Article ID: CJ-18-0743
Background: It has been suggested that protein directly activated by cAMP (Epac), one of the downstream signaling molecules of β-adrenergic receptor (β-AR), may be an effective target for the treatment of arrhythmia. However, there have been no reports on the anti-arrhythmic effects or cardiac side-effects of Epac1 inhibitors in vivo.
Methods and Results: In this study, the roles of Epac1 in the development of atrial and ventricular arrhythmias are examined. In addition, we examined the usefulness of CE3F4, an Epac1-selective inhibitor, in the treatment of the arrhythmias in mice. In Epac1 knockout (Epac1-KO) mice, the duration of atrial fibrillation (AF) was shorter than in wild-type mice. In calsequestrin2 knockout mice, Epac1 deficiency resulted in a reduction of ventricular arrhythmia. In both atrial and ventricular myocytes, sarcoplasmic reticulum (SR) Ca2+ leak, a major trigger of arrhythmias, and spontaneous SR Ca2+ release (SCR) were attenuated in Epac1-KO mice. Consistently, CE3F4 treatment significantly prevented AF and ventricular arrhythmia in mice. In addition, the SR Ca2+ leak and SCR were significantly inhibited by CE3F4 treatment in both atrial and ventricular myocytes. Importantly, cardiac function was not significantly affected by a dosage of CE3F4 sufficient to exert anti-arrhythmic effects.
Conclusions: These findings indicated that Epac1 is involved in the development of atrial and ventricular arrhythmias. CE3F4, an Epac1-selective inhibitor, prevented atrial and ventricular arrhythmias in mice.
The sympathetic nervous system has been shown to play important roles in the development of cardiac diseases.1,2 Activation of the β-adrenergic receptor (β-AR) mediated pathway is one of the causes of arrhythmia, including atrial fibrillation (AF)2 and ventricular tachyarrhythmia.3 β-AR blockade therapy is widely accepted as a treatment for arrhythmias.4,5 Catecholaminergic polymorphic ventricular tachycardia (CPVT) is an inherited arrhythmogenic disorder that causes exercise- and emotion-induced polymorphic ventricular arrhythmias.6 In young individuals, CPVT is an important cause of sudden death. As the sympathetic activation has been recognized as a trigger for the arrhythmic events, full doses of β-AR blockers, if tolerable, are recommended as a first-choice therapy.7 The inhibitory effect of the β-AR blockers on arrhythmias is regarded as one of the main reasons for the treatment-induced improvement in the prognosis of patients with heart failure.8 However, the clinical usage of β-AR blockers has been limited by their serious side-effects on heart function. As the sympathetic nervous system regulates cardiac function, blockade of the system can cause heart failure. In addition, arrhythmias can be a serious problem, especially among patients with heart failure. Therefore, we need to pay close attention to patients’ cardiac function during treatment, and we may sometimes be forced to discontinue the therapy. Thus, the development of anti-sympathetic therapy without such side-effects is needed.
Exchanged protein directly activated by cAMP (Epac) has been reported to be a novel key player in the β-AR mediated signaling pathway.9 Epac is a direct target of cAMP, which is induced by β-AR signaling. There are 2 isoforms of Epac: Epac1 and Epac2. In Epac1 or Epac2 knockout (KO) mice, only slight or no suppression of cardiac function has been observed,10–12 indicating that Epac inhibition may not cause severe suppression of cardiac function. Recently, several reports have identified the roles of Epac in the development of cardiac diseases including arrhythmias.10,11,13–17 These findings indicate that inhibition of Epac function may be a novel and useful strategy for the treatment of arrhythmias.
However, there have been no reports on the anti-arrhythmic effects or cardiac side-effects of Epac inhibitors in vivo. In addition, the role of the Epac1 isoform in arrhythmogenesis remains controversial. In Epac1-KO mice on a C57BL/6 and CBA mixed background, the duration of AF induced by transesophageal rapid atrial pacing was decreased.10 In rat neonatal ventricular myocytes, Epac1 silencing with siRNA attenuated isoproterenol (ISO)-mediated increases of ryanodine receptor 2 (RyR2) phosphorylations on serine-2808 and serine-2814,10 which is considered to be a trigger of arrhythmias.18,19 In addition, 5,7-dibromo-6-fluoro 3,4-dihydro-2-methyl-1quinolinecarboxaldehyde (CE3F4), an Epac1-selective inhibitor, attenuated spontaneous Ca2+ waves induced by 8-(4-Chlorophenylthio)-2’-O-methyladenosine-3’,5’-cyclic monophosphate (8-CPT), an Epac activator, in rat adult ventricular myocytes.13 These findings indicate that Epac1 may mediate sympathetic activation-induced arrhythmogenic signaling.
However, the role of Epac1 in arrhythmogenesis has been questioned, especially in ventricular arrhythmia. In mouse ventricular myocytes, Epac activation-induced sarcoplasmic reticulum (SR) Ca2+ leak, which has been considered a potential arrhythmogenic trigger, was attenuated by disruption of the Epac2 gene, but not by the Epac1 gene. In addition, in a mouse model of pacing-induced ventricular tachycardia (VT), the sustained VT duration was decreased by the disruption of the Epac2 gene, but not by that of the Epac1 gene, indicating that Epac2 but not Epac1 plays an important role in the development of ventricular arrhythmias.11
It has been indicated that the discrepancy may be attributed to the difference of cell types and species used in these studies.20 In addition, the different genetic backgrounds of the mice might affect the results. It is reported that the vulnerability to arrhythmia varies among different mice strains.21 In addition, there have been no reports on the roles of Epac1 in the regulation of SR Ca2+ leak in atrial myocytes. Therefore, the role of Epac1 in the development of atrial and ventricular arrhythmias still remains unclear.
In this study, we examined the roles of Epac1 in the development of atrial and ventricular arrhythmias in mice on a C57BL/6 background. Moreover, we evaluated the usefulness of CE3F4, an Epac1-selective inhibitor, in the treatment of atrial and ventricular arrhythmia. To address the concerns mentioned above, we developed Epac1-KO mice on a C57BL/6 background. All mice and cardiomyocytes used in this study are on a C57BL/6 background. In addition, to examine the roles of Epac1 in ventricular arrhythmia, we generated Epac1 and Calsequestrin 2 (Casq2) double KO (DKO) mice by mating single-gene mutants. The Casq2 gene-KO (Casq2-KO) mouse is one of the mouse models of CPVT.22
We will demonstrate that Epac1 is involved in the development of atrial and ventricular arrhythmia in mice. Treatment with CE3F4, an Epac1-selective inhibitor, prevented atrial and ventricular arrhythmias without suppressing cardiac function. Inhibition of Epac1 function may be an effective therapeutic target for arrhythmogenesis.
More details about the Methods are available in the Supplementary Methods section online.
AnimalsThe previously reported Epac1-KO mice and Epac2-KO mice10 were repeatedly backcrossed more than 6-fold on a C57BL/6 background. The age-matched (12–18 weeks) wild-type (WT) mice served as controls. Casq2-KO mice were generated as described previously.22 Epac1 and Casq2 DKO mice were generated by intercrossing the Epac1-KO mice and Casq2-KO mice, and age-matched Casq2-KO mice from the same colony were used as a control. All procedures involving animals were approved by the Animal Care and Use Committee of Yokohama City University.
Drug TreatmentAll of the reagents used in this study were purchased from Sigma Aldrich (St. Louis, MO, USA) unless otherwise stated. The acetoxymethyl ester form of 8-CPT (8-CPT-AM) was purchased from Biolog Life Science Institute (Bremen, Germany). The treatment of mice with the pharmacological Epac1 inhibitor, CE3F4, a β1-AR-selective antagonist, metoprolol, or vehicle (30% dimethyl sulfoxide [DMSO]) was administered 20 min before the induction of AF via the internal jugular vein (bolus infusion). In the induction of ventricular arrhythmias, the Casq2-KO mice were injected intravenously through the tail vein (1-min slow infusion) with CE3F4, metoprolol or vehicle.
Induction of AFSimple and minimally invasive AF was induced in mice by transesophageal burst pacing, and the duration of AF (AFD) was measured, as previously reported.23 Briefly, the mice were anesthetized under isoflurane (1.5%) inhalation. For 10 s, the transesophageal atrial burst pacing was conducted at a stimulation of 1.5 mA with 10-ms cycle lengths and a pulse width of 3 mA. The AF was identified in a lead II body surface electrocardiogram (ECG) using the following criteria: (1) loss of P wave; (2) irregular R-R interval; and (3) duration greater than 2 s. Three trials of burst pacing at 3-min intervals were performed during the measurement, and the longest episode was recorded as the duration of AF.
Induction of Ventricular ArrhythmiasThe induction of catecholaminergic ventricular arrhythmias was performed based on a previous report.22 The mice were anesthetized by isoflurane inhalation and ISO (3 mg/kg) was injected intraperitoneally. The number of premature ventricular contractions (PVCs) was counted in the lead II body surface ECG for 20 min after ISO administration. We counted the number of PVCs in single, couplet, and triplet. To calculate the duration of VT (four or more contiguous PVCs), we summed the total time spent by each mouse in VT.24
Measurement of SR Ca2+ Leak and Spontaneous SR Ca2+ ReleaseThe measurement of SR Ca2+ leak and SR Ca2+ release (SCR) were performed as described previously, with some modifications.25–28 The isolated myocytes were loaded with 5 μmol/L fluo-4 AM (Dojindo, Kumamoto, Japan) in normal Tyrode solution (in mmol/L: 140 NaCl, 5 KCl, 1 MgCl2, 10 glucose, and 10 HEPES [pH 7.4 adjusted with NaOH]) containing 1.8 mmol/L Ca2+ for 15 min. The dye-loaded myocytes were placed on a Ti2000 confocal microscope system (Nikon, Tokyo, Japan) in a dark room. SCR was determined by rapidly switching from normal Tyrode to 0 Na+/0 Ca2+ Tyrode, and SCRs were counted for 30 s. A Ca2+ spike greater than 0.15 ∆F/F0 was defined as a SCR.29 Then, the diastolic Ca2+ leakage from the SR was estimated by measuring the downward shift in fluorescence after tetracaine treatment. The tetracaine-dependent shift of Ca2+ from the cytosol to the SR was considered to be proportional to the SR Ca2+ leakage.26 Finally, the SR Ca2+ content was estimated by applying 10 mmol/L caffeine. The magnitude of the diastolic Ca2+ leakage from the SR was expressed as a value relative to the caffeine-releasable SR Ca2+ content.30 For drug-treated samples, myocytes were pre-incubated with 10 μmol/L of CE3F4 for 30 min prior to the measurement.
Statistical AnalysisStatistical analysis was performed by using GraphPad Prism software (GraphPad Software, Inc., La Jolla, CA, USA). A Mann-Whitney test was used for comparisons between 2 groups. For comparisons among multiple groups, a Kruskal-Wallis test was run followed by a Mann-Whitney post-hoc test. For analysis of two factors, a 2-way ANOVA test followed by a Bonferroni correction was used. Data were expressed as mean±SEM, and statistical significance was set to P<0.05.
Epac has been shown to play an important role in the development of arrhythmia.10,11,13,14 We purified Epac1-KO mice on a C57BL/6 background (Supplementary Figure 1A) and examined the role of Epac1 in the development of AF. The duration of AF was significantly shorter in Epac1-KO mice than in WT mice (WT vs. Epac1-KO: 184.83±60.27 s vs. 47.46±11.84 s, P<0.05) (Figure 1A–C). We previously demonstrated that the sympathetic activation by intraperitoneal administration of norepinephrine (NE) (1.5 mg/kg) strikingly elongated the duration of AF in mice.23 The sympathetic activation-induced elongated AF was also shorter in Epac1-KO mice (WT vs. Epac1-KO: 716.49±174.20 s vs. 215.19±70.94 s, P<0.05) (Figure 1D). These findings indicate that Epac1 plays an important role in the development of AF in mice. Recently, it was reported that Epac1-activation reduced left atrium (LA) fibrosis in a mouse heart failure model.31 Therefore, we examined LA fibrosis in the Epac1-KO mice. There were no significant differences in LA fibrosis between WT and Epac1-KO mice (Supplementary Figure 2), indicating that the regulation of atrial fibrosis by Epac1 might not affect the difference in the duration of AF.
A protein directly activated by cAMP (Epac)1 deficiency shortened the duration of atrial fibrillation (AF). (A,B) Representative body surface ECG (lead II) of wild-type (WT) and Epac1-knockout (KO) mice. The asterisk indicates a P wave; BP, burst pacing. (C) Transesophageal burst pacing was performed to induce AF, and we compared the AF duration (AFD) between Epac1-KO and WT mice (n=12; *P<0.05). (D) Ten minutes after the sympathetic activation (NE; 1.5 mg/kg), AF was induced by transesophageal burst pacing. *P<0.05. n=11–12.
We next examined whether Epac1 also plays an important role in the development of ventricular arrhythmias. We generated DKO mice on a C57BL/6 background by mating single-gene mutants (Supplementary Figure 1B). The Casq2-KO mouse is one of the mouse models of CPVT.22 The number of PVCs was counted for 20 min after ISO administration. The incidence of PVC was significantly lower in DKO mice than in Casq2-KO mice (Casq2-KO vs. DKO: 91.81±18.36 vs. 32.40±7.10, P<0.05). Further, the VT duration was significantly shorter in Casq2-KO and DKO mice (Casq2-KO vs. DKO: 3.39±1.02 vs. 0.19±0.14 s, P<0.05) (Supplementary Figure 3, Supplementary Table 1). These findings indicate that Epac1 plays an important role in the development of ventricular arrhythmias (Figure 2A–C). Epac1 deletion in Casq2-KO did not appreciably affect the expression of the other important molecules involved in cAMP signaling such as β1-Adrenergic receptor (AR), β2-AR, β-adrenergic receptor kinase (βARK), Gs, Giα, Gβ, Gq, PKA regulatory IIα subunit (RIIα), regulatory Iα subunit (RIα) and Epac2 (Supplementary Figure 4). The Epac2 mRNA level was also not significantly different between the Casq2-KO mice and DKO mice (Supplementary Figure 5).
A protein directly activated by cAMP (Epac)1 deficiency reduced the incidence of sympathetic activation-induced ventricular arrhythmias. (A,B) Representative body surface ECG (lead II) recording of calsequestrin 2 (Casq2)-knockout (KO) mice. Premature ventricular contractions (PVCs) were induced by an intraperitoneal administration of isoproterenol (ISO; 3 mg/kg, arrowhead). The asterisk shows PVC. (C) The number of PVCs was counted on a body-surface ECG for 20 min after the administration of ISO and compared between Casq2-KO and DKO mice. *P<0.05. n=10–11.
To further examine the mechanism of Epac1 deficiency-induced inhibition of arrhythmias, we next examined the effect of Epac1 deficiency on sympathetic activation-induced SR Ca2+ leak and SCR, which are recognized as a major source of ectopic activity triggering arrhythmia, in cardiac myocytes. Cardiac myocytes were isolated from WT and Epac1-KO mice, and the calcium concentration was assessed based on a previously reported protocol.26 In the atrial myocytes of Epac1 WT mice, sympathetic activation by NE, which we also used in the AF model experiment, significantly increased the SR Ca2+ leak and rate of SCR. However, the effect was drastically blunted in Epac1-KO myocytes. The sympathetic activation-induced SR Ca2+ leak and rate of SCR were significantly lower in Epac1-KO myocytes (Figure 3E,F). Similarly, also in the Epac1-KO ventricular myocytes, the sympathetic activation-induced SR Ca2+ leak and rate of SCR were lower compared with those in the WT myocytes (Figure 3G,H). In the ventricular myocytes, to induce sympathetic activation, we used ISO, which we also used in the mouse ventricular arrhythmia model. These findings suggest that Epac1 may be involved in the sympathetic activation-induced SR Ca2+ leak and the rate of SCR.
A protein directly activated by cAMP (Epac)1 deficiency attenuated sympathetic activation-induced sarcoplasmic reticulum (SR) Ca2+ leak and SR Ca2+ release (SCR) in cardiac myocytes. Representative Ca2+ traces of ventricular myocytes in the absence (A,B) or presence (C,D) of isoproterenol (ISO) in wild-type (WT) (A,C) and Epac1-knockout (KO) (B,D) mice. The effect of sympathetic activation by NE on SCR and Ca2+ leakage from the SR in atrial (E,F) myocytes was compared between WT and Epac1-KO mice. The SCR was counted for 30 s. The SR Ca2+ leak (below red baseline) was estimated by measuring the downward shift in fluorescence after tetracaine (1 mmol/L) treatment. Finally, caffeine (10 mmol/L) was administered to estimate the SR Ca2+ content. The effect of sympathetic activation by ISO on SCR and the diastolic SR Ca2+ leak in ventricular myocytes (G,H). Cardiac myocytes prepared from WT and Epac1-KO mice were loaded with fluo-4 AM (2.5 μmol/L). (E,F) n=16–19; (G,H) n=32–43; ***P<0.001, **P<0.01, *P<0.05. The myocytes in each group were from 3 or more animals.
As our results indicated that Epac1 played an important role in the development of atrial and ventricular arrhythmias, we then examined the usefulness of CE3F4, an Epac1-selective inhibitor, in the treatment of atrial and ventricular arrhythmias in mice. We induced AF after 20 min of CE3F4 (3 mg/kg and 1 mg/kg) administration through a catheter in the internal jugular vein. A previous report showed that intravenous infusion of CE3F4 (10 mg/kg) improved cardiac function after myocardial infarction in mice.32 The CE3F4 (3 mg/kg) shortens the duration of the pacing-induced AF (control vs. CE3F4: 62.49±3.62 s vs. 25.32±5.76 s, P<0.05) to a level similar to that of the metoprolol-treated (3 mg/kg) mice (Figure 4A). Further, we observed that the sympathetic activation-induced AF duration was also significantly shortened by the CE3F4 treatment (control vs. CE3F4: 553.26±121.71 s vs. 133.73±75.49 s, P<0.01; Figure 4B).
An Epac1-selective inhibitor (CE3F4) inhibited atrial fibrillation (AF) and ventricular arrhythmias. (A) CE3F4 (3 mg/kg and 1 mg/kg) and metoprolol (3 mg/kg) were administered through the catheter in the internal jugular vein in wild-type (WT) mice. Twenty minutes later, AF was induced by transesophageal burst pacing. *P<0.05, n=9–11. (B) Effect of CE3F4 (3 mg/kg) and metoprolol (3 mg/kg) on the sympathetic activation-induced AF duration (AFD). Transesophageal burst pacing was performed to induce AF after 10 min of intraperitoneal injection of NE (1.5 mg/kg). **P<0.01, n=9. (C) Effect of CE3F4 on the premature ventricular contractions (PVC) induced in anesthetized calsequestrin 2 (Casq2)-KO mice. CE3F4 (3 mg/kg and 1 mg/kg) and metoprolol (3 mg/kg) were administered through the tail vein into conscious Casq2-KO mice. After 20 min of drug administration, PVC was induced by an intraperitoneal administration of isoproterenol (ISO; 3 mg/kg). A body surface ECG was recorded for 20 min, and the number of PVCs was counted. *P<0.05, n=8–12.
We next evaluated the effect of CE3F4 on sympathetic activation-induced ventricular arrhythmias using Casq2-KO mice. PVC was induced by ISO injection 20 min after intravenous CE3F4 (3 mg/kg) administration. We found that CE3F4 (3 mg/kg) treatment reduced the incidence of sympathetic activation-induced ventricular arrhythmias to a degree equivalent to metoprolol (3 mg/kg) treatment in Casq2-KO mice (control vs. CE3F4: 85.5±20.5 vs. 11.13±3.51, P<0.05; Figure 4C).
CE3F4 Inhibited SR Ca2+ Leak and SCR in Cardiac MyocytesWe examined the effect of CE3F4 treatment on the SR Ca2+ leak and the rate of SCR in cardiac myocytes. In atrial myocytes, the SR Ca2+ leak and the rate of SCR were significantly increased by sympathetic activation (NE; 1 μmol/L), and those increases were eliminated by CE3F4 (10 μmol/L) treatment (Figure 5D,E). Further, in ventricular myocytes, the sympathetic activation with ISO (1 μmol/L) significantly increased the SR Ca2+ leak and the rate of SCR. CE3F4 treatment also attenuated the sympathetic activation-induced SR Ca2+ leak and SCR in ventricular myocytes (Figure 5F,G). Taken together, the sympathetic activation-induced SR Ca2+ leak and rate of SCR were attenuated by CE3F4 in mouse atrial and ventricular myocytes.
An Epac1-selective inhibitor (CE3F4) inhibited sarcoplasmic reticulum (SR) Ca2+ leak and SR Ca2+ release (SCR) in cardiac myocytes. Representative Ca2+ traces of ventricular myocytes in the absence (A) or presence (B,C) of isoproterenol (ISO). The influence of pharmacological inhibition of Epac1 on SCR and Ca2+ leak (below red baseline) from the SR in atrial (D,E) and ventricular myocytes (F,G). The atrial myocytes were treated with NE (1 μmol/L) for 5 min, and the ventricular myocytes were treated with ISO (1 μmol/L) for 10 min prior to the measurement of SR Ca2+ leak and SCR. (D,E) n=19–25; (F,G) n=20–24. ***P<0.001, **P<0.01, *P<0.05. The myocytes in each group were from 5 or more animals.
To examine whether CE3F4 suppressed the SR Ca2+ leak and the rate of SCR through its inhibitory effect on Epac1, we examined the effect of CE3F4 on the SR Ca2+ leak and the rate of SCR induced by 8-CPT, an Epac-specific activator, in WT and Epac2-deficient cardiac myocytes (Supplementary Figure 6). CE3F4 significantly inhibited the SR Ca2+ leak and SCR in both WT and Epac2-deficient myocytes, indicating that the inhibitory effect of CE3F4 on the arrhythmogenic mechanisms results, at least in part, from its effect on Epac1.
CE3F4 Did Not Affect Cardiac FunctionIn Epac1-KO mice, only slight or no suppression of cardiac function was observed,10,11 indicating that Epac1 may not be mainly involved in the maintenance of basic cardiac function by catecholamines. Therefore, CE3F4, an Epac1 inhibitor, may have less of a suppressive effect on cardiac function compared with β-AR blockers. We examined the effects of CE3F4 and metoprolol on cardiac functions by echocardiography and hemodynamic measurements in WT mice. CE3F4 or metoprolol was intravenously administered through the right internal jugular vein. Ten minutes after injection, we evaluated their effects on cardiac function. CE3F4 (at 3 mg/kg, the amount that is sufficient to exert an anti-arrhythmic effect) did not affect the indexes of cardiac function (LVEF: 74.57±0.81% vs. 74.18±0.69%; HR: 497.80±3.0 vs. 467.60±9.06 beats/min; maximum dP/dt: 10,464.03±966.89 mmHg/s vs. 11,136.40±2,337.89 mmHg/s; and minimum dP/dt: −9,242.05±1,956.30 mmHg/s vs. −8,346.50±1,796.80 mmHg/s). In contrast, metoprolol (0.3 mg/kg), the one-tenth dose that shortened AF duration in mice to a degree equivalent to CE3F4 (3 mg/kg) therapy, significantly decreased the indexes of cardiac function (LVEF: 74.39±0.96% vs. 49.93±1.48%, P<0.001; HR: 498.05±7.24 vs. 333.20±12.84 beats/min, P<0.001; maximum dP/dt: 11,861.60±1,763.76 mm Hg/s vs. 5,046.66±1,035.34 mmHg/s, P<0.001; and minimum dP/dt: −10,106.38±1,628.20 mmHg/s vs. −4,388.72±1,302.60 mmHg/s, P<0.001; Figure 6A–D, Supplementary Table 2).
An Epac1-selective inhibitor (CE3F4) did not affect cardiac function. (A) The left ventricular ejection fraction (LVEF) and (B) heart rate (HR) measured by echocardiography. CE3F4 (3 mg/kg) and metoprolol (0.3 mg/kg) were administered via the internal jugular vein. ***P<0.001 vs. pre; ###P<0.001 control vs. metoprolol; +++P<0.001 CE3F4 vs. metoprolol. n=5 in each group. (C,D) The maximum and minimum rate of rise in left ventricular pressure (dP/dt max and dP/dt min respectively). Mice were injected with CE3F4 (3 mg/kg) and metoprolol (0.3 mg/kg) via the internal jugular vein. ***P<0.001 vs. pre; ###P<0.001, ##P<0.01 control vs. metoprolol; +++P<0.001, ++P<0.01 CE3F4 vs. metoprolol. n=5 in each group.
It has been suggested that Epac, a direct effector of cAMP, may be an effective target for the treatment of arrhythmia.9,10,20 However, the role of Epac1 in arrhythmogenesis remains controversial.10,11,13 The discrepancy in these reports was thought to be due to the difference of cell types,20 species, and genetic backgrounds of the mice or cardiomyocytes used in these studies. To address these concerns, in this study, we developed Epac1-KO mice and Epac1-Casq2 DKO mice on a C57BL/6 background. In these mice lines, we demonstrated that Epac1 plays an important role in the development of atrial and ventricular arrhythmias. Epac1 deficiency results in reduced susceptibility to AF and ventricular arrhythmia (Figures 1,2). In addition, since there has been no report on the effect of Epac1 inhibitors on the arrhythmogenesis in vivo, we examined the usefulness of CE3F4, an Epac1-selective inhibitor, in the prevention of those arrhythmias. We showed that CE3F4 prevented atrial and ventricular arrhythmias in mice (Figure 4). In the cardiac myocytes purified from our mice lines on a C57BL/6 background, 8-CPT, an Epac-specific activator, significantly increased SCR and SR Ca2+ leak in both WT and Epac2-deficient cardiomyocytes, suggesting the involvement of Epac1 in the arrhythmogenic mechanisms. Moreover, CE3F4 significantly inhibited the 8-CPT-induced effects in both WT and Epac2-deficient cardiomyocytes, indicating that CE3F4 suppressed the arrhythmogenic mechanism at least in part through Epac1 inhibition. As the Epac1-inhibiting therapy suppressed arrhythmias in a mouse model of AF and CPVT, Epac1, a downstream signaling molecule of β-AR, may represent a novel potential therapeutic target for these arrhythmias.
Epac1 expression is reported to be increased in the failing human heart.33 The enhanced Epac1 expression may be involved in the high susceptibility of heart failure patients to arrhythmias. SR Ca2+ leak has been shown to be one of the mechanisms of cardiac contractile dysfunction.30 As the Epac1 inhibitor attenuated the SR Ca2+ leak in cardiomyocytes, the inhibition of Epac1 may be effective also in the treatment of heart failure.
The differential function of the Epac isoforms has been reported. For example, in the pancreas, Epac2 has been reported to play an important role in insulin secretion.34 In addition, Epac2 deficiency caused disorders in social interactions and ultrasonic vocalizations in mice.35 Considering the side-effects of Epac-inhibiting therapies, it is quite important to clarify the detailed involvement of each isoform in the development of arrhythmias. Isoform-specific, Epac-inhibiting therapy should be applied depending on the type of arrhythmias and the pathophysiological state of the patients.
Although a previous report indicated the arrhythmogenic effects of Epac2,11 a recent report showed that an Epac2 inhibitor, ESI-05, increased early after-depolarization arrhythmia (EAD)-like Ca2+ oscillations and the frequency of spontaneous Ca2+ sparks in adult rat ventricular myocytes, suggesting that Epac2 inhibitors may be pro-arrhythmic.36 The inhibition of Epac2-induced attenuation of mitochondrial ROS production is thought to be the mechanism. CE3F4 should also be examined in a similar assay. Although our results do not negate the role of Epac2 in the arrhythmogenesis, they suggest that Epac1 may be a promising candidate as a novel therapeutic target of arrhythmia treatment.
As β-AR-mediated signaling is important for the maintenance of the circulatory system, the suppressive effects of β-AR blockers on cardiac function often become a critical problem.37 As the patients who need anti-arrhythmic therapy are likely to suffer from cardiac dysfunction, such cardiac side-effects are quite important issues. In patients with heart failure, we need to pay special attention to their cardiac function in the initiation of the therapy.37 To avoid such critical side-effects, there have been attempts to establish a method for more specific inhibition of arrhythmogenic signaling among β-AR activation-induced effects.29,38,39 In this study, we focused on Epac1, one of the downstream molecules of β-AR. Importantly, CE3F4 did not cause significantly adverse effects on cardiac function, at least under our experimental conditions (Figure 6). Indexes of cardiac function including left ventricular ejection fraction, dP/dT, and heart rate were not significantly affected by a dosage of CE3F4 sufficient to exert an anti-arrhythmic effect. Consistent with previous reports that showed that Epac1 deficiency does not cause severe cardiac dysfunction,10,11 Epac1 may not be centrally involved in the maintenance of basic cardiac function. Thus, Epac1-inhibiting therapy may be useful and more secure than a β-AR blocker in the treatment of arrhythmias. However, in this study we examined only the acute effects of CE3F4 treatment. In order to evaluate the usefulness of CE3F4 treatment compared with β-AR blocker therapy in the management of cardiac diseases, studies on the long-term effects of CE3F4 would be required.
Taken together, our findings indicate that Epac1 was involved in the development of atrial and ventricular arrhythmias. CE3F4, an Epac1-selective inhibitor, prevented atrial and ventricular arrhythmias in mice. Epac1-inhibiting therapy may be a useful and secure treatment for arrhythmias.
This work was supported, in part, by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant (25460296, 16K08501 to T.F., 17K15560 to W.C., 24390200, 25670131, 16H05300, and 16K15205 to Y.I.); the Ministry of Education, Culture, Sports, Science and Technology (MEXT) KAKENHI Grant (22136009 to Y.I.); the New Energy and Industrial Technology Development Organization (NEDO) (60890021 to Y.I.); the National Cerebral and Cardiovascular Center (NCVC) (22-2-3 to Y.I.); the Japan Agency for Medical Research and Development (AMED) (66890005, 66890011, 66890001, 66890023, 66890007, and 66891153 to Y.I.); Yokohama Foundation for Advancement of Medical Science (W.C.); a Grant for Strategic Research Promotion of Yokohama City University (T.F.); The Ichiro Kanehara Foundation for the Promotion of Medical Sciences and Medical Care (W.C.); and the Tokyo Biochemical Research Foundation (W.C. and Y.I.).
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-18-0743
