Differential Effects of the Betablockers Carvedilol, Metoprolol and Bisoprolol on Cardiac Kv4.3 (Ito) Channel Isoforms

Cardiac Kv4.3 channels contribute to the transient outward K+ current, Ito, during early repolarization of the cardiac action potential. Two different isoforms of Kv4.3 are present in the human ventricle and exhibit differential remodeling in heart failure (HF). Cardioselective betablockers are a cornerstone of HF with reduced ejection fraction therapy as well as ventricular arrhythmia treatment. In this study we examined pharmacological effects of betablockers on both Kv4.3 isoforms to explore their potential for isoform-specific therapy. Kv4.3 isoforms were expressed in Xenopus laevis oocytes and incubated with the respective betablockers. Dose-dependency and biophysical characteristics were examined. HEK 293T-cells were transfected with the two Kv4.3 isoforms and analyzed with Western blots. Carvedilol (100 µM) blocked Kv4.3 L by 77 ± 2% and Kv4.3 S by 67 ± 6%, respectively. Metoprolol (100 µM) was less effective with inhibition of 37 ± 3% (Kv4.3 L) and 35 ± 4% (Kv4.3 S). Bisoprolol showed no inhibitory effect. Current reduction was not caused by changes in Kv4.3 protein expression. Carvedilol inhibited Kv4.3 channels at physiologically relevant concentrations, affecting both isoforms. Metoprolol showed a weaker blocking effect and bisoprolol did not exert an effect on Kv4.3. Blockade of repolarizing Kv4.3 channels by carvedilol and metoprolol extend their pharmacological mechanism of action, potentially contributing beneficial antiarrhythmic effects in normal and failing hearts.


Introduction
The cardiac K v 4.3 channel is the main carrier of the transient outward potassium current (I to ) during the early repolarization phase of the cardiac action potential [1].Changes in expression and function of this channel are associated with cardiac conditions such as Brugada syndrome, atrial fibrillation, and early repolarization syndrome.Moreover, K v 4.3 expression changes have been reported after myocardial infarction and heart failure (HF) [2][3][4].A consistent electrophysiological feature in HF is a prolongation of the ventricular action potential [5,6].In animal HF models, as well as in human cardiac myocytes, a reduction of the transient outward potassium current in the early repolarization phase mediated by K v 4.3 was observed [7][8][9][10][11].The cardiac K v 4.3 channel, which is encoded by the KCND3 gene, is activated voltage dependently.There are two different isoforms that result from the alternative splicing of exon 6.The longer K v 4.3 L variant and the shorter isoform K v 4.3 S differ in a 19 amino acid long sequence at the intracellular C-terminus [12,13].In patients with dilated cardiomyopathy expression levels of K v 4.3, isoforms are differentially expressed: K v 4.3 L is upregulated, whereas K v 4.3 S is downregulated.As a result, the long isoform is predominantly expressed in the insufficient heart [4].
The aim of this study was to examine differential effects of carvedilol, bisoprolol, and metoprolol on K v 4.3 isoforms to provide the basis for isoform-specific HF drug therapy.

Effects of Carvedilol on the Function of K v 4.3 Channel Isoforms
Carvedilol caused time-dependent inhibition of the K v 4.3 peak current.Both isoforms were similarly affected.Maximum inhibitory effects observed after 50 min yielded a current block of 76.6 ± 2.0% (K v 4.3 L, n = 8, p < 0.001) and 67.1 ± 5.7% (K v 4.3 S, n = 8, p < 0.001).K v 4.3 currents before and after application of carvedilol, as well as the development of the block, are shown in Figure 1A-D.

Biophysical Characteristics of Kv4.3 Inhibition by Carvedilol
Activation and inactivation kinetics were assessed by applying depolarizing steps from −100 mV to +50 mV (2000 ms, 10 mV increments) from a holding potential of −80mV.
There was a return pulse to +50 mV after the first voltage step.The double-step voltage protocol as well as typical current traces in the absence and the presence of 100 µM carvedilol are depicted in Figure 2A,B.
To examine the effects of carvedilol on Kv4.3 channel activation, peak currents meas- The inhibition of K v 4.3 L and K v 4.3 S currents was partially reversible.The peak current returned to 33.8 ± 2.6% for the L isoform, and 49.1 ± 6.8% for the S isoform after 30 min wash out.Blockade was concentration dependent with IC 50 values of 57.1 ± 12.6 µM (K v 4.3 L, n = 8-12) and 58.2 ± 6.2 µM (K v 4.3 S, n = 8-13) (Figure 1E,F).To examine the effects of carvedilol on K v 4.3 channel activation, peak currents measured during the first part of the voltage protocol were normalized to the peak current of the last trace and then plotted against the test pulse voltage.Application of carvedilol led to a shift of the half-maximal activation voltages of both K v 4.3 isoforms towards more positive voltages.The half-maximal activation voltage in the presence of carvedilol (K v 4.3 L: V 1/2 = 31.3± 4.0 mV, n = 11, p < 0.0001, K v 4.3 S: V 1/2 = 32.0 ± 3.1 mV, n = 11, p < 0.0001) differed significantly from the measurements under control conditions (K v 4.3 L: The current peak evoked by the second voltage step reflects channel inactivation.Inactivation of the two K v 4.3 isoforms was analyzed by plotting the peak current during the second step of the voltage protocol against the respective test pulse potential.Halfmaximal inactivation voltages after application of 100 µM carvedilol were numerically more positive compared to control conditions without reaching statistical significance.K v 4.3 L half-maximal inactivation voltage was −39.5 ± 1.3 mV under control conditions and −38.3 ± 1.3 mV after incubation with carvedilol (n = 11, p = 0.016), whereas values obtained with K v 4.3 S yielded −40.2 ± 2.1 mV under control conditions and −39.5 ± 1.6 mV after carvedilol administration (n = 11, p = 0.398) (Figure 2E,F).
To reflect ventricular tachyarrhythmias more closely with shorter cycle lengths, activation and inactivation kinetics were additionally analyzed by applying shorter depolarizing steps of 200 ms (Figure 3).Similar to the protocol described above, steps ranging from −100 mV to +50 mV were used from a holding potential of −80 mV with increments of 10 mV.Typical current traces are depicted in Figure 3A,B.Activation and inactivation voltages were analyzed as described before.Half-maximal activation voltages were again significantly different between control conditions (K v 4.3 L: V 1/2 = 25.1 ± 2.1 mV, K v 4.3 S: V 1/2 = 24.6 ± 2.8 mV) and after 50 min incubation with 100 µM carvedilol (K v 4.3 L: V 1/2 = 34.8± 4.8 mV, n = 11, p < 0.0001, K v 4.3 S: V 1/2 = 39.9 ± 15.2 mV, n = 11, p = 0.010) (Figure 3C,D).Furthermore, half-maximal inactivation voltages were slightly more positive after carvedilol administration compared to drug-free conditions, resembling findings obtained with longer voltage pulses.The difference was not significant for both isoforms: The half-maximal inactivation voltage of K v 4.3 was −22.5 ± 1.6 mV (control) and −21.7 ± 1.2 mV after incubation with carvedilol (n = 11, p = 0.123).For K v 4.3 S, values of −23.6 ± 3.1 mV under control conditions and −22.7 ± 1.7 mV after treatment with carvedilol (n = 11, p = 0.464) were calculated (Figure 3E,F).
Next, the recovery from K v 4.3 inactivation was assessed using a multistep protocol.After a depolarization step from a holding potential of −80 mV to +50 mV for 500 ms, a return pulse to the −80 mV holding potential was applied with durations ranging from 10 to 200 ms in 10 ms increments.The variable duration allowed the channel to recover from inactivation and was then followed by a 250 ms long depolarizing step from −80 mV to +50 mV.The peak currents recorded during the final variable voltage step were normalized to the highest peak current and then plotted against the respective duration of the preceding step.Representative traces are depicted in Figure 4A,B.Relative peak currents normalized to the highest peak current after the longest duration of the −80 mV holding potentials (200 ms) did not differ between control measurements and recordings after 100 µM carvedilol activation for 50 min.Time constants of recovery from inactivation were calculated after applying a single exponential fit (Figure 4C,D).Significant differences between time constants under control conditions (τ = 267.0± 25.8 ms) and after 50 min incubation with carvedilol (τ = 505.2± 57.4 ms, n = 10, p = 0.002) were observed with K v 4.3 L. For K v 4.3 S, the increase of the time constant was not significant.Time constants of 229.1 ± 45.3 ms under control conditions and 295.9 ± 20.2 ms in the presence of carvedilol were obtained (n = 10, p = 0.094).Finally, channel deactivation was analyzed by briefly depolarizing the oocyte from −80 mV to +50 mV for 7.5 ms, followed by second 250 ms-long voltage steps ranging from −60 mV to −30 mV in 10 mV increments.Representative traces are shown in Figure 5A,B.Time constants of deactivation were calculated by applying single exponential fits to deactivating currents of the second part of the protocol and were then plotted against the voltage of the second pulse of the voltage protocol.Time constants showed a linear behavior.For control conditions, as well as for the measurement after incubation with 100 µM carvedilol, channel deactivation accelerated with more negative membrane potentials (Figure 5C,D).Carvedilol caused further acceleration of channel deactivation compared to control conditions.Time constants differed significantly between the control measurements and after betablocker incubation for both channel isoforms (n = 11, p < 0.001) (Figure 5C,D).
activating currents of the second part of the protocol and were then plotted against t voltage of the second pulse of the voltage protocol.Time constants showed a linear b havior.For control conditions, as well as for the measurement after incubation with 1 µM carvedilol, channel deactivation accelerated with more negative membrane potenti (Figure 5C,D).Carvedilol caused further acceleration of channel deactivation compar to control conditions.Time constants differed significantly between the control measu ments and after betablocker incubation for both channel isoforms (n = 11, p < 0.001) (Figu 5C,D).

Effects of Bisoprolol on Kv4.3 Channel Isoforms
The impact of the bisoprolol on currents carried by Kv4.3 isoforms was analyzed described earlier using a single step depolarizing voltage protocol.Example traces a

Effects of Bisoprolol on K v 4.3 Channel Isoforms
The impact of the bisoprolol on currents carried by K v 4.3 isoforms was analyzed as described earlier using a single step depolarizing voltage protocol.Example traces are depicted in Figure 6A,B

Effects of Betablockers on the Expression of Kv4.3 Isoforms in HEK Cells
To study the effects of the betablockers on the expression of Kv4.3 protein, bo isoforms were individually expressed in HEK-293T cells (Figure 8A,B).Expression Kv4.3 L and Kv4.3 S whole cell protein was not significantly reduced after incubation w betablockers bisoprolol, carvedilol, or metoprolol, respectively (Figure 8C,D).Prote

Discussion
Carvedilol and metoprolol significantly blocked two cardiac Kv4.3 channels isoforms that contribute to repolarization of the cardiac action potential.By contrast, bisoprolol did not affect Kv4.3 currents, indicating drug-specific actions of carvedilol and metoprolol.

Mechanisms of Block
The rapid development of the block after application of metoprolol and carvedilol indicates a direct interaction between the drug molecules and the respective Kv4.3 isoforms as the primary molecular mechanism of action.This assumption is reinforced by changes in biophysical characteristics of the Kv4.3 current after application of carvedilol.It has been shown that pre-pulse duration affects Kv4.3 inactivation kinetics [23].Thus, we compared longer (2000 ms) pre-pulses to shorter (200 ms) voltage steps assuming faster heart rates during arrhythmia, revealing rate-independent inhibition of Kv4.3 L and Kv4.3 S isoforms by carvedilol.
Kv4.3 L is upregulated and Kv4.3 S is downregulated in HF [4].Therefore, an Ito-targeting antiarrhythmic drug therapy should ideally target the Kv4.3 L isoform.There were no pronounced differences between the effects on Kv4.3 L and Kv4.3 S currents for any of the studied betablockers and biophysical parameters.We conclude from these findings that metoprolol and carvedilol cause Kv4.3 current inhibition of both isoforms via the same molecular mechanism(s).The Kv4.3 L type isoform differs from the S isoform in a 19 amino acid long sequence at the C-terminal, intracellular end of the channel [12,13].This portion of the protein harbors a PKC phosphorylation site.We cannot exclude those indirect differential effects of betablockers on Kv4.3 isoforms via adrenergic signaling pathways under adrenergic stimulation that are beyond scope of this study.

Discussion
Carvedilol and metoprolol significantly blocked two cardiac K v 4.3 channels isoforms that contribute to repolarization of the cardiac action potential.By contrast, bisoprolol did not affect K v 4.3 currents, indicating drug-specific actions of carvedilol and metoprolol.

Mechanisms of Block
The rapid development of the block after application of metoprolol and carvedilol indicates a direct interaction between the drug molecules and the respective K v 4.3 isoforms as the primary molecular mechanism of action.This assumption is reinforced by changes in biophysical characteristics of the K v 4.3 current after application of carvedilol.It has been shown that pre-pulse duration affects K v 4.3 inactivation kinetics [23].Thus, we compared longer (2000 ms) pre-pulses to shorter (200 ms) voltage steps assuming faster heart rates during arrhythmia, revealing rate-independent inhibition of K v 4.3 L and K v 4.3 S isoforms by carvedilol.K v 4.3 L is upregulated and K v 4.3 S is downregulated in HF [4].Therefore, an I totargeting antiarrhythmic drug therapy should ideally target the K v 4.3 L isoform.There were no pronounced differences between the effects on K v 4.3 L and K v 4.3 S currents for any of the studied betablockers and biophysical parameters.We conclude from these findings that metoprolol and carvedilol cause K v 4.3 current inhibition of both isoforms via the same molecular mechanism(s).The K v 4.3 L type isoform differs from the S isoform in a 19 amino acid long sequence at the C-terminal, intracellular end of the channel [12,13].This portion of the protein harbors a PKC phosphorylation site.We cannot exclude those indirect differential effects of betablockers on K v 4.3 isoforms via adrenergic signaling pathways under adrenergic stimulation that are beyond scope of this study.
Bisoprolol and metoprolol counteracted the decreased K v 4.3 expression and the concomitant reduction of I to in different HF animal models [24,25].In our in vitro analysis, the incubation of transfected HEK cells with the respective betablockers did not affect expression levels of the respective K v 4.3 isoforms.It should be noted that HEK-293T cells endogenously express alpha-1 adrenoreceptors [26], and even potentially possible adrenoreceptor-blocking effects did not change expression levels.

Clinical Significance
Betablockers bisoprolol, carvedilol, and metoprolol exert a class effect during treatment of HF patients with reduced ejection fraction, with no apparent evidence for the superiority of any single agent over the others [14,27].However, additional inhibition of potassium channels may suppress cardiac arrhythmias through prolongation of the action potential and by preventing electrical reentry.Multi-channel blocking effects of carvedilol may contribute to these antiarrhythmic effects [18][19][20][21][22]. Indeed, carvedilol was superior to metoprolol in small studies in reducing ICD therapies for ventricular arrhythmias [28] and in avoiding inappropriate ICD therapies [29].
Inhibition of K v 4.3 channels by carvedilol or to lesser extent metoprolol could thus exert beneficial antiarrhythmic effects.The drug concentrations used in our study are apparently higher when compared to the maximum therapeutic plasma concentrations: IC 50 values of 57.1 ± 12.6 µM (L isoform) and 58.2 ± 6.2 µM (S isoform) were obtained for carvedilol.During therapeutic application of carvedilol, maximum plasma concentrations ranging from 0.1-0.6 µM were measured [20] (Table S1 in Supplementary Material).However, compared to mammalian cells, the concentrations for pharmacological ion channel inhibition in Xenopus laevis oocytes tend to be about 5 to 10 times higher [30], indicating that carvedilol effects observed here may be physiologically relevant during drug use in humans.

Potential Limitations
This study focused on acute, direct effects of betablockers on K v 4.3 channels.Potential pharmacological effects on other subunits of I to such as K chip2 were not analyzed.In addition, other mechanisms beyond direct channel binding were not assessed and need to be investigated in future studies, and cardiac cell lines must be modified to assess isoform specific K v 4.3 characteristics [31,32].In contrast to human ventricular cardiomyocytes in heart failure [4,33], differential K v 4.3 isoform expression and remodeling has not been assessed in cardiac cell lines so far.Finally, clinical consequences and the potential differential effect on arrhythmias would have to be investigated in clinical trials with head-to head comparisons.

Drugs
Bisoprolol was dissolved in 87% distilled water and 13% dimethyl sulfoxide, carvedilol in 24% distilled water and 76% dimethyl sulfoxide, and metoprolol in distilled water.The betablockers were stored as 100 mM stock solutions for electrophysiological measurements, and as 10 mM stock solutions for the experiments involving HEK cells at room temperature.For the experiments, stock solutions were diluted to the required concentrations.

Animal Handling and Ethics Statement
Animal studies were performed in compliance with the Guide for the Care and Use of Laboratory Animals, as approved and published by the U.S. National Institutes of Health (NIH publication No. 85 -23, revised 1985) as well as the current version of the German Law on the Protection of Animals.The investigation conforms to the Directive 2010/63/EU of the European parliament.Surgical procedures on female Xenopus laevis frogs were institutionally approved .81/G-270/17)and performed as previously reported [34].

Expression of K v 4.3 Channel Isoforms in Xenopus Laevis Oocytes
DNA encoding K v 4.3 isoforms L and S were introduced into DH5-α bacteria with the help of the plasmid vector pMAX − .Plasmid DNA was then isolated using the QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany) and linearized using PmeI.Next, the DNA was transcribed using T7 DNA polymerase and the mMessage mMachine Kit (Ambion, Austin, TX, USA).The concentration of the transcribed RNA was determined using Nan-oDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA).The RNA was then injected into stages V and VI defolliculated Xenopus laevis oocytes using a Nanoinjector (Nanoject II, H. Saur, Reutlingen, Germanys).The injected volume was 46 nl and the concentration of injected cRNA was 10 ng.The electrophysiological measurements were carried out 2-3 days after the injection.

Expression of K v 4.3 Channel Isoforms in HEK Cell Line
Human embryonic kidney (HEK-293T) cells were cultured at 37 • C with 5% CO 2 .HEK cells were transfected with the 1 µg DNA of K v 4.3 L and S per well using the Lipofectamine TM 3000 Transfection reagent (Invitrogen, Thermo Fisher Scientific Inc. Carlsbad, CA, USA).Twenty-four hours after transfection, the cells were incubated with 10 µM of the different betablockers for another 24 h.

Voltage-Clamp Electrophysiology
Two to three days after RNA injection, the electrophysiological measurements were performed using the two-voltage electrode clamp technique as described before [35].Currents were recorded using an Oocyte Clamp amplifier (Warner OC-725A, Warner Instruments, Hamden, CT, USA) and Pclamp software version 8.2 (Axon Instruments, Foster City, CA, USA).Data were sampled at 2 kHz and filtered at 1 kHz.Voltage clamp electrodes were pulled from glass capillaries (GB100F-10, Science Products GmbH, Hofheim, Germany) using a micropipette puller (P-1000 Next Generation Micropipette Puller, Sutter Instrument, Novato, CA, USA) filled with 3 M of KCl solution; the electrodes hat tip resistances were 5-10 MΩ.The standard extracellular bath solution contained 96 mM NaCl, 4 mM KCl, 1.1 mM CaCl 2 , 1 mM MgCl 2 und 5 mM HEPES and was adjusted to pH 7.4 with NaOH.Before the measurements, the oocytes were preincubated in this solution for at least 20 min.The experiments were carried out under steady gravity-driven perfusion at room temperature.K v 4.3 currents were induced with a single depolarizing voltage step from a holding potential of −80 mV to +50 mV for 250 ms.To evaluate the effect of 100 µM carvedilol on K v 4.3 current, oocytes were first treated with the betablocker for 50 min, followed by washout with a 4 mM K + solution to assess reversibility.Values of measured peak amplitudes were normalized to the value of the peak amplitude obtained during the last measurement prior to drug application.

Data Analysis and Statistics
Data was analyzed using Origin2022 software (OriginLab, Northhampton, MA, USA) and Microsoft Excel software 2021 (Microsoft, Redmond, WA, USA).Data are expressed as mean ± standard error of the mean (SEM).The concentration response curve was fitted with a Hill1 function (y = START + ((END − START) x n )/(k n + x n )).Curves for activation and inactivation were fitted with a Boltzmann function (y = A2 + (A1 − A2)/(1 + exp((x − x0)/dx))).Current traces for the deactivation measurements were fitted with a one-phase exponential decay function with time constant parameter (ExpDec1) (y = y 0 + Ae −x/t ).Curves for the recovery measurements were fitted with a one-phase exponential association equation (ExpAssoc1) (y = Yb + A × (1 − e −(x−TD)/Tau) ).Kolmogorov-Smirnov tests were used to confirm normal distribution of the data.To test the statistical significance, paired Student s t-tests were applied for all the electrophysiological measurements.For the statistical analysis of the Western Blots, ANOVA tests were used.p < 0.05 was considered statistically significant.

Conclusions
Widely used cardioselective betablockers exert differential effects on cardiac repolarizing K v 4.3 channels underlying the I to current.Carvedilol has a strong inhibitory effect on the K v 4.3 isoforms, whereas metoprolol was less effective.Concentrations required for blockade were within upper physiological ranges.Bisoprolol did not have any effect on K v 4.3 currents.Specific electropharmacological actions of carvedilol and metoprolol may be considered when choosing betablockers for HF therapy.

Figure 2 .
Figure 2. Effects of carvedilol (100 µM) on activation and inactivation of the Kv4.3 isoforms.(A,B) Representative current traces of Kv4.3 L (A) and S (B) evoked by the indicated voltage protocol prior to (gray) and after (black) carvedilol application (50 min).(C,D) I-V-plots for activation of Kv4.3 L

Figure 3 .
Figure 3. Effects of carvedilol (100 µM) on the activation and inactivation of the two Kv4.3 isoforms with short depolarizing steps.(A,B) Representative current traces of Kv4.3 L (A) and S (B) induced by the indicated voltage protocol prior to (gray) and after (black) carvedilol application (50 min).

Figure 3 .
Figure 3. Effects of carvedilol (100 µM) on the activation and inactivation of the two K v 4.3 isoforms with short depolarizing steps.(A,B) Representative current traces of K v 4.3 L (A) and S (B) induced by the indicated voltage protocol prior to (gray) and after (black) carvedilol application (50 min).(C,D) I-V plots for activation of K v 4.3 L (C) and K v 4.3 S (D) (n = 11).(E,F) I-V-plots for inactivation of K v 4.3 L (E) and K v 4.3 S (F) (n = 11).* p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4 .
Figure 4. Effects of carvedilol (100 µM) on the recovery from inactivation of the two Kv4.3 isoform (A,B) Representative Kv4.3 L (A) and Kv4.3 S (B) current traces of induced by the indicated volta protocol prior to (gray) and after (black) carvedilol application (50 min).(C,D) Recovery fro

Figure 4 .
Figure 4. Effects of carvedilol (100 µM) on the recovery from inactivation of the two K v 4.3 isoforms.(A,B) Representative K v 4.3 L (A) and K v 4.3 S (B) current traces of induced by the indicated voltage protocol prior to (gray) and after (black) carvedilol application (50 min).(C,D) Recovery from inactivation curves were calculated by plotting peak current amplitudes against the duration of the preceding repolarizing step for K v 4.3 L (C) and K v 4.3 S (D) (n = 10).* p < 0.05, ** p < 0.01.

Figure 5 .
Figure 5. Effects of carvedilol (100 µM) on deactivation of K v 4.3 isoforms.(A,B) Representative current traces of K v 4.3 L (A) and K v 4.3 S (B) induced by indicated voltage protocol prior to (gray) and after (black carvedilol application (50 min)).(C,D) Deactivation time constants for K v 4.3 L (C) and K v 4.3 S (D) (n = 11).*** p < 0.001.

Figure 6 .
Figure 6.Effects of bisoprolol (100 µM) on K v 4.3 isoforms.(A,B) Representative K v 4.3 L (A) and K v 4.3 S (B) current traces of prior to (gray) and after (black) bisoprolol application (50 min).(C,D) Relative peak current carried by K v 4.3 L (n = 11) (C) or K v 4.3 S (n = 9) (D) during application of 100 µM bisoprolol.

Figure 8 .
Figure 8. Effects of betablockers on K v 4.3 isoform expression in HEK-293T cells.(A) Representative Western blots after transfection with K v 4.3 L (A) or K v 4.3 S (B) before and after incubation with respective betablockers (100 µM) for 24 h (n = 6 each).(C,D) Protein quantification of K 4.3 L (C) and K v 4.3 S (D) protein relative to respective controls and normalized to GAPDH.