Introduction

Reactive oxygen species (ROS) are increasingly generated under various pathological conditions, such as ischemia/reperfusion, myocardial infarction, and heart failure. Increased oxidative stress causes a series of detrimental effects on the heart1. Hydrogen peroxide (H2O2), which can cause oxidative stress, is significantly elevated in hearts exposed to ischemia/reperfusion and predisposes the heart to ischemia/reperfusion injury and arrhythmias. In myocardial infarction, increasing ROS contribute to left ventricular remodeling and arrhythmias1,2. Increased oxidative stress is also associated with reduced left ventricular function in heart failure3. The underlying mechanisms may involve an impairment of Na and Ca homeostasis4,5, which subsequently causes mechanical dysfunction, arrhythmias and even cell death6. Recently, Ca/calmodulin-dependent protein kinase II (CaMKII) has been suggested to be an important mediator of ROS-induced detrimental effects6,7. ROS activates CaMKII via autophosphorylation- and oxidation-dependent mechanisms8, which in turn enhance late sodium current (INa,L) and L-type calcium current (ICa,L) in cardiac myocytes6,9. Augmentation of INa,L and ICa,L not only causes a prolongation of action potential duration (APD) and the occurrence of early afterdepolarizations (EADs) but also leads to sodium and calcium overload, contractile dysfunction, the occurrence of delayed afterdepolarizations (DADs) and triggered activities6,7,10. Despite the serious detrimental effects caused by increased oxidative stress in structural heart diseases, few drugs exist that can be used for clinical treatment.

Resveratrol (trans-3,4′,5-trihydroxystilbene) is a natural compound found in abundance in grapes, mulberries, peanuts and red wine11. Over the past two decades, it has been demonstrated from both in vitro and in vivo studies to have cardioprotective effects, including anti-inflammatory, antioxidative, and anti-hyperlipidemic properties as well as the prevention of platelet aggregation and cardiac hypertrophy12,13,14. These beneficial effects of resveratrol may provide explanations for the “French paradox”, the finding that the consumption of red wine is associated with a decreased incidence of cardiovascular diseases15. Recently, studies have revealed that resveratrol can reduce ventricular arrhythmias in myocardial infarction2,16, ischemia/reperfusion17, heart failure and other pathological conditions18. Accumulating evidence indicates that increased oxidative stress is an important factor predisposing the diseased heart to calcium overload and lethal arrhythmias6,7,10. However, it is unclear whether resveratrol has protective effects against oxidative stress-induced arrhythmias. Therefore, the present study aims to investigate the effects of resveratrol on exogenous H2O2-induced arrhythmogenic activity and calcium overload and explore the underlying mechanisms.

Materials and methods

Animals

Six-month-old New Zealand White male rabbits weighing 2.0 to 3.0 kg were used for experiments. Animal care and handling procedures were approved by the Animal Care and Use Committee, Research Institute of Medicine, Shanghai Jiao Tong University, in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institute of Health (NIH Publication No 85–23, revised 1996).

Materials

H2O2 and resveratrol were purchased from Sigma Chemical (St Louis, MO, USA). Resveratrol was dissolved in dimethyl sulfoxide (DMSO) to make a stock solution of 50 mmol/L with the final concentration of DMSO at less than 0.1%. The same amount of DMSO was added to all groups to exclude the effects of DMSO on myocytes. All experiments were performed at a temperature of 25±1 °C, unless otherwise mentioned.

Cell isolation

Ventricular myocytes were enzymatically isolated from the hearts of New Zealand White rabbits as previously described10.

Measurement of cellular arrhythmias

Ca2+ transients and cell shortening were simultaneously detected as previously described19. Freshly isolated rabbit ventricular myocytes were incubated with a Ca2+ indicator, Fura-2 AM (2 μmol/L; Molecular Probes, Carlsbad, CA, USA), at 25 °C for 10 min. Loaded cells were electrically stimulated at a pacing cycle length (PCL) of 6 s or 2 s. Cell shortening was continuously monitored and Ca2+ transients were recorded every 3 min or when the cellular arrhythmias emerged. Cellular arrhythmias were induced upon addition of H2O2 (200 μmol/L). It is generally accepted that after-contractions and after-transients were defined as cellular proarrhythmogenic events, in which early after-transients/contractions (EATs/EACs) may correspond to EADs and delayed after-transients/contractions (DATs/DACs) may correspond to DADs7,20,21,22,23. Therefore, EATs/EACs and DATs/DACs can be used as proximal direct indices of EADs-like and DADs-like arrhythmias, respectively. The probabilities of EATs/EACs occurrence were assessed by calculating the percentage of calcium transients or cell shortenings that developed EATs/EACs within 1 min after the treatments (H2O2 and resveratrol) reached steady state. The calcium transient duration (CaTD) was measured as the time from the upstroke to 80% recovery (ie, CaTD80).

Measurement of [Ca2+]i and cell shortening

Myocytes were paced at a frequency of 0.5 Hz (PCL=2 s). The ratio of emitted fluorescence at 340 and 380 nm was recorded as an indicator of [Ca2+]i. Cell shortening was detected using an optical edge detector and collected using a charge-coupled device camera (IonOptix; Milton, MA, USA). The data were analyzed using IonWizard 6.0 software (IonOptix; Milton, MA, USA).

Electrophysiological study

Whole cell currents (INa,L, ICa,L) were recorded under voltage clamp mode with an EPC10 amplifier (Heka Electronic, Lambrecht, Pfalz, Germany) as previously described10,24. Current signals were filtered at 1 kHz and digitized at 10 kHz. For INa,L recording, pipettes were filled with an internal solution containing (in mmol/L): 120 CsCl2, 1.0 CaCl2, 5 MgCl2, 5 Na2ATP, 10 TEACl, 11 EGTA, and 10 HEPES (pH 7.2, adjusted with CsOH). Myocytes were bathed with a modified Tyrode's solution containing (in mmol/L): 135 NaCl, 5.4 CsCl2, 1.8 CaCl2, 1 MgCl2, 0.3 BaCl2, 0.33 NaH2PO4, 10 glucose, 10 HEPES, and 0.001 nicardipine at pH 7.3. INa,L was elicited by 300 ms depolarizing pulses from -120 to -20 mV at a PCL of 6 s. The amplitude of INa,L was measured at 200 ms after the initiation of the depolarization step.

To record ICa,L, patch pipettes were filled with an internal solution containing (in mmol/L): 110 Cs-Aspartate, 30 CsCl, 5 NaCl, 10 HEPES, 0.1 EGTA, 5 MgATP at pH 7.2 adjusted with CsOH, and the cells were superfused with a modified Tyrode's solution in which KCl was replaced with CsCl. To inactivate the Na+ current, the myocytes were stimulated by a 100 ms prepulse to -40 mV from the holding potential of -80 mV. Subsequently, ICa,L was elicited by a test depolarization step to 0 mV for 300 ms. The current density was calculated by dividing the amplitude by the cell capacitance.

Detection of intracellular ROS

The myocytes were incubated with 5 μmol/L C-DCDHF-DA-AM (Invitrogen, Grand Island, NY, USA) for 30 min. After loading, myocytes were washed twice with Tyrode's solution. Loaded myocytes were treated with H2O2 (200 μmol/L) in the absence or presence of resveratrol (50 μmol/L) for 20 min. Myocytes unexposed to H2O2 were set as the control group. C-DCDHF-DA can be oxidized by ROS to dichlorofluorescein (DCF), which was used to determine ROS production. Cellular DCF fluorescence intensities were determined by fluorescence microscopy with excitation and emission spectra of 488 and 525 nm, respectively.

Measurement of CaMKII activity

Isolated myocytes were treated with H2O2 (200 μmol/L) in the absence or presence of resveratrol (50 μmol/L) for 10 min. Myocytes unexposed to H2O2 were set as the control group. The CaMKII activity of myocytes was measured with the SignaTECT Calcium/Calmodulin-Dependent Protein Kinase Assay system (Promega, Madison, WI, USA) following the manufacturer's instructions. [γ-32P]ATP was purchased from PerkinElmer, Inc (Waltham, MA, USA).

Protein preparation and Western blotting

After intravenous anesthetization with sodium pentobarbital (30 mg/kg), rabbit hearts were immediately removed and mounted on a Langendorff apparatus and retrogradely perfused through the aorta (30 mL/min) with specific solutions depending on the condition for 10 min at 37 °C: Tyrode's solution containing vehicle, Tyrode's solution containing 200 μmol/L H2O2 and vehicle, and Tyrode's solution containing 200 μmol/L H2O2 and 50 μmol/L resveratrol. The left ventricular tissue was then isolated and flash frozen in liquid nitrogen. Protein preparations were performed using methods as described previously25 and the protein contents of CaMKII (antiphospho-CaMKII, 1:1000, Abcam, Cambridge, UK) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 1:1000, Beyotime, Haimen, China) in total protein preparations were analyzed by standard Western blotting.

Statistical analysis

The data are expressed as the mean±SEM. Statistical significance was assessed using Student's t-tests or ANOVA analysis followed by the Student-Newman-Keuls test. The frequencies of arrhythmia were compared using a chi-squared test. P<0.05 was considered statistically significant.

Results

Resveratrol suppressed and prevented H2O2-induced arrhythmogenic activity in rabbit ventricular myocytes

Cell shortening and calcium transients were simultaneously recorded from isolated rabbit ventricular myocytes. To stably induce EAD-like arrhythmias, the myocytes were paced at a low frequency (0.16 Hz, PCL=6 s) as previously described7. After reaching steady state, the myocytes were continuously perfused with 200 μmol/L H2O2. As shown in Figure 1A, the exposure of myocytes to H2O2 led to prolongation of CaTD80 and induction of EATs/EACs, the EAD-like arrhythmias. Most of the cells developed EATs/EACs (90.5%, 19 out of 21 cells from 10 rabbits) after an average exposure time of 7.4±0.5 min with H2O2. Prolonged exposure to H2O2 caused continuous EATs/EACs followed by DATs/DACs, the DAD-like arrhythmias. The other two cells (2 of 21 cells, 9.5%) did not induce EATs /EACs, but directly developed DATs/DACs.

Figure 1
figure 1

Effects of resveratrol on H2O2-induced arrhythmogenic activity at 0.16 Hz. In each panel, the top recordings are cell shortening and the bottom ones are calcium transients (PCL=6 s, 0.16 Hz). (A) Arrhythmogenic activity induced by H2O2 (200 μmol/L) in a rabbit ventricular myocyte. Continuous EATs/EACs developed after 10 min of H2O2 perfusion. At 15 min, continuous DATs/DACs developed. (B) After EATs/EACs were induced by 200 μmol/L H2O2, addition of resveratrol (Res, 50 μmol/L) suppressed EATs/EACs after approximately 4 min. (C) In the presence of resveratrol (Res, 50 μmol/L), a myocyte exposed to H2O2 (200 μmol/L) did not develop EATs/EACs until 48 min and no DATs/DACs were induced within 48 min. (D) Summarized histogram showing dose-dependent inhibitory effects of resveratrol (Res, 30 and 50 μmol/L) on the incidence of EATs/EACs induced by H2O2. (E) Summarized histogram shows the time to develop EATs/EACs in the presence of H2O2 and H2O2+Res (50 μmol/L). (F) Summarized histogram shows the proportion of cells displaying DATs/DACs in the presence of H2O2 and H2O2+Res (50 μmol/L). Mean±SEM. bP<0.05, cP<0.01 vs H2O2.

PowerPoint slide

Full size image

Next, we studied the effects of resveratrol on H2O2-induced arrhythmogenic activity. When EATs/EACs were induced by H2O2, addition of resveratrol (30 or 50 μmol/L) resulted in a significant suppression of the EATs/EACs within 3 to 5 min (Figure 1B). As summarized in Figure 1D, the probability of EATs/EACs occurrence was significantly reduced by resveratrol at both 50 and 30 μmol/L, in a dose-dependent manner (92%±2.8% vs 36%±1.2% and 6%±0.8%, P<0.05 and P<0.01, respectively). Accordingly, the CaTD80 prolongation induced by H2O2 was also attenuated by resveratrol. The CaTD80 was increased by H2O2 from 1055±72 to 1892±154 ms; resveratrol (50 μmol/L) decreased the CaTD80 to 1340±116 ms and attenuated by 55%±11% the prolongation of CaTD caused by H2O2 (P<0.05, n=8 cells from 4 rabbits).

Co-treatment of myocytes with resveratrol (50 μmol/L) blunted the prolongation of CaTD caused by H2O2 and significantly prevented the induction of EATs/EACs (Figure 1C). Although resveratrol could not completely prevent the induction of H2O2-induced EATs/EACs, it markedly delayed the time for the development of EATs/EACs from 7.4±0.5 min (H2O2 group) to 31±3 min (P<0.01, n=8 cells from 5 rabbits) (Figure 1E). In addition to preventing the induction of EATs/EACs caused by H2O2, resveratrol also prevented the myocytes from developing DATs/DACs (Figure 1C). All myocytes treated with H2O2 developed DATs/DACs (100%, 21 of 21 cells from 10 rabbits), while none of the myocytes treated with resveratrol during exposure to H2O2 developed these DAD-like arrhythmias within 40 min (0%, 0 of 10 cells from 6 rabbits, P<0.01 vs H2O2) (Figure 1F). Interestingly, resveratrol completely prevented H2O2-induced EATs/EACs and DATs/DACs when the myocytes were paced at a higher frequency (0.5 Hz, PCL=2 s). At 0.5 Hz, H2O2 induced EATs/EACs in 6 out of 11 myocytes (54.5%, P<0.05 vs at 0.16 Hz) at an average exposure time of 14±2.7 min (P<0.01 vs at 0.16 Hz) (Figure 2A). Meanwhile, H2O2 induced DATs/DACs in 11 of 11 myocytes (100%) within 30 min. Treatment with resveratrol (50 μmol/L) prevented H2O2-induced EATs/EACs and DATs/DACs for at least 40 min (P<0.05 and P<0.01 vs H2O2 group, respectively, n=10 cells from 6 rabbits) (Figures 2B–2D).

Figure 2
figure 2

Effects of resveratrol on H2O2-induced arrhythmogenic activity at 0.5 Hz. In each panel, the top recordings are cell shortening and the bottom ones are calcium transients (PCL=2 s, 0.5 Hz). (A) Arrhythmogenic activity induced by H2O2 (200 μmol/L) in rabbit ventricular myocytes. (B) In the presence of resveratrol (Res, 50 μmol/L), myocytes exposed to H2O2 (200 μmol/L) did not develop EATs/EACs and DATs/DACs for at least 40 min. (C) Summarized histogram shows proportion of cells displaying EATs/EACs in the presence of H2O2 and H2O2+Res (50 μmol/L). (D) Summarized histogram shows proportion of cells displaying DATs/DACs in the presence of H2O2 and H2O2+Res (50 μmol/L). Mean±SEM. bP<0.05, cP<0.01 vs H2O2.

PowerPoint slide

Full size image

Resveratrol prevented H2O2-induced calcium overload and cell death

H2O2-induced calcium overload is an important factor causing arrhythmias and cell death. In this series of experiments, the effects of 200 μmol/L H2O2 on [Ca2+]i and the contractions of rabbit ventricular myocytes were determined in the absence or presence of 50 μmol/L resveratrol (at a PCL of 2 s). As shown in Figures 3A and 3B, H2O2 led to time-dependent increases of diastolic [Ca2+]i. Treatment with resveratrol significantly blunted the time-dependent increases of diastolic [Ca2+]i during exposure to H2O2. After a 21 min incubation of myocytes with H2O2, diastolic [Ca2+]i was increased to 143%±2% and 114%±3% of the control in the absence and presence of resveratrol, respectively (P<0.01; Figure 3B).

Figure 3
figure 3

Effects of resveratrol on H2O2-induced calcium overload and hypercontracture. In this part, cells were paced at 0.5 Hz (PCL=2 s). (A) Original traces of Ca2+ transient in the absence and presence of 50 μmol/L resveratrol (Res) before (0 min) and after 9 and 21 min exposures to 200 μmol/L H2O2. (B) Time course of changes in diastolic [Ca2+]i, which was normalized as a percentage of the control (before application of drug). Each point represents data collected from 5 to 10 cells (n=5–10). (C) Representative cell shortening curve of myocytes exposed to H2O2 in the absence (top) or presence (bottom) of resveratrol (Res, 50 μmol/L). (D) Kaplan-Meier analysis of hypercontracture development. cP<0.01 vs H2O2 group.

PowerPoint slide

Full size image

Consistent with this finding, resveratrol prevented hypercontracture caused by H2O2. As shown in Figure 3C, H2O2-induced calcium overload led to the development of hypercontracture, which is a permanent reduction (<50%) of the longitudinal cell length attributable to a persistent activation of myofilaments6. H2O2 caused the development of hypercontracture within 21±1.3 min (n=43 cells from 10 rabbits). However, in the presence of resveratrol, myocytes (n=22 cells from 8 rabbits) exposed to H2O2 never developed hypercontracture within 40 min. A Kaplan-Meier survival analysis (Figure 3D) revealed that resveratrol effectively prevented ROS-induced hypercontracture and cell death.

Resveratrol inhibited ROS-induced INa,L augmentation

It has been reported previously6,7,10 that augmentation of INa,L plays a key role in ROS-induced cellular Na+ and Ca2+ overload and arrhythmias. Therefore, we first examined whether resveratrol could inhibit ROS-induced INa,L augmentation. When myocytes were exposed to H2O2, INa,L gradually increased in a time-dependent manner and reached a steady state at approximately 10–14 min after H2O2 treatment. Treatment with resveratrol markedly prevented the time-dependent increase of INa,L induced by H2O2 (Figure 4A). At the point of 12 min after H2O2 exposure, INa,L increased from -0.31±0.014 pA/pF to -1.06±0.037 pA/pF in the H2O2 group, increased by 245%±23% (n=6 cells from 4 rabbits). While in the resveratrol treatment group, INa,L increased from -0.31±0.032 pA/pF to -0.53±0.028 pA/pF, only a 75%±11% increase of INa,L (n=6 cells from 4 rabbits, P<0.01, Figure 4B). Resveratrol also reversed the H2O2-induced INa,L augmentation (Figure 4C). INa,L increased from -0.31±0.013 pA/pF to -1.07±0.045 pA/pF after the incubation of myocytes with H2O2 (P<0.05, n=6 cells from 5 rabbits). The addition of resveratrol reduced the current to -0.53±0.031 pA/pF in the presence of H2O2 (P<0.05; Figure 4D), a 178%±17% decrease of the H2O2-induced INa,L.

Figure 4
figure 4

Effects of resveratrol on H2O2-induced INa,L augmentation. INa,L was measured at a PCL of 6 s (protocol in inset) in isolated rabbit myocytes. Original traces (A) and mean data (B) showed that H2O2 significantly increased INa,L in myocytes. This increase was significantly prevented by resveratrol (Res, 50 μmol/L). Original traces (C) and mean data (D) showed that resveratrol (Res, 50 μmol/L) reversed the H2O2-induced INa,L augmentation. bP<0.05, cP<0.01 vs H2O2 group. eP<0.05 vs control. Mean±SEM. n=6 for each group.

PowerPoint slide

Full size image

Resveratrol inhibited ROS-induced ICa,L augmentation

Because reactivation of ICa,L contributes to the H2O2-induced EAD in rabbit ventricular myocytes7, we next investigated the potential involvement of ICa,L in the inhibitory effect of resveratrol on H2O2-induced EATs/EACs. As shown in Figures 5A and 5B, H2O2 increased ICa,L from -8.2±0.8 pA/pF to -12.3±1.4 pA/pF (P<0.05, n=6 cells from 4 rabbits) and resveratrol reduced the current to -9.1±0.8 pA/pF (P<0.05), indicating that resveratrol markedly inhibited H2O2-induced ICa,L augmentation.

Figure 5
figure 5

Effects of resveratrol on H2O2-induced ICa,L augmentation. (A) Representative ICa,L traces under control conditions, in the presence of H2O2, and H2O2+resveratrol (Res, 50 μmol/L). (B) A bar graph summarizing the H2O2-induced increase of ICa,L, which is significantly suppressed by resveratrol (Res, 50 μmol/L). bP<0.05 vs H2O2. eP<0.05 vs control. Mean±SEM. n=6.

PowerPoint slide

Full size image

Resveratrol reduced H2O2-induced ROS production

To determine whether resveratrol inhibits the arrhythmogenic activity and calcium overload via decreasing intracellular ROS, the effect of resveratrol on intracellular ROS levels was measured. ROS production was monitored by detecting the fluorescence from the reaction of intracellular ROS with C-DCDHF-DA using fluorescence microscopy. As shown in Figures 6A and 6B, DCF fluorescence intensity was significantly increased after a 20 min exposure with H2O2. Treatment with resveratrol significantly decreased the generation of ROS (P<0.05 vs H2O2 group).

Figure 6
figure 6

Effects of resveratrol on H2O2-induced ROS production. (A) DCF fluorescence images of myocytes exposed to H2O2 (200 μmol/L) in the absence and presence of resveratrol (Res, 50 μmol/L). (B) DCF fluorescence intensity measured as the image optical density (IOD) per unit area. bP<0.05 vs H2O2. eP<0.05 vs control. Mean±SEM. n=3.

PowerPoint slide

Full size image

Resveratrol prevented H2O2-induced CaMKII activation

It has been reported that ROS can markedly activate CaMKII6. Having confirmed that resveratrol reduced ROS production, we next tested whether ROS-induced CaMKII activation could be prevented by resveratrol. Using a CaMKII assay kit, we found that H2O2 (200 μmol/L) effectively increased CaMKII activity and resveratrol (50 μmol/L) significantly prevented this activation in isolated myocytes (P<0.05 vs H2O2 group; Figure 7A). With a Western blotting analysis using the antibody against p-CaMKII, we further confirmed that resveratrol prevented H2O2-induced CaMKII activation. As shown in Figures 7B and 7C, CaMKII phosphorylation was significantly increased in hearts perfused with 200 μmol/L H2O2 for 10 min (P<0.05 vs control), and resveratrol decreased the activation of CaMKII (P<0.05 vs H2O2 group).

Figure 7
figure 7

Effects of resveratrol on H2O2-induced CaMKII activation. (A) Histograms illustrating the CaMKII activity in response to a 10 min exposure to 200 μmol/L H2O2 in the absence or presence of resveratrol (Res, 50 μmol/L). bP<0.05 vs H2O2. eP<0.05 vs control. n=4. (B) Western blotting analysis showing the levels of phosphorylated CaMKII in response to a 10 min exposure to 200 μmol/L H2O2 in the absence or presence of resveratrol (Res, 50 μmol/L). (C) Mean data of CaMKII phosphorylation normalized to the total amount of GAPDH from 3 experiments. bP<0.05 vs H2O2. eP<0.05 vs control. Mean±SEM. n=3.

PowerPoint slide

Full size image

Discussion

The cardioprotective effects of resveratrol are complex and are not completely understood. In the present study, we provide the first evidence showing that resveratrol prevents and suppresses oxidative stress-induced arrhythmogenic activity. It also prevents oxidative stress-induced calcium overload and cell death. Furthermore, our study suggests that the underlying mechanisms involve: (1) inhibition of INa,L/ICa,L; (2) reduction of ROS generation; and (3) prevention of CaMKII activation.

EAD-mediated triggered activity plays an important role in ROS-induced arrhythmias26. In this study, we observed that resveratrol significantly suppressed and prevented H2O2-induced EATs/EACs, ie, EAD-like arrhythmias. At 0.16 Hz (PCL 6 s), resveratrol markedly delayed the time to develop EATs/EACs. It also reversed H2O2-induced EATs/EACs when added after EATs/EACs were induced. Recent studies7,10 have implicated that ROS-activated CaMKII mediates EAD formation. Downstream targets of ROS include CaMKII, ICa,L, and INa,L. ROS-activated CaMKII causes ICa,L and INa,L augmentation through the regulation of calcium and sodium channel phosphorylation sites27. Xie et al7,10 have demonstrated that both ICa,L and INa,L are required for ROS-induced EAD formation: activation of late INa to reduce the repolarization reserve (ie, to prolong APD) and the modification of ICa,L to enhance its reactivation properties to generate the EAD upstroke. In the present study, we found that resveratrol effectively reduced H2O2-induced ROS generation. With a CaMKII assay kit, we showed that resveratrol significantly prevented H2O2-induced CaMKII activation in isolated myocytes. Using CaMKII phosphorylation at threonine-287 as a marker of CaMKII activity, we further confirmed that resveratrol significantly prevented H2O2-induced CaMKII activation in rabbit hearts. By reduction of ROS generation and CaMKII activation, resveratrol may indirectly inhibit INa,L and ICa,L, and subsequently prevent and suppress H2O2-induced EATs/EACs. Previous studies16,28 have reported that resveratrol can directly modulate ion channels. For instance, resveratrol can inhibit ICa,L and enhance the ATP-sensitive potassium current (IK,ATP) in isolated ventricular myocytes16. It can also inhibit the sodium current (INa), transient (Ito) and sustain (Iss) outward potassium currents17. Therefore, we cannot rule out the possibility that a direct inhibition of INa,L and ICa,L by resveratrol contributes to its inhibition of EATs/EACs.

Interestingly, when myocytes were paced at a higher frequency (0.5 Hz, at a PCL of 2 s), resveratrol completely prevented the H2O2-induced EATs/EACs for at least 40 min, indicating that resveratrol has enhanced anti-arrhythmias properties at a higher frequency. This may be due to a larger INa,L at lower frequencies24,29, which makes myocytes more prone to EATs/EACs. In sinus rhythm, INa,L is much smaller than it is at 0.16 Hz. Thus, resveratrol may be expected to be more effective in preventing H2O2-induced EATs/EACs under sinus rhythm.

Calcium overload is an important factor causing DAD-induced arrhythmias and cell death6. In this study, we showed that resveratrol significantly blunted time-dependent [Ca2+]i increase induced by H2O2. As a result, resveratrol prevented H2O2-induced hypercontracture and cell death. Accordingly, resveratrol completely prevented the development of DATs/DACs, the DAD-like arrhythmias. Recent studies indicate that CaMKII activation and INa,L augmentation play important roles in the ROS-induced Ca2+ overload and arrhythmias4. Under various cardiac pathological conditions, there exists a lethal cycle between CaMKII and INa,L30,31. ROS dramatically enhances INa,L via CaMKII activation; the enhanced INa,L further activates CaMKII, leading to CaMKII over-activation6,30. Consequently, on the one hand, the ROS-induced over-activation of CaMKII phosphorylates RyR2 and causes sarcoplasmic reticulum (SR) calcium leak32; on the other hand, the CaMKII activation-enhanced INa,L leads to sodium overload and subsequently calcium overload via the NCX4. Calcium overload and SR calcium leak always induce arrhythmogenic calcium wave and cause DAD-like arrhythmias22. Additionally, calcium overload leads to the development of hypercontracture and cell death6. Therefore, elevated ROS leads to the development of a lethal cycle between CaMKII and INa,L in diseased hearts. In the present study, we showed that resveratrol prevented H2O2-induced CaMKII activation and augmentation of INa,L, and effectively blocked the lethal cycle between them. Thus, resveratrol prevented the H2O2-induced calcium overload and DAD-like arrhythmias. In addition, ROS can directly oxidize RyR233, leading to an increased calcium spark, DAD-like arrhythmias and calcium overload. By reducing ROS generation, resveratrol may reduce RyR2 oxidization and also suppress DAD-like arrhythmias and calcium overload.

A limitation of the present study was that we did not investigate the anti-arrhythmias properties of resveratrol at the organ/whole heart level. This requires further study in the future.

In conclusion, resveratrol significantly suppressed and prevented oxidative stress-induced arrhythmogenic activity and calcium overload by inhibition of INa,L/ICa,L, reduction of ROS generation, and prevention of CaMKII activation. This study provides a new way to treat arrhythmias and calcium overload induced by ROS. It also provides a novel drug to prevent CaMKII over-activation, a hallmark of pathological conditions, such as heart failure.

Author contribution

Yi-gang LI and Yue-peng WANG designed the research; Wei LI, Ling GAO, Peng-pai ZHANG, Qing ZHOU, and Quan-fu XU performed the research; Zhi-wen ZHOU, Kai GUO, and Ren-hua CHEN analyzed data; Wei LI wrote the paper; Yi-gang LI, Yue-peng WANG, and Huang-tian YANG revised the manuscript.