Human Cardiac Inwardly-Rectifying K+ Channel Kir2.1b Is Inhibited by Direct Protein Kinase C-Dependent Regulation in Human Isolated Cardiomyocytes and in an Expression System
Background— Protein kinases A (PKA) and C (PKC) are activated in ischemic preconditioning and heart failure, conditions in which patients develop arrhythmias. The native inward rectifier potassium current (IK1) plays a central role in the stabilization of the resting membrane potential and the process of arrhythmogenesis. This study investigates the functional relationship between PKC and IK1.
Methods and Results— In whole-cell patch-clamp experiments with isolated human atrial cardiomyocytes, the IK1 was reduced by 41% when the nonspecific activator of PKC phorbol 12 myristate 13-acetate (PMA; 100 nmol/L) was applied. To investigate the effects of PKC on cloned channel underlying parts of the native IK1, we expressed Kir2.1b heterologously in Xenopus oocytes and measured currents with the double-electrode voltage-clamp technique. PMA decreased the current by an average of 68%, with an IC50 of 0.68 nmol/L. The inactive compound 4-α-PMA was ineffective. Thymeleatoxin and 1-oleolyl-2-acetyl-sn-glycerol, 2 specific activators of PKC, produced effects similar to those of PMA. Inhibitors of PKC, ie, staurosporine and chelerytrine, could inhibit the PMA effect (1 nmol/L) significantly. After mutation of the PKC phosphorylation sites (especially S64A and T353A), PMA became ineffective.
Conclusions— The human IK1 in atrial cardiomyocytes and one of its underlying ion channels, the Kir2.1b channel, is inhibited by PKC-dependent signal transduction pathways, possibly contributing to arrhythmogenesis in patients with structural heart disease in which PKC is activated.
Received February 25, 2002; revision received June 21, 2002; accepted June 26, 2002.
In human cardiomyocytes, repolarization at the end of the action potential is caused by many different potassium currents, which can be classified as delayed outward and inward rectifiers. The inward rectifier potassium current (IK1) reveals inwardly rectifying properties responsible for the terminal repolarization at the end of the cardiac action potential. The native IK1 is composed of several different potassium channels with distinct single channel conductances of 20 pS, 35 pS and 10 pS.1 Likewise, different gene families (Kir2.1, Kir2.2, and Kir2.3) have been found in human heart encoding IK1.2 The native IK1 may be involved in arrhythmogenesis of coronary artery disease3 and dilated cardiomyopathy.4 Mutations of Kir2.1 channels are associated with Anderson’s syndrome, an inherited disease with an arrhythmia characterized by QT-prolongation, periodic paralysis, and dysmorphic features.5
Within the Kir2.1 family, several distinct channels have been cloned in chronological order. The first channel, named IRK1 or MM-IRK1, was found in mouse tissue.6 The rabbit channel RBHIK1 revealed a 97% amino acid homology to MM-IRK1.7 The human HH-IRK1, which has a 98% homology to MM-IRK1, was the first channel cloned from the human heart.8 With the help of primers derived from IRK1, 2 channel sequences were identified in human atrial tissue; one was nearly identical to HH-IRK1, and the other (hIRK) was 70% homologous.9 The hIRK channel,9,10⇓ which is called Kir2.1b according to the new terminology that uses a “b” to differentiate it from the other HH-hIRK1 (Kir2.1a) channel,8 is the subject of this study. This differentiation makes sense, as we will see in this study, because Kir2.1b has functional protein kinase C (PKC) phosphorylation sites, whereas Kir2.1a lacks these sites.
Emotional stress and exercise may trigger ventricular arrhythmias with consequent sudden cardiac death,11 especially in patients with coronary artery disease12 or inherited forms of long-QT syndrome or ion channel disease (torsade de pointes tachycardias).13
Catecholamines like norepinephrine, which are released under stress, may finally activate PKC,14 revealing phosphorylation targets on many regulatory proteins and ion channels and thereby changing the action potential duration of cardiomyocytes.15 Alternative signal transduction pathways activating PKC in atrial cardiomyocytes exist via the parasympathetic/muscarinic system.16
In this study, we investigated the regulation of IK1 in isolated human cardiomyocytes and the regulation of a cloned inward rectifier potassium channel Kir2.1b (hIRK) from human atrium by PKC.
Solutions and Drug Administrations
Patch-clamp measurements of human atrial cardiomyocytes were performed in a bath solution containing (in mmol/L) 3.5 KCl, 140 NaCl, 1.5 CaCl2, 1.4 MgSO4, and 10 HEPES (pH 7.4), as well as 10 μmol/L nisoldipine to block calcium currents, 10 μmol/L chromanol 293B to inhibit the slow component IKs of delayed rectifier K+ channels, 1 μmol/L dofetilide to block the rapid component of the delayed rectifier potassium current IKr, and 1 μmol/L glibenclamide to inhibit ATP-dependent K+ channels (KATP). The pipette solution contained (in mmol/L) 140 KCl, 0.5 CaCl2, 1.5 MgCl2, 5 KATP, 1 EGTA, and 10 HEPES (pH 7.4). The high concentration of ATP should contribute to full inhibition of KATP.
Two microelectrode voltage clamp measurements of Xenopus oocytes were performed in a low K+ solution containing (in mmol/L) 5 KCl, 100 NaCl, 1.5 CaCl2, 2 MgCl2, and 10 HEPES (pH 7.3). Current and voltage electrodes were filled with 3 mol/L KCl solution. Phorbol 12 myristate 13-acetate (PMA; Calbiochem), 4α-PMA (Biomedicals Inc), thymeleatoxin (Calbiochem), staurosporine (Calbiochem), chelerythrine (Calbiochem) and chromanol 293B (Aventis) were dissolved in DMSO to a stock solution of 10 mmol/L and stored at −20° C. We then dissolved 1-oleolyl-2-acetyl-sn-glycerol (OAG; Calbiochem) in DMSO to a stock solution of 100 mmol/L and stored it at −20° C. Nisoldipine and glibenclamide (Sigma) were dissolved in 70% ethanol to a stock solution of 10 mmol/L and were stored at 4° C. BaCl2 (Sigma) was dissolved in water. On the day of experiments, aliquots of the stock solutions were diluted to the desired concentration with the bath solution. Vehicle control experiments with ethanol or DMSO in a final concentration of 0.1% did not reveal any effects on currents measured with human atrial cardiomyocytes or Xenopus oocytes. All measurements were made at room temperature (20° C), and experimental conditions were identical to those described by Karle et al.17
Electrophysiology and Data Analysis
Human cardiomyocytes were isolated from atrial appendages derived from patients undergoing heart surgery and cannulation for the heart-lung machine. Single cells were enzymatically dispersed by use of collagenase and papain and were measured with the whole-cell patch-clamp technique as described for portal vein smooth muscle cells.18 Each single measurement within a series of experiments was done with a cardiomyocyte from a separate patient. The 2 microelectrode voltage-clamp configuration was used to record currents from Xenopus laevis oocytes.17 No leak subtraction was done during the experiments; only recordings with <5% leak current were considered for data analysis. Statistical data are presented as mean±SD. Statistical significance for data in Figures 1 and 2⇓ was evaluated using the paired (time course of current run-up) and unparied (comparison of drug effects with time-dependent run-up) Student’s t test. Differences were considered significant when a probability value <0.05 was reached. Statistical significance for data in Figures 3 and 4⇓ was evaluated using the 1-way ANOVA. Differences were considered significant when a probability value <0.05 was reached.
The Kir2.1b wild type (hIRK wt) clone9 (Dr B.A. Wible, Case Western Reserve University, Cleveland, Ohio, GenBank accession number L36069) contains the hIRK potassium channel coding region in the pCR II transcription vector (Invitrogen). Complementary RNA was prepared as described by Kiehn et al.10 Kir2.1b contains 4 consensus sites for PKC phosphorylation (Thr-38, Ser-64, Thr-353, and Ser-357). The serine or threonine residues of these sites were replaced by alanine to eliminate potential protein kinase-mediated phosphorylation at these sites. These mutations were generated by polymerase chain reaction (PCR) with synthetic mutant oligonucleotide primers using the QuikChange site-directed mutagenesis kit (Stratagene). This resulted in the mutated channels Kir2.1b T38A, Kir2.1b S64A, Kir2.1b T353A, and Kir2.1b S357A. The PCR products were then sequenced (SeqLab Goettingen). The restriction fragments ApaI/ClaI contained the mutations T38A and S64A, and ClaI/BamHI contained the mutations T353A and S357A (all restriction enzymes from Roche Diagnostics). The restriction fragments were subcloned into the original Kir2.1b plasmid and sequenced again.
To generate clones that contained combinations of mutations, the T38A mutation was introduced into Kir2.1b S64A and S357A into Kir2.1b T353A by site-directed mutagenesis as described above. This resulted in mutated clones containing T38A+S64A and T353A+S357A. By subcloning as described above, the following clones were produced: Kir2.1b 2 mol/L (S64A, T353A); Kir2.1b 3 mol/L (T38A, S64A, T353A); and Kir2.1b 4 mol/L (all 4 mutations). Finally, cDNA of all clones that were used for in vitro transcription was verified by additional sequencing.
The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (Bethesda, Md: NIH publication No. 85 to 23, revised 1996).
Inhibition of Native Human Atrial IK1 by Protein Kinase Activation With PMA
The role of the serine/threonine PKA and PKC in the regulation of cardiac inward rectifier potassium channels was investigated by measuring the native IK1 in human atrial cardiomyocytes with the whole-cell patch clamp technique.
Starting with a holding potential of −40 mV, test pulses from −100 mV to 50 mV were applied to human atrial cardiomyocytes in steps of 10 mV (400 ms). This evoked time-independent inward currents at potentials from −100 to −60 mV, which were conducted by IK1, and outward currents at potentials from −30 to 50 mV that showed incomplete inactivation and resulted from delayed rectifier potassium currents (Figure 1A).
After application of PMA, a nonspecific activator of serine/threonine and tyrosine protein kinases, the inward portion of current, ie, IK1, was inhibited (Figure 1A). The drug-sensitive current showed the classic IK1 current-voltage behavior with strong inward rectification and was part of the larger current component inhibited by low concentrations of BaCl2 (50 μmol/L, Figure 1D and 1E). PMA inhibited IK1 in a concentration-dependent manner, with an IC50 of 4.95 nmol/L (Figure 1C). With 100 nmol/L PMA, the current at −100 mV was decreased from 94.3±28.0 pA to 47.1±12.5 pA ( 41.0±4.7%; significantly different from the predrug control with t=2.776, P=0.017, at the 0.05 level for n=12; Figure 1A). The inactive phorbol ester 4-α-PMA (100 nmol/L) showed no significant effect, indicating that the effects of PMA are due to protein kinase activation and are not nonspecific (Figure 1B). The current was 106.4±34.8 pA under control conditions and 104.9±33.4 pA after application of 4-α-PMA (not significantly different; n=8; Figure 1B).
Inhibition of Kir2.1b by PMA
To study the inhibition of IK1 in more detail, we investigated the effects of protein kinases on one of the cloned counterparts of IK1, the Kir2.1b channel, heterologously expressed in Xenopus oocytes, with the double-electrode voltage-clamp technique. Starting with a holding potential of −80 mV, test pulses from −120 mV to 40 mV were applied in steps of 10 mV lasting 400 ms. Over 30 minutes of observation time, the Kir2.1b current at −120 mV showed a small run-up phenomenon, thereby increasing the measurable current from 100% to 114.1±3.3% (Figure 2C). After application of 100 nmol/L PMA, the current decreased steadily, reaching a steady-state inhibition after 30 minutes (Figure 2E). The current decreased significantly to 31.9±7.8% (n=6; Figure 2A and 2B). Inhibition of Kir2.1b by PMA was dose-dependent, with an IC50 of 0.68 nmol/L (Figure 2D). The control drug 4-α-PMA (100 nmol/L) did not have any significant effect on the current; Kir2.1b again showed a weak run-up to 109.6±8.8% (n=5), similar to control (Figure 3).
Pharmacological Evidence That PKC Mediates the Effects of PMA on Kir2.1b
PMA is nonspecific and activates various types of protein kinases. Thus, we chose a more specific pharmacological approach. As an activator of the conventional (cPKC α, β1/2, and γ) and novel type PKC isoenzymes (nPKC δ, ε, η, and θ), OAG at a concentration of 10 μmol/L was used, thereby inhibiting the current at −120 mV significantly to 44.9±2.0% (n=4) after 30 minutes (Figure 3). Thymeleatoxin (100 nmol/L), a specific activator of the conventional protein kinase C isoenzymes,19 was even more effective and caused a strong inhibition of hIRK to 18.6±0.4% (n=5) after 30 minutes (Figure 3). The co-application of 1 μmol/L staurosporine, a nonspecific inhibitor of several serine/threonine protein kinases, attenuated the effect of 1 nmol/L PMA. The current remained almost unchanged, with a run-up to 125.6±3.9% after 30 minutes (significantly different from 1 nmol/L PMA alone; n=5; Figure 3). Co-application of 10 μmol/L chelerythrine, a specific inhibitor of protein kinase C, also attenuated the PMA effect. Under these experimental conditions, a slight but significant decrease to 95.3±12.6% (n=6; Figure 3) was observed.
Site-Directed Mutagenesis of PKC Phosphorylation Sites Attenuates PMA Effects
To identify PKC phosphorylation sites in the sequence of Kir2.1b, we searched with the computer program HUSAR Prosite (DKFZ, Heidelberg, Germany) for the key sequence for PKC phosphorylation (amino acids S/T-X-R/K) and identified 4 putative PKC phosphorylation sites (T38, S64, T353, S357; Figure 5). We created mutated channels by site-directed mutagenesis in which each phosphorylatable serine or threonine was replaced by the non-phosphorylatable alanine, resulting in the channels Kir2.1b T38A, S64A, T353A, or S357A. The effects of PMA were then investigated on each individual mutation. We found that the 2 mutations, S64A and T353A, lead to a strong attenuation of the PMA effect; the current at −120 mV remained almost unchanged after application of 100 nmol/L PMA and came to 86.3±1.2% (n=7) for S64A, and 109.5±2.0% (n=6) for T353A, significantly different from 100 nmol/L PMA applied to the wild-type. The 2 other mutations, T38A and S357A, were less effective, and the current was changed to 51.4±0.9% (n=5) for T38A, significantly different from 100 nmol/L PMA applied to the wild type, and to 46.9±2.4% (n=5) for S357A, not significantly different from 100 nmol/L PMA applied to the wild type after application of 100 nmol/L PMA (Figure 4).
In all combined mutations, the Kir2.1b current showed a run-up rather than any inhibition by PMA. In Kir2.1b 2 mol/L (S64A combined with T353A), the current increased to 181.1±25.4% (n=5) after application of 100 nmol/L of PMA, and to 186.1±14.1% (n =5) after application of 1000 nmol/L PMA, significantly different compared with the control run-up. In Kir2.1b 3 mol/L (2 mol/L combined with T38A), the current changed significantly to 179.6±7.8% (n=5) after application of 100 nmol/L PMA. In Kir2.1b 4 mol/L (all 4 PKC sites were mutated), the current changed significantly to 171.8±15.1% (n=6) compared with the control run-up (Figure 4).
We made an alignment with other Kir2 channels to compare the presence of the functional PKC phosphorylation sites in these channels. Interestingly, in human Kir2.1b, the functionally most relevant PKC phosphorylation sites (S64 and T353) are present, although they are not present in human Kir2.1a and mouse Kir2.1 (IRK1). In IRK1, the major PKC sites S64 and T353 are lacking, but the PKC phosphorylation site S357 is present, which might mediate a weak inhibitory effect of PKC, thereby explaining the controversial data existing in the literature (Figure 6).
This is the first study demonstrating regulation of the cloned human cardiac inward rectifier potassium channel Kir2.1b by PKC. Within the Kir2.1 family of inward rectifier potassium channels, Kir2.1b is the only member regulated by PKC. Kir2.1b is of particular interest because this channel has unique PKC phosphorylation sites that are not present in other members of the Kir2.1 family, which explains why Kir2.1a is not regulated by PKC (Figure 6, alignment).
Activation of serine/threonine or tyrosine kinases by PMA resulted in a decrease of Kir2.1b inwardly rectifying K+ currents, and this effect could be blocked by co-application of specific PKC inhibitors. Likewise, the Kir2.1b (hIRK) current could be blocked by specific activators of PKC. These results suggest a predominant role of PKC in PMA-mediated inhibition of hIRK, a major component of the human cardiac IK1. The physiological relevance of these results could be confirmed by PMA-dependent inhibition of native cardiac IK1 in isolated human atrial cardiomyocytes measured with the patch-clamp technique. In our experiments, the PMA-dependent portion of current accounts for 78.2% of the BaCl2-sensitive IK1 in human atrial cardiomyocytes. The IC50 of the PMA effects is 7-fold higher in human atrial cardiomyocytes. This difference can be explained by the composition of native IK1 by additional channels (21.8% of total current) that are not PMA-sensitive.
Four putative PKC phosphorylation sites have been identified in the Kir2.1b protein, but only 2 sites, S64A and T353A, have functional effects that lead to a strong inhibition of PMA-mediated effects. The fact that these mutations prevent the effects of PMA suggests that PKC regulates the Kir2.1b protein at these sites directly.
Contradictory reports exist in the literature about the regulation of members of the Kir2 family by PKC. Fakler et al20 found a PKC-dependent inhibition and PKA-dependent increase of the mouse inward rectifier IRK1 initially cloned by Kubo et al.6 Likewise, the PMA-dependent run-up of current in all combined mutations of PKC-phosphorylation sites in our study might be explained by a putative PMA-dependent activation of resting PKA-sites.
In contrast, experiments by Henry et al21 argue against any regulation of IRK1 by PKC; rather, they suggest the inward rectifier Kir2.3 as a target for PKC-mediated phosphorylation, at threonine 53, as supported by the findings of Zhu et al.22 The inward rectifier IRK1, however, was not significantly affected by PMA. Likewise, after inserting the N-terminal region of Kir2.1 into the Kir2.3 channel, the resulting chimerical channel lost its PMA sensitivity. Interestingly, creation of the threonine residue known from Kir2.3 at the corresponding position (I79T) in Kir2.1 gave the mutant channel a PMA sensitivity almost identical to the wild-type Kir2.3. The alignment of our study in Figure 6 explains some of these results and demonstrates the unique role of Kir2.1b within the human Kir2 clones that have functional relevant PKC sites S64 and T353.
The important finding of this study is that the functional effects of PKC on Kir2.1b could be linked to the IK1 in human cardiomyocytes that are composed by many IK1 type currents. Some are regulated by PKC and others are not. Functional current inhibition of native IK1 by PMA in isolated human cardiomyocytes, as presented in this study, suggests that this pathway has functional relevance in the human heart and may be important in arrhythmogenesis.
This study was supported by a grant of the “Deutsche Forschungsgemeinschaft” Ki 6631/1 to Dr Kiehn and by grants of the University of Heidelberg to Drs Karle, Kathöfer, and Kiehn. The skillful technical assistance of Klara Güth and Sonja Lück is gratefully acknowledged.
- ↵Whang Z, Yue L, White M, et al. Differential distribution of inward rectifier potassium channel transcripts in human atrium versus ventricle. Circulation. 1998; 98: 2422–2428.
- ↵Aimond F, Alvarez JL, Rauzier JM, et al. Ionic basis of ventricular arrhythmias in remodeled rat heart during long-term myocardial infarction. Cardiovasc Res. 1999; 42: 402–415.
- ↵Koumi S, Backer CL, Arentzen CE. Characterization of inwardly rectifying K+ channel in human cardiac myocytes: alterations in channel behavior in myocytes isolated from patients with idiopathic dilated cardiomyopathy. Circulation. 1995; 92: 164–174.
- ↵Wible BA, De Biasi M, Majumber K, et al. Cloning and functional expression of an inwardly rectifying K+ channel from human atrium. Circ Res. 1995; 76: 343–350.
- ↵Kiehn J, Wible B, Ficker E, et al. Cloned human inward rectifier K+ channel as a target for class III methanesulfonanilides. Circ Res. 1995; 77: 1151–1155.
- ↵Lown B. Sudden cardiac death: biobehavioral perspective. Circulation. 1987; 76(1 pt 2): I-186–I-196.
- ↵Leenhardt A, Lucet V, Denjoy I, et al. Catecholaminergic polymorphic ventricular tachycardia in children: a 7-year follow-up of 21 patients. Circulation. 1995; 91: 1512–1519.
- ↵Dirksen RT, Sheu SS. Modulation of ventricular action potential by α1-adrenoceptors and protein kinase C. Am J Physiol. 1990; 258(3 pt 2): H907–H911.
- ↵Karle CA, Kreye VAW, Thomas D, et al. Antiarrhythmic drug carvedilol inhibits HERG potassium channels. Cardiovasc Res. 2001; 49: 361–370.
- ↵Karle CA, Yao X, Kreye VAW. A K+ single channel and whole-cell clamp study on the effects of levcromakalim in guinea pig portal vein cells. Naunyn-Schmiedeberg‘s Arch Pharmacol. 1998; 358: 374–381.
- ↵Zhu G, Qu Z, Cui N, et al. Suppression of Kir2.3 activity by protein kinase C phosphorylation of the channel protein at threonine 53. J Biol Chem. 1999; 274: 11643–11646.
- ↵Perier F, Radeke CM, Vandenberg CA. Primary structure and characterization of a small-conductance inwardly rectifying potassium channel from human hippocampus. Proc Natl Acad Sci U S A. 1994; 91: 6240–6244.