Epac1-dependent phospholamban phosphorylation mediates the cardiac response to stresses

PKA phosphorylates multiple molecules involved in calcium (Ca²⁺) handling in cardiac myocytes and is considered to be the predominant regulator of β-adrenergic receptor–mediated enhancement of cardiac contractility; however, recent identification of exchange protein activated by cAMP (EPAC), which is independently activated by cAMP, has challenged this paradigm. Mice lacking Epac1 (Epac1 KO) exhibited decreased cardiac contractility with reduced phospholamban (PLN) phosphorylation at serine-16, the major PKA-mediated phosphorylation site. In Epac1 KO mice, intracellular Ca²⁺ storage and the magnitude of Ca²⁺ movement were decreased; however, PKA expression remained unchanged, and activation of PKA with isoproterenol improved cardiac contractility. In contrast, direct activation of EPAC in cardiomyocytes led to increased PLN phosphorylation at serine-16, which was dependent on PLC and PKCε. Importantly, Epac1 deletion protected the heart from various stresses, while Epac2 deletion was not protective. Compared with WT mice, aortic banding induced a similar degree of cardiac hypertrophy in Epac1 KO; however, lack of Epac1 prevented subsequent cardiac dysfunction as a result of decreased cardiac myocyte apoptosis and fibrosis. Similarly, Epac1 KO animals showed resistance to isoproterenol- and aging-induced cardiomyopathy and attenuation of arrhythmogenic activity. These data support Epac1 as an important regulator of PKA-independent PLN phosphorylation and indicate that Epac1 regulates cardiac responsiveness to various stresses.

β-Adrenergic receptor (β-AR) signaling is well established as a primary defense mechanism against acute stress or changes in hemodynamic load; however, its role in cardiac pathogenesis, although studied extensively, remains poorly understood (1). β-AR blockade, rather than stimulation, is beneficial in patients with heart failure, and pharmacological inhibition of the entire β-AR signaling pathway at the receptor level is one of the accepted treatments for heart failure (2). Studies in mouse models overexpressing β-AR, Gsα, or PKA support the usefulness of β-AR blockade in heart failure (3–5). Recently, however, we have developed a mouse model in which type 5 adenylyl cyclase (AC), a major cardiac subtype, is disrupted (Ac5 KO), and we found that Ac5 KO showed resistance to the development of heart failure and exhibited increased longevity (6–9), indicating that inhibition of cAMP signaling at the level of AC, not β-AR, may also result in cardiac protection.

Systolic contraction follows activation of sarcolemmal voltage-gated L-type calcium (Ca²⁺) channel during an action potential, resulting in Ca²⁺-influx, which activates cardiac ryanodine receptor (RyR2) Ca²⁺ release channels on the sarcoplastic reticulum (SR) (major intracellular Ca²⁺ store). Diastolic relaxation occurs with cessation of Ca²⁺ release, and Ca²⁺ sequestration by the SR Ca²⁺-transporting adenosine triphosphatase (SERCA2a) and its regulator, phospholamban (PLN), indicate that phosphorylations of PLN and RyR2 have a central role in modulating Ca²⁺ homeostasis and, therefore, cardiac function (10, 11). PLN is a low–molecular weight phosphoprotein in cardiac SR, and dephosphorylated PLN is an inhibitor of SERCA2a-mediated transport of Ca²⁺. Following β-AR activation, and thus production of cAMP by AC, PKA phosphorylates PLN on serine-16 (10) in addition to RyR2 on serine-2808 and serine-2814 (11). PKA-mediated serine-16 phosphorylation (and thus SERCA2a activation) is the major mechanism of the lusitropic and also inotropic (12) effects of β-AR stimulation in regulating cardiac function (13). In failing hearts, in contrast, decreased PLN phosphorylation, and thus a decrease in Ca²⁺ uptake by SERCA2a, is a central feature (14, 15). Indeed, decreased inhibition of SERCA2a by PLN ablation can prevent progression of heart failure (16, 17). Others have found that decreased inhibition of SERCA2a through PLN ablation or hyperphosphorylation of PLN exaggerated heart failure and arrhythmogenic activity (18–22). Thus, it remains controversial whether or not enhancement of Ca²⁺ uptake through PLN ablation or PLN hyperphosphorylation can inhibit the progression of heart failure. Nevertheless, it is well accepted that β-AR–mediated phosphorylation of PLN plays a central role in regulating cardiac function and also in the pathogenesis of cardiac failure.

But PKA is not the only molecule activated by cAMP. Exchange protein activated by cAMP (EPAC) was recently identified as a new target of cAMP signaling that is activated independently of PKA (23, 24). EPAC has 2 isoforms (Epac1 and Epac2), and Epac1 is predominantly expressed in the heart. EPAC modulates various cellular functions, including migration, proliferation, exocytosis, and apoptosis, via regulation of RAP1 (25). In cardiac myocytes, EPAC activation induced hypertrophy (26). In adult mouse cardiac myocytes, pharmacological EPAC activation with an EPAC-selective but not isoform-selective cAMP analogue was reported to enhance evoked Ca²⁺ transients through the PLCε/PKCε/CaMKII pathway, and activation of CaMKII via EPAC induced store depletion and enhancement of Ca²⁺ sparks through increased phosphorylation of PLN on threonine-17 and RyR2 on serine-2815 by EPAC via phospholipase C (PLC) ε/PKC ε/Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) (27–29). EPAC2 was also reported to be closely associated with CaMKII, and EPAC2 mediated β₁-AR–induced cardiac arrhythmia via CaMKIIδ (a major isoform of cardiac CaMKII) and RyR2 phosphorylation on serine-2815 (30). However, the isoform-specific role of EPAC in cardiac function and pathogenesis remains poorly understood.

Here, we show that EPAC1, in an additive and independent manner with respect to PKA, phosphorylates PLN and RyR2 to regulate cardiac function. Loss of EPAC1 slightly decreased basal cardiac function, but afforded greater protection against various stresses, including arrhythmogenic stress, whereas loss of EPAC2 did not show cardioprotective effects. Accordingly, selective EPAC1 inhibition may prevent hyperphosphorylation of PLN and RyR2, and this may be an alternative strategy to current β-AR blocker therapy for the treatment of established cardiac failure or arrhythmia.

Cardiac function is decreased in Epac1 KO. Because EPAC activation induced cardiac myocyte hypertrophy in culture (26), we originally thought that EPAC might play a role in regulating cardiac myocyte growth. Therefore, we genetically disrupted Epac1 in mice (Epac1 KO) (31), but found that these mice did not show any change in the size of the heart. Instead, Epac1 KO exhibited a modest, but significant, decrease in cardiac function. LV ejection fraction (LVEF) was significantly decreased (WT versus Epac1 KO: 70% ± 0.8% versus 60% ± 1.1%, n = 18–31, P < 0.01), and LV end-systolic diameter (LVESD) was increased (WT versus Epac1 KO: 4.3 ± 0.03 versus 4.4 ± 0.1 mm, n = 18–31, P < 0.01) (Table 1). Hemodynamic measurements demonstrated that Max dP/dt was significantly decreased (WT versus Epac1 KO: 9091 ± 240 versus 7246 ± 226 mmHg, n = 16–17, P < 0.01), and Min dP/dt was significantly increased in Epac1 KO (WT versus Epac1 KO: –9110 ± 240 versus –6201 ± 125 mmHg, n = 14–18, P < 0.01) (Table 1). Because disruption of Epac1 might alter the expression of other molecules involved in cAMP signaling, such as PKA subunits, we examined the protein expression of β₁- AR, β₂-AR, β-AR kinase (βARK), Gsα, Giα, Gβ, Gq, and type 5/6 AC in addition to PKA subunits by means of Western blotting (Supplemental Figure 1A; supplemental material available online with this article; doi:10.1172/JCI64784DS1). There was no difference in expression of these molecules between WT and Epac1 KO. Also, there was no compensatory increase of Epac2 isoform in Epac1 KO at either the mRNA or protein level (Supplemental Figure 1, A and B).

Table 1

Heart size and cardiac function in Epac1 KO

Decreased cardiac function in Epac1 KO was readily compensated by isoproterenol (ISO) infusion. Responses of LVEF and LVESD to ISO (0, 0.13, 0.27, 0.40 μg/kg/min for 5 minutes) were well preserved and, indeed, reached the same level as those of WT (Supplemental Figure 2, A and B). Changes in cardiac contractility were further examined using isolated segments of myocardium (Supplemental Figure 3). The basal contractile force was significantly smaller in Epac1 KO (WT versus Epac1 KO: 112 ± 14.7 versus 73 ± 7.1 mg/mm², n = 6–7, P < 0.05) (Supplemental Figure 3A). ISO dose dependently enhanced the contractile force in WT and Epac1 KO. The magnitude of increase in response to ISO was similar in WT and Epac1 KO (Supplemental Figure 3B). The sensitivity to β-adrenergic stimulation with ISO as represented by the –log EC₅₀ value for the enhancement of contractile force was 7.04 ± 0.07 in WT and 7.06 ± 0.01 in Epac1 KO (P = NS, n = 6–7). The time required to reach peak tension was also similar in Epac1 KO and WT at baseline (WT versus Epac1 KO: 50 ± 1.1 versus 52 ± 1.5 ms, n = 6–7, P = NS) and decreased similarly in both WT and Epac1 KO (Supplemental Figure 3C). However, the time required for 90% relaxation was significantly longer in Epac1 KO at baseline (WT versus Epac1 KO: 54 ± 1.5 versus 61 ± 2.0 ms, n = 6–7, P < 0.05), but became similar in response to ISO (Supplemental Figure 3D). Thus, Epac1 KO showed decreased cardiac function, exemplified most clearly by delayed relaxation, but the dysfunction was readily compensated by β-AR stimulation.

PLN phosphorylation on serine-16 is decreased in Epac1 KO. Phosphorylation of PLN occurs, following β-adrenergic stimulation, on serine-16 and threonine-17. Indeed, PKA-mediated serine-16 phosphorylation is a key event in increasing cardiac function (32). Unexpectedly, phosphorylation on serine-16 was significantly decreased in Epac1 KO (WT versus Epac1 KO: 100% ± 6.5% versus 57% ± 7.1%, n = 6, P < 0.01) (Figure 1A) despite the absence of any difference in expression of the PKA catalytic and regulatory units (RIα, RIIα) (Supplemental Figure 1A). PLN phosphorylation on threonine-17 (Figure 1B), which is mediated by CaMKII, was similar in WT and Epac1 KO. Phosphorylation on threonine-286, and thus activation of CaMKII, was also similar (Supplemental Figure 4).

Figure 1

Phosphorylation of PLN on serine-16 and threonine-17 and SERCA2a expression in the heart of Epac1 KO at baseline. (A) Phosphorylation on serine-16 (Ser16) was significantly decreased in Epac1 KO compared with WT (WT versus Epac1 KO: 100% ± 6.5% versus 57% ± 7.1%, n = 6, *P < 0.01). The ratio of phosphorylated/total protein expression of PLN in WT was taken as 100% in each determination. (B) Phosphorylation on threonine-17 (Thr17) was not different in Epac1 KO and WT (WT versus Epac1 KO: 100% ± 5.7% versus 92% ± 11.5%, n = 7–10, P = NS). The ratio of phosphorylated/total protein expression of PLN in WT was taken as 100% in each determination. (C) The expression of SERCA2a protein was not different in WT and Epac1 KO at baseline (WT versus Epac1 KO: 100% ± 11.1% versus 105% ± 3.5%, n = 6–7, P = NS). The ratio of SERCA2a protein/GAPDH in WT was taken as 100% in each determination. (D) Representative immunoblotting results of phosphorylation of PLN on serine-16 (upper) and threonine-17 (middle) and SERCA2a (lower) are shown. T-PLN, total PLN; p-PLN, phosphorylated PLN.

EPAC-selective cAMP analogue increases PLN phosphorylation on serine-16 and threonine-17 in neonatal rat cardiac myocytes. The above findings indicated that disruption of EPAC, but not PKA, decreased PLN phosphorylation. When we examined the consequences of EPAC activation for PLN phosphorylation in neonatal rat cardiac myocytes, we used a new membrane-permeable EPAC-selective but not isoform-selective agonist, i.e., 8-(4-chlorophenylthio)-2′-O-Me-cAMP-AM (8-CPT-AM) (33). We found that 10 μM 8-CPT-AM was necessary to obtain a significant PLN phosphorylation signal with our detection system in neonatal rat cardiac myocytes (Figure 2D), and thus this concentration was used. This concentration of 8-CPT-AM is in line with those used in previous studies to examine EPAC-mediated signaling, i.e., 10 μM for adult rat cardiac myocytes (34) or 50 μM for 293T, JAR, and BeWo cells (35). Furthermore, Brette et al. reported very recently that 8-CPT-AM at 10 μM did not exert an inhibitory effect on phosphodiesterase (PDE) activity (34). On the other hand, non-AM ester 8-CPT (8-CPT) does inhibit PDE, which is undesirable, since we wish to examine EPAC-specific effects (36, 37).

Figure 2

Effects of EPAC activation on PLN phosphorylation in neonatal rat cardiac myocytes. (A and B) Treatment of neonatal rat cardiac myocytes with 8-CPT-AM (10 μM). PLN phosphorylation on serine-16 was significantly increased at 15 minutes and remained significantly (*P < 0.05 or **P < 0.01 versus 0 minutes, n = 4) greater at 120 minutes than at 0 minutes (A). PLN phosphorylation on threonine-17 was also significantly increased at 15 minutes after treatment. However, increase fell below significance at 60 minutes versus 0 minutes and remained unchanged at 120 minutes (P = NS, versus 0 minutes, n = 4) (B). Ratio of phosphorylated/total protein expression of PLN at baseline (0 min: Ctrl) was taken as 1-fold. (C) EPAC-mediated PLN phosphorylation on serine-16 was examined in neonatal rat cardiac myocytes transfected with PKCε siRNA or control siRNA (*P < 0.01, n = 5–7). Ratio of phosphorylated/total protein expression of PLN in cells transfected with control siRNA at baseline was taken as 100%. (D) PLN phosphorylation on serine-16 was examined in cells treated with Bnz-cAMP (50 μM) and/or 8-CPT-AM (10 μM). (^##P < 0.001 versus Bnz-cAMP [50 μM] alone; ^††P < 0.001 versus 8-CPT-AM [10 μM] alone). A similar tendency was observed when 10 μM Bnz-cAMP and 5 μM 8-CPT-AM were used together (^#P < 0.001 versus Bnz-cAMP [10 μM] alone; ^†P < 0.001 versus 8-CPT-AM [5 μM] alone, n = 4–8; *P < 0.01 versus baseline, n = 4–8). Ratio of phosphorylated/total protein expression of PLN at baseline was taken as 100%.

We found that 8-CPT-AM significantly increased the PLN phosphorylation on serine-16 by approximately 89-fold from baseline and that on threonine-17 by approximately 19-fold from baseline at 15 minutes after the treatment with 8-CPT-AM (n = 4, P < 0.01) (Figure 2, A and B). PLN phosphorylation on serine-16 was decreased gradually, but was significantly increased even at 120 minutes after 8-CPT-AM treatment. However, PLN phosphorylation on threonine-17 decreased rapidly and did not show a significant increase at 60 minutes after 8-CPT-AM treatment. Thus, activation of EPAC with 8-CPT-AM, in addition to PKA and CaMKII, could induce PLN phosphorylation on serine-16 and threonine-17, but PLN phosphorylation on serine-16 persists for a long time, compared with that on threonine-17, which has been reported to be phosphorylated by EPAC via PLCε/PKCε/CaMKII (27, 28).

Increase of PLN phosphorylation at serine-16 was attenuated in neonatal cardiac myocytes prepared from Epac1 KO after the treatment of 8-CPT-AM. We examined the effects of 8-CPT-AM on PLN phosphorylation at serine-16 using neonatal cardiac myocytes prepared from Epac1 KO and WT. In WT neonatal mouse cardiac myocytes, we found that 1 μM 8-CPT-AM was sufficient to obtain a significant PLN phosphorylation signal with our detection system (Supplemental Figure 5). Since the magnitude of the increase was similar to that obtained with 10 μM 8-CPT-AM in neonatal rat cardiac myocytes, we used this concentration (1 μM) (Figure 2D and Supplemental Figure 5). Epac activation with 8-CPT-AM (1 μM) significantly increased PLN phosphorylation at serine-16 from baseline in WT, and the level was also significantly greater than that in Epac1 KO (P < 0.05). In contrast with WT, the increase of PLN phosphorylation at serine-16 was not significant (WT: from 100% ± 11% to 1534% ± 528%, n = 4–6, P < 0.01; Epac1 KO: from 61% ± 22% to 476% ± 141%, n = 4–6, P = NS). These results, together with the data shown in Figure 1A and Figure 2A, indicate that EPAC1 is an important regulator of PLN phosphorylation at serine-16, independently of PKA. More importantly, these findings also suggest that 8-CPT-AM at the concentration of 1 μM phosphorylates PLN at serine-16 via EPAC, not other signaling pathways, such as PKA, cGMP-dependent PKG, or PDE (34, 38).

EPAC phosphorylates PLN at serine-16 via PLC/PKCε. It is established that PKC phosphorylates PLN and that the phosphorylation via PKC is additive to that via PKA and CaMKII, but the significance of these findings and the mechanisms involved remain poorly understood (39–41). It has been suggested that EPAC mediates a novel pathway of regulatory crosstalk between the cAMP and the PLC/PKC pathways (42), and so we hypothesized that EPAC mediates PLN phosphorylation on serine-16 via PLC/PKC. Indeed, in the presence of PLC inhibitor (U73122) or PKC inhibitor (Ro-31-7549), EPAC-mediated phosphorylation of serine-16 was negated in neonatal mouse and rat cardiac myocytes (Supplemental Figure 6). Twelve PKC isoforms are known to exist; conventional PKC subtypes, which include PKCα, PKCβ, and PKCγ, require both Ca²⁺ and phospholipid for activation, whereas novel PKC subtypes, which include PKCδ and PKCε, require phospholipid, but not Ca²⁺ (43). We examined the effect of silencing each PKC isoform with siRNA on PLN phosphorylation (Supplemental Figure 7A). Silencing of PKCα, PKCβ, PKCγ, and PKCδ did not affect the EPAC-mediated PLN phosphorylation on serine-16 with 8-CPT-AM (Supplemental Figure 8). However, when PKCε was silenced with siRNA (its specificity was confirmed: Supplemental Figure 7B), EPAC-mediated PLN phosphorylation on serine-16 was significantly attenuated compared with the control (control siRNA versus PKCε siRNA: 410% ± 44% versus 158% ± 10%, n = 5–7, P < 0.01), suggesting that PKCε plays a role in PLN phosphorylation on serine-16 by EPAC (Figure 2C).

EPAC phosphorylates PLN in an additive and independent manner with respect to PKA. In order to compare the roles of EPAC and PKA in PLN phosphorylation, we examined the effects of N⁶-benzoyladenosine-cAMP (Bnz-cAMP), a PKA-selective cAMP analogue, and 8-CPT-AM, an EPAC-selective analogue (Figure 2D). The degree of PLN phosphorylation achieved with 8-CPT-AM (10 μM) was approximately 60% of that achieved with Bnz-cAMP (50 μM) (Bnz-cAMP versus 8-CPT-AM: 3324% ± 289% versus 1851% ± 202%, n = 4–8, P < 0.001). However, when 8-CPT-AM and Bnz-cAMP were used together, PLN phosphorylation was additively and significantly increased (7688% ± 497%) relative to that with Bnz-cAMP alone (3324% ± 289%) or with 8-CPT-AM alone (1851% ± 202%) (P < 0.001 versus Bnz-cAMP or 8-CPT-AM alone). A similar tendency was observed when 5 μM 8-CPT-AM and 10 μM Bnz-cAMP were used together, suggesting that EPAC-mediated PLN phosphorylation may be independent of PKA-mediated PLN phosphorylation. Taken together, these results indicate that PLN phosphorylation via EPAC is additive to that via PKA, and both EPAC and PKA might be required for maximal PLN activation via phosphorylation on serine-16.

Effects of infection with Ad-PKI-GFP on PLN phosphorylation in neonatal cardiac myocytes transfected with negative or EPAC1 siRNA. In order to confirm the role of EPAC1 in PKA-independent PLN phosphorylation, we generated recombinant adenovirus of mouse protein kinase inhibitor α–GFP (Pkia-GFP) infusion gene (Ad-PKI-GFP) and adenovirus of GFP (Ad-GFP) as controls and examined the effects of Ad-PKI-GFP infection on PLN phosphorylation in neonatal rat cardiac myocytes transfected with control or EPAC1 siRNA (Supplemental Figure 9A, Supplemental Figure 10, and ref. 44).

We first confirmed that more than 98% of cells appeared green and that GFP fluorescence was evenly distributed in the cytoplasm of cardiomyocytes infected with Ad-PKI-GFP or Ad-PKI at MOI 100 (Supplemental Figure 10B).

We next performed Western blotting and found that PLN phosphorylation on serine-16 was significantly increased in cells infected with Ad-GFP or Ad-PKI-GFP in addition to control siRNA transfection after ISO treatment (10 μM) for 15 minutes. However, the magnitude of the increase was significantly smaller in cells transfected with Ad-PKI-GFP than with Ad-GFP (Ad-GFP [lane 2] versus Ad-PKI-GFP [lane 4]: 60 ± 2.6 versus 45 ± 3.1-fold, P < 0.05, n = 4) (Supplemental Figure 10C). In order to examine the role of EPAC1 in PKA-independent PLN phosphorylation on serine-16, we examined the effect of EPAC1 siRNA transfection and found that silencing EPAC1 significantly decreased the ISO-promoted PLN phosphorylation on serine-16 in cells infected with Ad-PKI-GFP (lane 5: 22 ± 0.6-fold, P < 0.01, n = 4). However, pretreatment with KN-93 (10 μM for 30 minutes), a specific CaMKII inhibitor, did not decrease ISO-promoted PLN phosphorylation on serine-16 in cells transfected with control siRNA (lane 6: 37 ± 3.1-fold, P = NS, n = 8) (Supplemental Figure 10C).

We also examined whether PLN phosphorylation on threonine-17 was increased in cells transfected with Ad-GFP or Ad-PKI-GFP in addition to control siRNA after ISO treatment (10 μM) for 15 minutes. However, the magnitudes of the increase were similar in both cases (Ad-GFP [lane 2] versus Ad-PKI-GFP [lane 4]: 59 ± 0.04 versus 56 ± 0.05-fold, P = NS, n = 8) (Supplemental Figure 10D). Transfection of EPAC1 siRNA significantly decreased ISO-promoted PLN phosphorylation on threonine-17 in cells transfected with Ad-PKI-GFP (lane 5: 41 ± 0.04, P < 0.05, n = 8) as in the case of serine-16. However, KN-93 pretreatment (10 μM for 30 minutes) abrogated the ISO-promoted PLN phosphorylation on serine-17 in cells transfected with control siRNA (lane 6: 1.4 ± 0.003, P < 0.01, n = 8) (Supplemental Figure 10D).

Together with the data in Figure 2D, these data obtained with Ad-PKI-GFP clearly demonstrated that EPAC1 and PKA regulate PLN phosphorylation on serine-16 in an additive and an independent manner. Also, EPAC1 regulates PLN phosphorylation at both serine-16 and threonine-17 in vitro, in addition to PKA and CaMKII.

The increase of PLN phosphorylation on serine-16 with ISO was decreased in Epac1 KO. We next examined the phosphorylation of PLN in total homogenate prepared from WT and Epac1 KO hearts perfused with the Langendorf method. Hearts were perfused with Krebs solution at constant pressure with or without the treatment of ISO (0.1 μM) for 5 minutes (45), then rapidly frozen in liquid nitrogen for Western blotting (Figure 3).

Figure 3

PLN phosphorylation on serine-16 and threonine-17 and CaMKII phosphorylation on threonine-286 in isolated WT or Epac1 KO heart perfused according to Langendorf method with or without subsequent ISO (0.1 μM) for 5 minutes. (A) PLN phosphorylation on serine-16 was significantly increased (*P < 0.05, **P < 0.01) in response to ISO in WT and Epac1 KO (WT: from 100% ± 7.0% to 351% ± 32%, n = 6–8; Epac1 KO: from 60 ± 2.6 to 153% ± 9%, n = 5), but increase was significantly smaller in Epac1 KO (153% ± 9%) compared with WT (351% ± 32%, **P < 0.01, n = 5–6). Ratio of phosphorylated/total protein expression of PLN in WT at baseline was taken as 100%. (B) PLN phosphorylation on threonine-17 was similar in WT and Epac1 KO at baseline and was significantly increased in response to ISO in WT (from 100% ± 2.3% to 183% ± 14%, **P < 0.01, n = 6–8) and Epac1 KO (from 94% ± 6.7% to 173% ± 13%, **P < 0.01, n = 6–8). Magnitudes of the increase were similar (P = NS). Ratio of phosphorylated/total protein expression of PLN in WT at baseline was taken as 100%. (C) CaMKII phosphorylation on threonine-286 was similar in WT and Epac1 KO at baseline and was significantly increased (**P < 0.01) in response to ISO in WT (from 100% ± 12% to 297% ± 37%, n = 5) and Epac1 KO (from 82% ± 7.7% to 278% ± 58%, n = 5). Magnitudes of increase were similar (P = NS). Ratio of phosphorylated/total protein expression of CaMKII in WT at baseline was taken as 100%. (D) Representative immunoblotting results of phosphorylation of PLN on serine-16 (upper) and threonine-17 (middle) and CaMKII on threonine-286 (lower). p-CaMKII, phosphorylated CaMKII; T-CaMKII, total CaMKII.

PLN phosphorylation on serine-16 was significantly decreased in Epac1 KO compared with that in WT at baseline (WT versus Epac1 KO: 100% ± 7.0% versus 60% ± 2.5%, P < 0.05, n = 5–8). It was significantly increased in response to ISO, but the magnitude of the increase was significantly smaller in Epac1 KO (WT versus Epac1 KO: 351% ± 32% versus 153% ± 9%, n = 5–6, P < 0.01) (Figure 3A). PLN phosphorylation on threonine-17 was similar in both WT and Epac1 KO at baseline (100% ± 3.0% versus 94% ± 6.7%, n = 8, P = NS). However, it was significantly increased in both WT and Epac1 KO in response to ISO (P < 0.01), and the magnitudes of the increase were similar (WT versus Epac1 KO: 183% ± 14% versus 173% ± 13%, n = 6, P = NS) (Figure 3B). CaMKII phosphorylation on threonine-286 was also similar in WT and Epac1 KO at baseline (WT versus Epac1: 100% ± 12% versus 82% ± 7.7%, n = 5, P = NS). It was significantly increased in both WT and Epac1 KO in response to ISO (P < 0.01), and the magnitudes of the increase were similar (WT versus Epac1 KO: 297% ± 37% versus 278% ± 58%, n = 5, P = NS) (Figure 3C). These data indicate that PLN phosphorylation on serine-16, the major site of PKA-mediated phosphorylation, is also regulated by EPAC1. However, EPAC1 does not alter PLN phosphorylation on threonine-17, the major site of CaMKII-mediated phosphorylation, at baseline or in response to ISO.

Silencing EPAC1 in cardiac myocytes attenuates ISO-mediated increases of RyR2 phosphorylations on serine-2808 and serine-2814. PKA and CaMKII-mediated phosphorylation of RyR2 play important roles in modulating cardiac contractility and arrhythmogenesis. It was reported that RyR2 phosphorylation on serine-2808 is mediated by PKA, while RyR2 phosphorylation on serine-2814 is mediated by CaMKII (46). More recently, both sites have been reported to be phosphorylated by PKA and CaMKII, and their phosphorylation is required for adequate RyR2 functional activity (11). We thus examined ISO-induced (10 μM for 3 minutes) phosphorylation on serine-2808 and serine-2814 of RyR2 in neonatal rat cardiac myocytes transfected with control or EPAC1 siRNA because we found that phosphorylated and total RyR2 antibodies did not work well in cardiac homogenate prepared from mouse heart perfused with the Langendorf method (Supplemental Figure 9A and Supplemental Figure 11). RyR2 phosphorylation levels on serine-2808 and serine-2814 were similar in cells transfected with control or EPAC1 siRNA at baseline (serine-2808: control siRNA versus EPAC1 siRNA: 100% ± 1.9% versus 130% ± 4.9%, n = 4, P = NS; serine-2814: control siRNA versus EPAC1 siRNA: 100% ± 16% versus 116% ± 16%, n = 4–6, P = NS). They were significantly increased in response to ISO, but the magnitudes of the increase on both serine-2808 (Supplemental Figure 11B) and serine-2814 (Supplemental Figure 11C) were significantly smaller in cells transfected with EPAC1 siRNA than those in cells transfected with control siRNA (serine-2808: control siRNA versus EPAC1 siRNA 268% ± 8.3% versus 199% ± 13%; P < 0.01, n = 6; serine-2814: control siRNA versus EPAC1 siRNA 220% ± 13% versus 170% ± 2.3%, P < 0.05, n = 4–6). These data indicate that EPAC1 regulates RyR2 phosphorylation on serine-2808 and serine-2814 in addition to PKA and CaMKII.

Intracellular Ca²⁺ concentration is decreased in Epac1 KO myocytes. Cardiac contraction and relaxation are influenced by the intracellular increase in Ca²⁺ during systole and its decrease during diastole. Sarcoplasmic reticulum Ca²⁺ uptake is regulated via PLN phosphorylation. PLN phosphorylation has recently been concluded to play a prominent role in the regulation of myocardial contraction as well as relaxation based on findings in Pln knockout mice (12).

We thus examined the intracellular Ca²⁺ concentration in isolated cardiac myocytes of Epac1 KO (Figure 4, A and B). The basal and peak Ca²⁺ concentrations were significantly decreased in Epac1 KO compared with WT (basal Ca²⁺ in WT versus Epac1 KO 139 ± 6.8 versus 99 ± 11.8 nM, n = 32, P < 0.01; peak Ca²⁺ in WT versus Epac1 KO: 883 ± 35.3 versus 559 ± 33.2 nM, n = 32, P < 0.01) (Figure 4C). The decay time constant (τ) was significantly larger in Epac1 KO than that in WT controls (WT versus Epac1 KO: 108 ± 6.6 versus 187 ± 13.2 ms, n = 32, P < 0.01) (Figure 4D). We also examined the cytoplasmic Ca²⁺ concentration after addition of caffeine (10 mM) as a measure of Ca²⁺ content in SR in isolated cardiac myocytes of WT and Epac1 KO (Figure 4, E–G). The amplitude of the caffeine-induced increase in cytoplasmic Ca²⁺ concentration was significant in WT compared with Epac1 KO (WT versus Epac1 KO: 1210 ± 143 versus 835 ± 87.6 nM, n = 7, P < 0.05), indicating that the Ca²⁺ content in SR was decreased in Epac1 KO compared with WT.

Figure 4

Ca²⁺ transient of adult isolated cardiac myocytes from Epac1 KO. (A and B) Typical recordings of Ca²⁺ transients in cardiac myocytes from WT (A) and Epac1 KO (B). Note that the basal and peak Ca²⁺ transient amplitude and decay rate are smaller in Epac1 KO (n = 32 cells from 4 animals each). (C and D) Ca²⁺ transient parameters of isolated cardiac myocytes from Epac1 KO and WT. The basal Ca²⁺ concentration (WT versus Epac1 KO: 139 ± 6.8 versus 99 ± 11.8 nM) and peak Ca²⁺ concentration (WT versus Epac1 KO: 883 ± 35.3 versus 559 ± 33.2 nM) were significantly lower in Epac1 KO than in WT (C). The decay time constant (τ) was significantly larger in Epac1 KO than in WT (WT versus Epac1 KO: 108 ± 6.6 versus 187 ± 13.2 ms) (n = 32 cells, *P < 0.05) (D). (E and F) Typical recordings of Ca²⁺ transients after caffeine (10 mM) treatment of cardiac myocytes from WT (E) and Epac1 KO (F). (G) Peak Ca²⁺ concentration after caffeine (10 mM) treatment was significantly decreased in Epac1 KO compared with WT (WT versus Epac1 KO: 1210 ± 143 versus 835 ± 87.6 nM, n = 7, *P < 0.05).

Activities of L-type Ca²⁺ channel and NCX channel in Epac1 KO are similar to those in WT. We measured the L-type Ca²⁺ channel current (I_CaL) and Na⁺-Ca²⁺ exchange current (I_NCX) using the cell patch-clamp technique in isolated cardiac myocytes of Epac1 KO and confirmed that the density and current-voltage relationships of I_CaL (Supplemental Figure 12) and I_NCX (Supplemental Figure 13) were not different. We also examined the response of I_CaL, a PKA-dependent target, in response to ISO (10^–7 M) and confirmed that it was not different between WT and Epac1 KO (Supplemental Figure 12B and ref. 6).

Disruption of EPAC1 does not affect development of cardiac hypertrophy, but prevents heart failure in pressure overload stress. In order to examine the role of EPAC1 in the development of heart failure, we performed aortic banding in Epac1 KO and WT for 3 weeks. The pressure gradients were similar in WT and Epac1 KO at 3 weeks after banding (Figure 5A). At baseline, there was no difference between WT and Epac1 KO in the LV weight (mg)/tibial length (mm) (LV/tibial length ratio in WT versus Epac1 KO: 5.2 ± 0.2 versus 5.2 ± 0.2 mg/mm, n = 18–20, P = NS). The degree of cardiac hypertrophy was increased at 3 weeks, but remained similar in Epac1 KO and WT (LV/tibial length ratio in WT versus Epac1 KO: 7.3 ± 0.2 versus 7.0 ± 0.2, n = 18–20, P = NS) (Figure 5, B, D, and E). The change of LV/body weight ratio, another index of cardiac hypertrophy, confirmed the finding based on LV/tibial length ratio ( WT versus Epac1 KO: from 3.2 ± 0.1 to 4.4 ± 0.1 versus from 3.3 ± 0.2 to 4.4 ± 0.2 mg/g, n = 18–20, P = NS) (Figure 5C).

Figure 5

Comparison of cardiac hypertrophy after aortic banding (TAC) in WT and Epac1 KO. (A) Pressure gradients were not different in WT and Epac1 KO at 3 weeks after TAC (n = 13-16, P = NS). (B and C) LV weight (mg)/tibial length (mm) ratio (B) and LV weight (mg)/body weight (BW, g) ratio (C) were determined at 3 weeks. The degree of cardiac hypertrophy was increased at 3 weeks, but was similar in WT and Epac1 KO (LV/tibial length ratio for WT versus Epac1 KO: 7.3 ± 0.2 versus 7.0 ± 0.2; LV/BW ratio for WT versus Epac1 KO: 4.4 ± 0.1 versus 4.4 ± 0.2, n = 18–20, P = NS). (D and E) Representative gross LV section of Masson-trichrome staining in sham-operated and TAC-operated WT and Epac1 KO heart. Note that the degree of cardiac hypertrophy was similar, but the fibrotic area after aortic banding was smaller in Epac1 KO than that in WT.

At 3 weeks after banding, however, cardiac function (LVEF) was significantly decreased in WT (P < 0.01), while it remained unchanged in Epac1 KO (WT versus Epac1 KO: from 70% ± 0.8% to 54% ± 2.0% versus from 60% ± 1.1% to 60% ± 0.7%, n = 17–30) (Figure 6, A and B). LVESD was increased significantly in WT at 3 weeks after banding (P < 0.01), while it remained unchanged in Epac1 KO (WT versus Epac1 KO: from 2.9 ± 0.04 to 3.3 ± 0.1 versus from 3.4 ± 0.1 to 3.1 ± 0.1 mm, n = 17–31) (Figure 6, C and D). Hemodynamic measurements demonstrated that Max dP/dt was significantly decreased in WT, but not in Epac1 KO (WT versus Epac1 KO: from 9091 ± 240 to 6359 ± 393 mmHg [P < 0.01] versus from 7246 ± 226 to 6811 ± 250 mmHg, n = 13–17) and the magnitude of the decrease was much smaller in Epac1 KO (Supplemental Figure 14, A and D). Min dP/dt was significantly increased in WT, while it remained unchanged in Epac1 KO (WT versus Epac1 KO: from –9110 ± 240 to –4389 ± 469 versus from –6201 ± 125 to –5131 ± 334 mmHg, n = 13–18, P = NS) and the magnitude of the increase was much smaller in Epac1 KO (Supplemental Figure 14, B and E). End-diastolic pressure (EDP) was significantly increased in both WT and Epac1 KO (WT versus Epac1 KO: from 7.2 ± 0.3 to 16.8 ± 1.4 versus from 6.8 ± 0.3 to 11.5 ± 1.1 mmHg, n = 13–22, P < 0.01), but the magnitude of the increase was significantly smaller in Epac1 KO (Supplemental Figure 14, C and F).

Figure 6

Changes in LV function at 3 weeks after aortic banding (TAC) in WT and Epac1 KO. Echocardiographic measurements of LV function were performed at 3 weeks after TAC of WT and Epac1 KO and in sham-operated controls. (A and B) LVEF was significantly decreased in WT (*P < 0.01), but not in Epac1 KO (P = NS) at 3 weeks (WT versus Epac1 KO: from 70% ± 0.8% to 54% ± 2.0% versus from 60% ± 1.1% to 59% ± 0.7%, n = 17–30) (A). The data were compared with those from sham-operated controls at 3 weeks in each mouse. Change of LVEF from sham-operated controls at 3 weeks after TAC was significantly greater in WT than that in Epac1 KO (B) (n = 17–31, *P < 0.01). (C and D) LVESD was significantly increased in WT, but not in Epac1 KO at 3 weeks after TAC (C). The data were compared with those from sham-operated controls at 3 weeks. Change of LVESD from sham-operated controls at 3 weeks after banding was greater in WT than that in Epac1 KO (n = 17–31, *P < 0.01) (D). The decrease of LVESD in response to intravenous acute ISO infusion (0, 0.13, 0.27, 0.40 mg/kg/min for 5 minutes) was depressed in WT, but not in Epac1 KO (n = 4–5, *P < 0.01). (E and F) The increase of LVEF (E) and the decrease of LVESD (F) in response to intravenous acute ISO infusion (0, 0.13, 0.27, 0.40 mg/kg/min for 5 minutes) were depressed in WT, but not in Epac1 KO (n = 4–5, *P < 0.01) (E).

We also examined the effect of long-term chronic pressure overload on cardiac hypertrophy and cardiac function at 5 weeks after banding in Epac1 KO (Supplemental Figure 15A). At 5 weeks after banding, cardiac hypertrophy was increased in both WT and Epac1 KO and the magnitude of the increase was similar (WT versus Epac1 KO, LV/tibia length ratio: 7.7 ± 0.4 versus 7.7 ± 0.2 mg/mm; LV/body weight ratio: 4.4 ± 0.3 versus 4.6 ± 0.2 mg/g, n = 15, P = NS) (Supplemental Figure 15, B and C). However, cardiac function was significantly decreased in WT (P < 0.01), while it remained unchanged in Epac1 KO (from 60% ± 1.1% to 61% ± 0.7%, n = 15). LVESD was increased significantly in WT at 5 weeks after banding (P < 0.01), while it remained unchanged in Epac1 KO (WT versus Epac1 KO: from 2.9 ± 0.04 to 3.3 ± 0.1 versus from 3.3 ± 0.1 to 3.2 ± 0.1 mm, n = 15–31) (Supplemental Figure 15, D–F). These results indicate that Epac1 KO developed hypertrophy similar to that of WT in response to pressure overload, but did not develop cardiac failure.

Cardiac fibrosis and apoptosis are attenuated in Epac1 KO. There was no difference in degree of fibrosis between WT and Epac1 KO at baseline. Thus, the development of cardiac fibrosis was unaffected in Epac1 KO (WT versus Epac1 KO: 0.3% ± 0.09% versus 0.3% ± 0.06%, n = 5, P = NS). Aortic banding increased the area of fibrosis in both WT and Epac1 KO (Figure 5, D and E), but the magnitudes of increase were much smaller in Epac1 KO (WT versus Epac1 KO: 7.1% ± 0.9% versus 1.7% ± 0.8%, n = 4, P < 0.01) (Figure 7, A and B).

Figure 7

Accelerated morphological deterioration after aortic banding (TAC) was attenuated in Epac1 KO. (A) Representative images of Masson-trichrome–stained sections of sham-operated and TAC-operated heart of WT and Epac1 KO at 3 weeks. Scale bars: 100 μm. (B) Quantitative analysis of the fibrotic area in sham-operated and TAC-operated heart at 3 weeks after TAC of WT and Epac1 KO. Cardiac fibrosis was significantly increased after TAC in WT and Epac1 KO, but magnitude of the increase was much smaller in Epac1 KO (n = 4, *P < 0.01). (C) TUNEL-positive myocytes in LV myocardium were counted in WT and Epac1 KO and expressed as percentage of total myocytes. Number of TUNEL-positive myocytes was significantly smaller in Epac1 KO than in WT at 3 weeks after aortic banding (n = 4–6, *P < 0.05). (D) Expression of BAX protein was compared between WT and Epac1 KO. Protein expression of BAX was determined by Western blot analysis, which showed greater expression in WT than in Epac1 KO at 3 weeks after TAC (n = 4, *P < 0.05). Expression of BAX protein in the heart of sham-operated control WT was taken as 100%. Representative immunoblotting results are shown. (E) Representative images of double-immunostaining for dystrophin (brown) and PECAM (blue) of WT and Epac1 KO at baseline and 3 weeks after aortic banding. Scale bars: 50 μm. (F) Number of microvessels per cardiomyocyte was compared in WT and Epac1 KO at baseline and at 3 weeks after aortic banding (n = 4, *P < 0.01).

Similarly, there was no difference in the number of TUNEL-positive cells between WT and Epac1 KO at baseline (WT versus Epac1 KO: 0.08% ± 0.01% versus 0.06% ± 0.007%, n = 6, P = NS). Aortic banding increased the number of TUNEL-positive cells in both WT and Epac1 KO, but the magnitudes of increase were much smaller in Epac1 KO (WT versus Epac1 KO: 0.49% ± 0.07% versus 0.24% ± 0.05%, n = 4–6, P < 0.05) (Figure 7C). To examine changes in the molecules involved in apoptosis signaling, we quantitated BAX protein, an accelerator of apoptosis (31), in WT and Epac1 KO and found that its expression after banding was significantly increased, by 2.1-fold in WT, but not in Epac1 KO (n = 4–6, P < 0.05) (Figure 7D).

Myocardial blood flow is preserved in Epac1 KO. We next examined cardiac response to intravenous ISO infusion, which is a hallmark of transmural myocardial blood flow, at 3 weeks after aortic banding (47). The increase of LVEF and decrease of LVESD in response to ISO were blunted in WT, but they were preserved in Epac1 KO (n = 4–5, P < 0.01 versus WT), suggesting that cardiac angiogenesis during aortic banding might be preserved in Epac1 KO (Figure 6, E and F, and ref. 48).

We also examined the number of microvessels per cardiomyocyte. There was no difference in the number of microvessels between WT and Epac1 KO at baseline (WT versus Epac1 KO: 0.41 ± 0.02 versus 0.42 ± 0.04, n = 4, P = NS). However, the number was significantly decreased in WT compared with Epac1 KO at 3 weeks after aortic banding (WT versus Epac1 KO: 0.29 ± 0.01 versus 0.42 ± 0.04, n = 4, P < 0.01) (Figure 7, E and F).

Disruption of EPAC1 results in resistance to long-term ISO infusion. We also examined the effect of long-term ISO infusion on LV function (Supplemental Figure 16A). At 1 week after chronic ISO infusion, LVEF was significantly decreased in WT (P < 0.01), while it remained unchanged in Epac1 KO (WT versus Epac1 KO: from 70 ± 0.8 to 60 ± 1.1 versus from 62% ± 1.4% to 60% ± 0.9%, n = 14–31).

Cardiac fibrosis after long-term ISO infusion was also significantly decreased in Epac1 KO (WT versus Epac1 KO: 4.6% ± 1.6% versus 1.6% ± 0.2%, n = 5, P < 0.01) (Supplemental Figure 16, B and C). Similarly, cardiac apoptosis was also significantly attenuated in Epac1 KO after long-term ISO infusion (WT versus Epac1 KO: 0.49% ± 0.07% versus 0.24% ± 0.05%, n = 4–5, P < 0.05) (Supplemental Figure 16D).

The degree of cardiac hypertrophy was increased at 1 week, but remained similar in Epac1 KO and WT (LV/tibial length ratio for WT versus Epac1 KO: 5.7 ± 0.2 versus 5.9 ± 0.1, n = 14–15, P = NS; LV/BW ratio for WT versus Epac1 KO: 3.5 ± 0.2 versus 3.7 ± 0.1, n = 14–15, P = NS) (Supplemental Figure 17, A and B).

Silencing of EPAC1 in cardiac fibroblasts did not alter cell proliferation. Since the mouse model is a global Epac1 KO, the phenotype of less fibrosis after long-term ISO infusion or aortic banding may result from altered cell proliferation or signaling in cardiac fibroblasts. We thus examined the effect of silencing EPAC1 on neonatal rat cardiac fibroblast proliferation in response to ISO, using 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Supplemental Figure 9B, Supplemental Figure 18, and ref. 49). Cardiac fibroblasts were cultured in medium with 10% fetal bovine serum for 24 hours and were starved for 48 hours without serum prior to stimulation. At 24 hours after ISO (1 μM or 10 μM) stimulation, MTT assay was performed. Cell proliferation was significantly increased (P < 0.01) by ISO (10 μM) from baseline (control siRNA versus EPAC1 siRNA: from 100% ± 5.7% to 137% ± 4.6% versus from 108% ± 1.7% to 140% ± 4.6%, n = 6–8), but the magnitudes of the increase were similar (P = NS) in cells transfected with control or EPAC1 siRNA (ref. 49 and Supplemental Figure 18A).

We also examined p44/42 MAPK (ERK1/2) phosphorylation on tyrosine- 202/threonine-204 and SRC phosphorylation on tyrosine-416 because phosphorylations of ERK1/2 and SRC are important for G protein–coupled receptor–mediated cardiac fibroblast proliferation (50). ERK1/2 phosphorylation and SRC phosphorylation were similar in both control and EPAC1 siRNA–treated cells (Supplemental Figure 18, B–D). However, both ERK1/2 and SRC phosphorylations were significantly (P < 0.01) increased in response to ISO (10 μM) in cells transfected with control or EPAC1 siRNA, and the magnitudes of their increase were similar in the 2 cases (ERK1/2: control siRNA versus EPAC1 siRNA: from 100 ± 2.5 to 151 ± 2.0 versus from 108% ± 3.7% to 154% ± 6.7%, n = 4; SRC: control siRNA versus EPAC1 siRNA: from 100% ± 2.5% to 233% ± 12% versus from 107% ± 9.7% to 235% ± 23%, n = 7-8). These data indicate that silencing EPAC1 in cardiac fibroblasts did not alter the cell-proliferative response to β-AR–signaling stimulation.

Disruption of EPAC1 results in resistance to aging-related cardiomyopathy. Key features of aging-related cardiomyopathy were examined, i.e., LV weight/tibial length ratio, LV function, apoptosis, and fibrosis. The LV weight/tibial length ratio was not different between old WT (5.6 ± 0.5) and old Epac1 KO (5.6 ± 0.5) at baseline (24–32 months). However, aging-related changes of cardiac function (LVEF in WT versus Epac1 KO: from 70% ± 0.8% [young] to 58% ± 1.3% [old] [P < 0.01] versus from 60% ± 1.1% to 58% ± 2.2%, P = NS) (Supplemental Figure 16E), fibrosis (old WT versus old Epac1 KO: 3.2% ± 0.2% versus 1.2% ± 0.2%, n = 5–8, P < 0.01) (Supplemental Figure 16F) and apoptosis (old WT versus old Epac1 KO: 0.38% ± 0.04% versus 0.22% ± 0.03%, n = 5, P < 0.05) (Supplemental Figure 16G) were all significantly decreased in Epac1 KO.

Incidence of atrial fibrillation after rapid atrial pacing is attenuated in Epac1 KO. It has been reported that phosphorylations of RyR2 on serine-2814 and PLN on threonine-17 are increased after rapid atrial pacing in atrial myocytes and genetic inhibition of RyR2 phosphorylation on serine-2814 attenuates the incidence of atrial fibrillation (AF) induced by rapid atrial pacing (51). Accordingly, we examined the effects of Epac1 deletion on the initiation and maintenance of AF in the mouse model (Figure 8 and ref. 52). Transesophageal rapid atrial pacing was performed at a cycle length of 30 ms for 1 minute, followed by measurement of the duration of pacing-induced AF (Figure 8A). In all WT control mice, AF was consistently induced for 50 to 60 seconds. In striking contrast, little (less than 4 seconds) or no AF was induced in Epac1 KO. These data indicate that Epac1 deletion may suppress the initiation and maintenance of AF after transesophageal rapid atrial pacing, probably through inhibition of hyperphosphorylation of RyR2 and/or PLN in atrial myocytes (AF duration for WT versus Epac1 KO: 58 ± 3.8 versus 1.4 ± 1.4 seconds, n = 4–5, P < 0.01).

Figure 8

The duration of AF induced by transesophageal rapid atrial pacing was decreased in Epac1 KO. (A and B) Transesophageal pacing was performed at a cycle length of 30 ms for 1 minute, and the duration of pacing-induced AF was examined in Epac1 KO and WT (A). The duration of AF was significantly decreased in Epac1 KO compared with WT controls. AF was induced in all WT mice and its duration was 59 ± 3.4 seconds (n = 4). In Epac1 KO, we found that AF was hardly inducible (1.4 ± 1.0 seconds, n = 5, **P < 0.01 versus WT) (B).

Catecholamine-mediated spontaneous activity in the pulmonary vein cardiomyocytes is attenuated in Epac1 KO. High-frequency focal activity in the pulmonary vein cardiomyocytes may contribute to arrhythmogenic activity, which is induced mostly via an increase of intracellular Ca²⁺ derived from SR (53, 54). The incidence of the spontaneous activity was not different in WT (4/17:24%) and Epac1 KO (7/24:29%) at baseline (P = NS) (Supplemental Figure 19B). However, it was increased after noradrenaline (1 μM) treatment in both WT and Epac1 KO, though its incidence was less in Epac1 KO (12/17:71%) than in WT (15/15:100%) (P < 0.05 evaluated by Fisher’s exact probability test). These finding indicated that catecholamine-mediated spontaneous activity in the pulmonary vein cardiomyocytes is attenuated in Epac1 KO (Supplemental Figure 19, A and C).

Disruption of EPAC2 does not prevent the cardiac hypertrophy and the development of heart failure in response to pressure overload stress. In order to examine the effects of EPAC2, another cardiac isoform of EPAC, on the development of heart failure, we generated Epac2 KO (Supplemental Figure 20). Epac2 KO did not show any change in the size of the heart or in cardiac function, despite the smaller body weight, compared with that in WT (Supplemental Table 1). We examined PLN phosphorylation on serine-16 and on threonine-17 in Epac2 KO, but the levels were similar in both WT and Epac2 KO (Supplemental Figure 21).

In order to examine the role of EPAC2 in the development of heart failure, we performed aortic banding in Epac2 KO and WT for 3 weeks. At baseline, there was no difference between WT and Epac2 KO in the LV weight (mg)/tibial length (mm) (LV/tibial length ratio in WT versus Epac2 KO: 5.7 ± 0.1 versus 5.2 ± 0.2 mg/mm, n = 5–7, P = NS). The degree of cardiac hypertrophy was significantly (P < 0.01) increased at 3 weeks, but the magnitudes of the increase were similar in both WT and Epac2 KO (LV/tibial length ratio in WT versus Epac2 KO: 7.1 ± 0.4 versus 6.8 ± 0.2, n = 4, P = NS) (Supplemental Figure 22A).

At 3 weeks after banding, cardiac function (LVEF) was significantly decreased in both WT and Epac2 KO from baseline (WT versus Epac2 KO: from 73% ± 1.1% to 55% ± 3.1% [P < 0.01, n = 4–5] versus from 72% ± 1.1% to 58% ± 5.2% [P < 0.05, n = 5–7] ), but the magnitudes of the decrease were similar (P = NS) (Supplemental Figure 22B). LVESD was increased in both WT and Epac2 KO at 3 weeks after banding (WT versus Epac2 KO: from 2.7 ± 0.07 to 3.1 ± 0.2 [P < 0.05, n = 4–5] versus from 2.6 ± 0.05 to 3.1 ± 0.2 mm [P < 0.05, n = 4–13]), but the magnitudes of the increase were similar (P = NS) (Supplemental Figure 22C).

There was no difference in the degree of fibrosis between WT and Epac2 KO at baseline (Supplemental Figure 22, D and E). Thus, the development of cardiac fibrosis was unaffected in Epac2 KO (WT versus Epac2 KO: 0.4% ± 0.08% versus 0.4% ± 0.09%, n = 5, P = NS). Aortic banding increased the area of fibrosis in both WT and Epac2 KO, but the magnitudes of the increase were similar (WT versus Epac2 KO: 22% ± 3.6% versus 18% ± 5.3%, n = 4, P =NS) (Supplemental Figure 22, D and E).

Similarly, there was no difference in the number of TUNEL-positive cells between WT and Epac2 KO at baseline (WT versus Epac2 KO: 0.09% ± 0.02% versus 0.09% ± 0.01%, n = 5, P = NS) (Supplemental Figure 22F). Aortic banding significantly (P < 0.01) increased TUNEL-positive cells in both WT and Epac2 KO, but the magnitudes of increase were similar (WT versus Epac2 KO: 0.8% ± 0.04% versus 0.8% ± 0.05%, n = 4–5, P = NS) (Supplemental Figure 22F). These results indicate that Epac2 KO did not show decreased cardiac contractility and PLN phosphorylation at baseline and did not show resistance after aortic banding, indicating that the cardioprotective effects of EPAC under stress such as pressure overload (Figures 5–7, Supplemental Figure 14, and Supplemental Figure 15), chronic catecholamine stress (Supplemental Figure 16, A–D), aging stress (Supplemental Figure 16, E–G), and AF susceptibility (Figure 8) are Epac1 KO–specific phenotypes.

RyR2 is a prominent regulator of systolic contraction and PLN is a prominent regulator of myocardial diastolic relaxation via modulation of the activity of the SERCA2a. Specifically, RyR2 in its phosphorylated state following the Ca²⁺ influx through the L-type Ca²⁺ channel is an accelerator of Ca²⁺ leakage from SR, and PLN in its dephosphorylated state is an inhibitor of Ca²⁺ transport to SR via SERCA2a, while phosphorylation of PLN removes this inhibition, indicating that both molecules have central roles in modulating Ca²⁺ homeostasis and, therefore, cardiac function (10, 11, 46).

Early studies of enhanced SR Ca²⁺ leakage in heart failure focused on PKA, a classic cAMP target. Recently, it has been shown that β-AR stimulation can also induce PKA-independent, i.e., CaMKII- or EPAC-mediated SR Ca²⁺ leakage in vitro by increasing RyR2 phosphorylation on serine-2814/2815 and PLN phosphorylation on threonine-17 via PLCε/PLCε/CaMKII signaling (27–30, 55). Nevertheless, key functional issues remain to be clarified because most previous studies relied on pharmacological stimulation with 8-CPT, an EPAC-selective but not isoform-selective cAMP analogue, which could have off-target effects (37). First, it is unknown whether these effects are mediated by EPAC1 or EPAC2. Second, it is unknown whether EPAC-regulated Ca²⁺ homeostasis is important for cardiac function and arrhythmogenesis under physiological conditions. Third, it remains unknown whether EPAC regulates PLN phosphorylation on serine-16 in addition to threonine-17, which is known to be a major regulator of Ca²⁺ cycling in the heart and closely associated not only with an increase in cardiac function (16, 17, 56, 57), but also with arrhythmia and cardiomyopathy after chronic catecholamine infusion, aortic banding, or ischemia (18, 20, 21). We thus generated Epac1 KO and Epac2 KO in addition to silencing EPAC1 with siRNA and selectively inhibiting PKA with Ad-PKI-GFP in cardiac myocytes. The results clearly demonstrated that EPAC1-regulated Ca²⁺ homeostasis via PLN phosphorylation on serine-16, together with RyR2 phosphorylation on serine-2808 and serine-2814, may cause transition from hypertrophy to heart failure and evoke susceptibility to arrhythmia (Figure 9 and refs. 21, 58).

Figure 9

A schematic model of cAMP/EPAC signaling as opposed to cAMP/PKA signaling in the heart. Ca²⁺ stored in the SR is released into the cytosol to activate cardiac muscle contraction and subsequently reaccumulated to promote relaxation. PLN phosphorylation on serine-16 as well as threonine-17 occurs via the EPAC/PLC/PKCε/CaMKII pathway (27, 28). However, under physiological conditions, PLN phosphorylation on serine-16 by PKA rather than on threonine-17 by CaMKII is the major regulator of Ca²⁺ cycling in the heart (56, 57). Our current study indicates that PLN on serine-16 and RyR2 on serine-2808 and serine-2814 are phosphorylated by EPAC1 in addition to and independently of PKA or CaMKII. More importantly, hyperphosphorylation of PLN on serine-16 was recently reported to be associated not only with an increase in cardiac function in young animals (16, 17), but also with arrhythmia and cardiomyopathy after adrenergic stress, aortic banding, or ischemia (18, 20, 21). Our results suggest that Epac1-mediated hyperphosphorylation of PLN and RyR2 might be required for the development of heart failure as well as arrhythmia, in addition to PKA- or CaMKII mediated activation.

Pln-null mice show no gross developmental abnormalities, but exhibit enhanced myocardial contractility owing to increased SERCA2a pump activity and higher SR Ca²⁺ load in cardiomyocytes (12). Ablation of PLN restored contractility in some cardiomyopathic mouse models, such as cardiac-specific calsequestrin overexpression, targeted disruption of muscle-specific LIM protein, or cardiac-specific β₁-AR overexpression (16, 17, 59). Thus, various therapeutic approaches targeting or inhibiting PLN have been examined to rescue the failing heart (60, 61).

However, it was subsequently reported that ablation of PLN shows no cardioprotective effect against chronic pressure overload or ischemic injury and, indeed, exacerbates cardiac dysfunction in mice with cardiac CaMKIIδc overexpression, even if the Ca²⁺ transient is improved (19). More recently, it was reported that hyperphosphorylation of PLN in mice with constitutively active phosphatase inhibitor-1 increased contractile function in young animals, but also resulted in increased susceptibility to arrhythmias and cardiomyopathy in response to chronic catecholamine stress and aging, probably through increased Ca²⁺ leakage, leading to arrhythmogenicity and cardiomyocyte apoptosis (21). Our current data obtained in Epac1 KO and Epac2 KO demonstrated that EPAC1 plays an important role in PLN phosphorylation on serine-16 and EPAC1-mediated PLN phosphorylation rather than PKA-mediated PLN phosphorylation might have a pivotal role in the development of heart failure in response to chronic pressure overload, catecholamine stress, and aging, because the PKA pathway remains intact in Epac1 KO.

For many decades, it has been believed that the major target of catecholamine/cAMP signaling is PKA. Although acute sympathetic stimulation and activation of the cAMP/PKA pathway represent a major mechanism of improvement of cardiac function, previous studies by us and other groups using transgenic models have demonstrated that chronic activation of these pathways arising from specific cardiac overexpression of β-AR, Gsα, and PKA results in cardiomyopathy (3–5). On the other hand, disruption of type 5 AC, a major cardiac AC isoform, protects the heart from chronic pressure overload and catecholamine stress without affecting baseline cardiac function in mice (6–8). Interestingly, our results indicate that cardiac function is preserved in response to chronic pressure overload, chronic catecholamine, and aging stress in Epac1 KO, indicating that both phosphorylation of PLN on serine-16 and phosphorylation of RyR2 on serine-2808 and serine-2814 by EPAC1, rather than PKA, might be essential for the transition of hypertrophy to heart failure in response to chronic activation of β-AR signaling (11, 46, 58, 62).

Increased Ca²⁺ leakage in atrial myocytes has been demonstrated to be a cause and consequence of AF, but the mechanism remains poorly understood (63). Recently, it was demonstrated that PLN phosphorylation on serine-16, rather than threonine-17, was significantly increased in the atrium of chronic AF patients while PKA phosphorylation of inhibitory troponin subunit was unaltered (22). Similarly, it was demonstrated that rapid atrial pacing induces phosphorylation of RyR2 on serine-2814 and PLN on threonine-17 and genetic inhibition of RyR2 phosphorylation on serine-2814 prevents AF after rapid atrial pacing (64). These results indicate that altered Ca²⁺ homeostasis in atrial myocytes through EPAC-mediated hyperphosphorylation of PLN and RyR2 may cause susceptibility to AF. We thus hypothesized that EPAC1 plays an important role in the occurrence of AF, and we confirmed this idea using a transesophageal pacing-induced AF mouse model. American College of Cardiology, American Heart Association, and European Society of Cardiology guidelines strongly recommend β-blockers for the management of adrenergically induced AF, but it is difficult to use β-blockers in patients with chronic obstructive pulmonary disease or severe heart failure. Therefore, EPAC may be available as a new target of pharmacotherapy for adrenergically induced AF (65).

PLN was demonstrated to be phosphorylated by PKC in vitro in addition to PKA and CaMKII (39, 41), but the mechanism of PKC-dependent PLN phosphorylation remains unclear and its importance for cardiac function is controversial (66, 67). Our current in vitro studies, in agreement with earlier studies (27–29), indicate that EPAC regulates PLN phosphorylation on both serine-16 and threonine-17 through PLC/PKCε signaling. However, our in vivo studies using Epac1 KO show that EPAC1 regulates PLN phosphorylation on serine-16 at baseline and under stress, but it does not affect the phosphorylation on threonine-17. Previous in vitro studies have shown that serine-16 of PLN and threonine-17 of PLN can be readily and independently phosphorylated by PKA and CaMKII, respectively. However, previous in vivo studies indicated that PLN phosphorylation on threonine-17 can be induced by β-AR stimulation that is sufficient to increase cytosolic Ca²⁺ concentration, which in turn would activate CaMKII and/or inhibit the phosphatase that dephosphorylates PLN. The apparent difference between in vitro and in vivo PLN phosphorylation is yet to be reconciled (68), but a possible explanation of the difference between our in vitro and in vivo data on PLN phosphorylation at threonine-17 is that the extent of CaMKII activation was not enough to produce a significant difference between WT and Epac1 KO under β-AR stimulation in response to ISO in the Langendorf-perfused heart system used in this study. We also cannot exclude the possibility that this might be due to a difference in the specificity of the gene-silencing technique in vitro with siRNA and in vivo with genetic KO because complete gene silencing can be achieved by genetic KO (31), but not by using siRNA (Supplemental Figure 9 and ref. 69).

In contrast with Epac1 KO, disruption of EPAC2 does not alter the cardiac function or PLN phosphorylation at baseline and has no protective effects after aortic banding. These data indicate that EPAC1, but not EPAC2, is a major regulator of cardiac function at baseline and under stress.

More importantly, cardiac function and the Ca²⁺ transient are similar in global Plce KO and WT at baseline even if the ISO-stimulated Ca²⁺ transient is smaller in Plce KO (70). Global Plce KO does not show a protective phenotype against chronic catecholamine stress (70) in contrast with Epac1 KO. Otherwise, global Plce KO is more susceptible than WT to the development of cardiac hypertrophy and fibrosis after chronic ISO infusion (70). In contrast with global Plce KO, myocyte-specific Plce KO is not susceptible, but is rescued from cardiac hypertrophy as well as cardiac dysfunction after TAC (71). Also, PLCε was demonstrated to generate a multifunctional complex with muscle-specific A-kinase–anchoring protein at or near the nuclear envelope along with Epac1, PKCε, PKD, and RyR2 in cardiac myocytes (71). Pkce KO does not show a protective phenotype and is not susceptible to development of cardiac hypertrophy after TAC, in contrast with global Plce KO and myocyte-specific Plce KO (72). These data indicate that disruption of EPAC1 modulates or antagonizes the downstream PLCε/PKCε pathway, which in turn is associated with the cardioprotective phenotype of Epac1 KO under stress.

During the revision of this manuscript, another group reported that cardiac function was not different between WT and Epac1 KO at baseline in contrast with our findings. The reason for the difference of the cardiac phenotype between our Epac1 KO and that generated by the other group is not clear (30). However, the genetic backgrounds of Epac1 KO generated by us (CBA-C57BL/6) and by the other group (R1-C57BL/6) were different, and this might cause the difference in the mouse phenotype (73, 74), as in the case of type 5 AC KO (129/SvJ-C57BL/6) generated by us (6, 7) and another group (C57BL/6) (75) and Plce KO generated by 2 different groups: 129/Sv-C57BL/6 (76) versus 129/S6-C57B/6 (70). Also, environmental influences have been reported to alter phenotype, as in the case of Tgfb KO (77, 78), and we cannot exclude this as another possibility.

In summary, we have demonstrated that disruption of EPAC1 protects the heart from chronic pressure overload, chronic catecholamine stress, aging-related cardiomyopathy, and AF susceptibility through the inhibition of PLN phosphorylation on serine-16 in addition to the inhibition of RyR2 phosphorylation on serine-2808 and serine-2814. The cAMP/PKA pathway is believed to be a major pathway for the development of heart failure as well as arrhythmia (5, 21). Our present data show that the cAMP/EPAC1 pathway plays an important role independently of the cAMP/PKA pathway in the development of heart failure and arrhythmia.

View Supplemental data

This study was supported in part by grants from the Ministry of Health, Labor and Welfare (to Y. Ishikawa), and a Grant-in-Aid for Scientific Research on Innovative Areas (22136009) as well as grants from the Kitsuen Kagaku Research Foundation (to Y. Ishikawa), the Japanese Ministry of Education, Culture, Sports, Science, and Technology (to Y. Ishikawa [24390200, 259670131], S. Okumura [60233475], T. Fujita [40468202], M. Jin [postdoctoral fellowship for foreign researchers 2011.9-2013.8], I. Namekata [25860194], Y. Mototani [22791147], M. Sato [24590280, 25136721], U. Yokoyama [25293236], S. Suzuki [21790208], R. Kurotani [24591151], Y. Bai [22790719], T. Tsunematsu [22590811], Y. Ichikawa [25860614], S. Adachi-Akahane [23659142], H. Tanaka [24590334]), the Japan Space Forum (to Y. Ishikawa), the Takeda Science Foundation (to Y. Ishikawa and S. Okumura), the Yokohama Foundation for Advancement of Medical Science (to S. Okumura and Y. Ohnuki), a grant for a 2006–2007 Strategic Research Project (no. K19027) from Yokohama City University, Japan (to S. Okumura), the Mitsubishi Pharma Research Foundation (to S. Okumura), Research for Promoting Technological Seeds A (discovery type) (to S. Okumura), the Yokohama Academic Foundation (to S. Okumura and Y. Ohnuki), the 2010 Commercialization Promotion Program for Biotechnology-Related Studies (to S. Okumura), and Grants for Research and Development Project II (nos. 8 and 14) from Yokohama City University (to Y. Ishikawa and S. Okumura), and the Research Foundation for Community Medicine (to S. Okumura).

Address correspondence to: Satoshi Okumura, Department of Physiology, Tsurumi University School of Dental Medicine, 2-1-2 Tsurumi, Tsurumi-ku, Yokohama 230-8501, Japan. Phone: 81.45.580.8476; Fax: 81.45.585.2889; E-mail: okumura-s@tsurumi-u.ac.jp. Or to: Yoshihiro Ishikawa, Cardiovascular Research Institute, Yokohama City University Graduate School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama 236-0004, Japan. Phone: 81.45.787.2575; Fax: 81.45.788.1470; E-mail: yishikaw@yokohama-cu.ac.jp.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J Clin Invest. 2014;124(6):2785–2801. doi:10.1172/JCI64784.

1. Ishikawa Y, Homcy CJ. The adenylyl cyclases as integrators of transmembrane signal transduction. Circ Res. 1997;80(3):297–304.
   View this article via: PubMed CrossRef Google Scholar
2. Vatner DE, et al. β-adrenergic receptor-G protein-adenylyl cyclase signal transduction in the failing heart. Am J Cardiol. 1999;83(12A):80H–85H.
   View this article via: PubMed CrossRef Google Scholar
3. Engelhardt S, Hein L, Wiesmann F, Lohse MJ. Progressive hypertrophy and heart failure in β₁-adrenergic receptor transgenic mice. Proc Natl Acad Sci U S A. 1999;96(12):7059–7064.
   View this article via: PubMed CrossRef Google Scholar
4. Asai K, et al. β-adrenergic receptor blockade arrests myocyte damage and preserves cardiac function in the transgenic Gsα mouse. J Clin Invest. 1999;104(5):551–558.
   View this article via: JCI PubMed CrossRef Google Scholar
5. Antos CL, et al. Dilated cardiomyopathy and sudden death resulting from constitutive activation of protein kinase a. Circ Res. 2001;89(11):997–1004.
   View this article via: PubMed CrossRef Google Scholar
6. Okumura S, et al. Type 5 adenylyl cyclase disruption alters not only sympathetic but also parasympathetic and calcium-mediated cardiac regulation. Circ Res. 2003;93(4):364–371.
   View this article via: PubMed CrossRef Google Scholar
7. Okumura S, et al. Disruption of type 5 adenylyl cyclase gene preserves cardiac function against pressure overload. Proc Natl Acad Sci U S A. 2003;100(17):9986–9990.
   View this article via: PubMed CrossRef Google Scholar
8. Okumura S, et al. Disruption of type 5 adenylyl cyclase enhances desensitization of cyclic adenosine monophosphate signal and increases Akt signal with chronic catecholamine stress. Circulation. 2007;116(16):1776–1783.
   View this article via: PubMed CrossRef Google Scholar
9. Yan L, et al. Type 5 adenylyl cyclase disruption increases longevity and protects against stress. Cell. 2007;130(2):247–258.
   View this article via: PubMed CrossRef Google Scholar
10. Koss KL, Kranias EG. Phospholamban: a prominent regulator of myocardial contractility. Circ Res. 1996;79(6):1059–1063.
    View this article via: PubMed CrossRef Google Scholar
11. Li J, et al. β-adrenergic stimulation increases RyR2 activity via intracellular Ca²⁺ and Mg²⁺ regulation. PLoS One. 2013;8(3):e58334.
    View this article via: PubMed CrossRef Google Scholar
12. Luo W, et al. Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of β-agonist stimulation. Circ Res. 1994;75(3):401–409.
    View this article via: PubMed CrossRef Google Scholar
13. Kuschel M, Karczewski P, Hempel P, Schlegel WP, Krause EG, Bartel S. Ser¹⁶ prevails over Thr¹⁷ phospholamban phosphorylation in the β-adrenergic regulation of cardiac relaxation. Am J Physiol. 1999;276(5 pt 2):H1625–H1633.
    View this article via: PubMed Google Scholar
14. Sordahl LA, McCollum WB, Wood WG, Schwartz A. Mitochondria and sarcoplasmic reticulum function in cardiac hypertrophy and failure. Am J Physiol. 1973;224(3):497–502.
    View this article via: PubMed Google Scholar
15. Whitmer JT, Kumar P, Solaro RJ. Calcium transport properties of cardiac sarcoplasmic reticulum from cardiomyopathic Syrian hamsters (BIO 53.58 and 14.6): evidence for a quantitative defect in dilated myopathic hearts not evident in hypertrophic hearts. Circ Res. 1988;62(1):81–85.
    View this article via: PubMed CrossRef Google Scholar
16. Minamisawa S, et al. Chronic phospholamban-sarcoplasmic reticulum calcium ATPase interaction is the critical calcium cycling defect in dilated cardiomyopathy. Cell. 1999;99(3):313–322.
    View this article via: PubMed CrossRef Google Scholar
17. Sato Y, et al. Rescue of contractile parameters and myocyte hypertrophy in calsequestrin overexpressing myocardium by phospholamban ablation. J Biol Chem. 2001;276(12):9392–9399.
    View this article via: PubMed CrossRef Google Scholar
18. Cross HR, Kranias EG, Murphy E, Steenbergen C. Ablation of PLB exacerbates ischemic injury to a lesser extent in female than male mice: protective role of NO. Am J Physiol Heart Circ Physiol. 2003;284(2):H683–H690.
    View this article via: PubMed Google Scholar
19. Zhang T, et al. Phospholamban ablation rescues sarcoplasmic reticulum Ca²⁺ handling but exacerbates cardiac dysfunction in CaMKIIδc transgenic mice. Circ Res. 2010;106(2):354–362.
    View this article via: PubMed CrossRef Google Scholar
20. Kiriazis H, et al. Hypertrophy and functional alterations in hyperdynamic phospholamban-knockout mouse hearts under chronic aortic stenosis. Cardiovasc Res. 2002;53(2):372–381.
    View this article via: PubMed CrossRef Google Scholar
21. Wittkopper K, et al. Constitutively active phosphatase inhibitor-1 improves cardiac contractility in young mice but is deleterious after catecholaminergic stress and with aging. J Clin Invest. 2010;120(2):617–626.
    View this article via: JCI PubMed CrossRef Google Scholar
22. El-Armouche A, et al. Molecular determinants of altered Ca²⁺ handling in human chronic atrial fibrillation. Circulation. 2006;114(7):670–680.
    View this article via: PubMed CrossRef Google Scholar
23. Kawasaki H, et al. A family of cAMP-binding proteins that directly activate Rap1. Science. 1998;282(5397):2275–2279.
    View this article via: PubMed CrossRef Google Scholar
24. de Rooij J, et al. Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature. 1998;396(6710):474–477.
    View this article via: PubMed CrossRef Google Scholar
25. Gloerich M, Bos JL. Epac: defining a new mechanism for cAMP action. Annu Rev Pharmacol Toxicol. 2010;50:355–375.
    View this article via: PubMed CrossRef Google Scholar
26. Morel E, et al. cAMP-binding protein Epac induces cardiomyocyte hypertrophy. Circ Res. 2005;97(12):1296–1304.
    View this article via: PubMed CrossRef Google Scholar
27. Oestreich EA, et al. Epac-mediated activation of phospholipase Cε plays a critical role in β-adrenergic receptor-dependent enhancement of Ca²⁺ mobilization in cardiac myocytes. J Biol Chem. 2007;282(8):5488–5495.
    View this article via: PubMed CrossRef Google Scholar
28. Oestreich EA, et al. Epac and phospholipase Cε regulate Ca²⁺ release in the heart by activation of protein kinase Cε and calcium-calmodulin kinase II. J Biol Chem. 2009;284(3):1514–1522.
    View this article via: PubMed CrossRef Google Scholar
29. Smrcka AV, Oestreich EA, Blaxall BC, Dirksen RT. EPAC regulation of cardiac EC coupling. J Physiol. 2007;584(pt 3):1029–1031.
    View this article via: PubMed CrossRef Google Scholar
30. Pereira L, et al. Epac2 mediates cardiac β1-adrenergic-dependent sarcoplasmic reticulum Ca²⁺ leak and arrhythmia. Circulation. 2013;127(8):913–922.
    View this article via: PubMed CrossRef Google Scholar
31. Suzuki S, et al. Differential roles of Epac in regulating cell death in neuronal and myocardial cells. J Biol Chem. 2010;285(31):24248–24259.
    View this article via: PubMed CrossRef Google Scholar
32. Lindemann JL, Jones LR, Hathaway DR, Henry BG, Watanabe AM. β-adrenergic stimulation of phospholamban phosphoralation and Ca²⁺-ATPase activity in guinea pig ventricle. J Biol Chem. 1983;258(1):464–471.
    View this article via: PubMed Google Scholar
33. Vliem MJ, et al. 8-pCPT-2′-O-Me-cAMP-AM: an improved Epac-selective cAMP analogue. Chembiochem. 2008;9(13):2052–2054.
    View this article via: PubMed CrossRef Google Scholar
34. Brette F, Blandin E, Simard C, Guinamard R, Salle L. Epac activator critically regulates action potential duration by decreasing potassium current in rat adult ventricle. J Mol Cell Cardiol. 2013;57:96–105.
    View this article via: PubMed CrossRef Google Scholar
35. Chang CW, Chang GD, Chen H. A novel cyclic AMP/Epac1/CaMKI signaling cascade promotes GCM1 desumoylation and placental cell fusion. Mol Cell Biol. 2011;31(18):3820–3831.
    View this article via: PubMed CrossRef Google Scholar
36. Laxman S, Riechers A, Sadilek M, Schwede F, Beavo JA. Hydrolysis products of cAMP analogs cause transformation of Trypanosoma brucei from slender to stumpy-like forms. Proc Natl Acad Sci U S A. 2006;103(50):19194–19199.
    View this article via: PubMed CrossRef Google Scholar
37. Poppe H, et al. Cyclic nucleotide analogs as probes of signaling pathways. Nat Methods. 2008;5(4):277–278.
    View this article via: PubMed CrossRef Google Scholar
38. Hammond J, Balligand JL. Nitric oxide synthase and cyclic GMP signaling in cardiac myocytes: from contractility to remodeling. J Mol Cell Cardiol. 2012;52(2):330–340.
    View this article via: PubMed CrossRef Google Scholar
39. Movsesian MA, Nishikawa M, Adelstein RS. Phosphorylation of phospholamban by calcium-activated, phospholipid-dependent protein kinase. Stimulation of cardiac sarcoplasmic reticulum calcium uptake. J Biol Chem. 1984;259(13):8029–8032.
    View this article via: PubMed Google Scholar
40. Simmerman HK, Collins JH, Theibert JL, Wegener AD, Jones LR. Sequence analysis of phospholamban. J Biol Chem. 1986;261(28):13333–13341.
    View this article via: PubMed Google Scholar
41. Iwasa Y, Hosey MM. Phosphorylation of cardiac sarcolemma proteins by the calcium-activated phospholipid-dependent protein kinase. J Biol Chem. 1984;259(1):534–540.
    View this article via: PubMed Google Scholar
42. Borland G, Bird RJ, Palmer TM, Yarwood SJ. Activation of protein kinase Calpha by EPAC is required for the ERK- and CCAAT/Enhancer-binding protein β-dependent induction of the SOCS-3 gene by cyclic AMP in COS1 cells. J Biol Chem. 2009;284(26):17391–17403.
    View this article via: PubMed CrossRef Google Scholar
43. Braz JC, et al. PKC-α regulates cardiac contractility and propensity toward heart failure. Nat Med. 2004;10(3):248–254.
    View this article via: PubMed Google Scholar
44. Luo J, et al. A protocol for rapid generation of recombinant adenoviruses using the AdEasy system. Nat Protoc. 2007;2(5):1236–1247.
    View this article via: PubMed CrossRef Google Scholar
45. Kim S-J, et al. Differential regulation of inotropy and lusitropy in overexpressed Gsα myocytes through cAMP and Ca²⁺ channel pathways. J Clin Invest. 1999;103(7):1089–1097.
    View this article via: JCI PubMed CrossRef Google Scholar
46. Marx SO, Marks AR. Dysfunctional ryanodine receptors in the heart: new insights into complex cardiovascular diseases. J Mol Cell Cardiol. 2013;58:225–231.
    View this article via: PubMed CrossRef Google Scholar
47. Hittinger L, et al. Isoproterenol-induced alterations in myocardial blood flow, systolic and diastolic function in conscious dogs with heart failure. Circulation. 1989;80(3):658–668.
    View this article via: PubMed CrossRef Google Scholar
48. Sano M, et al. p53-induced inhibition of Hif-1 causes cardiac dysfunction during pressure overload. Nature. 2007;446(7134):444–448.
    View this article via: PubMed CrossRef Google Scholar
49. Pan Z, et al. MicroRNA-101 inhibited postinfarct cardiac fibrosis and improved left ventricular compliance via the FBJ osteosarcoma oncogene/transforming growth factor-β₁ pathway. Circulation. 2012;126(7):840–850.
    View this article via: PubMed CrossRef Google Scholar
50. Kim J, Eckhart AD, Eguchi S, Koch WJ. β-adrenergic receptor-mediated DNA synthesis in cardiac fibroblasts is dependent on transactivation of the epidermal growth factor receptor and subsequent activation of extracellular signal-regulated kinases. J Biol Chem. 2002;277(35):32116–32123.
    View this article via: PubMed CrossRef Google Scholar
51. Li N, et al. Inhibition of CaMKII phosphorylation of RyR2 prevents induction of atrial fibrillation in FKBP12.6 knockout mice. Circ Res. 2012;110(3):465–470.
    View this article via: PubMed CrossRef Google Scholar
52. Temple J, et al. Atrial fibrillation in KCNE1-null mice. Circ Res. 2005;97(1):62–69.
    View this article via: PubMed CrossRef Google Scholar
53. Chen YJ, et al. Effects of rapid atrial pacing on the arrhythmogenic activity of single cardiomyocytes from pulmonary veins: implication in initiation of atrial fibrillation. Circulation. 2001;104(23):2849–2854.
    View this article via: PubMed CrossRef Google Scholar
54. Chen YJ, Chen SA, Chang MS, Lin CI. Arrhythmogenic activity of cardiac muscle in pulmonary veins of the dog: implication for the genesis of atrial fibrillation. Cardiovasc Res. 2000;48(2):265–273.
    View this article via: PubMed CrossRef Google Scholar
55. Pereira L, et al. The cAMP binding protein Epac modulates Ca²⁺ sparks by a Ca²⁺/calmodulin kinase signalling pathway in rat cardiac myocytes. J Physiol. 2007;583(pt 2):685–694.
    View this article via: PubMed CrossRef Google Scholar
56. Chu G, Lester JW, Young KB, Luo W, Zhai J, Kranias EG. A single site (Ser¹⁶) phosphorylation in phospholamban is sufficient in mediating its maximal cardiac responses to β-agonists. J Biol Chem. 2000;275(49):38938–38943.
    View this article via: PubMed CrossRef Google Scholar
57. Brixius K, Wollmer A, Bolck B, Mehlhorn U, Schwinger RH. Ser¹⁶-, but not Thr¹⁷-phosphorylation of phospholamban influences frequency-dependent force generation in human myocardium. Pflugers Arch. 2003;447(2):150–157.
    View this article via: PubMed CrossRef Google Scholar
58. Ling H, et al. Requirement for Ca²⁺/calmodulin-dependent kinase II in the transition from pressure overload-induced cardiac hypertrophy to heart failure in mice. J Clin Invest. 2009;119(5):1230–1240.
    View this article via: JCI PubMed CrossRef Google Scholar
59. Engelhardt S, Hein L, Dyachenkow V, Kranias EG, Isenberg G, Lohse MJ. Altered calcium handling is critically involved in the cardiotoxic effects of chronic β-adrenergic stimulation. Circulation. 2004;109(9):1154–1160.
    View this article via: PubMed CrossRef Google Scholar
60. He H, et al. Effects of mutant and antisense RNA of phospholamban on SR Ca²⁺-ATPase activity and cardiac myocyte contractility. Circulation. 1999;100(9):974–980.
    View this article via: PubMed CrossRef Google Scholar
61. Eizema K, et al. Adenovirus-based phospholamban antisense expression as a novel approach to improve cardiac contractile dysfunction: comparison of a constitutive viral versus an endothelin-1-responsive cardiac promoter. Circulation. 2000;101(18):2193–2199.
    View this article via: PubMed CrossRef Google Scholar
62. Marks AR. Calcium cycling proteins and heart failure: mechanisms and therapeutics. J Clin Invest. 2013;123(1):46–52.
    View this article via: JCI PubMed CrossRef Google Scholar
63. Greiser M, Lederer WJ, Schotten U. Alterations of atrial Ca²⁺ handling as cause and consequence of atrial fibrillation. Cardiovasc Res. 2011;89(4):722–733.
    View this article via: PubMed CrossRef Google Scholar
64. Voigt N, et al. Enhanced sarcoplasmic reticulum Ca²⁺ leak and increased Na⁺-Ca²⁺ exchanger function underlie delayed afterdepolarizations in patients with chronic atrial fibrillation. Circulation. 2012;125(17):2059–2070.
    View this article via: PubMed CrossRef Google Scholar
65. Fuster V, et al. ACC/AHA/ESC Guidelines for the Management of Patients With Atrial Fibrillation: Executive Summary A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation) Developed in Collaboration With the North American Society of Pacing and Electrophysiology. Circulation. 2001;104(17):2118–2150.
    View this article via: PubMed Google Scholar
66. Edes I, Kranias EG. Phospholamban and troponin I are substrates for protein kinase C in vitro but not in intact beating guinea pig hearts. Circ Res. 1990;67(2):394–400.
    View this article via: PubMed CrossRef Google Scholar
67. Yamamura K, Steenbergen C, Murphy E. Protein kinase C and preconditioning: role of the sarcoplasmic reticulum. Am J Physiol Heart Circ Physiol. 2005;289(6):H2484–H490.
    View this article via: PubMed CrossRef Google Scholar
68. Grimm M, Brown JH. β-adrenergic receptor signaling in the heart: role of CaMKII. J Mol Cell Cardiol. 2010;48(2):322–330.
    View this article via: PubMed CrossRef Google Scholar
69. Metrich M, et al. Epac mediates beta-adrenergic receptor-induced cardiomyocyte hypertrophy. Circ Res. 2008;102(8):959–965.
    View this article via: PubMed CrossRef Google Scholar
70. Wang H, et al. Phospholipase C ε modulates β-adrenergic receptor-dependent cardiac contraction and inhibits cardiac hypertrophy. Circ Res. 2005;97(12):1305–1313.
    View this article via: PubMed CrossRef Google Scholar
71. Zhang L, et al. Phospholipase Cepsilon hydrolyzes perinuclear phosphatidylinositol 4-phosphate to regulate cardiac hypertrophy. Cell. 2013;153(1):216–227.
    View this article via: PubMed CrossRef Google Scholar
72. Klein G, et al. Increased collagen deposition and diastolic dysfunction but preserved myocardial hypertrophy after pressure overload in mice lacking PKCε. Circ Res. 2005;96(7):748–755.
    View this article via: PubMed CrossRef Google Scholar
73. Doetschman T. Influence of genetic background on genetically engineered mouse phenotypes. Methods Mol Biol. 2009;530:423–433.
    View this article via: PubMed Google Scholar
74. Yoshiki A, Moriwaki K. Mouse phenome research: implications of genetic background. ILAR J. 2006;47(2):94–102.
    View this article via: PubMed CrossRef Google Scholar
75. Tang T, et al. Adenylyl cyclase type V deletion increases basal left ventricular function and reduces left ventricular contractile responsiveness to β-adrenergic stimulation. Basic Res Cardiol. 2006;101(2):117–126.
    View this article via: PubMed CrossRef Google Scholar
76. Tadano M, et al. Congenital semilunar valvulogenesis defect in mice deficient in phospholipase C epsilon. Mol Cell Biol. 2005;25(6):2191–2199.
    View this article via: PubMed CrossRef Google Scholar
77. Engle SJ, Hoying JB, Boivin GP, Ormsby I, Gartside PS, Doetschman T. Transforming growth factor β1 suppresses nonmetastatic colon cancer at an early stage of tumorigenesis. Cancer Res. 1999;59(14):3379–3386.
    View this article via: PubMed Google Scholar
78. Engle SJ, et al. Elimination of colon cancer in germ-free transforming growth factor β 1-deficient mice. Cancer Res. 2002;62(22):6362–6366.
    View this article via: PubMed Google Scholar
79. Yagi T, et al. A novel ES cell line, TT2, with high germline-differentiating potency. Anal Biochem. 1993;214(1):70–76.
    View this article via: PubMed CrossRef Google Scholar
80. Ishikawa Y, et al. Downregulation of adenylylcyclase types V and VI mRNA levels in pacing-induced heart failure in dogs. J Clin Invest. 1994;93(5):2224–2229.
    View this article via: JCI PubMed Google Scholar
81. Iwatsubo K, et al. Direct inhibition of type 5 adenylyl cyclase prevents myocardial apoptosis without functional deterioration. J Biol Chem. 2004;279(39):40938–40945.
    View this article via: PubMed CrossRef Google Scholar
82. Sato M, et al. Activator of G protein signaling 8 (AGS8) is required for hypoxia-induced apoptosis of cardiomyocytes. J Biol Chem. 2009;284(45):31431–31440.
    View this article via: PubMed CrossRef Google Scholar
83. Yokoyama U, Patel HH, Lai NC, Aroonsakool N, Roth DM, Insel PA. The cyclic AMP effector Epac integrates pro- and anti-fibrotic signals. Proc Natl Acad Sci U S A. 2008;105(17):6386–6391.
    View this article via: PubMed CrossRef Google Scholar
84. LaRocca TJ, et al. β2-Adrenergic receptor signaling in the cardiac myocyte is modulated by interactions with CXCR4. J Cardiovasc Pharmacol. 2010;56(5):548–559.
    View this article via: PubMed CrossRef Google Scholar
85. Nishimaru K, Kobayashi M, Matsuda T, Tanaka Y, Tanaka H, Shigenobu K. α-Adrenoceptor stimulation-mediated negative inotropism and enhanced Na⁺/Ca²⁺ exchange in mouse ventricle. Am J Physiol Heart Circ Physiol. 2001;280(1):H132–H141.
    View this article via: PubMed Google Scholar
86. Adachi-Akahane S, Lu L, Li Z, Frank JS, Philipson KD, Morad M. Calcium signaling in transgenic mice overexpressing cardiac Na+-Ca2+ exchanger. J Gen Physiol. 1997;109(6):717–729.
    View this article via: PubMed CrossRef Google Scholar
87. Namekata I, et al. Intracellular mechanisms and receptor types for endothelin-1-induced positive and negative inotropy in mouse ventricular myocardium. Naunyn Schmiedebergs Arch Pharamacol. 2008;376(6):385–395.
    View this article via: PubMed Google Scholar
88. Tanaka H, et al. Unique excitation-contraction characteristics of mouse myocardium as revealed by SEA0400, a specific inhibitor of Na+-Ca2+ exchanger. Naunyn Schmiedebergs Arch Pharmacol. 2005;371(6):526–534.
    View this article via: PubMed CrossRef Google Scholar
89. Takamatsu H, Nagao T, Ichijo H, Adachi-Akahane S. L-type Ca2+ channels serve as a sensor of the SR Ca2+ for tuning the efficacy of Ca2+-induced Ca2+ release in rat ventricular myocytes. J Physiol. 2003;552(pt 2):415–424.
    View this article via: PubMed CrossRef Google Scholar
90. Adachi T, et al. The mechanism of increased postnatal heart rate and sinoatrial node pacemaker activity in mice. J Physiol Sci. 2013;63(2):133–146.
    View this article via: PubMed CrossRef Google Scholar
91. Adachi-Akahane S, Cleemann L, Morad M. Cross-signaling between L-type Ca²⁺ channels and ryanodine receptors in rat ventricular myocytes. J Gen Physiol. 1996;108(5):435–454.
    View this article via: PubMed CrossRef Google Scholar
92. Takeshita D, et al. A new calpain inhibitor protects left ventricular dysfunction induced by mild ischemia-reperfusion in in situ rat hearts. J Physiol Sci. 2013;63(2):113–123.
    View this article via: PubMed CrossRef Google Scholar
93. Namekata I, et al. Involvement of the Na+/Ca2+ exchanger in the automaticity of guinea-pig pulmonary vein myocardium as revealed by SEA0400. J Pharmacol Sci. 2009;110(1):111–116.
    View this article via: PubMed CrossRef Google Scholar
94. Takahara A, et al. Electrophysiological and pharmacological characteristics of triggered activity elicited in guinea-pig pulmonary vein myocardium. J Pharmacol Sci. 2011;115(2):176–181.
    View this article via: PubMed CrossRef Google Scholar