Differential Effects of Ginsenoside Metabolites on HERG K⁺ Channel Currents

Abstract

The human ether-a-go-go-related gene (HERG) cardiac K⁺ channels are one of the representative pharmacological targets for development of drugs against cardiovascular diseases such as arrhythmia. Panax ginseng has been known to exhibit cardioprotective effects. In a previous report we demonstrated that ginsenoside Rg₃ regulates HERG K⁺ channels by decelerating deactivation. However, little is known about how ginsenoside metabolites regulate HERG K⁺ channel activity. In the present study, we examined the effects of ginsenoside metabolites such as compound K (CK), protopanaxadiol (PPD), and protopanaxatriol (PPT) on HERG K⁺ channel activity by expressing human α subunits in Xenopus oocytes. CK induced a large persistent deactivating-tail current (I_{deactivating-tail}) and significantly decelerated deactivating current decay in a concentration-dependent manner. The EC₅₀ for persistent I_{deactivating-tail} was 16.6±1.3 μM. In contrast to CK, PPT accelerated deactivating-tail current deactivation. PPD itself had no effects on deactivating-tail currents, whereas PPD inhibited ginsenoside Rg₃-induced persistent I_{deactivating-tail} and accelerated HERG K⁺ channel deactivation in a concentration-dependent manner. These results indicate that ginsenoside metabolites exhibit differential regulation on I_{deactivating-tail} of HERG K⁺ channel.

INTRODUCTION

K⁺ channels play critical roles in a wide variety of physiological processes, including the regulation of neurotransmitter release, neuronal excitability, heart rate, muscle contraction, hormone secretion, epithelial electrolyte transport, cell volume, and cell proliferation [1]. Cardiomyocytes contain two kinds of delayed rectifier K⁺ channels, which are important for cardiac repolarization after cardiac action potential and shorten the action potential duration [2]. Thus, the human ether-a-go-go-related gene (HERG, I_Kr) and KVLQT (KCNQ) (I_Ks) K⁺ channels are mainly responsible for repolarization of the heart cardiac action potential [3]. The genetic or pathological dysfunctions of HERG or KCNQ K⁺ channels are one of the primary triggers of cardiac diseases such as arrhythmias. Since arrhythmias are a major cause of sudden cardiac death in the world [2], HERG as well as KCNQ K⁺ channels are the principal pharmacological targets for development of therapeutic drugs against cardiovascular diseases including arrhythmias.

Ginseng, the root of Panax ginseng Meyer, has been used as a general tonic to promote longevity and enhance bodily functions against stress, fatigue, diseases, cancer and diabetes mellitus [4]. Among various ginseng components, ginsenosides (also called ginseng saponins) exhibit anti-hypertension and cardio-protective effects [5]. In a previous study, we demonstrated that ginsenoside Rg₃ (Rg₃) enhanced cardiac I_Ks channel currents, which consist of KCNQ1 plus KCNE1 subunits in both concentration- and voltage-dependent manners [6]. In addition, we also demonstrated that Rg₃ enhanced outward currents (I_HERG) and transient tail currents (I_tail). Rg₃ induced a large persistent deactivating-tail current (I_{deactivataing-tail}) and significantly decelerated deactivating current decay [7].

On the other hand, ginsenosides administered via the oral route can pass into the large intestine without being broken down by either gastric juices or digestive enzymes [8]. But protopanaxadiol (PPD) ginsenosides are metabolized by intestinal microorganisms into compound K (CK) with one glucose at the C-20 position and are further metabolized to form PPD without glucose, whereas protopanaxatriol (PPT) ginsenosides are metabolized to PPT that only maintains the backbone structure of ginsenosides without any carbohydrate component (Fig. 1). The purpose of this study was to investigate how ginsenoside metabolites affect HERG K⁺ channel activity. We report here that CK decelerated HERG K⁺ channel deactivation, whereas PPT accelerated HERG K⁺ channel deactivation. PPD itself had no effect on HERG K⁺ channel activity but antagonized Rg₃-mediated HERG K⁺ channel regulations. These results indicate that ginsenoside metabolites exhibit differential regulation on HERG K⁺ channel activity.

Fig. 1. Chemical structures of ginsenoside Rg₃ and ginsenoside metabolites used in this study. CK, compound K; PPD, protopanaxadiol; PPT, protopanaxatriol; Glc, glucopyranoside.

MATERIALS AND METHODS

Materials

The individual ginsenoside Rg₃ and ginsenoside metabolites such as PPD, PPT, and CK were provided by AMBO Institute (Seoul, Korea) (Fig. 1A). The cDNAs for human HERG K⁺ channels (accession no. U04270) were kindly provided by Dr. Pongs (University of Hamburg, Germany). All other reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Preparation of Xenopus oocytes and microinjection

Xenopus laevis frogs were purchased from Xenopus I (Ann Arbor, MI, USA). Their care and handling were in accordance with the highest standards of institutional guidelines of Konkuk University. For isolation of oocytes, frogs were anesthetized with an aerated solution of 3-amino benzoic acid ethyl ester followed by removal of ovarian follicles. The oocytes were treated with collagenase and then agitated for 2 h in Ca²⁺-free medium containing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl₂, 5 mM HEPES, 2.5 mM sodium pyruvate, 100 units/mL penicillin and 100 μg/mL streptomycin. Stage V-VI oocytes were collected and stored in ND96 medium (in mM: 96 NaCl, 2 KCl, 1 MgCl₂, 1.8 CaCl₂, and 5 HEPES, pH 7.5) supplemented with 50 μg/mL gentamicin. The oocyte-containing solution was maintained at 18℃ with continuous gentle shaking and renewed daily. Electrophysiological experiments were performed within 5-6 days of oocyte isolation, with ginsenoside Rg₃ or ginsenoside metabolites added to the bath. For HERG K⁺ channel experiments, HERG K⁺ channel-encoding cRNAs (40 nL) were injected into the animal or vegetal pole of each oocyte one day after isolation, using a 10 μl microdispenser (VWR Scientific, San Francisco, CA, USA) fitted with a tapered glass pipette tip (15-20 μm in diameter) [9].

Data recording

A custom-made Plexiglas net chamber was used for two-electrode voltage-clamp recordings as previously reported [7]. The oocytes were impaled with two microelectrodes filled with 3M KCl (0.2-0.7 MΩ), and electrophysiological experiments were carried out at room temperature using an Oocyte Clamp (OC-725C; Warner Instruments, Hamsden, CT, USA). Stimulation and data acquisition were controlled with a pClamp 8 (Axon Instruments, Union City, CA, USA). For most electrophysiological experiments, oocytes were initially perfused with ND96 solution (in mM: 96 NaCl, 3 KCl, 2 CaCl₂, 5 HEPES, pH 7.4 with NaOH) to obtain control current recordings. The oocytes were then clamped at a holding potential of –90 mV, the membrane potential was depolarized to 0 mV for 4 s, followed by repolarization to –60 mV at 20 s intervals, and current values were recorded.

Data analysis

To obtain the concentration-response curve of the respective ginsenoside metabolites effects on the K⁺ current from the HERG K⁺ channel, the peak amplitudes at various concentrations of the respective ginsenoside metabolites were plotted, and Origin software (Origin, Northampton, MA, USA) was used to fit the plot to the Hill equation: y/y_max=[A]^nH/([A]^nH+[IC₅₀]^nH), where y is the peak current at a given concentration of ginsenoside metabolite, y_max is the maximal peak current, EC₅₀ is the concentration of the respective ginsenoside metabolite producing a half-maximal effect, [A] is the concentration of the respective ginsenoside metabolite, and nH is the Hill coefficient. All values were presented as the mean±SEM. The significance of differences between mean control and treatment values was determined using Student’s t-test, where p<0.05 was considered statistically significant.

RESULTS AND DISCUSSION

Effects of CK on HERG K⁺ channel currents

We first examined the effect of intermediate ginsenoside metabolite CK on the HERG K⁺ channel currents using a Xenopus oocyte gene expression system. HERG K⁺ channel currents were recorded using the two-electrode voltage-clamp recording technique at room temperature [10]. Throughout these experiments, the holding potential was maintained at –90 mV and repolarized to –60 mV after depolarizing pulses from 0 mV, unless otherwise indicated. Fig. 2A gives an example of a voltage-clamp recording with the representative current traces of I_HERG at end of depolarization, transient peak I_tail, and slow I_{deactivating-tail} as indicated by arrows in the absence or presence of various concentrations of CK. CK had almost no effect on I_HERG and I_tail (Fig. 2A). Interestingly, CK showed a major effect on slow I_{deactivating-tail}. Thus in the presence of CK, I_{deactivating-tail} failed to decay, and instead, a large persistent outward current developed. CK (100 μM) increased slow I_HERG by an average of 97.8±8.13% compared to control (Fig. 2A). CK response was completely reversible upon washing the oocytes with ND96 (data not shown). Next, we examined the concentration-dependent effect of CK on I_HERG, I_tail, and slow I_{deactivating-tail}. As shown in Fig. 2B, CK had a negligible effect on both I_HERG and I_tail. CK also increased a persistent I_{deactivating-tail} in a concentration-dependent manner. The EC₅₀ value was16.6±1.3 μM for slow I_{deactivating-tail}.

Fig. 2. Effects of compound K (CK) on I_HERG, I_tail, and I_{deactivating-tail}. (A) Representative current traces on human ether-a-go-go-related gene (HERG) K⁺ channel enhancements by various concentrations of CK. Currents were in response to 4-s voltage steps to 0 mV from a holding potential of –90 mV, followed by repolarization to –60 mV. I_HERG was obtained at the end of depolarization; I_tail was obtained at beginning of repolarization; slow I_{deactivating-tail} was obtained at the end of repolarization as indicated by arrow in (A). (B) Concentration-response curves for the activation of HERG K⁺ currents by CK on I_HERG, I_tail, and I_{deactivating-tail}. Solid lines were fitted to the Hill equation. CK was potent for the enhancement of I_{deactivating-tail}. Bars represent the means±SEM (n=5-7). (C) The representative control (upper traces) and 100 μM CK-mediated currents (down traces) in I-V relationship. Currents were recorded at test potential from –60 to + 50 mV. I_tail and I_{deactivating-tail} were recorded after repolarization to –60 mV. (D) I-V relationships for HERG K⁺ currents measurement at the end of the 4-s test pulse before and after application of 3, 10, and 30 μM CK (n=5). Currents were normalized to the control current at 0 mV for each oocyte. Data are represented by the means±SEM (n=7). (E) Effects of CK on the steady-state activation curve for HERG K⁺ channel. I_tail were normalized to the peak current under each condition, and the data were fitted with a Boltzmann function. Treatment of CK (10, 30, and 100 μM) did cause a leftward shift. Con, control.

Next, we investigated the HERG K⁺ current-voltage (I-V) relationship using CK (Fig. 2C, D). The amplitude of I_HERG measured at the end of depolarization increased with increasing positive voltage steps, reaching a maximum at –10 mV and then progressively decreased with further voltage steps in the control, indicating that HERG K⁺ channel is activated with rectified manner (Fig. 2C, D). When the amplitude of I_HERG was normalized to the maximum amplitude of the I_HERG obtained under the control conditions and was plotted against the potential of the step depolarization (Fig. 2C, D), the presence of various concentrations of CK did not show a significant increase of I_HERG even at high concentrations of CK, indicating that CK mainly affects slow I_{deactivating-tail}.

We next examined whether CK affects steady-state activation (Fig. 2E). Treatment of 10 and 30 μM CK caused significant leftward shifts in steady-state activation curves compared to control oocytes. The V_1/2 values were –25.5±0.39 mV in the control and –24.6±0.43, –30.9±0.50 and –32.4±0.82 mV in the presence of 3, 10, and 30 μM CK-treated groups, respectively (p<0.05 at 10 and 30 μM Rg₃ compared to control, n=5). The slope factors (kg) were not significantly changed, yielding values of 6.7±0.34 in control and 6.3±0.37, 6.1±0.44 and 6.5±0.70 in the presence of 3, 10 and 30 μM CK-treated groups, respectively. We also examined CK (30 μM) and observed that it did not affect steady-state inactivation (data not shown). Therefore, these results indicate that CK affects the activation gating rather than inactivation gating of HERG K⁺ channels.

Effects of PPT on HERG K⁺ channel currents

PPT had no effects on I_HERG and I_tail as observed with CK. However, PPT showed a significant effect on slow I_{deactivating-tail}. As a result, in the presence of PPT, I_{deactivating-tail} accelerated to decay in contrast to CK. PPT (100 μM) decreased slow I_{deactivating-tail} by an average of 88.2±1.3% compared to control (Fig. 3A). PPT response was also completely reversible upon washing the oocytes with ND96 (data not shown). Next, we examined the concentration-dependent effect of PPT on I_HERG, I_tail, and slow I_{deactivating-tail}. As shown in Fig. 3B, PPT showed a slight inhibition on I_tail but a negligible effect on I_HERG, however PPT primarily inhibited I_{deactivating-tail} in a concentrationdependent manner. Interestingly, PPT had a considerable effect on I_{deactivating-tail} compared to I_HERG and I_tail. The IC₅₀ value was 27.5±2.3 μM for slow I_tail, while the IC₅₀ value was 10.6±0.5 μM for slow I_{deactivating-tail}.

Fig. 3. Effects of protopanaxatriol (PPT) on I_HERG, I_tail, and I_{deactivating-tail}. (A) Representative current traces on human ether-a-go-go-related gene (HERG) K⁺ channel inhibitions by various concentrations of PPT. Currents were in response to 4-s voltage steps to 0 mV from a holding potential of –90 mV, followed by repolarization to –60 mV. I_HERG was obtained at the end of depolarization; I_tail was obtained at beginning of repolarization; slow I_{deactivating-tail} was obtained at the end of repolarization as indicated by the arrow. (B) Concentration-response curves for the inhibition of HERG K⁺ currents by PPT on I_HERG, I_tail, and I_{deactivating-tail}. Solid lines were fitted to the Hill equation. PPT was potent for the inhibition of I_{deactivating-tail}. Bars represent the means±SEM (n=5-7). (C) The representative control (upper traces) and 100 μM PPT-mediated currents (down traces) in I-V relationship. Currents were recorded at test potential from –60 to +50 mV. I_tail and I_{deactivating-tail} were recorded after repolarization to –60 mV. (D) I-V relationships for HERG K⁺ currents measurement at the end of the 4-s test pulse before and after application of 3, 10, and 30 μM PPT (n=5). Currents were normalized to the control current at 0 mV for each oocyte. Data are represented by the means±SEM (n=7). (E) Effects of PPT on the steady-state activation curve for HERG K⁺ channel. I_tail were normalized to the peak current under each condition, and the data were fitted with a Boltzmann function. Treatment of PPT (3, 10, and 30 μM) did not cause a leftward shift. Con, control.

Next, we examined the HERG K⁺ current-voltage (I-V) relationship using PPT (Fig. 3C, D). The amplitude of I_HERG measured at the end of depolarization increased with increasing positive voltage steps, reaching a maximum at –10 mV and then progressively decreased with further voltage steps in the control, indicating that HERG K⁺ channel rectifies (Fig. 3C, D). When the amplitude of I_HERG was normalized to the maximum amplitude of the I_HERG obtained under the control conditions and was plotted against the potential of the step depolarization (Fig. 3C, D), the presence of various concentrations of PPT did not show a significant increase of I_HERG even at high concentrations of PPT, indicating that PPT mainly affects slow I_{deactivating-tail}.

We next set forth to determine whether PPT affects steady-state activation (Fig. 3E). Treatment of 10 and 30 μM PPT did induce slightly but not significant leftward shifts in steady-state activation curves compared to control oocytes. The V_1/2 values were –22.6±0.37 mV in the control and –28.7±1.35, –29.1±1.43 and –30.1±0.46 mV in the presence of 3, 10, and 30 μM PPT-treated groups, respectively (p<0.08 at 10 and 30 μM Rg₃ compared to control, n=5). The slope factors (kg) were not significantly changed, yielding values of 7.9±0.33 in control and 4.8±0.09, 5.2±1.29 and 6.3±0.39 in the presence of 3, 10 and 30 μM PPT-treated groups, respectively. We also examined PPT (100 μM) and observed that it did not affect steady-state inactivation (data not shown). Therefore, these results indicate that PPT affects the activation gating rather than inactivation gating of HERG K⁺ channels.

Effects of PPD on HERG K⁺ channel currents

PPD itself had nearly no effects on I_HERG, I_tail, and slow I_{deactivating-tail} (Fig. 4A). However, PPD showed an inhibition on Rg₃-mediated enhancement on slow I_{deactivating-tail}. Hence in the presence of PPD, Rg₃-induced persistent I_{deactivating-tail} was greatly attenuated. PPD effect on Rg₃-induced persistent I_{deactivating-tail} was completely reversible upon washing the oocytes with ND96 (data not shown). We subsequently examined the concentration-dependent effect of PPD on Rg₃-induced persistent I_{deactivating-tail}. As shown in Fig. 4B, PPD showed a negligible effect on I_HERG and I_tail but PPD greatly attenuated Rg₃-induced persistent I_{deactivating-tail} in a concentration-dependent manner. Interestingly, PPD caused a rightward shift of Rg₃-induced persistent I_{deactivating-tail} from 0.5±0.04 to 2.9±0.57 μM for slow I_{deactivating-tail} (Fig. 4C).

Fig. 4. Effects of protopanaxadiol (PPD) on Rg₃-mediated I_HERG, I_tail, and I_{deactivating-tail}. (A) Representative current traces on human ether-a-go-go-related gene (HERG) K⁺ channel inhibitions by various concentrations of PPD itself (left panel). Currents were in response to 4-s voltage steps to 0 mV from a holding potential of –90 mV, followed by repolarization to –60 mV. I_HERG was obtained at the end of depolarization; I_tail was obtained at beginning of repolarization; slow I_{deactivating-tail} was obtained at the end of repolarization as indicated by the arrow. Concentration-dependent inhibitions of HERG K⁺ currents by PPD on Rg₃-mediated I_HERG, I_tail, and I_{deactivating-tail} (right panel). (B) I-V relationships for HERG K⁺ currents measurement at the end of the 4-s test pulse before (upper traces) and after application of 1 μM Rg₃ plus 100 μM PPD (down traces). Currents were normalized to the control current at 0 mV for each oocyte (right panel). Data are represented by the means±SEM (n=7). (C) Concentration-response curves for the activation of HERG K⁺ currents by Rg₃ in the absence or presence of PPD on I_HERG, I_tail, and I_{deactivating-tail}. Solid lines were fitted to the Hill equation. PPD significantly affected I_{deactivating-tail}. Bars represent the means±SEM (n=5-7). (D) Effects of PPD on Rg₃-mediated the steady-state activation curve for HERG K⁺ channel. I_tail were normalized to the peak current under each condition, and the data were fitted with a Boltzmann function. Treatment of 100 μM PPD in the presence of 0.3, 1, or 3 μM Rg₃ did not cause a leftward shift. Con, control.

We then studied the HERG K⁺ current-voltage (I-V) relationship using PPD on Rg₃-induced enhancement of I_HERG (Fig. 4B). The amplitude of I_HERG measured at the end of depolarization increased with increasing positive voltage steps, reaching a maximum at –10 mV and then progressively decreased with further voltage steps in the control, indicating that HERG K⁺ channel rectifies (Fig. 4B). When the amplitude of I_HERG was normalized to the maximum amplitude of the I_HERG obtained under the control conditions and was plotted against the potential of the step depolarization (Fig. 4B), the presence of various concentrations of PPD did not show a significant increase of I_HERG even at high concentrations of PPD, indicating that PPD mainly affects Rg₃-induced persistent I_{deactivating-tail}.

Our following goal was to examine whether PPD affects steady-state activation (Fig. 3E). In contrast to CK or PPT, co-treatment of 10 and 30 μM PPD with Rg₃ did not affect on Rg₃-caused leftward shifts in steady-state activation curves compared to control oocytes. The V_1/2 values were –21.7±0.18 mV in the control and –21.2±0.35, –24.6±0.27 and –24.6±0.28 mV in the presence of 3, 10, and 30 μM PPD-treated groups, respectively (p<0.07 at 10 and 30 μM Rg₃ compared to control, n=5). The slope factors (kg) were not significantly changed, yielding values of 7.9±0.16 in control and 8.7±0.31, 7.5±0.24 and 7.5±0.25 in the presence of 3, 10 and 30 μM PPD-treated groups, respectively. We also examined PPD (100 μM) and observed that it did not affect steady-state inactivation (data not shown). These results suggest PPD affects the activation gating rather than inactivation gating of HERG K⁺ channels.

Ginsenosides consist of aglycone and carbohydrates portions. The aglycone is the backbone of the ginsenoside, with a hydrophobic four-ring steroid-like structure that may be non-polar, whereas the carbohydrates on carbons-3, 6, and 20 of the backbone are polar (Fig. 1A). In vitro and in vivo studies have shown that ginsenosides administered by an oral route were metabolized and finally become an agylcone such as CK which has one glucose at the C-20, PPD and PPT [8]. Recent report showed that these ginsenoside metabolites might also induce apoptosis of the cancer cells and play a role as an anti-cancer agent [8,11], suggesting that ginsenosides are pro-drugs of these metabolites. However, relatively little is known regarding how ginsenoside metabolites regulate ion channel or receptor activity.

In the present study, we report three major findings. First, we observed that CK decelerated deactivating time constants in both concentration- and voltage-dependent manners, whereas CK had no effects on I_HERG and I_tail, Second, we found that PPT accelerated I_{deactivating-tail} deactivation in both concentration- and voltage-dependent manners without affecting on I_HERG and I_tail. Third, PPD itself had no effect on I_HERG, I_tail, and I_{deactivating-tail}. However, co-treatment of PPD with Rg₃ blocked Rg₃-mediated decelerating effects on I_{deactivating-tail}. Thus, the major findings of the present study are that ginsenoside metabolites had no significant effects on I_HERG and I_tail of HERG K⁺ channel but they did affect I_{deactivating-tail} in differing manners, indicating that ginsenoside metabolites show a differential regulation of I_{deactivating-tail} with respect to the HERG K⁺ channel.

On the other hand, we demonstrated in previous studies that ginsenoside metabolites such as CK, PPD and PPT regulate ion channels and receptors. For example, CK but not PPT caused strong inhibition of the voltage-dependent α1G-type Ca²⁺ channel [12]. Similarly, we have also found that CK but not M4 (PPT) inhibited a neuronal Na⁺ (Nav1.2) channel [13]. We demonstrated that M4, but not CK, exhibited an inhibitory effect on 5-HT_3A receptor-mediated currents [14]. In addition, CK and PPT exhibited an inhibitory effect on α3β4 nicotinic acetylcholine receptor-mediated currents [15]. All these findings indicate that ginsenoside metabolites as well as ginsenosides have regulatory effects on voltage-dependent ion channel and receptor activities.

Since the HERG K⁺ channel regulators are clinically important for treatment of cardiac diseases such as arrhythmia, there are numerous reports on the development of HERG K⁺ channel regulators from natural compounds [16]. In the previous study, we found that Rg₃ increased I_HERG and peak I_tail. In addition, Rg₃ profoundly delayed deactivation kinetics by inducing a persistent I_{deactivating-tail} and caused a leftward shift of steady-state voltage-dependent activation but not inactivation, leading to the possibility that Rg₃ is an activator to open HERG K⁺ channel at a more negative voltage step than control and that Rg₃ mainly works by slowing closure of the HERG K⁺ channel once it has opened. In the present study, when we examined the effects of ginsenoside metabolites on HERG K⁺ channel currents, we found that CK induced a persistent I_{deactivating-tail} and caused a leftward shift of steady-state voltage-dependent activation (Fig. 2). However, the effective concentration of CK was higher than that of Rg₃. In contrast to CK, PPT caused an acceleration of I_{deactivating-tail} decay (Fig. 3). Interestingly, co-treatment of PPD with Rg₃ blocked Rg₃-mediated deceleration of deactivation process. It is noteworthy that since Rg₃ became PPD after metabolic process, PPD-mediated antagonism of Rg₃ action on HERG K⁺ channel activity show the possibility that aglycone, PPD, ginsenoside Rg₃ and other PPD ginsenosides might interfere with in vivo Rg₃ actions on HERG K⁺ channel activity.

Taken together, we found that CK decelerated the deactivation process by inducing a persistent I_{deactivating-tail} with delayed deactivation in HERG K⁺ channel without affecting I_HERG and I_tail. In contrast, PPT accelerated the deactivation process by inducing an inhibition of I_{deactivating-tail} without affecting I_HERG and I_tail. PPD itself had no observable effects on I_HERG, I_tail, and I_{deactivating-tail} but co-treatment of PPD with Rg₃ blocked Rg₃-mediated a persistent I_{deactivating-tail} with delayed deactivation in HERG K⁺ channel. These novel findings provide an insight into the molecular action of ginsenoside metabolites on HERG K⁺ channel activity.