Resveratrol, a Red Wine Antioxidant, Reduces Atrial Fibrillation Susceptibility in the

Failing Heart by PI3K/AKT/eNOS Signaling Pathway Activation

Eric Chong1,2*, MD; Shih-Lin Chang1,2*, MD PhD; Ya-Wen Hsiao2*, PhD; Rahul Singhal1,2, MD; Shuen Hsin Liu1,2, MD; Trung Leha1,2, MD; Wen-Yu Lin1,2, MD; Chiao-Po Hsu1,3,MD, PhD; Yao-Chang Chen4, MSc; Yi-Jen Chen5,6, MD, PhD; Tsu-Juey Wu1,7, M.D., PhD; Satoshi Higa8, M.D., PhD; Shih-Ann Chen1,2, MD

1Department of Medicine, National Yang-Ming University School of Medicine, Taipei, Taiwan; 2Division of Cardiology, Taipei Veterans General Hospital, Taipei, Taiwan; 3Division of Cardiovascular Surgery, Taipei Veterans General Hospital, Taipei, Taiwan; 4Department of Biomedical Engineering, National Defense Medical Center, Taipei, Taiwan; 5Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan; 6Division of Cardiovascular Medicine, Department of Internal Medicine, Wan-Fang Hospital, Taipei Medical University, Taipei, Taiwan; 7Division of Cardiology, Taichung Veterans General Hospital, Taichung, Taiwan; 8Cardiac Electrophysiology and Pacing Laboratory, Division of Cardiovascular Medicine, Makiminato Central Hospital, Okinawa, Japan.*Drs Eric Chong, Shih-Lin Chang and Ya-Wen Hsiao contributed equally to this paper.

Running title: Resveratrol Reduces AF

Conflict of interest: none declared

Total word count: 4993

Address for correspondence to: Shih-Ann Chen, MDDivision of Cardiology, Veterans General Hospital-Taipei, 201 Sec 2Shih-Pai Road, Taipei, TaiwanTel: 886-2-2875-7156, Fax: 886-2-2873-5656E-Mail: epsachen@ms41.hinet.net

1

Abstract

Background: Resveratrol has shown benefits in reducing ventricular remodeling and

arrhythmias.

Objective: This study aimed to assess the therapeutic efficacy of resveratrol in reducing

atrial fibrillation (AF) in a heart failure (HF) model and explore underlying mechanisms.

Methods: HF rabbits were created 4 weeks after undergoing coronary ligation. Group 1 (n =

6) included (a) normal; (b) HF sham; (c) HF rabbits treated for 1 week with intraperitoneal

injections of resveratrol; (d) resveratrol + wortmannin, and (e) resveratrol +

diphenyleneiodonium chloride (DPI). All rabbits underwent epicardial catheter stimulation.

Collagen content, mRNA and protein expression in ionic channels, and PI3K/AKT/eNOS

signaling pathways were studied in left atrial appendage (LAA) preparations. To investigate

acute drug effects on left atrial (LA) electrophysiology, Groups 2 a–e (n = 6) were subjected

to Langendorff perfusion.

Results: Higher AF inducibility was found in the HF group and groups that were

administered PI3K and eNOS inhibitors than in the normal and resveratrol-treated groups (P

< 0.001). Histological analysis of LAA revealed a decrease in fibrosis in resveratrol-treated

groups compared with the HF group (8.95% ± 1.53% vs. 26.62% ± 2.19%, P < 0.001). In

real-time PCR analysis, ionic channels including Kv1.4, Kv1.5, KvLQT1, Kir2.1, Nav1.5,

Cav1.2, NCX, SERCA2a, and PLB were upregulated by resveratrol. PI3K, AKT, and eNOS

2

mRNA and protein expressions were upregulated by resveratrol but were inhibited by the co-

administration of wortmannin and DPI.

Conclusion: Resveratrol decreases LA fibrosis and regulates variation in ionic channels to

reduce AF through the PI3K/AKT/eNOS signaling pathway.

Key words: Resveratrol; atrial fibrillation; ionic channel; signaling pathway

3

Abbreviations

AF = atrial fibrillation

HF = heart failure

SR = sarcoplasmic reticulum

PI3K = phosphoinositide 3-kinases

eNOS = endothelial nitric oxide synthase

DPI = diphenyleneiodonium chloride

LAA = left atrial appendage

LVEF = left ventricular ejection fraction

MAP = monophasic action potential

EAD = early after depolarization

DAD = delayed after polarization

4

Introduction

Drinking red wine in moderation has long been known to reduce the risk of heart

disease. Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a natural phenol, and there has been

growing interest in its pharmaceutical property in treating cardiac arrhythmias1. Atrial

fibrillation (AF) is one of the fastest growing cardiac arrhythmias, that is largely attributable

to an increase in the prevalence of heart failure (HF) population2, 3. HF frequently develops

after myocardial infarction, and it is one of the most important risk factors for the

development of AF. HF induces atrial Ca2+ overload and leads to spontaneous Ca2+ leakage,

triggered activity, and eventually AF3. Effective therapy to prevent AF in patients with HF is

limited because of insufficient understanding of molecular mechanisms involved in its

development. Recent studies have shown that phosphoinositide 3-kinases (PI3Ks), a

cardioprotective protein, plays an important role in the inhibition of pathological signaling

cascades in AF4. Loss of PI3K activity directly contributes to AF phenotypic expression,

whereas increase in its activity markedly decreases atrial fibrosis5. Resveratrol can enhance

endothelial nitric oxide synthase (eNOS) function. Although reduction in eNOS has been

linked to the development of AF via the calcium overload mechanisms6, the molecular

mechanism involved in suppression of AF during HF with resveratrol therapy has not been

previously explored. Based on the fragmentary evidence, we hypothesized that resveratrol

could suppress AF in HF hearts via enhancement of the PI3K/AKT/eNOS pathway. The aims

5

of our study were therefore, (1) to investigate the therapeutic effect of resveratrol in AF

prevention in an ischemic HF model and (2) to testify the hypothesis that the AF suppression

mechanism works along the PI3K/AKT/eNOS pathway.

Methods

The detailed and expanded methodology is provided in the online supplement.

HF model and animal preparation by coronary ligation

The study protocol was reviewed and approved by the Institutional Animal Care and Ethics

Committee of the Taipei Veterans General Hospital. The study was conducted in accordance

with the US National Institutes of Health and the European Commission guidelines. The

procedure for the creation of the HF rabbit model by myocardial ischemia was modified from

a previous study7.

Grouping

Group 1 included 5 subgroups: (a) normal rabbits (b) HF sham rabbits (c) HF rabbits

receiving 1 week of intraperitoneal (ip) resveratrol injection (d) HF rabbits receiving 1 week

of ip resveratrol and wortmannin injection (e) HF rabbits receiving 1 week of ip resveratrol

and diphenyleneiodonium chloride (DPI) injection, and there were 6 rabbits in each

subgroup. The purpose of designing Group 1 was to evaluate the chronic therapeutic effects

of resveratrol administration.

Group 2 also included 5 subgroups as in group 1, with 6 rabbits in each subgroup. The rabbit

6

hearts were hanged on Langendorff perfusion to investigate AF inducibility and

electrophysiological changes following treatment by acute perfusion of drugs. The purpose of

designing Group 2 was to evaluate the acute therapeutic effects of resveratrol.

Rationale for resveratrol dosage

In previous Langendorff perfusion studies, the dosage of resveratrol ranged from 1 to 100

uM. Hung et al. demonstrated the benefit of using a dosage of 10 uM for cardioprotection

following acute ischemic reperfusion8, and this dosage was also used in a similar animal

experiment9. We considered that a dosage of 10 uM would be adequate with regard to

studying the acute therapeutic effects of resveratrol, and this dosage was used for the

Langendorff perfusion experiment in the acute treatment group.

Evaluation of chronic effects of resveratrol in Group 1

Drug administration

The HF rabbit model was created 4 weeks after coronary ligation when drug administration

was initiated for the chronic treatment group. Resveratrol (2.5 mg/kg, i.p.) was administered

daily for 1 week. For Groups 1d and 1e, DPI (3 mg/kg) and wortmannin (0.6 mg/kg) were

simultaneously injected with resveratrol in order to study the antagonistic effects of these

pathway inhibitors during resveratrol treatment. After 1 week of drug administration, we

performed open-chest AF provocation test, which was followed by sacrificing the animals for

left atrial tissue sample collection.

AF provocation test

7

Prior to sacrificing Group 1 rabbits, open-chest epicardial catheter stimulation was performed

for AF provocation. The rabbits were administered general anesthesia, intubated, and

ventilated with ambient air. A central chest incision was made to expose the beating heart.

The pericardial space was then entered by careful dissection. A 4-mm pacing catheter was

placed and manually stabilized on the left atrial appendage (LAA) surface, and AF

provocation was performed via rapid atrial pacing (5–20 Hz, 4 × threshold current, 10 s). The

percentage of inducible AF episodes was calculated as the ratio of induced episodes of

duration >1 s to the total number of pacing stimuli applied (10 attempts were made). This

induction protocol was standardized across all experiments.

Real-time polymerase chain reaction (PCR)

Total RNA was isolated from rabbit LAA tissue (Group 1) using the RNeasy® Fibrous Tissue

Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s manual. The detailed

methodology is provided in the online supplement.

Western blotting analysis

Tissues from LAA isolated from Group 1 rabbits were suspended in a lysis buffer. Protein

samples were separated on a 10% or 15% SDS-polyacrylamide electrophoresis gel and then

transferred to polyvinylidene difluoridemembranes (Millipore, Bedford, MA, USA). The

membrane was blocked in 5% BSA buffer for 1 h and then incubated with primary antibodies

against phospholamban (PLB), sarcoplasmic reticulum Ca2+-ATPase (SERCA2a) (Thermo

8

Scientific), PLB phosphor Ser-16 and Thr-17 (Badrilla Ltd., Leeds, UK), PI3K-p85, PL3K-

p110, phosphor-AKT, and phosphor-eNOS (Santa Cruz, CA, USA), cleaved caspase 3

(Millipore) and caspase 3 (Thermo Scientific). Proteins were detected by Immobilon™

Western chemiluminescent HRP substrate (Millipore).

Evaluation of acute effects of resveratrol Group 2

Langendorff perfusion experiment

The group 2 rabbits were subjected to the Langendorff perfusion experiment to study acute

effects of resveratrol.

Electrophysiological study

Electrophysiological study was performed during the Langendorff whole-heart perfusion

experiment in Group 2 rabbits.

Statistical analysis

Quantitative data were expressed as mean ± SD. Two-way non-repeat analysis of variance

(ANOVA) was used to compare differences between subgroups, and two-way repeat

ANOVA was used to compare differences before and after drug administration in subgroups.

Multiple comparisons were analyzed by Bonferroni test, and chi-square test or Fisher exact

test were used for analysis of categorical data. The value of p <0.05 was considered

statistically significant. A senior biostatistician performed the statistical analysis using the

SPSS statistical software (Version 17.0).

Results

9

Echocardiography

Echocardiography findings are presented in Table 1. Although HF rabbits had lower left

ventricular ejection fraction (LVEF) compared with normal rabbits, those administered with

resveratrol showed improved LVEF compared with HF rabbits, and the beneficial effect of

resveratrol on LVEF was abolished after addition of either PI3K or eNOS inhibitor. HF

rabbits had larger left atrial (LA) diameter compared with normal rabbits and HF rabbits with

resveratrol. Co-administration of PI3K or eNOS inhibitor abolished the beneficial effect of

resveratrol in reducing atrial size.

AF episodes during open-chest provocation in Group 1

AF episodes were significantly increased in HF rabbits compared with normal controls (38.0

13.2% vs. 0%, P = 0.001), and were significantly reduced in the resveratrol-administered

group (38.0±13.2% vs. 5.0±5.8%, P = 0.001). The AF episodes were significantly higher in

rabbits co-administered with resveratrol and PI3K or eNOS inhibitors compared with rabbits

administered with resveratrol alone (28.0±16.3% and 26.0±8.94%, respectively vs. 5.0±5.8%,

P = 0.002).

Electrophysiological characteristics during Langendorff perfusion in Group 2

Figure 1 shows various electrophysiological parameters including APD90, APD50, APD20, and

activation time during LA endocardial pacing at cycle lengths of 250 and 300 ms. The results

from MAP analysis show that LA in HF developed a significantly more prolonged APD

compared with LA in control. Resveratrol treatment reduced APD toward the normal level.

10

Co-administration of resveratrol and PI3K or eNOS inhibitor opposed the effects of

resveratrol, leading to APD prolongation. There was no difference in local activation time

among the 5 subgroups.

Arrhythmogenesis during Langendorff perfusion in Group 2

The incidence of early after depolarization (EAD), delayed after polarization (DAD), and AF

in the HF group significantly increased compared with the control group (Figure 2).

Resveratrol treatment reduced the incidence of triggered activities and AF, but the beneficial

AF suppressive effect was abolished by co-administration of either PI3K or eNOS inhibitor.

Tissue fibrosis inhibited by resveratrol

Histological and immunohistochemical staining of LAA in control showed minimal collagen

formation and fibrosis (Figure 3A and 3B), while those LAA in HF showed extensive

interstitial fibrosis, prominent myocyte hypertrophy, and cell loss. Resveratrol administration

resulted in significant attenuation of interstitial fibrosis; demonstrating that tissue fibrosis in

HF rabbits administered with resveratrol was significantly less comparing with HF controls.

However, this beneficial result was abolished after co-administration of either wortmannin or

DPI. To investigate the effect of resveratrol on apoptosis, we examed the cleaved caspase 3

expressions in LAA after resveratrol treatment. We found that the protein expression of

cleaved caspase 3 was increased in HF rabbits and decreased after treatment with resveratrol,

11

and this enhanced effect was reversed by co-administration of PI3K or eNOS inhibitor

(Figure 3C).

Ion channel expression increased after resveratrol treatment in Group 1

Repolarization delays caused by ion channel remodeling are common in AF. To determine

the mechanism of reduced APD after resveratrol administration in our animal model, we

investigated mRNA expression in various ion channels in the LAA samples by real-time PCR

(Figures 4 and 5A). The mRNA levels of ionic channels Kv1.4, Kv1.5, KvLQT1, Nav1.5,

NCX, Cav1.2, ERG SERCA2a, and PLB significantly decreased in LAA of HF rabbits in

comparison with those in LAA of control rabbits. However, all ionic channels except ERG

were significantly up regulated after resveratrol administration. Wortmannin, a PI3K

inhibitor, abolished the effects of resveratrol on mRNA expression in Cav1.2, Nav1.5, NCX,

Kv1.4, Kv1.5, KvLQT1, SERCA2a, and PLB. DPI, an eNOS inhibitor, abolished the effects

of resveratrol on mRNA expression in Cav1.2, Nav1.5, NCX, Kv1.4, and Kv1.5 but not in

KvLQT1, SERCA2a, and PLB. Furthermore, we determined the protein levels of SERCA2a,

PLB, and PLB phosphor-Ser-16 and Thr-17 in LAA of HF rabbits (Figure 5B). The data

show that protein levels of SERCA2a, PLB, and PLB phosphor-Ser-16 and Thr-17 decreased

in HF compared with those in controls, and these protein levels increased after resveratrol

administration. Moreover, wortmannin and DPI abolished the effects of resveratrol on the

protein expression.

12

PI3K, AKT, and eNOS upregulated after resveratrol treatment

Previous studies have demonstrated that decreased PI3K activity altered ion channel gene

expression in ventricular tissue, whereas increased PI3K activity markedly decreased atrial

fibrosis. AKT, PI3K’s downstream cardiac target, promoted growth and survival of cardiac

myocytes10. In addition, eNOS has been suggested to play an important role in AF pathways.

To investigate the mechanisms of resveratrol, we investigated the levels of mRNA and

proteins in PI3K, AKT, and eNOS. The results of real-time PCR and immune-blotting

analysis showed that mRNA expression of PI3K, AKT, and eNOS significantly reduced in

HF rabbits compared with controls (Figure 6). We also found that the protein expression of

PI3K-p110, PI3K -p85, p-AKT, and p-eNOS was down regulated in HF rabbits. However,

mRNA and protein expressions significantly increased after treatment with resveratrol, and

this enhanced effect was reversed by co-administration of PI3K or eNOS inhibitor.

Discussion

Resveratrol modulates ionic channels and suppresses AF

Resveratrol, a phytoalexin polyphenol, possesses diverse cardioprotective benefits, and it is

attributable to its ability to induce preconditioning. However, there is very limited

information on resveratrol therapy in HF and AF. Our present study demonstrated that

resveratrol administration significantly changed the abnormal signaling of ionic channel in

HF as compared with normal rabbits. These changes were corresponding to the correction of

action potential duration and the suppression of AF. Moreover, the effects of resveratrol on

the changes of ionic channels and electrophysiology, including AF prevention were offset by

the inhibitors of resveratrol. Therefore, there could be a causal relation between

13

downregulation of ion channels and induction of AF. HF decreases sodium and potassium

currents while increasing NCX and INa-Late, which may be responsible for APD prolongation11.

HF, in both human and animal models, is associated with APD prolongation because of down

regulation of K+ currents, predominantly the calcium-independent Ito (Kv1.4)12. Animal

studies have also demonstrated reduction of Kv1.5 and KvLQT1 in a rabbit HF model13.

From our study, resveratrol administration significantly reversed some of the abnormal

potassium signaling channel changes in HF. Gating dysfunction and changes in the number of

functional channels of cardiac sodium channels were linked to several cardiac arrhythmias14.

Therapies directed at reducing the activation of abnormal mRNA or their upstream induction

stimuli may prevent the reduction in cardiac Na+ channels seen in HF and reduce risks of

arrhythmias in this condition15. In our study, Nav1.5 was down regulated in HF rabbits, while

administration of resveratrol reversed this change toward normal controls, a result consistent

with previous findings.

When APD is excessively prolonged during HF, cell membrane Ca2+ currents recover from

inactivation and allow the inward movement of Ca2+, leading to EAD. Abnormal cells

generating EAD adjacent to normally repolarizing cells raise the latter’s threshold, causing

them to fire and initiate focal activity16. Abnormal calcium regulation plays an important role

in AF genesis. Cellular Ca2+overload, leading to spontaneous diastolic Ca2+ release,

arrhythmogenic diastolic membrane potential oscillations, and triggered activity account for

14

atrial arrhythmogenesis in HF17. HF increases atrial SR Ca2+ load, thereby promoting

spontaneous SR Ca2+ release3. Ischemic atrial substrates cause Ca2+-release events, increased

Na+-Ca2+ exchange current and sustained reentry, which could trigger and maintain AF18. A

recent study using rat myocytes demonstrated that the expression of RyR-associated proteins

including SERCA, PLB phosphorylation at Ser16 or Thr17, and L-type Ca2+ channel

(Cav1.2) significantly reduced in HF19. Systolic SR reduced Ca2+ release and diastolic SR

Ca2+ leak along with reduced SR Ca2+ uptake and defective Ca2+-induced Ca2+ release may

contribute to HF and AF inducibility. In agreement with previous studies, the present study

showed down regulation of SERCA and Cav1.2 in LA in HF, whereas PLB phosphorylation

at Ser16 and Thr17 decreased. Resveratrol administration increased mRNA expression of

SERCA, PLB, and Cav1.2, thereby enhancing cardiac function and reducing atrial

arrhythmogenesis.

Resveratrol activates PI3K/AKT/eNOS signaling pathway

Normal cardiac calcium transients are initiated by Cav1.2 channels at the cardiac T-tubules,

while PI3K p110 is required to maintain the organized network of T-tubules that is vital for

efficient Ca2+-induced Ca2+ release and ventricular contraction20. Cardiac pathophysiology is

significantly linked to receptor-mediated signal transduction, and recent progress in research

on such transduction events in the heart highlighted the role of a family of lipid kinases

known as PI3Ks21. Currently, there is no data linking resveratrol, PI3K/AKT, eNOS, and AF,

15

although a recent study on resveratrol and cancer suggested that resveratrol targets multiple

components of the PI3K/AKT/mTOR signaling pathways22. Another recent study, on

hemorrhagic traumatized rats, showed that traumatic hemorrhage decreased cardiac AKT

phosphorylation (p-AKT), and resveratrol administration following traumatic hemorrhage

prevented decrease in cardiac p-AKT. Co-administration of the PI3K inhibitor, wortmannin,

abolished the beneficial effects of resveratrol. eNOS has also been suggested to play an

important role in AF pathways, and it will be very interesting to know whether resveratrol

can indeed suppress AF during HF via the PI3K/AKT/eNOS pathway.

PI3K signaling in cardiomyocytes suggests that systemic myocardial PI3K inhibition has the

potential to cause cardiac dysfunction by decreasing cardiac contractility and increasing

susceptibility to cardiac arrhythmias21. Mouse heart, lacking the PI3K p110α catalytic

subunit, exhibited prolonged APD and QT intervals23. Through phosphatidylinositol 3-

triphosphate (PIP3), PI3K activates AKT, which inhibits apoptosis mediated by eNOS as the

final messenger. AKT also activates Ca2+-activated K+ channels and may lead to reduction in

APD24. Activation of PI3K has also been shown to prevent myocardial infarction-induced

heart failure in mice25. It is therefore of value to explore the relationships between AF and the

PI3K/AKT pathway. Our data showed that AF episodes decreased in HF rabbits treated with

resveratrol, and protein levels of PI3K-p110, p85, and p-AKT were up regulated in HF

rabbits after administration of resveratrol for 1 week. Co-administration of PI3K inhibitor

16

abolished the protective effect of resveratrol and increased AF episodes during Langendorff

perfusion. Electrophysiological parameters were altered toward those of normal controls

during Langendorff perfusion with resveratrol, but these changes were reversed after co-

administration of the PI3K inhibitor. Evidence from this study suggests that PI3K

suppression occurred in HF and contributed to AF genesis. Enhancement of the PI3K

signaling pathway by resveratrol treatment proved protective against AF in HF.

In a human study, eNOS expression and activity were down regulated in acute HF26. The

regulation of myocardial function by eNOS is important for the maintenance of myocardial

Ca2+ homeostasis, relaxation, and distensibility, and for protection from arrhythmia and

abnormal stress stimuli27. We found that down regulation of eNOS played an essential part in

the development of AF during HF, and thus by enhancing eNOS expression and activity, AF

can be effectively suppressed.

Potential upstream therapy of AF

AF is a growing problem in the HF population and has been described as “epidemic28.”

Current therapeutic options for prevention of AF in patients with HF are not only limited but

also lacking in efficacy. Consequently, a variety of novel, mechanism-based approaches are

under development29. One approach has been employed to target atrial remodeling, with the

use of “upstream therapy” interventions that interfere with putative signaling pathways.

Resveratrol treatment enhanced both mRNA and protein expressions of PI3K/AKT/eNOS.

17

The reversal of electrophysiological phenotypes and the changes of mRNA and protein levels

of ionic channels following inhibition of PI3K/AKT/eNOS, suggesting that the changes in

mRNA and protein levels of ionic channels caused by resveratrol could contribute the

alteration of EP phenotype. By intervening through the upstream PI3K/AKT/eNOS pathway,

resveratrol can effectively reverse abnormal Ca2+ signaling and APD change and thereby

significantly suppress AF, DAD, and EAD episodes. The proposed pathway is shown in

Figure 7.

Resveratrol administration in ischemic HF resulted in improvement of LV function,

attenuation of arrhythmogenic atrial remodeling and reduction of LA fibrosis, thereby

reducing vulnerability to AF. AF protection was related to significant drug-induced effects on

the enhancement of eNOS and upstream PI3K/AKT. Our study illustrates the importance of

the PI3K/AKT/eNOS pathway in AF genesis and opens the door for pathway-enhancing

therapy in the novel armamentarium of AF treatment.

Limitation

Although wortmannin is not highly selective for PI3K isoforms, it has been widely used as

pharmaceutical tool for elucidating the role of PI3Ks30, 31. There is unfortunately a lacking of

a more specific PI3K isoform inhibitor for the studying of cardiac mechanism. Wortamnnin is

a reasonable choice as the initial PI3K inhibitor. Furthermore, the potential mechanism of

resveratrol in suppression of AF was investigated at the mRNA and protein levels. The

18

inhibitors of PI3K pathway were used to validate our hypothesis. Our study showed the

transcriptional regulation, however, the possibility of presence of a theoretical bystander

pathway within the complex HF-induced remodeling can not be fully excluded, and it can be

difficult to demonstrate the putative effects of resveratrol. The key intermediate mediators

involved in remodeling and the reversal by reseveratrol will need further study. We did not

perform resveratrol, wortmannin and DPI treatment in normal rabbit. The effect of these

drugs in the normal hearts is unclear. Moreover, the acute effect of resveratrol might not be

related to the protective effect in long-term use.

19

Conclusion

Resveratrol regulates ionic channels and attenuates LA fibrosis through activation of the

PI3K/AKT/eNOS signaling pathway, which may suppress AF in the failing heart.

Acknowledgments

This study was supported by grants NSC 100-2314-B-075-046-MY3, 101-2314-B-010-057-

MY3, 101-2314-B-075-056-MY3, Taipei Veterans General Hospital grant V102C-054,

V102E7-002, VGHUST102-G1-1-1, International Center of Excellence in Advanced Bio-

engineering sponsored by the Taiwan National Science CouncilI-RiCE Program under Grant

Number NSC-99-2911-I-009 -101; NSC 99-2314-B-075-032, NSC 99-2314-B-016-034-

MY3, and Foundation for the Development of Internal Medicine in Okinawa (26-02-001, 26-

02-002, 26-02-003).

20

References

1. Hung LM, Chen JK, Huang SS, Lee RS, Su MJ. Cardioprotective effect of

resveratrol, a natural antioxidant derived from grapes. Cardiovasc Res Aug 18

2000;47:549-555.

2. Chen YJ, Chen SA, Chen YC, Yeh HI, Chan P, Chang MS, Lin CI. Effects of rapid

atrial pacing on the arrhythmogenic activity of single cardiomyocytes from pulmonary

veins: implication in initiation of atrial fibrillation. Circulation Dec 4 2001;104:2849-

2854.

3. Chang SL, Chen YC, Yeh YH, Lin YK, Wu TJ, Lin CI, Chen SA, Chen YJ. Heart

failure enhanced pulmonary vein arrhythmogenesis and dysregulated sodium and

calcium homeostasis with increased calcium sparks. J Cardiovasc Electrophysiol Dec

2011;22:1378-1386.

4. McMullen JR, Amirahmadi F, Woodcock EA, et al. Protective effects of exercise and

phosphoinositide 3-kinase(p110alpha) signaling in dilated and hypertrophic

cardiomyopathy. Proc Natl Acad Sci U S A Jan 9 2007;104:612-617.

5. Pretorius L, Du XJ, Woodcock EA, et al. Reduced phosphoinositide 3-kinase

(p110alpha) activation increases the susceptibility to atrial fibrillation. Am J Pathol

Sep 2009;175:998-1009.

6. Cai H, Li Z, Goette A, Mera F, Honeycutt C, Feterik K, Wilcox JN, Dudley SC, Jr.,

Harrison DG, Langberg JJ. Downregulation of endocardial nitric oxide synthase

21

expression and nitric oxide production in atrial fibrillation: potential mechanisms for

atrial thrombosis and stroke. Circulation Nov 26 2002;106:2854-2858.

7. Pye MP, Black M, Cobbe SM. Comparison of in vivo and in vitro haemodynamic

function in experimental heart failure: use of echocardiography. Cardiovasc Res Jun

1996;31:873-881.

8. Hung LM, Su MJ, Chu WK, Chiao CW, Chan WF, Chen JK. The protective effect of

resveratrols on ischaemia-reperfusion injuries of rat hearts is correlated with

antioxidant efficacy. Br J Pharmacol Apr 2002;135:1627-1633.

9. Dernek S, Ikizler M, Erkasap N, Ergun B, Koken T, Yilmaz K, Sevin B, Kaygisiz Z,

Kural T. Cardioprotection with resveratrol pretreatment: improved beneficial effects

over standard treatment in rat hearts after global ischemia. Scand Cardiovasc J Aug

2004;38:245-254.

10. Shioi T, McMullen JR, Kang PM, Douglas PS, Obata T, Franke TF, Cantley LC,

Izumo S. Akt/protein kinase B promotes organ growth in transgenic mice. Mol Cell

Biol Apr 2002;22:2799-2809.

11. Li D, Melnyk P, Feng J, Wang Z, Petrecca K, Shrier A, Nattel S. Effects of

experimental heart failure on atrial cellular and ionic electrophysiology. Circulation

Jun 6 2000;101:2631-2638.

12. Kaab S, Dixon J, Duc J, Ashen D, Nabauer M, Beuckelmann DJ, Steinbeck G,

McKinnon D, Tomaselli GF. Molecular basis of transient outward potassium current

22

downregulation in human heart failure: a decrease in Kv4.3 mRNA correlates with a

reduction in current density. Circulation Oct 6 1998;98:1383-1393.

13. Birner C, Husser O, Jeron A, Rihm M, Fredersdorf S, Resch M, Schmid P, Endemann

D, Riegger G, Luchner A. Differential expression of potassium channels and

abnormal conduction in experimental tachycardia-induced heart failure. Naunyn

Schmiedebergs Arch Pharmacol May 2012;385:473-480.

14. Wang Q, Shen J, Splawski I, Atkinson D, Li Z, Robinson JL, Moss AJ, Towbin JA,

Keating MT. SCN5A mutations associated with an inherited cardiac arrhythmia, long

QT syndrome. Cell Mar 10 1995;80:805-811.

15. Gao G, Xie A, Huang SC, et al. Role of RBM25/LUC7L3 in abnormal cardiac sodium

channel splicing regulation in human heart failure. Circulation Sep 6 2011;124:1124-

1131.

16. Nattel S. New ideas about atrial fibrillation 50 years on. Nature Jan 10 2002;415:219-

226.

17. Yeh YH, Wakili R, Qi XY, Chartier D, Boknik P, Kaab S, Ravens U, Coutu P,

Dobrev D, Nattel S. Calcium-handling abnormalities underlying atrial

arrhythmogenesis and contractile dysfunction in dogs with congestive heart failure.

Circ Arrhythm Electrophysiol Jun 1 2008;1:93-102.

23

18. Nishida K, Qi XY, Wakili R, et al. Mechanisms of atrial tachyarrhythmias associated

with coronary artery occlusion in a chronic canine model. Circulation Jan 18

2011;123:137-146.

19. Hu ST, Liu GS, Shen YF, Wang YL, Tang Y, Yang YJ. Defective Ca(2+) handling

proteins regulation during heart failure. Physiol Res 2011;60:27-37.

20. Wu CY, Jia Z, Wang W, Ballou LM, Jiang YP, Chen B, Mathias RT, Cohen IS, Song

LS, Entcheva E, Lin RZ. PI3Ks maintain the structural integrity of T-tubules in

cardiac myocytes. PLoS One 2011;6:e24404.

21. Ghigo A, Morello F, Perino A, Damilano F, Hirsch E. Specific PI3K isoform

modulation in heart failure: lessons from transgenic mice. Curr Heart Fail Rep Sep

2011;8:168-175.

22. Wu Y, Liu F. Targeting mTOR: evaluating the therapeutic potential of resveratrol for

cancer treatment. Anticancer Agents Med Chem Sep 2013;13:1032-1038.

23. Lu Z, Wu CY, Jiang YP, Ballou LM, Clausen C, Cohen IS, Lin RZ. Suppression of

phosphoinositide 3-kinase signaling and alteration of multiple ion currents in drug-

induced long QT syndrome. Sci Transl Med Apr 25 2012;4:131ra150.

24. Cantley LC. The phosphoinositide 3-kinase pathway. Science May 31

2002;296:1655-1657.

24

25. Lin RC, Weeks KL, Gao XM, et al. PI3K(p110 alpha) protects against myocardial

infarction-induced heart failure: identification of PI3K-regulated miRNA and mRNA.

Arterioscler Thromb Vasc Biol Apr 2010;30:724-732.

26. Speranza L, Pesce M, Franceschelli S, Bucciarelli T, Gallina S, Riccioni G, Patruno

A, Felaco M. The biological evaluation of ADMA/SDMA and eNOS in patients with

ACHF. Front Biosci (Elite Ed) 2013;5:551-557.

27. Carnicer R, Crabtree MJ, Sivakumaran V, Casadei B, Kass DA. Nitric oxide

synthases in heart failure. Antioxid Redox Signal Mar 20 2013;18:1078-1099.

28. Chen LY, Shen WK. Epidemiology of atrial fibrillation: a current perspective. Heart

Rhythm Mar 2007;4:S1-6.

29. Dobrev D, Nattel S. New antiarrhythmic drugs for treatment of atrial fibrillation.

Lancet Apr 3 2010;375:1212-1223.

30. Wen Q, Zhou L, Zhou C, Zhou M, Luo W, Ma L. Change in hepatocyte growth factor

concentration promote mesenchymal stem cell-mediated osteogenic regeneration. J

Cell Mol Med Jun 2012;16:1260-1273.

31. Zhang Q, Huang WD, Lv XY, Yang YM. Puerarin protects differentiated PC12 cells

from H(2)O(2)-induced apoptosis through the PI3K/Akt signalling pathway. Cell Biol

Int May 1 2012;36:419-426.

25

Clinical Perspectives

Our study unravels one of the possible mechanisms in which AF is mediated. The

mechanistic pathway involves PI3K and eNOS signaling, and it is a novel discovery in AF

management during HF. In ischemic HF model, PI3K and eNOS are down regulated, which

lead to a myriad of signaling changes in the downstream ion channel proteins and eventually

lead to increased susceptibility to atrial arrhythmogenesis. The P13K and eNOS also play a

role in mediating the process of atrial fibrosis and ventricular function preservation post

ischemic injury. Moreover, we study the healing property of anti-oxidant resveratrol, via the

enhancement of the PI3K and eNOS expressions and its downstream signaling pathway, both

acute infusion and chronic therapy of resveratrol have been shown to reduce AF occurrence

during open-heart electrophysiological stimulation. The reduction in AF susceptibility can be

linked to improvement of action potential duration and reversal of ions channel protein levels

post myocardial infarction. This knowledge can be applied to improve care in patients with

HF post myocardial infarction. Drugs that enhance P13K and eNOS pathway can potentially

reduce the development of AF in HF patients. Our study hence paves the way for novel

therapies in HF. Further human trials involving HF patients can be designed to address the

electrophysiological and histological modulations during resveratrol or other PI3K/eNOS

enhancing drug treatment. Randomized controlled trials involving resveratrol therapy in HF

patients can be conducted to evaluate its potential beneficial effects.

26

Figure Legends

Figure 1: Monophasic action potentials (MAPs) recorded in the left atrial appendage. Panel A

shows an example of MAP during a pacing cycle length of 300 ms at baseline and after drug

perfusion. Panel B shows the comparison of action potential duration (APD) at 90%, 50%,

and 20% (APD90, APD50, APD20, respectively), and activation time at baseline and after

perfusion of drugs. *P < 0.05, **P < 0.01, ***P < 0.001

Figure 2: Triggered activities and induction of atrial fibrillation (AF). Panel A shows

examples of AF induction (left arm) and programmed electrical stimulation (middle and right

arms) in different groups. In normal LA (1st line), no AF or early after depolarization (EAD)/

delayed after depolarization (DAD) were inducible. In HF LA (2nd line), AF, EAD (asterisk)

and DAD (arrow) were induced during pacing. In HF LA with resveratrol (3rd line), no AF or

EAD/DAD were inducible. In HF LA with resveratrol +wortmannin (4th line) or resveratrol

+DPI (5th line), AF, EAD (asterisk) and DAD (arrow) were induced during pacing. Panel B:

Bar chart shows the incidence of AF, EAD, and DAD in normal model, heart failure model

and after perfusion of respective drugs (HF+ resveratrol, HF+ resveratrol+ wortamnnin, and

HF+ resveratrol+ DPI). *P < 0.05, **P < 0.01, ***P < 0.001

Figure 3: Histology and apoptosis of left atrial appendage (LAA). Panel A showed the

histology of LAA stained with trichrome stain at baseline and after chronic intraperitoneal

drugs administration. Blue denotes collagenous fibrosis and red denotes normal atrial tissue.

Quantification of collagen is shown in the bar chart. Panel B revealed the immunostaining of

collagen I in LAA at baseline and after drugs administration. Panel C demonstrated the

expression of cleaved caspase 3 in LAA at baseline and after drugs administration. Samples

were obtained from normal, heart failure (HF), resveratrol, and resveratrol + eNOS or PI3K

inhibitor in HF rabbits (n = 6, each). *P < 0.05, **P < 0.01, ***P < 0.001

27

Figure 4: mRNA expression of various ion channels in the left atrial appendage determined

by real-time polymerase chain reaction (PCR). mRNA was obtained from normal, heart

failure (HF), resveratrol, and resveratrol + eNOS or PI3K inhibitor in HF rabbits (n = 6,

each). *P <0.05, **P <0.01, ***P <0.001

Figure 5: Expression of SERCA2a and PLB in left atrial appendage after resveratrol

administration. Samples were obtained from normal, HF, resveratrol, and resveratrol + eNOS

or PI3K inhibitor in heart failure (HF) rabbits (n = 6, each). (A) mRNA expression of

SERCA2a and PLB as determined by real-time polymerase chain reaction (PCR). (B) Protein

expression of SERCA2a, PLB, PLB-phosphor-Ser-16, and PLB-phosphor-Thr-17 as

determined by western blotting. *P < 0.05, **P < 0.01, ***P < 0.001

Figure 6: Expression of PI3K, AKT, and eNOS in left atrial appendage after resveratrol

administration. Samples were obtained from normal, heart failure (HF), resveratrol, and

resveratrol + eNOS or PI3K inhibitor in HF rabbits (n = 6, each). (A) mRNA expression of

PI3K, AKT, and eNOS as determined by real-time polymerase chain reaction (PCR). (B)

Protein expression of PI3K-p110, p85, p-AKT, and p-eNOS as determined by western

blotting. *P < 0.05, **P < 0.01, ***P < 0.001

Figure 7: Proposed pathway of resveratrol effects on atrial fibrillation (AF).

28

Table 1 Echo parameters in group 1 rabbits that underwent 1 week of intraperitoneal drug therapy