Epac1 inhibition as a novel cardioprotective strategy: lights and shadows on GRK5 canonical and non-canonical functions

β1-adrenergic receptors (β1AR) are members of the super-family of G protein-coupled receptors (GPCR) and major regulators of cardiac function. β1AR stimulation leads to G protein-mediated generation of the second messenger cyclic adenosine monophosphate (cAMP) and activation of effectors such as cAMP-dependent Protein Kinase A (PKA) or Exchange Proteins Activated by cAMP (Epac) (Figure 1). In addition, β1AR stimulation activates GPCR kinases (GRK) ‘canonical’ pathways,1 regulating β1AR phosphorylation, desensitization, down-regulation, degradation and recycling as well as activation of β-arrestin-dependent, G protein-independent signalling (Figure 1).2 GRK5 non-canonical activities in the nucleus have been also recently demonstrated, and new complexes/interactions between GRK5 and other proteins are being continuously identified.2 Epac1 and GRK5 canonical/non-canonical signalling. β1-adrenergic receptors (β1AR) ligand binding promotes (a) G protein-dependent stimulation of adenylyl cyclase (AC), cyclic adenosine monophosphate (cAMP) generation and Exchange Protein Activated by cAMP 1 (Epac1) activation. β1AR stimulation also activates GRK5 canonical signalling (b) and non-canonical activities in the nucleus (c), including modulation of MEF2, NFAT, and nuclear factor-κB (NFκB). Laudette et al. demonstrated that AM-001 inhibited non-canonical GRK5 effects on Histone Deacetylase type 5 (HDAC5) and MEF2 transcription (AM-001, red lines). However, after β1AR stimulation Epac1 also binds β-arrestin and is recruited at the plasma membrane and promotes transcription through nuclear factor of activated T cells (NFAT) in the nucleus. Whether Epac1 inhibition by AM-001 exerted effects on other GRK5 nuclear targets or GRK5 canonical signalling is still unknown and will deserve further investigation (Figure 1, AM-001?). Pathological cardiac overload produces profound abnormalities in β1AR signalling through several different mechanisms, affecting the transmission of cAMP signals from the plasma membrane to different intracellular locations, including the nucleus.3,4 Novel molecular insights into post-receptor mechanisms might lay the grounds for innovative therapeutic approaches acting beyond the plasma membrane that might have great scientific and clinical implications for cardiovascular diseases. Among downstream intracellular signals of β1AR, Epac1 represents an attractive new candidate. Genetically modified mouse models have demonstrated that Epac1 deletion prevents cardiac dysfunction in response to pressure overload or ischaemia/reperfusion.5,6 Laudette et al. report the identification and testing of a novel Epac1 non-competitive antagonist, AM-001, blocking Epac1 conformational changes induced by cAMP. This small molecule inhibited Epac1-dependent cardiomyocyte hypertrophy and death, reduced infarct size after myocardial ischaemia/reperfusion injury, and attenuated cardiac hypertrophy, inflammation and fibrosis induced by chronic βAR stimulation with isoprenaline (ISO). Although, as Laudette et al. conclude, this study represents a ‘proof-of-concept’ for the therapeutic effectiveness of Epac1 inhibitors in cardiovascular diseases, several questions remain unanswered and will require further investigation. The important role of Epac1 in other cell types will surely deserve further investigation in order to maximize the advantages of Epac1 antagonism and minimize the rate of possible unwanted side effects. Epac1 has been shown to regulate several aspects of fibrosis, including atrial fibroblast function and cardiac remodelling.7 In addition, Epac1 activation has been proposed as an effective therapeutic approach for inflammatory cardiac diseases, and it has been even proposed that Epac1 might be a candidate drug target in vascular endothelial cells due to its ability to attenuate proinflammatory cytokine signalling normally associated with cardiovascular diseases, including atherosclerosis.8 Although Laudette et al. did not find significant differences in fibrosis or inflammation in AM-001-treated mice undergoing ISO administration, these results should be confirmed in other model systems taking into account possible differences between atria and ventricles, and the risk of arrhythmias.7 Furthermore, Epac1 plays important roles in energy balance and glucose homeostasis, and adipose tissue-specific Epac1 knockout mice are more susceptible to high-fat diet-induced obesity and impaired glucose tolerance.9 Interestingly, Epac1 has been recently shown to modulate haemostasis and platelets function,10 and therefore possible additional anti-platelets effects of AM-001 or other Epac1 antagonists should be investigated. Similar to PKA, Epac1 biological actions strongly depend on its subcellular localization.3 In addition to nuclear targeting and regulation of gene expression, Epac1 has been also reported to act on the myofilament compartment, mitochondria, and plasma membrane.3 Thus, Epac1 antagonists might have other important translational implications in addition to cardiac hypertrophy and function. For example, treatment of both atrial and ventricular myocytes with the Epac1 inhibitor CE3F4 significantly inhibited sarcoplasmic reticulum Ca2+ release, and significantly prevented atrial fibrillation and ventricular arrhythmias in mice.11 It would be interesting to determine what is the relative relevance of specific compartmentalized Epac1 pools and how AM-001 affects them. Undoubtedly, the most novel and intriguing results in the article by Laudette et al. are related to the newly discovered relation between Epac 1 and GRK5. Here, the authors show that ISO increased Epac1-GRK5 interaction and induced GRK5 nuclear import, while AM-001 prevented GRK5 nuclear translocation, supporting the concept that Epac1 is required for non-canonical nuclear functions of GRK5 (Figure 1, red lines). In addition to regulating Histone Deacetylase type 5 (HDAC5) nuclear localization, GRK5 can affect nuclear activity through different mechanisms, including its capability to directly bind DNA. In addition, GRK5 can modulate expression and activity of Nuclear Factor-κB (NF-κB), even if underlying mechanisms and effects are controversial.12,13 Moreover, GRK5 can act as a facilitator of nuclear factor of activated T cells (NFAT), being part of a DNA binding complex responsible for pathological hypertrophic gene transcription.14 Whether Epac1/GRK5 interaction is also involved in these latter processes, and how these pathways were affected by AM-001 treatment remains obscure (Figure 1, question marks). It is conceivable that GRK5 might have additional unknown partners in the nucleus, potentially involved in DNA binding and activation of transcription. In order to design effective therapeutic strategies, it would be crucial to determine the relative importance of these non-canonical activities and focus on the most crucial partners of GRK5 to be targeted. In addition to the reported interaction between Epac1 and GRK5 in the nucleus, Epac1 has been shown to bind β-arrestin and to be recruited at the plasma membrane after β1AR stimulation.3 GRK-mediated β1AR phosphorylation resulting in β-arrestin recruitment might also stabilize Epac1 at the plasma membrane.15 Whether AM-001 might also affect canonical GRK5 functions and β-arrestin-mediated signalling was not investigated (Figure 1, question mark). This information would be particularly important, since GRK5 role in cardiac remodelling and function is still somewhat controversial, and while some studies suggest that GRK5 non-canonical nuclear functions may trigger pathological cardiac remodelling and function, other studies have proposed that canonical GRK5 activation of β-arrestin-mediated signalling might indeed be beneficial.1 In conclusion, results of this study are interesting and intriguing, opening new avenues of research in the field. Although many mechanistic issues still need to be addressed regarding the relations between Epac1 and GRK5, these proteins and related inhibitory compounds are expected to play a major role in future cardiovascular therapeutics. Conflict of interest: none declared. This work was supported, in part, by the Ministero dell'Istruzione, Università e Ricerca Scientifica (RBFR124FEN; 2015583WMX) to C.P. and was also carried out in the frame of Programma STAR, financially supported by the Federico II University (Unina) and Compagnia di San Paolo. The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.