Inhibitory Effects of Glycyrrhetinic Acid on the Delayed Rectifier Potassium Current in Guinea Pig Ventricular Myocytes and HERG Channel

Background. Licorice has long been used to treat many ailments including cardiovascular disorders in China. Recent studies have shown that the cardiac actions of licorice can be attributed to its active component, glycyrrhetinic acid (GA). However, the mechanism of action remains poorly understood. Aim. The effects of GA on the delayed rectifier potassium current (I K), the rapidly activating (I Kr) and slowly activating (I Ks) components of I K, and the HERG K+ channel expressed in HEK-293 cells were investigated. Materials and Methods. Single ventricular myocytes were isolated from guinea pig myocardium using enzymolysis. The wild type HERG gene was stably expressed in HEK293 cells. Whole-cell patch clamping was used to record I K (I Kr, I Ks) and the HERG K+ current. Results. GA (1, 5, and 10 μM) inhibited I K (I Kr, I Ks) and the HERG K+ current in a concentration-dependent manner. Conclusion. GA significantly inhibited the potassium currents in a dose- and voltage-dependent manner, suggesting that it exerts its antiarrhythmic action through the prolongation of APD and ERP owing to the inhibition of I K (I Kr, I Ks) and HERG K+ channel.


Introduction
Cardiac arrhythmias are associated with significant morbidity and mortality in developed and developing countries. Although nonpharmacologic approaches, ablative therapy, and implantable defibrillators, for example, are being used more and more commonly as complementary and alternative interventions for the treatment of cardiac arrhythmias, drug therapy was traditionally the mainstay of arrhythmia treatment. Improved understanding of the cardiomyocyte ion channels holds the promise of identifying novel targets for the treatment of cardiac arrhythmias. Furthermore, the limited efficacy and the potency of provoking life-threatening arrhythmias of present drugs have generated interest in finding new antiarrhythmic agents [1].
Potassium channels constitute the most abundant family of ion channels involved in cardiac physiological or pathophysiological processes, disruptions of which could give rise to prolongation of the action potential and potentiate occurrence of lethal arrhythmias subsequently [2,3]. In the hearts of many mammalian species including humans, delayed rectifier K + current ( K ), the major outward potassium current responsible for ventricular repolarization, can be divided into at least two different components, rapidly activating ( Kr ) and slowly activating ( Ks ). The human ethera-go-go-related gene (HERG) encodes the -subunit for the rapid delayed rectifier potassium channel ( Kr ) in cardiac myocytes. Kr and Ks are pivotal in cardiac repolarization, especially in the later phases of the action potential [4,5]. Although the structural and functional aspects of potassium channels have been widely examined, only a few K + channel modulators are being used clinically at present for the unwanted adverse effects [2,6]. As an alternative to find new antiarrhythmic agents, there has been increased interest in investing natural compounds that are effective against cardiac arrhythmia.

Evidence-Based Complementary and Alternative Medicine
Glycyrrhiza radix is a commonly prescribed herb to prevent palpitations in Chinese traditional medicine for about 2000 years, derived from the dried roots and rhizomes of Glycyrrhiza uralensis, G. glabra, and G. inflata. Glycyrrhizin, the major constituent of G. glabra, is a glycoside, which occurs as an admixture of sodium, potassium and calcium salts [7]. Orally administered, glycyrrhizin is poorly absorbed by the intestinal tract and is hydrolyzed by -D-glucuronidasecontaining intestinal bacteria to yield two molecules of Dglucuronic acid and the aglycone glycyrrhetinic acid (GA), a pentacyclic triterpene [8]. If intravenously administered, GA is metabolized in the liver by lysosomal -D-glucuronidase to the 3-monoglucuronide of glycyrrhetinic acid. This metabolite is excreted via the bile into the intestine, where it is transformed by bacteria into GA, which can be reabsorbed, causing a pronounced delay in terminal plasma clearance [9].
The use of Glycyrrhiza uralensis as a pharmacological remedy dates back far into the past [10]. Various pharmacological properties of licorice have been proved including cardioprotective [11,12], antiulcer, anti-inflammatory, spasmolytic, antioxidative, contravariant, antiviral, anticancer, and hepatoprotective effects, as well as eliminating phlegm and reinforcing memory [7,[13][14][15]. Many components have been isolated from licorice including triterpene saponins, flavonoids, isoflavonoids, and chalcones. Triterpene saponins are the main components of Glycyrrhiza radix and its pharmacological activities are comparatively well understood and clear.
Recently, wide-ranging studies have provided evidence that GA is cardioprotective; this action involves different pathways. In rat cardiac mitochondria [16], GA was shown to increase permeability and concomitant release of proapoptotic factors. In particular, GA, acting as a gap junction inhibitor, influences connexin 43, the major gap junctionforming protein in adult cardiac ventricles and a regulator of mitochondrial function [17]. Studies have demonstrated that GA and its derivatives affect the inotropic, lusitropic, chronotropic, and coronary performances of the mammalian heart and the signal transduction pathways that could be involved [9]. Furthermore, research has demonstrated that GA reduces cardiac sodium currents [18], particularly the late Na . These findings might help to elucidate the traditional use of licorice in therapy for cardiovascular disorders [19]. It has also been reported that 18 -GA has significant potential for development as a novel antiarrhythmic agent and for treating myocardial ischemia by preferentially blocking the Na,L [20]. However, the effects of GA on the potassium channel, especially the rapid and slow components of the delayed rectifier potassium current, are not clearly defined.
In this study, we investigated the effects of GA on the K ( Kr , Ks ) in guinea pig ventricular myocytes. We also extended our study to investigate the effects of GA on the human ether-à-go-go-related gene (HERG) K + channel current expressed in HEK-293 cells. Our study could provide theoretical support for developing the significant potential of GA as a novel antiarrhythmic agent. To record HERG activity in the HEK 293 cells, the internal solution contained (mM) KCl 130, MgCl 2 1, HEPES 10, EGTA 10, and MgATP 5 (adjusted to pH 7.25 with KOH). The external solution contained (mM) NaCl 136, KCl 5.4, CaCl 2 1, MgCl 2 1, HEPES 10, and glucose 10 (adjusted to pH 7.36-7.38 with NaOH).

Isolation of Single Ventricular Myocytes.
Adult guinea pigs were fully anesthetized by intraperitoneal injection of urethane (40 mg/kg). The hearts were quickly removed and mounted on a Langendorff column, and cardiac myocytes were isolated as follows. The heart was dissected and rinsed in cold oxygenated Ca 2+ -free Tyrode's solution and then perfused in the Langendorff apparatus at 37 ∘ C. Perfusion with Ca 2+ -free Tyrode's solution for 5 min was followed first by 25 ± 5 min perfusion with low Ca 2+ (0.1 mmol/L) Tyrode's solution containing 0.04-0.06 g/L collagenase and 0.5-0.8 g/L BSA and then 5 min perfusion with collagenasefree Tyrode's solution containing 0.5-0.8 g/L BSA. The heart was then minced and the cells were filtered through 200 m nylon mesh, resuspended in Kraft-Brühe (KB) solution, and stored at room temperature (22-25 ∘ C) until use. (DMEM, Hyclone) supplemented with 10% (v/v) fetal bovine serum (FBS, Gibco) and 1 mg/mL geneticin (G-418, Gibco), the cultures were passed every 3-5 days by use of a brief trypsin treatment. The cells were maintained at 37 ∘ C in 5% CO 2 and plated on a glass culture dish 2-3 days before electrophysiological experiments.

Electrophysiological
Recordings. All currents were recorded using the conventional whole-cell patch-clamp technique. Borosilicate glass electrodes had tip resistances of 1-3 MΩ when filled with the pipette solution. All experiments were performed at room temperature (22-23 ∘ C) using an Axopatch 200B amplifier (Axon Instrument, USA). Adequate series resistances (less than five times the pipette resistances) were usually attained within 10 min after the gigaohm seal was formed. Measurements were taken using an Axopatch 200B amplifier (Axon Instruments). The current signals were filtered via a 4 kHz, 4-pole low-pass filter and digitized with an AD-DA converter (Digidata 1440, Axon Instruments) for subsequent analysis using pCLAMP 10.0 software.

Statistical Analysis.
The data were analyzed with the use of Data Processing System (Version 7.05). All data are expressed as means ± SEM. Paired Student's t-tests were used for statistical comparisons when appropriate, and differences were considered significant at < 0.05.

Effect of GA on the of Guinea Pig Ventricular Myocytes.
We first tested the effect of GA on K using guinea pig ventricular myocytes. K were recorded by applying voltage pulses ranging from −10 to +80 mV for 5 s from the holding potential of −40 mV, and the repolarization potential was maintained at a constant −30 mV for the K,tail analysis. Figure 1(a) shows an example of a voltage-clamp recording from a single ventricular myocyte, with representative current traces given under control conditions and after exposure to 10 M GA. Under control conditions, the depolarizing steps activated time-dependent outward currents. The amplitude of the outward currents measured at the end of the pulse ( K ) increased with increasingly positive voltage steps. The current-voltage relationships for K,step and K,tail obtained at various concentrations of GA are shown in Figures 1(b) and 1(c). As the concentration of GA increased, the amplitude of K,step and K,tail decreased dose dependently (Figures 1(d) and 1(e)). The K,step measured at +80 mV was 6.97 ± 0.34 pA/pF under control conditions and decreased to 4.93 ± 0.51 pA/pF, 3.35 ± 0.55 pA/pF, and 2.12 ± 0.29 pA/pF after application of 1, 5, and 10 M GA, respectively ( = 5, < 0.05). Meanwhile, the K,tail evoked after repolarization to −30 mV was 1.57±0.11 pA/pF under control conditions and decreased to 1.24 ± 0.12 pA/pF, 0.67±0.15 pA/pF and 0.48±0.06 pA/pF in the presence of 1, 5 and 10 M GA, respectively ( = 5, < 0.05). These results show that GA dose-dependently blocked K,step and K,tail .

Effect of GA on I and I in Guinea Pig Ventricular
Myocytes. To investigate the effects of GA on the rapid and slow components of the delayed rectifier currents in guineapig ventricular myocytes, we used a voltage clamp protocol designed to separate the currents electrophysiologically. The holding potential was maintained at −40 mV. Our results revealed that depolarization to +60 mV activated both Kr and Ks . Repolarization to 0 mV revealed Ks as a deactivating Ks,tail , and subsequent repolarization to −40 mV resulted in deactivation of Kr . We confirmed that 2 M E-4031, a selective blocker of Kr , blocked the rapid component of the delayed rectifier K + current but had no effect on Ks ( Figure 2). As shown in Figures 2(b) and 2(c), 1, 5, and 10 M GA dose dependently inhibited Ks,tail and Kr,tail . In the cells examined, Ks,tail measured at 0 mV was 0.59 ± 0.03 pA/pF under control conditions and decreased to 0.53 ± 0.06 pA/pF, 0.33 ± 0.06 pA/pF, and 0.16 ± 0.03 pA/pF after application of 1, 5, and 10 M GA, respectively ( = 5, < 0.05). The Kr,tail evoked after repolarization to −40 mV was 1.57 ± 0.11 pA/pF under control conditions and decreased to 0.57 ± 0.06 pA/pF, 0.29 ± 0.02 pA/pF and 0.21 ± 0.05 pA/pF in the presence of 1, 5 and 10 M GA, respectively ( = 5, < 0.05). This result shows that GA blocks the rapid and slow components of delayed rectifier K + current.

Inhibition of HERG K + Currents and Concentration-
Dependent Block by GA. The HERG current was elicited from the holding potential of −80 mV by test pulses ranging from −60 to +30 mV in 10 mV steps. Each test pulse was followed by a repolarization step to −50 mV, which evoked large, slowly decaying outward tail currents. Currents were recorded first under control conditions, and then GA (5 M) was washed into the bath for 10 min, with the cell kept at the holding potential, before current recording commenced in the presence of the drug (Figure 3(a)). In the absence of drug, the I-V relationship exhibited the characteristic bellshaped curve increasing from −40 to 0 mV, and owing to the fast C-type inactivation of HERG channels, it decreased with further depolarization. GA reduced both the HERG current ( step ) during the test potentials and the tail current ( tail ) after a test pulse to 0 mV (Figures 3(c) and 3(d)). GA inhibited the tail current at all potentials, although voltage dependence was evident with a significantly weaker block at more negative potentials (Figure 3(d)).

Effects of GA on HERG Channel
Kinetics. Drugs that block ion channels often alter the voltage dependence or kinetics of channel gating. Therefore, we examined the effects of GA on the voltage dependence of activation and rectification and on the kinetics of inactivation and deactivation. The activation curves were constructed by normalizing the tail currents recorded with the protocol used in Figure 3(a). The activation curve showed that the threshold voltage for HERG current activation was close to −50 mV and that it was fully activated with voltage steps to −10 mV. The rate of activation  was similar before and after exposure to 5 M GA. 1/2 values were −19.83 ± 2.36 mV in the control and −22.56 ± 1.84 mV in GA ( > 0.05, = 5). Thus, 5 M GA had little effect on the voltage dependence of activation (Figure 3(b)).
To measure inactivation, a special protocol was used that inactivated the channel at a holding potential of 40 mV, recovered the channel from inactivation at various potentials from −120 to 20 mV in 10 mV steps, and measured the resulting peak outward current at constant 20 mV as a measure of steady-state inactivation ( Figure 5 The effect of GA on the onset of inactivation of the HERG current was investigated using a three-pulse protocol. The channels were first inactivated by clamping the membrane at 40 mV followed by a prepulse to −100 mV. This prepulse was sufficiently long to allow rapid recovery of channels from inactivation but short enough to prevent significant channel deactivation. Following recovery from the prepulse, a series of test pulses were delivered to potentials ranging from −120 to 20 mV, resulting in outward inactivating currents (Figure 6(a)). The time constants for the onset of inactivation were obtained by fitting exponential functions to the decaying current traces during the third pulse of the protocol and were significantly decreased following perfusion with 5 M GA ( < 0.05, = 5) (Figure 6(c)).
To determine recovery from inactivation, the fully activated I-V protocol shown in Figure 6(b) was used: a depolarizing pulse to 40 mV to inactivate the HERG channels, followed by different repolarizing pulses to test potentials between −120 and 20 mV in 10 mV steps. The prepulse potential at 40 mV was positive enough to induce full conductance of the HERG channels but also inactivated many of the channels. The rate of recovery from inactivation was obtained by fitting a single exponential to the initial increase in tailcurrent amplitude, whereas the time constant of deactivation was ascertained by fitting a single exponential to the decay   of the tail current. However, 5 M GA did not change the deactivation rate significantly ( > 0.05, = 5) (Figure 6(d)).

Discussion
Cardiac K + channels play a pivotal role in maintaining normal cardiac electrical activity. They regulate the resting membrane potential and excitability, participate in repolarization, and determine the shape and duration of cardiac action potential. A malfunction of the K + channels due to either gene mutations or drug blockade alters not only cardiomyocyte excitability but also the electrical balance of depolarization and repolarization, which causes a long or short QT interval in the electrocardiogram (ECG) and underlies different types of cardiac arrhythmia [21,22]. Therefore, cardiac K + channels are important targets for antiarrhythmic drugs.
In the present study, we have firstly provided the evidence that the antiarrhythmic ionic mechanism of GA is related to the inhibition of potassium currents ( Kr , Ks ) in guinea pig ventricular myocytes. The cardiac ion channel gene products  that are targets for GA are unknown. However, previous studies have demonstrated that the peak and late Na , studied in the Xenopus oocytes expressing either human Na V 1.5 or mutant Na V 1.5-ΔKPQ of the -subunit channel, were strongly reduced by GA [19], resulting in a prolongation of the action potential. We therefore tested GA for a possible block of HERG expressed in HEK293 cells. The results confirmed that GA blocks HERG.
One of the main objectives of studying ionic channels is to provide a theoretical basis for the clinical treatment of tachyarrhythmia. Many active ingredients of Chinese medicine could block the HERG channel, decrease K , and lead to the acquired long QT syndrome, which makes the channel the therapeutic target for anti-arrhythmia [23]. It is thought that HERG channel inactivation is important in channel blocking by some (but not all) drugs, either by increasing drugbinding affinity or by facilitating the optimal orientation of the S6 aromatic residues to which drugs bind. In this study, the extent of channel inactivation was significantly altered ( Figure 5(c)), and the time course of inactivation seemed to be accelerated ( Figure 5(b)). Furthermore, the time constants for the onset of inactivation were significantly smaller following perfusion with GA ( Figure 6(c)). These effects of GA on channel kinetics were consistent with affinity for the inactivated state, suggesting that GA blocks the HERG channel by affecting its inactivation but not its activation. Our results demonstrated that GA has inhibitory effects on Kr , Ks in guinea pig ventricular myocytes, and the HERG potassium channel, and the inhibition was in a concentration-dependent manner. The increase in action potential duration induced by GA is mainly due to its blocking effects on Kr , Ks , and HERG and are its major mechanisms of antiarrhythmic action. Kr and Ks are important in cardiac repolarization [24]. Kr channels open rapidly upon depolarization of the action potential but are quickly inactivate. The channel inactivation is released following repolarization with a slow deactivation [25]. Owing to this inward rectification property, Kr contributes a little during the plateau of the cardiac action Evidence-Based Complementary and Alternative Medicine 9 potential and progressively increases at phase 3 repolarization of the action potential [26]. Therefore, Kr is pivotal in cardiac repolarization, especially in the later phases of the action potential, due to its unique kinetics. Ks activates slowly with almost no inactivation after activation [25,27] and contributes to the phase 2 slow repolarization of cardiac action potential. Ks has been demonstrated in cardiac tissues/myocytes from various species including human [24,26,27]. The physiological contribution of Ks to the human ventricular action potential is limited; however, during tonic sympathetic stimulation or when the cardiac repolarization reserve is attenuated, Ks becomes important in limiting APD prolongation owing to its slow deactivation [26]. Ks is expressed heterogeneously in different regions of the heart. In the canine ventricle, Ks density is greater in epicardial and endocardial cells than in the M cells [28,29]. This lower Ks density in the M cells is considered to be related to the steeper APD rate relationship and their greater tendency to display longer APD and to develop EADs at slow heart rates or in response to QT-prolonging drugs [29].
Any abnormality in channel density or function (upor downregulation) may result in changes of currents and APD, even inducing arrhythmias. The inhibition of Kr , Ks , and HERG by GA would induce a prolongation of APD, which could contribute to its antiarrhythmic actions because prolongation of APD could prevent or terminate the reentrant excitation and prolong the refractory period. Class III antiarrhythmic agents, blockers of Kr such as dsotalol, exert a proarrhythmic effect with reverse frequencydependent manner, which is largely attenuated at fast rates and enhanced at lower stimulation frequencies. This reverse frequency-dependent effect [30] could lead to an increase in the dispersion of repolarization and favor the occurrence of cardiac arrhythmias. Jurkiewicz and Sanguinetti [31] proposed that the reverse frequency-dependent effect on APD of typical class III agents is a consequence of selective blockade of Kr . Kr blockers prolong atrial and ventricular APD and the QT interval and can cause TdP, which can degenerate into ventricular fibrillation and sudden cardiac death [2]. Proarrhythmia induced by Kr blockers is related to [32][33][34][35] (a) excessive prolongation of APD near plateau voltages, which favor the development of early after-depolarizations; (b) a more marked prolongation of the APD in M cells than in subepicardial or subendocardial ventricular muscle, possibly because of the relative scarcity of Ks in M cells [29]. Thus, triggered focal activity and ventricular reentry associated with increased inhomogeneity of repolarization across the ventricular wall would lead to the development of TdP [3,4]. It has been suggested that Ks accumulation at increased frequencies decreases the relative importance of Kr , reducing the impact of Kr blockade on APD prolongation. The authors suggested that the compounds which inhibit Ks might be devoid of reverse use-dependence. Actually, the agent that blocks both components of K might have a more consistent effect on action potentials at different frequencies and a better safety profile than a specific Kr blocker [36].
Control of cardiac electrical activity is well organized by an array of ion channels activated with a delicate balance between inward and outward ion currents [37]. Upon receiving an incoming impulse, cardiac cells are excited with rapid membrane depolarization followed by a relatively slow repolarization. Repolarization disorders, either excessive slowing or acceleration of the rate, can cause electrical perturbations resulting in cardiac arrhythmias, while excessive blockade of the HERG channel might increase the risk of arrhythmogenic activities. A good antiarrhythmic drug should affect multiple channels and keep invalids free from further episodes of arrhythmia [23]. Taking this into consideration, the drugs currently available in clinics are not satisfactory. Previous studies have revealed that GA can block peak and late Na [19,20]. In our task group, the results suggested that GA not only blocks Kr (HERG) and Ks but also inhibits Ca-L . The effect of GA on multiple channels might make it a promising antiarrhythmic that can lead the cardiac cell to restore normal sinus rhythm and prevent further arrhythmia. Further basic and clinical studies will be needed to explore whether GA has proarrhythmic actions.

Conclusion
In conclusion, our study demonstrated that GA has an inhibitory effect on K ( Kr , Ks ) and the HERG potassium channel expressed in HEK293 cells. The results indicate that the antiarrhythmic activity and prolonged action potential of GA could be due to the blocking of K ( Kr , Ks ) and the HERG channel. It is thought that a good antiarrhythmic drug should affect multiple channels and be able to restore normal sinus rhythm and keep patients free from further episodes of arrhythmia. GA not only blocks the HERG channel, K ( Kr , Ks ), but also inhibits Na , so the results reveal that it has significant potential for development as a novel antiarrhythmic agent, particularly targeting the genesis of arrhythmias. Therefore, our findings might help to elucidate the traditional use of licorice in the treatment of cardiovascular disorders. Nonetheless, further evaluation of the therapeutic potential of GA is warranted.