control cells (Figure 8D).
We have shown that continuous exposure to the IKr inhibitor dofetilide causes upregulation of IKs and its subunit proteins in canine cardiomyocytes activated at physiological rates without changing several other currents that contribute to APD regulation (IKr, IK1, Ito, ICa,L, and late INa). Associated physiological consequences include reduced APD and increased repolarization reserve.
Drug-induced long-QT syndrome is an important clinical problem.1,2 Although the occurrence of drug-induced long-QT syndrome is unpredictable, the concept of repolarization reserve, first proposed by Roden1 in 1998, has provided important insights into the underlying determinants. An intrinsic ability of cardiomyocytes to protect themselves from excess APD prolongation depends on the intactness of a variety of functional K+ channels through which repolarizing current can increase in response to a repolarization-suppressing challenge. The phenomenon of repolarization reserve has been demonstrated clearly in a variety of experimental paradigms that show that the loss of more than one K+ current produces much more repolarization impairment than expected on the basis of observations of the effects of the loss of individual currents.2–6,20 In the present study, we provide evidence for a novel mechanism contributing to the control of repolarization reserve, feedback regulation of ion channel expression. Sustained reductions in IKr, which increase APD acutely, lead to enhanced expression of IKs subunit proteins and increased IKs density. IKs acts as a “safety mechanism” for cardiac repolarization, becoming larger when APD is increased by reduced outward or increased inward current and preventing excess repolarization delays.4,6 Increased IKs density resulting from chronic IKr block will limit the associated APD prolongation and create increased repolarization reserve to counteract the effects of repolarization inhibition, limiting the potential risk of torsade de pointes. This effect was illustrated by our experiment showing that cells treated with dofetilide long term showed smaller APD increases on IKr-inhibiting (dofetilide) challenge than control cells.
Our observations provide potential new insights into phenomena of clinical relevance. Loss-of-function mutations in the genes encoding IKs subunits increase the severity and risk of arrhythmias in congenital and acquired long-QT syndromes caused by IKr abnormalities.21,22 This observation has generally been attributed to decreased functional ability to increase IKs in response to reduced IKr, for which much experimental evidence exists. However, our results suggest that an additional contributory mechanism may be a reduced effectiveness of the upregulation of IKs currents in response to sustained IKr decreases.
The phenomena we observed in the present study have precedents. The loss of a K+ current in genetically engineered mice results in compensatory upregulation of other cardiac K+ channel subunits.8–10 A limitation of the available observations in mouse models is the fact that the dominant repolarizing currents in mice are quite different from those in humans. The findings of the present study suggest that the principles regulating K+ current expression in mouse models also may apply to K+ current systems much closer to the dominant repolarizing currents in the human heart.
Our results imply the presence of feedback systems capable of responding to changes in cardiac electric function by inducing adaptive changes in cardiac ion channel expression, as recently suggested on the basis of theoretical considerations.7 The detailed molecular mechanisms underlying IKs upregulation in response to IKr inhibition remain to be established. We observed an increase in KvLQT1 and minK protein without a change in their corresponding mRNA, which indicates the existence of potential changes in posttranscriptional regulation of the protein expression of subunits underlying IKs that occur with sustained dofetilide incubation. MicroRNAs can mediate gene regulation at the posttranscriptional level and have been considered an important novel component of the mechanisms regulating the expression of genes and their protein products.23,24 Limited studies have been performed to explore the role of microRNA in ion channel subunit expression regulation. The available information shows that microRNAs can regulate ion channel protein expression and transmural distribution and that they may be important in posttranscriptional regulation of ion channel expression in the heart.16–18,25 Luo et al17 showed that miR-133 can repress KvLQT1 protein expression without affecting corresponding mRNA expression, whereas miR-1 represses minK protein but not mRNA expression. Our study suggests that repolarization-delaying interventions can alter the regulation of microRNA expression. To study microRNA regulation of KvLQT1 and minK expression, we had to first clone the 3′UTRs of dog KCNQ1 and KCNE1 genes, which had not been reported in the literature. Having done so, we identified 3 putative binding sites for miR-133 on the canine KCNQ1 3′UTR and 1 putative binding site for miR-1 on dog KCNE1. Interestingly, we found that the expression levels of the muscle-specific miRNA miR-133 were reduced on long-term dofetilide culture. The 3 miR-133 binding sites on dog KCNQ1 3′UTR could cooperate with each other to confer enhanced miR-133 binding, therefore restricting KvLQT1 protein expression. When miR-133 expression is reduced during long-term dofetilide exposure, reduced miR-133 binding might result in higher KvLQT1 protein levels, as observed in the present study.
The mechanisms responsible for reduced miR-133 expression after sustained dofetilide exposure remain unknown, requiring further detailed study that is beyond the scope of the present study. On the other hand, we found miR-1 expression to be unchanged in cells exposed long term to dofetilide. In addition, we found only 1 putative miR-1 binding site on the KCNE1 3′UTR, which raises questions about the importance of miR-1 in regulating KCNE1 expression in the dog. The changes that we observed in minK protein expression thus remain unexplained. Whether the observed higher minK protein expression is caused by regulation through other unknown miRNAs is not clear and will be interesting to explore in the future. Xiao et al16 have shown that miR-133 represses ERG protein expression in diabetic rabbit hearts. We did not observe any changes in IKr density after sustained dofetilide exposure, as we might have expected if canine ERG expression were regulated by miR-133. The potential role of miR-133 in determining canine ERG expression is presently unknown and would be interesting to investigate further.
We cannot exclude additional contributors, besides microRNA changes, to IKs increases in dofetilide-exposed cardiac cells. Sustained APD prolongation could increase cellular Ca2+ entry and cellular Ca2+ content.26 Calmodulin binding to KvLQT1 is necessary for proper channel assembly and for conferring Ca2+-sensitive stimulation of IKs.27 Therefore, remodeling of IKs after prolonged dofetilide exposure might be partially due to changes in intracellular [Ca2+] and effects mediated by Ca2+/calmodulin regulation of KvLQT1 protein assembly. In addition, IKs subunits are subject to important regulation of expression at the level of cellular trafficking and membrane insertion and significant functional regulation by cell phosphorylation machinery.28,29 Posttranslational regulatory changes may explain why IKs increased by ≈40% despite only 20% to 25% increases in underlying subunit protein expression. Changes in intracellular [Ca2+]i could certainly be involved because increased [Ca2+]i can enhance IKs. Better insights into the mechanisms underlying feedback regulation of K+ channel expression might lead to novel strategies for the treatment of patients at risk of long-QT syndromes and to potential new insights into the determinants of long-QT syndrome occurrence.
We used an in vitro system to assess the effects of sustained IKr inhibition on repolarization reserve and on the expression of various cardiac ion channels. Pacing cells in vitro allows them to display cardiac electric activity at rates equivalent to normal heart rates in vivo, providing results under well-controlled conditions that may be more relevant to in vivo physiological function than those obtained in quiescent cell systems. Follow-up studies in other relevant systems would be of interest.
We used CdCl2 to inhibit ICa,L when studying Ito and IK1. Although Cd2+ effectively inhibits ICa,L and prevents K+ current contamination, it positively shifts voltage dependence by neutralizing membrane surface charges.30 However, this effect should apply equally to control and dofetilide cells and would not explain the lack of change in these currents. For IKr and IKs measurements, which can be affected in complex ways by divalent cations,30 we used nimodipine to inhibit ICa,L.
We thank Dr Jacques Barhanin, Université de Nice Sophia Antipolis, France, for kindly providing us the anti-minK primary antibody. We also thank Chantal St-Cyr, Chantal Maltais, and Nathalie L’Heureux for technical assistance and France Thériault for secretarial help with the manuscript.
Sources of Funding
This work was supported by the Canadian Institutes of Health Research (operating grant MOP 68929) and the Quebec Heart and Stroke Foundation.
Roden DM, Yang T. Protecting the heart against arrhythmias: potassium current physiology and repolarization reserve. Circulation. 2005; 112: 1376–1378.
Silva J, Rudy Y. Subunit interaction determines IKs participation in cardiac repolarization and repolarization reserve. Circulation. 2005; 112: 1384–1391.
Jost N, Virag L, Bitay M, Takacs J, Lengyel C, Biliczki P, Nagy Z, Bogats G, Lathrop DA, Papp JG, Varro A. Restricting excessive cardiac action potential and QT prolongation: a vital role for IKs in human ventricular muscle. Circulation. 2005; 112: 1392–1399.
Rosati B, McKinnon D. Regulation of ion channel expression. Circ Res. 2004; 94: 874–883.
Guo W, Li H, London B, Nerbonne JM. Functional consequences of elimination of Ito,f and Ito,s: early afterdepolarizations, atrioventricular block, and ventricular arrhythmias in mice lacking Kv1.4 and expressing a dominant-negative Kv4 α subunit. Circ Res. 2000; 87: 73–79.
Zhou J, Kodirov S, Murata M, Buckett PD, Nerbonne JM, Koren G. Regional upregulation of Kv2.1-encoded current, IK,slow2, in Kv1DN mice is abolished by crossbreeding with Kv2DN mice. Am J Physiol Heart Circ Physiol. 2003; 284: H491–H500.
Xiao L, Zhang L, Han W, Wang Z, Nattel S. Sex-based transmural differences in cardiac repolarization and ionic-current properties in canine left ventricles. Am J Physiol Heart Circ Physiol. 2006; 291: H570–H580.
Brundel BJ, Shiroshita-Takeshita A, Qi X, Yeh YH, Chartier D, van Gelder IC, Henning RH, Kampinga HH, Nattel S. Induction of heat shock response protects the heart against atrial fibrillation. Circ Res. 2006; 99: 1394–1402.
Maltsev VA, Reznikov V, Undrovinas NA, Sabbah HN, Undrovinas A. Modulation of the late sodium current by Ca2+, calmodulin, and CaMKI I in normal and failing dog cardiomyocytes: similarities and differences. Am J Physiol Heart Circ Physiol. 2008; 294: H1597–H1608.
Yang T, Roden DM. Extracellular potassium modulation of drug block of IKr: implications for torsade de pointes and reverse use-dependence. Circulation. 1996; 93: 407–411.
Xiao J, Luo X, Lin H, Zhang Y, Lu Y, Wang N, Zhang Y, Yang B, Wang Z. MicroRNA miR-133 represses HERG K+ channel expression contributing to QT prolongation in diabetic hearts. J Biol Chem. 2007; 282: 12363–12367.
Luo X, Xiao J, Lin H, Li B, Lu Y, Yang B, Wang Z. Transcriptional activation by stimulating protein 1 and posttranscriptional repression by muscle-specific microRNAs of IKs-encoding genes and potential implications in regional heterogeneity of their expressions. J Cell Physiol. 2007; 212: 358–367.
Berthet M, Denjoy I, Donger C, Demay L, Hammoude H, Klug D, Schulze-Bahr E, Richard P, Funke H, Schwartz K, Coumel P, Hainque B, Guicheney P. C-terminal HERG mutations: the role of hypokalemia and a KCNQ1-associated mutation in cardiac event occurrence. Circulation. 1999; 99: 1464–1470.
Yang P, Kanki H, Drolet B, Yang T, Wei J, Viswanathan PC, Hohnloser SH, Shimizu W, Schwartz PJ, Stanton M, Murray KT, Norris K, George AL Jr, Roden DM. Allelic variants in long-QT disease genes in patients with drug-associated torsades de pointes. Circulation. 2002; 105: 1943–1948.
Luo X, Lin H, Pan Z, Xiao J, Zhang Y, Lu Y, Yang B, Wang Z. Downregulation of MIRNA-1/MIRNA-133 contributes to re-expression of pacemaker channel genes HCN2 and HCN4 in hypertrophic heart. J Biol Chem. 2008; 283: 20045–20052.
Shamgar L, Ma L, Schmitt N, Haitin Y, Peretz A, Wiener R, Hirsch J, Pongs O, Attali B. Calmodulin is essential for cardiac IKS channel gating and assembly: impaired function in long-QT mutations. Circ Res. 2006; 98: 1055–1063.
Seebohm G, Strutz-Seebohm N, Birkin R, Dell G, Bucci C, Spinosa MR, Baltaev R, Mack AF, Korniychuk G, Choudhury A, Marks D, Pagano RE, Attali B, Pfeufer A, Kass RS, Sanguinetti MC, Tavare JM, Lang F. Regulation of endocytic recycling of KCNQ1/KCNE1 potassium channels. Circ Res. 2007; 100: 686–692.
Repolarization reserve is an important determinant of the response to contexts involving action potential prolongation, including potassium channel–blocking drugs (both antiarrhythmic agents and compounds with collateral effects on potassium channels), bradycardic drugs, and electrolyte disturbances like hypokalemia and hypomagnesemia. Repolarization reserve is generally attributed to functionally based compensatory increases in potassium currents in response to action potential duration–prolonging tendencies that minimize the resulting action potential prolongation. The possibility that sustained action potential prolongation by potassium channel inhibition may induce remodeling of ion current expression has not been tested. Accordingly, we assessed the effects of sustained rapid delayed-rectifier potassium current (IKr) inhibition with dofetilide on various cardiac ionic currents in isolated paced/cultured dog cardiomyocytes. Sustained dofetilide exposure led to abbreviated action potential duration and increased repolarization reserve (reduced action potential duration–prolonging response to IKr blockade), along with increased slow delayed-rectifier current (IKs) density. A variety of other currents remained unchanged. The mRNA expression of IKs subunits was unchanged, but their protein expression was increased, suggesting posttranscriptional regulation. We quantified the muscle-specific microRNA subtypes miR-133a and miR-133b, which can posttranscriptionally repress protein expression of the IKs α-subunit KvLQT1. Expression levels of miR-133a and miR-133b were decreased in cells cultured in dofetilide, possibly accounting for KvLQT1 protein upregulation. We conclude that chronic action potential prolongation can cause compensatory upregulation of potassium currents, possibly mediated (at least in part) by microRNA changes, adding the regulation of ion channel expression to the potential mechanisms governing repolarization reserve. Function and malfunction of this system could contribute to factors governing the occurrence of cardiac arrhythmias in repolarization dysfunction paradigms like congenital and acquired long-QT syndromes.
The online-only Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.107.758672/DC1.