Inhibitory Effects of Nobiletin on Voltage-Gated Na+ Channel in Rat Ventricular Myocytes Based on Electrophysiological Analysis and Molecular Docking Method

Nobiletin (NOB) has attracted much attention owing to its outstanding bioactivities. This study aimed to investigate its anti-arrhythmic effect through electrophysiological and molecular docking studies. We assessed the anti-arrhythmic effects of NOB using aconitine-induced ventricular arrhythmia in a rat model and the electrophysiological effects of NOB on rat cardiomyocytes utilizing whole-cell patch-clamp techniques. Moreover, we investigated the binding characters of NOB with rNav1.5, rNav1.5/QQQ, and hNaV1.5 via docking analysis, comparing them with amiodarone and aconitine. NOB pretreatment delayed susceptibility to ventricular premature and ventricular tachycardia and decreased the incidence of fatal ventricular fibrillation. Whole-cell patch-clamp assays demonstrated that the peak current density of the voltage-gated Na+ channel current was reversibly reduced by NOB in a concentration-dependent manner. The steady-state activation and recovery curves were shifted in the positive direction along the voltage axis, and the steady-state inactivation curve was shifted in the negative direction along the voltage axis, as shown by gating kinetics. The molecular docking study showed NOB formed a π-π stacking interaction with rNav1.5 and rNav1.5/QQQ upon Phe-1762, which is the homolog to Phe-1760 in hNaV1.5 and plays an important role in antiarrhythmic action This study reveals that NOB may act as a class I sodium channel anti-arrhythmia agent.


Introduction
Cardiac arrhythmias are one of the most common cardiovascular diseases, and cardiac arrest caused by organic heart disease with arrhythmias is a leading contributor to sudden cardiac death [1,2]. The prognosis and survival rates of patients who suffer from ventricular tachycardia or ventricular fibrillation remain unsatisfactory [3]. The resting potential of cell membranes is frequently disturbed in ventricular arrhythmias and organic heart disease, which leads to the disorder of ionic channel currents. In addition, malignant cardiovascular diseases may be caused by aberrant changes in the ion channel expression or gating kinetics in cardiomyocytes [4,5].
Ion channels are essential building blocks of various action potential currents in cardiac myocytes, and voltage-gated ion channels are crucial for understanding their structures, distributions, interactions, and functions [6]. The Na + channel current (I Na ), which begins with the activation of voltage-gated Na + channels encoded by SCN5A, is the most important

NOB Alleviated Fatal Ventricular Arrhythmia In Vivo
In the current study, a fatal cardiac arrhythmia model was established by administering aconitine intravenously. The onset times of ventricular premature contraction (VP) and ventricular tachycardia (VT), as well as the incidence of ventricular fibrillation (VF), were used to assess the severity of arrhythmia. Pretreatment with NOB postponed the onset of VP and VT and substantially reduced the morbidity of VF ( Figure 1 and Table 1). Compared to the normal saline group (Saline group), the therapeutic effects of NOB (10 mg/kg) were similar to those of amiodarone. Overall, NOB significantly alleviated aconitine-induced fatal ventricular arrhythmias. Saline NOB Amiodarone 95% 10% ## 5% ##

NOB Inhibited INa in a Concentration-Dependent Manner
Tetrodotoxin (TTX, 10 μmol/L) was used as a sodium-channel-specific blocker (F ure 2). The current was mostly restored after removing TTX in the extracellular enviro ment, indicating that the inward current of this stimulation protocol was INa. Before ev uating the administration of NOB, the current was maintained at a steady state for 5min. NOB at 5 and 10 μmol/L had no obvious effects on INa, compared with the contr group (Table 2). Furthermore, 25, 50, and 100 μmol/L NOB substantially inhibited INa in concentration-dependent manner. Compared with amiodarone, 100 μmol/L NOB had slightly weaker inhibitory effect. Amounts of 25, 50, and 100 μmol/L NOB and 24. μmol/L amiodarone reduced the peak sodium current density, demonstrating inhibito rates of 29.53% ± 2.90%, 42.04% ± 6.16%, 55.69 ± 3.66% (n = 10, p < 0.01), and 68.86% ± 5.28 (n = 10, p < 0.01), respectively, under −40 mV depolarization voltage. In the concentratio response curve, the fractional blockage was plotted against the appropriate concentratio of NOB. The Hill equation matched the curve [18]: where Y is the normalized response of the inhibitory rate, and X is the concentration NOB.

NOB Inhibited I Na in a Concentration-Dependent Manner
Tetrodotoxin (TTX, 10 µmol/L) was used as a sodium-channel-specific blocker ( Figure 2). The current was mostly restored after removing TTX in the extracellular environment, indicating that the inward current of this stimulation protocol was I Na . Before evaluating the administration of NOB, the current was maintained at a steady state for 5-10 min. NOB at 5 and 10 µmol/L had no obvious effects on I Na , compared with the control group (Table 2). Furthermore, 25, 50, and 100 µmol/L NOB substantially inhibited I Na in a concentrationdependent manner. Compared with amiodarone, 100 µmol/L NOB had a slightly weaker inhibitory effect. Amounts of 25, 50, and 100 µmol/L NOB and 24.24 µmol/L amiodarone reduced the peak sodium current density, demonstrating inhibitory rates of 29.53 ± 2.90%, 42.04 ± 6.16%, 55.69 ± 3.66% (n = 10, p < 0.01), and 68.86 ± 5.28% (n = 10, p < 0.01), respectively, under −40 mV depolarization voltage. In the concentration-response curve, the fractional blockage was plotted against the appropriate concentrations of NOB. The Hill equation matched the curve [18]: where Y is the normalized response of the inhibitory rate, and X is the concentration of NOB. Data are presented as means ± standard deviation (n = 10). pA/pF means membrane current/ itance. * p < 0.05, ** p < 0.01, ***p < 0.001 vs. Control group. Data are expressed as means ± standard deviation (n = 10, mea in six rats). (D) Effect of aconitine (3 μmol/L) and aconitine (3 μmol/L) with NOB (50 μmol/L) compared to Control group. Data are expressed as means ± standard deviation (n = 6, measu three rats); ns, no significant difference; * p < 0.05, ** p < 0.01, *** p < 0.001 vs. control group.

Effects of NOB on the Current-Voltage Curve (I-V curve) of INa
To explore the effects of NOB on the INa under different voltage conditions, ch in INa produced by the consecutive stimulations were gathered in the absence and ence of the administration of NOB and amiodarone. The results show that 25, 50, an μmol/L NOB and 24.24 μmol/L amiodarone decreased the depolarized current wi changing the form or trend of the curve ( Figure 3). Moreover, the I-V curve shifted the positive axis concentration-dependently. Overall, the current density was reduce der the protocol voltages, especially between applied voltages of −40 mV and −30 m Data are expressed as means ± standard deviation (n = 10, measured in six rats). (D) Effect of aconitine (3 µmol/L) and aconitine (3 µmol/L) with NOB (50 µmol/L) on I Na compared to Control group. Data are expressed as means ± standard deviation (n = 6, measured in three rats); ns, no significant difference; * p < 0.05, ** p < 0.01, *** p < 0.001 vs. control group. The half maximal inhibitory concentration (IC 50 ) of NOB on the peak sodium current was 54.57 µmol/L (49.10-61.46 µmol/L, 95% CI).

Effects of NOB on the Current-Voltage Curve (I-V Curve) of I Na
To explore the effects of NOB on the I Na under different voltage conditions, changes in I Na produced by the consecutive stimulations were gathered in the absence and presence of the administration of NOB and amiodarone. The results show that 25, 50, and 100 µmol/L NOB and 24.24 µmol/L amiodarone decreased the depolarized current without changing the form or trend of the curve ( Figure 3). Moreover, the I-V curve shifted along the positive axis concentration-dependently. Overall, the current density was reduced under the protocol voltages, especially between applied voltages of −40 mV and −30 mV.

Effects of NOB on the Steady-State INa Activation Curve
To investigate the effect of NOB on the activation and opening of sodium channels the variations in INa with and without the administration of NOB and amiodarone were recorded ( Figure 4). The data pertaining to the NaV1.5 channel were processed to draw the activation curve fitted by the Boltzmann equation [19]: where G is the electric conductance, and κ is the slope factor. The administration of NOB at different concentrations caused the steady-state activation curve to shift positively along the voltage axis, i.e., a depolarization trend, and the half activation voltage (V1/2-ac) increased. Additionally, 25, 50, and 100 μmol/L NOB and 24.24 μmol/L amiodarone changed the V1/2-ac value from −62.00 ± 0.53 to −56.17 ± 0.44 −53.77 ± 0.29, −49.89 ± 0.18 mV (n = 7, p < 0.01), and −47.52 ± 0.16 mV (n = 7, p < 0.01) respectively. The administration of NOB therefore suppressed the opening process of the Na + channel and raised the critical value for Na + channel activation, making the channel opening progress more difficult.

Effects of NOB on the Steady-State I Na Activation Curve
To investigate the effect of NOB on the activation and opening of sodium channels, the variations in I Na with and without the administration of NOB and amiodarone were recorded ( Figure 4). The data pertaining to the Na V 1.5 channel were processed to draw the activation curve fitted by the Boltzmann equation [19]: where G is the electric conductance, and κ is the slope factor.

Effects of NOB on the Steady-State INa Activation Curve
To investigate the effect of NOB on the activation and opening of sodium channels, the variations in INa with and without the administration of NOB and amiodarone were recorded ( Figure 4). The data pertaining to the NaV1.5 channel were processed to draw the activation curve fitted by the Boltzmann equation [19]: where G is the electric conductance, and κ is the slope factor. The administration of NOB at different concentrations caused the steady-state activation curve to shift positively along the voltage axis, i.e., a depolarization trend, and the half activation voltage (V1/2-ac) increased. Additionally, 25, 50, and 100 μmol/L NOB and 24.24 μmol/L amiodarone changed the V1/2-ac value from −62.00 ± 0.53 to −56.17 ± 0.44, −53.77 ± 0.29, −49.89 ± 0.18 mV (n = 7, p < 0.01), and −47.52 ± 0.16 mV (n = 7, p < 0.01), respectively. The administration of NOB therefore suppressed the opening process of the Na + channel and raised the critical value for Na + channel activation, making the channel opening progress more difficult.  The administration of NOB at different concentrations caused the steady-state activation curve to shift positively along the voltage axis, i.e., a depolarization trend, and the half activation voltage (V 1/2-ac ) increased. Additionally, 25, 50, and 100 µmol/L NOB and 24.24 µmol/L amiodarone changed the V 1/2-ac value from −62.00 ± 0.53 to −56.17 ± 0.44, −53.77 ± 0.29, −49.89 ± 0.18 mV (n = 7, p < 0.01), and −47.52 ± 0.16 mV (n = 7, p < 0.01), respectively. The administration of NOB therefore suppressed the opening process of the Na + channel and raised the critical value for Na + channel activation, making the channel opening progress more difficult.

Effects of NOB on the I Na Inactivation Curve
To analyze the effects of NOB on the I Na inactivation curve, the variation in I Na before and after the administration of NOB and amiodarone was elicited using double-pulse stimulation ( Figure 5). The inactivation data were fitted using the Boltzmann equation [19]: where κ is the slope factor. and 100 μmol/L NOB. (B) With or without 24.24 μmol/L amiodarone. (C) Effects of NOB and amiodarone on fitted activation curve. Data are expressed as means ± standard deviation (n = 7, measured from four rats); ** p < 0.01 vs. Control group.

Effects of NOB on Recovery Curves after Inactivation of INa
To study the effects of NOB on the INa recovery curve, the variation in INa was recorded using two-pulse stimulations, before and after the addition of different concentrations of NOB and amiodarone ( Figure 6). The recovery data for INa were fitted using the one-phase association equation [19]: where t is the time interval between two pulses, and τ is the time constant of reactivation after inactivation. The administration of NOB at different concentrations caused the steady inactivation curve to shift negatively along the voltage axis, i.e., a hyperpolarization trend; the half inactivation voltage (V 1/2-in ) decreased. Overall, the administration of 25, 50, and 100 µmol/L NOB and 24.24 µmol/L amiodarone reduced the V 1/2-in from −75.00 ± 0.57 to −79.59 ± 0.44, −82.72 ± 0.64, −87.94 ± 0.60 (n = 7, p < 0.01), and −95.69 ± 0.54 mV (n = 7, p < 0.01), respectively. Moreover, the κ value was altered from 8.03 ± 0.53 to 8.55 ± 0.56 (n = 7, p > 0.05), 9.23 ± 0.47, 9.39 ± 0.39 (n = 7, p < 0.01), and 10.80 ± 0.5 (n = 7, p < 0.01), respectively. The results suggest that NOB may have a strong affinity for binding to sodium channels during inactivation, slowing the time-dependent Na V 1.5 channel inactivation.

Effects of NOB on Recovery Curves after Inactivation of I Na
To study the effects of NOB on the I Na recovery curve, the variation in I Na was recorded using two-pulse stimulations, before and after the addition of different concentrations of NOB and amiodarone ( Figure 6). The recovery data for I Na were fitted using the one-phase association equation [19]: where t is the time interval between two pulses, and τ is the time constant of reactivation after inactivation. The recovery curve was positively shifted along the voltage axis following treatment with NOB and amiodarone. The results of the curve showed that 25, 50, and 100 μmol/L NOB and 24.24 μmol/L amiodarone changed the τ value from 21.51 ± 0.95 to 26.75 ± 0.94, 29.16 ± 0.34, 32.04 ± 0.78 (n = 7, p < 0.01), and 35.48 ± 1.12 ms (n = 7, p < 0.01), respectively. These results suggest that NOB and amiodarone significantly changed the recovery kinetics of INa after inactivation and procrastinated the process of INa from inactivation to activation. This also indicates that NOB has a considerable affinity for the recovery channel from inactivation to activation.

Molecular Docking Simulation
The molecular docking method was performed to explore how NOB interacted with the rat voltage-gated sodium channel (termed rNav1.5), the open state structure of rNav1.5 (termed rNav1.5/QQQ), and the human voltage-gated sodium channel (termed hNa V 1.5) by computer-based docking stimulation (Figure 7, Table 3). The docking analysis between aconitine and amiodarone with hNa V 1.5 was performed and compared with NOB ( Figure S1, Table S2). The docking model predicted that NOB has affinity binding energies of −6.655 and −6.562 kcal/mol to rNav1.5 and rNav1.5/QQQ, respectively. More specifically, NOB formed a π-π stacking interaction, a hydrogen bond with rNav1.5, and rNav1.5/QQQ upon Phe-1762 and Ser-1712 residues, respectively. In addition, there were five hydrophobic contacts that contributed to both the NOB-rNav1.5 and NOB-rNav1.5/QQQ binding, namely, Ile-1468, Leu-1464, Ser-1460, Phe-1420, and Lys-1421 ( Figure 7, Table 3). Although the NOB-rNav1.5 and NOB-rNav1.5/QQQ binding exhibited the same intermolecular binding interactions, their binding energy difference was due to the lower hydrophobic binding with Leu-1464 of NOB-rNav1.5/QQQ compared with NOB-rNav1.5 docking. The binding affinity of NOB, aconitine, and amiodarone, had the lowest energy poses, shown by the Gibbs free-energy values (∆G) as −5.693, −8.889, and −6.465 kcal/mol, respectively (Table 3, Tables S1 and S2), with hNa V 1.5. Furthermore, the molecular mechanics/generalized Born surface area (MM-GBSA) was used to predict the binding free energies of NOB, aconitine, and amiodarone to the human Na V 1.5 [20]. NOB, aconitine and amiodarone exhibited low glide energy at −51.71 kcal/mol, −62.91 kcal/mol, and −42.26 kcal/mol against hNa V 1.5, respectively, where molecular docking simulation found good binding affinity (Table S1). In particular, NOB and aconitine were mainly docked in the transmembrane helices of voltage-sensing domains (VSD) III and IV of hNa V 1.5, with different types of binding interactions (Figure 7 and Figure S1).  (Table S1). In particular, NOB and aconitine were mainly doc the transmembrane helices of voltage-sensing domains (VSD) III and IV of hNaV1. different types of binding interactions (Figures 7 and S1).

Discussion
The electrophysiological function of the heart is based on its dynamic ionic balance, which is frequently altered in aging or sick hearts and is regarded as a significant trigger for arrhythmias [21]. Aconitine is a highly cardiotoxic sodium channel agonist that is frequently utilized for modeling purposes in investigations. The mechanism of aconitine-induced arrhythmia relies on its ability to bind to Na V 1.5 and sodium-calcium exchanger and prolong their openings, which facilitates the entry of Na + into the cell and results in long-lasting depolarization and calcium overload [22,23]. These effects cause automaticity in fast-responding cells and are similar to the abnormally high inward flow of Na + flux that develops in the presence of enhanced triggered activity in ischemic heart diseases [24]. Moreover, aberrant mutations in SCN5A, which encodes the Na V 1.5 channel, can cause a variety of genetic arrhythmia syndromes, including LQT3, Brugada syndrome, progressive cardiac conduction disorder, sudden infant death syndrome, and different cardiac conduction abnormalities [17,25]. Therefore, the regulation of the Na V 1.5 current and the expression of the encoded Na V 1.5 channel SCN5A are crucial for the treatment of malignant arrhythmia [26].
In mammals, the nine subtypes of the human voltage-gated sodium (Na V 1.1-Na V 1.9) channels are responsible for the initiation and transmission of electrical impulses in different tissues and Na V 1.5 is the primary cardiac isoform. The topology of mammalian Na V channels comprised Domain I, Domain II, Domain III, IFM-motif, and Domain IV of the α-subunit, and the β1 and β2 subunits [27]. The sequence alignments of rNav1.5 (Uniprot ID: P15389) and hNa V 1.5 (Uniprot ID: Q14524) are compared in Figure S2 and exhibit more than 94% similarity. The open state structure of rNav1.5 was achieved by importing the IFM/QQQ mutations to remove the fast inactivation gate [28].
In this study, NOB protected against aconitine-induced ventricular arrhythmias and substantially slowed the onset of lethal ventricular fibrillation, similar to amiodarone. The peak I Na was substantially blocked by NOB in vitro with a dependent manner of voltage and concentration. Notably, NOB substantially shifted the activation and recovery curves along the voltage axis positively and the inactivation curve negatively. This suggests that NOB affects the kinetic gating characteristics of Na + channels, accelerating the inactivation and retarding the post-inactivation recovery of Na V 1.5 channels. In this experiment, for the first time NOB was docked with rNav1.5 and rNav1.5/QQQ. NOB, aconitine, and amiodarone were docked with hNa V 1.5, and their molecular effects were described. The molecular docking study showed that NOB formed a π-π stacking interaction with rNav1.5 and rNav1.5/QQQ upon Phe-1762, which is the homolog to Phe-1760 in hNa V 1.5 and plays an important role in antiarrhythmic action [29,30]. The computational modeling results have further verified the patch-clamp results of the inhibitory effects of NOB on Nav1.5 in rat ventricular myocytes in both closed and open states.
In the molecular docking stimulation, both NOB and aconitine formed hydrogen binding with Gln-371, which provides a reasonable explanation for how NOB and aconitine affected the kinetics of Na + gating in cardiomyocytes. Compared to aconitine, NOB forms additional hydrogen bonds with Lys-1419 and Ser-1458 of hNa V 1.5. In addition, NOB formed hydrophobic interactions with Phe-1418, Leu-1462, Phe-1760, and aconitine interacted with the Na V 1.5 channel by hydrophobic bonds such as Leu-931, Leu-1462, Ser-1458, Phe-1418, Lys-1419, Val-405, and Phe-1760. Several studies have demonstrated that residues Gln-371 and Leu-1462 of hNa V 1.5 exert a vital role in the dynamics of steady-state activation and inactivation [31]. In addition, amiodarone exhibited strong bindings as cation-π and π-π stacking interactions with Phe-934 and Phe-1760, respectively, when docking with hNa V 1.5. Amiodarone, possessing a stronger interaction with Phe-1760, which is the backbone of the bonding site of Na + with hNa V 1.5, showed excessive inhibitory activity against Na V 1.5 [32]. This excessive blockage is closely related to the most commonly reported adverse reactions in clinical application, such as conduction block and bradycardia [33]. It corresponded with the patch-clamp electrophysical results that amiodarone explicated greater inhibitory activity in I Na than aconitine and NOB. In summary, NOB inhibited the Na V 1.5 channel by substantially suppressing the peak I Na , reducing the channel opening levels, and retarding voltage-dependent channel inactivation and recovery via binding to inactivated channels. Consequently, NOB exhibited an inhibitory impact on the Na + channel in cardiomyocytes. Compared to amiodarone, which is already widely used in clinical practice in the treatment of arrhythmia, the inhibitory effect of NOB on cardiomyocyte Na V 1.5 was similar but more moderate. Exploring the mechanisms by which NOB (a potential antiarrhythmic drug in this study), amiodarone (the most widely used clinical antiarrhythmic drug), and aconitine (a widely used tool drug for ventricular arrhythmias) bind to Na V 1.5 by the molecular docking method will help us to gain more understanding of the structure-activity relationship, and to develop novel agents from natural sources for treating ventricular arrhythmias.
The SARS-CoV-2 infection is a high-risk-inducing factor for cardiovascular disease due to severe cytokine storms and systemic inflammatory responses [34][35][36][37][38]. Anti-arrhythmic drugs currently in clinical use, such as amiodarone, which has pulmonary toxicity, and quinidine, which has proven to show a pro-arrhythmia effect, may aggravate the damage to the cardiopulmonary circulatory system [39][40][41][42]. Therefore, the development of lowtoxicity and multi-target cardiovascular drugs is also urgently needed [43]. Previous studies have reported that NOB exerts anti-inflammatory, antioxidant, and anti-hyperlipidemic effects to prevent the development of cardiovascular disease [12,14,15,44,45]. We speculate that NOB-rich supplements or other forms of agents may represent promising multi-target therapeutic candidates for the prevention of arrhythmia and other cardiovascular diseases caused by COVID-19-induced myocardial injury.
The principal limitation of the current study is that we did not investigate the electrophysiological effects of NOB on human ventricular myocytes. Moreover, we did not investigate the potential interaction of NOB on the slow inactivation of the Na V 1.5 channel and other ion channels which demonstrated the target specificity of NOB. These results should be further validated in human ventricular myocytes to confirm the therapeutic value of NOB as a cardio-protective agent and in treating ventricular arrhythmias.

Animals and Ethics Statement
The Yangzhou University Comparative Medical Center provided Sprague-Dawley (male and female) rats weighing 200 ± 20 g. The rats were maintained in a room of constant temperature and humidity (25 ± 2 • C and 55 ± 5%), with a reverse 12 h light/12 h dark cycle. The Institutional Animal Care and Ethics Committee of Yangzhou University approved the animal protocol (SCXK (su) 2022-0044), and all experimental procedures were executed in compliance with the guidelines and regulations.

Establishment of Ventricular Arrhythmia In Vivo
The in vivo protocol for ventricular arrhythmia was based on an already existing protocol [46]. Thirty male rats were selected and fed for 3 days to ensure they had normal heart rates. The rats were divided into three parallel groups: the normal saline group (Saline group, n = 10), the NOB group (n = 10), and the amiodarone group (n = 10). Before the establishment of ventricular arrhythmia, anesthesia was induced by an intraperitoneal (i.p.) injection of chloral hydrate (0.38 g/kg). The rats were fixed in a supine position and a lead II electrocardiogram (ECG) was recorded simultaneously using an RM6240 Biological Data Acquisition and Analysis System (Chengdu Instrument Factory, Chengdu, China). After maintaining stable ECG recordings for 10 min, the rats in the three groups were intravenously injected with different solutions via the tail vein within 30 s: Saline group, 1.5 mL/kg saline; NOB group, 1.5 mL/kg saline containing NOB (10 mL/kg); and amiodarone group, 1.5 mL/kg saline containing amiodarone (5 mg/kg). After 10 min of stabilization, the rats were injected with aconitine (0.001%, 40 µg/kg) via the caudal vein. The ECGs of the three groups of rats were recorded for 1 h, focusing on the onset time of ventricular arrhythmia (i.e., ventricular premature (VP) and ventricular tachycardia (VT)) and the incidence of fatal ventricular fibrillation (VF).

Acute Single-Cell Isolation
Ventricular myocytes were isolated from Sprague-Dawley rats (male and female) weighing 200 ± 20 g, as previously described [47]. Ten minutes before anesthesia by i.p. injection with 2% sodium pentobarbital (40 mg/kg), the rats were anticoagulated by i.p. injection of heparin (2000 IU/kg). After successful anesthesia, the animals were fixed in a supine position, then quickly pinned and thoracotomized. The whole hearts were excised and rapidly immersed in Ca 2+ -free Tyrode's solution at 4 • C. After slight trimming of the redundant tissue, the hearts were anchored in a Langendorff device by ligating the aorta. The remaining blood was discharged by 30 s of Ca 2+ -saturated perfusion and 10 min of Ca 2+ -free Tyrode's solution. Finally, the hearts were digested for approximately 25 min in Ca 2+ -free Tyrode's solution containing type II collagenase (0.4 g/L), BSA (1 g/L), and taurine (0.4 g/L) until they increased in size. The hearts were seen to soften, with thick, turbid fluid exudate. All perfusion and digestion processes were kept at 37 • C with 100% oxygen saturation. The ventricular tissue was then removed, dissected, and filtered through a 100-mesh filter to obtain single cells. The single cells were washed thrice and reserved in KB solution for 0.5 h of perfusion with oxygen.

Stimulus Protocols
A single, square-wave pulse stimulation protocol was used to instruct voltage for 30 ms at −30 mV to induce an inward current from a holding potential of −80 mV.
The stimulation protocol for the Na + channel current-voltage curve (I-V curve) was as follows: the voltage clamp mode was sustained at 80 mV, and the single cells were stimulated with a series of square waves (−70 to 50 mV) with an amplitude step of 5 mV at 0.5 Hz for 30 ms.
The I Na activation current stimulation protocol was as follows: the ventricular myocytes were stimulated by a series of square waves (−80 to 50 mV) at a step of 5 mV with a holding potential at 80 mV of 0.5 Hz, for 30 ms.
The detection protocol for the I Na inactivation curve was a double-pulse method used at −80 mV in the voltage clamp mode. A pretreatment pulse was used to depolarize the cells (−140 to 40 mV) in 10 mV steps for 50 ms. Next, a −30-mV pulse was used to stimulate the ventricular cells for 25 ms.
A double-pulse stimulation protocol with a holding potential of −80 mV was used to test the recovery curve after I Na inactivation. The cells were stimulated with a −30 mV wave pulse lasting 30 ms for depolarization. After a set interval, 30 mV stimulatory pulses were applied, lasting 25 ms. A total of 15 pulses were applied to the ventricular myocytes, with the interval between pulses incrementing by 10 ms.

Whole-Cell Patch-Clamp Recording
Cells for patch-clamp assessment are required to be long and rod-shaped with a smooth, intact cytomembrane and distinct horizontal striation without spontaneous contract vibration. The cells were transferred from the suspension to a Petri dish at normal temperature (20-25 • C). The cells were left to stand for 10 min to allow them to adhere to the dish, and the dish was filled with the extracellular solution to remove the remaining KB solution. The MP-225 Micromanipulator System (Sutter Company, Novato, CA, USA), EPC-10 USB/Patchmaster Single Channel Patch-clamp amplifier (HEKA Company, Nordrhein-Westfalen, Germany), and an IX73 inverted scientific microscope (Olympus Company, Nagano, Japan) were used in the whole-cell patch-clamp system. Borosilicate glass microelectrodes (outer diameter 1.50 mm, inner diameter 1.14 mm; Wuhan Microprobe Co., Ltd., Wuhan, China) with a resistance of 2-6 MΩ were pulled with a P-97 microelectrode puller (Sutter Company, CA, USA). The application and information gathering of the stimulation protocol were based on the Patchmaster program (version v2x73.5; HEKA Company, Nordrhein-Westfalen, Germany).

Molecular Modeling and Computational Methods
The crystallography structures of the human voltage-gated sodium channel Na V 1.5 (PDB ID: 6LQA), the rat voltage-gate sodium channel Na V 1.5 (PDB ID: 6UZ3), and open state structure of rNav1.5/QQQ (PDB ID: 7FBS) [32] were retrieved from the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank. The chemical structures of the natural compounds were obtained from the PubChem compound database (http://pubchem.ncbi.nlm.nih.gov/ (accessed on 15 October 2022) with PubChem CIDs of 72344 (nobiletin), 245005 (aconitine), and 2157 (amiodarone). Then, molecular docking calculations were executed using AutoDock Vina software. For the molecular docking calculations, the pdbqt files for the proteins and ligands were prepared according to the AutoDock protocol. All docking parameters were conserved to their default values, except the maximum number of energy evaluations (eval) and the number of genetic algorithm (GA) runs. The docking grid was made as the binding site for the receptor with a grid size of 40 Å × 40 Å × 40 Å. The grid spacing value was adjusted to 0.375 Å. Gasteiger atomic partial charges were assigned for all investigated ligands. The PyMol (PyMOL Molecular Graphics System, version 1.7) program (https://pymol.org (accessed on 15 October 2022), was applied for visualization to obtain the hydrogen bond, hydrophobic, and electrostatic interactions [48].
The binding energies of the ligands with protein were calculated using the molecular mechanics/generalized Born surface area (MM-GBSA) approach with a GB model [20]. The binding energies (DG binding ) were computed using the molecular docking complex, given by: where the energy term (G) is estimated as: G x = E vdw + E ele + G GB + G SA (6) with E vdw , E ele , G GB , and G SA as the van der Waals, electrostatic, general Born solvation and surface area energies, respectively. For the inhibitors, entropy contributions were neglected.

Statistical Analysis
The data are expressed as the mean ± standard deviation (mean ± SD). Origin software (version 7.0; Microcal Software, Inc., Northampton, MA, USA) was used to analyze the experimental data. GraphPad Prism software (version 8.02, San Diego, CA, USA) was utilized to fit the curves. Two-tailed paired Student's t-tests were used for comparisons of two means, analysis of variance (ANOVA) was used for the comparison of multiple means, and the χ 2 test was used for the comparison of two incidences to assess the statistical significance.

Conclusions
In the present investigation, we preliminarily determined that NOB exerts an antiarrhythmic effect by studying the electrophysiological effects and through molecular docking analysis. NOB exerts anti-arrhythmic effects by inhibiting I Na and retarding the steadystate deactivation process. Thus, NOB is a potential option as a class I Na channel blocker in cardio-protective administration and the treatment of arrhythmia.

Informed Consent Statement: Not applicable.
Data Availability Statement: The raw data supporting the conclusions of this article will be madeavailable by the corresponding authors upon reasonable request.