Circulation Journal REVIEW Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Cyclic GMP-Dependent Signaling in Cardiac Myocytes Eiki Takimoto, MD, PhD

Cyclic GMP (cGMP) and its effector kinase PKG regulate diverse cellular functions. In cardiac myocytes, cGMP is produced by soluble and particulate guanylyl cyclases (GCs), the former stimulated by and the latter by natriuretic peptides, and is hydrolyzed to inactive 5′-GMP by cGMP-phosphodiesterases (PDEs). cGMP-PKG modulates cardiac contractility, hypertrophy and remodeling, and exerts cardioprotection. Although early research efforts have mostly focused on cGMP synthetic pathways, recent studies have revealed that cGMP degradation controlled by PDEs plays a critical role in the physiological action of cGMP. Among several cGMP-PDEs, cGMP- specific PDE5 has been intensively investigated. Studies in experimental animal models and humans consistently demonstrate benefits from PDE5 inhibitors in various cardiac pathologies. Several clinical trials are ongoing or planned to test the efficacy of PDE5 inhibitors in human heart disease, including a large multicenter clinical trial (RELAX) led by the NIH evaluating efficacy in heart failure with preserved ejection fraction. This review underscores the current understanding of cGMP-PKG signal regulation and its pathophysiological role in the heart, focusing on cardiac myocytes. (Circ J 2012; 76: 1819 – 1825)

Key Words: Cyclic GMP (cGMP); cGMP-dependent protein kinase; Myocytes; Phosphodiesterases

yclic GMP (cGMP) is the intracellular second messen- action of NO (NOS), and it can be released from ger that mediates the action of nitric oxide (NO) and such as glyceryl trinitrate ().5 C natriuretic peptides (NPs).1,2 It regulates a broad array There are 3 isozymes of NO , including neuronal NOS of physiologic processes in the cardiovascular system, includ- (nNOS or NOS1), inducible NOS (iNOS or NOS2) and endo- ing cardiac contractility, vascular tone, platelet function, and thelial NOS (eNOS or NOS3).6 eNOS and nNOS are expressed cardiac and vascular remodeling.1 NO activates soluble guany- constitutively and activated by Ca2+/, whereas iNOS lyl cyclase (GC) and NP activates membrane-bound GC, to is induced by (inflammatory) stimuli and its activity is Ca2+ produce cGMP.2 cGMP exerts its physiological action through independent. eNOS is highly expressed in endothelial cells, cGMP-dependent protein kinase (PKG), cGMP-regulated phos- whereas nNOS is abundant in neurons and . phodiesterases (PDE2, PDE3) and cGMP-gated cation chan- However, many tissues express all 3 isoforms, including car- nels, among which PKG might be the primary mediator. Impor- diac myocytes. NO activates soluble GC (sGC), a heterodi- tantly, the cGMP signal is compartmentalized within a cell so meric consisting of α- and β- subunits and a prosthetic that specific targeted proteins can be regulated by the same group with a ferrous iron, and produces cGMP.7 This “generic” cGMP to exert differential physiological effects.3 enzyme activity is critically affected by redox status: oxidation Key players in this regulation are the cyclic nucleotide degrad- of the heme moiety on the β-subunit renders the enzyme in- ing PDEs that are localized to specific subcellular space with sensitive to NO. Oxidative stress could compromise this cGMP confined activity.3 This review highlights the cardiac regulation synthetic pathway by other mechanisms as well. NO forms a of cGMP signaling and its role in cardiac pathophysiology, reactive oxidant peroxynitrite (ONOO−)8 in the presence of focusing on cardiac myocytes (Figure). superoxide (O2−), and NOS becomes dysfunctional (uncoupled) under oxidative stress to produce O2− rather than NO.9 cGMP Generation, Degradation, and NPs-pGC-cGMP Synthetic Pathway The NPs are a family of structurally related polypeptide hormones, including atrial Its Effector Kinase NP (ANP), B-type NP (BNP) and C-type NP (CNP).1,10 ANP and Generation Via 2 Distinct Pathways BNP are derived primarily from the atria and ventricles of the NO-sGC-cGMP Synthetic Pathway NO is a gaseous sig- heart, respectively, whereas CNP is secreted by endothelial naling molecule with a half-life of several seconds. Many key cells. NPs bind to membrane-associated GC receptors (pGC biological signals of NO are mediated by cGMP, while NO also or rGC) and produce cGMP. Among the 7 particulate GCs directly modifies protein function via covalent attachment of identified, GC-A (also called NP A receptor or NPR-A) and NO to reduced thiol groups, known as S-.4 NO is GC-B (also called NPR-B) are important in the cardiovascular endogenously synthesized from L- by the catalytic system. GC-A is the dominant form, expressed in various tis-

Received May 17, 2012; revised manuscript received June 7, 2012; accepted June 12, 2012; released online June 29, 2012 Division of Cardiology, Johns Hopkins University, Baltimore, MD, USA Mailing address: Eiki Takimoto, MD, PhD, FAHA, Division of Cardiology, Johns Hopkins University, 720 Rutland Avenue, Ross 850, Baltimore, MD 21205, USA. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-12-0664 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.76, August 2012 1820 TAKIMOTO E

Figure. Cardiac myocyte cGMP-PKG pathways. cGMP is synthesized by 2 pathways, NO-sGC and NP-pGC pathways, and is under compartmentalized regulation by PDEs. cGMP-PKG regulates cardiac function, hypertrophy/remodeling and cell survival response (see text for details). cGMP, cyclic GMP; NO, nitric oxide; NP, natriuretic peptide; PDE, phosphodiesterase; PKG, cGMP- dependent protein kinase. LTCC, L-type Ca2+ channel; TRPC, transient receptor potential canonical; GqPCR, Gq protein coupled receptor; Ang II, angiotensin II; PE, phenylephrine; ET1, endothelin 1; β3AR, β3 adrenergic receptor; cTnI, cardiac I; PLB, phospholamban; ○p , phosphorylation.

sues, including cardiac myocytes, and the principal receptor tivated by Ca+/calmodulin.15 There are 3 isoforms (PDE1A, for ANP and BNP.11 GC-B, a receptor for CNP, is mainly ex- PDE1B and PDE1C), all expressed in cardiac myocytes.16,17 pressed in the brain and vascular tissues, but some studies have PDE1A and PDE1B hydrolyze cGMP with higher affinity than suggested its presence and role in cardiac myocytes.12,13 NPs- cAMP, whereas PDE1C, the dominant isoform in human car- pGC-cGMP synthesis is compromised under disease condi- diac myocytes, hydrolyzes both with similar affinity.17 Although tions such as heart failure, because of the desensitization of PDE1 is a major esterase for both cAMP and cGMP in the GC-A.14 myocardium or cardiac myocytes in humans,16,18 its physio- logical effect, particularly in vivo, has not been fully clarified. Degradation by PDEs PDE2 PDE2 is a dual substrate esterase with similar affin- The PDEs are that hydrolyze cyclic nucleotides ity for cAMP and cGMP.15 PDE2 is expressed in cardiac myo- (cGMP, cAMP) to the inactive linear form (5′-GMP, 5′-AMP), cytes in various species, including humans,16 and localized at thereby regulating both the duration and the amplitude of cy- the plasma membrane and Z-line of cardiac myocytes.19 Due clic nucleotide signaling. PDEs comprise a 21- superfam- to the cGMP-sensitive GAF domain, the enzyme functions ily categorized into 11 families (PDE1–PDE11).15 PDEs 1, 2, mainly as a cGMP-stimulated cAMP esterase and thereby a 3, 4, 5, 8 and 9 are expressed in the heart (Table) and the roles cross-talk regulator between the 2 cyclic nucleotides.20 Through for PDEs 1, 2, 3 and 5 in cardiac cGMP regulation have been a cAMP modulating function, PDE2 regulates the β-adrener- investigated. PDE2 and PDE5 harbor allosteric cGMP-binding gic cardiac response in a subcellular compartment-selective sites, named GAF domains, located in the N-terminus portion. manner. cGMP-binding affinity to the GAF domain is much higher PDE3 PDE3 is also a dual substrate enzyme, known as a than to PDE catalytic sites, and cGMP binds to the GAF do- cGMP-inhibited cAMP esterase.15 Although both subtypes, main before cyclic nucleotide catalytic action to induce the PDE3A and PDE3B, exist in cardiac myocytes, PDE3A is the active state.15 main subtype and responsible for cardiac functional change PDE1 PDE1 is a dual substrate PDE and is potently ac- caused by PDE3 inhibition.21 PDE3 has similar affinity for

Circulation Journal Vol.76, August 2012 cGMP-Dependent Signaling in Cardiac Myocytes 1821

Table. PDEs in the Heart Km (μM) Heart PDE isoenzyme cAMP cGMP expression PDE1 Ca2+/CaM regulated, Dual PDE1A: 73–120, PDE1A: 2.6–5, Yes specificity PDE1B: 10–24, PDE1B: 1.2–5.9, PDE1C: 0.3–1.2 PDE1C: 0.6–2.2 PDE2 cGMP-stimulated, Dual 30–50 10–30 Yes specificity PDE3 cGMP-inhibited, Dual 0.02–0.15 0.18 Yes specificity PDE4 cAMP-specific 2.9–10 – Yes PDE5 cGMP-specific – 1–6.2 Yes PDE8 cAMP-specific 0.15 – Yes PDE9 cGMP-specific – 0.17–0.39 Yes PDEs, phosphodiesterases. cAMP and cGMP, but because of the higher catalytic rate for amban and .1 The role of PKG in the negative inotropy cAMP than for cGMP, cGMP acts as a competitive inhibitor induced by cGMP was demonstrated in a study of mice hearts against its cAMP catalytic action.2 lacking PKGI. Wegener et al reported that cGMP analogs (8- PDE5 PDE5A is a cGMP-specific esterase, abundantly ex- Br-cGMP, 8-pCPT-cGMP) have negative inotropic effects in pressed in vascular . Although PDE5A was not control hearts but not in hearts lacking PKGI,31 independent originally thought to be expressed in the heart, evidence has of cAMP signaling. accumulated that PDE5A protein is expressed in cardiac myo- Exogenous NO exerts a negative inotropic effect, which is cytes.1 The most recent study revealed that PDE5 represents mediated by cGMP-PKG-induced reduction in ~20% of total cGMP-esterase activity in human cardiac myo- calcium responsiveness. Layland et al32 demonstrated that a cytes,16 though there still remains some debate.18 PDE5 ex- NO donor induces negative inotropic effects and accelerates pression is upregulated in the diseased heart22–25 via an oxidant relaxation, without affecting the calcium transient, but with stress-dependent mechanism,26 and is activated by cGMP bind- increased phosphorylation, in rat ventricular myo- ing to the GAF domain and by PKG phosphorylation at Ser cytes. These effects were abrogated by soluble GC inhibitor 92.15 A PDE5 inhibitor such as sildenafil occupies its catalytic or PKG inhibitor. Endogenous NO has cGMP-dependent and site as a false substrate to increase cGMP and activate PKG, cGMP-independent effects, depending on the source.6 Studies leading to enhanced activity of the enzyme, and thus the inhib- have revealed that NO from eNOS is coupled to the cGMP- itor’s efficacy.2,15 PDE5 is localized to the Z-band of cardiac PKG signal pathway, regulating cardiac function,28,33 whereas myocytes under physiological conditions, but is diffusely dis- NO from nNOS directly modifies the ryanodine receptor (RyR) tributed in disease conditions.27,28 by S-nitrosylation and affects cardiac function.34 NO-cGMP derived from eNOS is functionally coupled to PDE5, modulat- cGMP Effector Kinase ing cardiac β-adrenergic stimulated contractility.28 PDE5 inhi- cGMP-Dependent Protein Kinase I (PKGI) cGMP-depen- bition suppresses β-adrenergic receptor-stimulated cardiac con- dent protein kinases (PKGs) are serine/threonine kinases that tractility in wild-type mice hearts/myocytes, an effect that is mediate many of the biological effects of cGMP. PKGI and abrogated in the presence of PKG inhibitor, but is absent in PKGII are encoded by 2 ,2 and PKGI is the primary hearts/myocytes lacking eNOS or β3 adrenergic receptor (β3AR) isotype in the cardiovascular system. PKGI is a homodimer of coupled to eNOS.33,35 This antiadrenergic effect might be at- identical subunits. Each subunit has 3 functional domains: (1) tributable to troponin I phosphorylation by PKG35 and/or inhi- an N-terminus domain containing a leucine/isoleucine zipper bition of the L-type Ca2+ channel (phosphorylation by PKG).33,36 and auto-inhibitory/pseudosubstrate region, (2) a regulatory The antiadrenergic response to PDE5 inhibitor is also observed domain with 2 allosteric cGMP biding sites, and (3) a catalytic in humans.37 Importantly, normal subcellular localization of domain. The N-terminus of PKGI is encoded by 2 alternative PDE5 to the Z-band is the key to this regulation. The anti- exons that produce 2 splice-variants, PKGIα and PKGIβ. This adrenergic effect of PDE5 inhibition is absent in failing canine N-terminal difference provides different cGMP sensitivity myocytes in which PDE5 is diffusely distributed.27 (PKGIα is more sensitive to cGMP than PKGIβ by 10-fold) Importantly, cGMP derived from pGC (GCA) by exoge- and protein targeting. PKGIα has a unique cysteine (Cys 43), nous ANP has no effect on cardiac systolic function or the β- the oxidative modification of which leads to the enzyme acti- adrenergic contractility response, suggesting the existence of vation independently of cGMP.29 This activation mechanism compartmentalized regulation.13,38,39 On the other hand, car- operates basally to control resistance vessel tone, but its role diac diastolic function is modified by the NP-pGC pathway in cardiac myocytes is unknown.30 similarly to the NO-sGC pathway. Exogenous BNP enhances the speed and extent of ventricular relaxation in normal and heart failure dogs.39 The PDE5 inhibitor, sildenafil, alone or cGMP Signal Regulation of Cardiac Myocytes in combination with BNP, increases left ventricular diastolic cGMP Regulation of Cardiac Contractility properties (distensibility) in both normal and old hypertensive cGMP negatively modulates contractility, accelerates relaxation dogs.40 This might be at least partially attributable to phosphory- and improves the stiffness of cardiac myocytes. These effects lation of titin, a sarcomeric “spring” that regulates myocardial might be mediated by direct PKG phosphorylation of proteins, passive stiffness.40 Reduction of titin-stiffness by cGMP-PKG including cardiac troponin I, L-type Ca2+ channel, phosphol- was also demonstrated in the human myocardium.41

Circulation Journal Vol.76, August 2012 1822 TAKIMOTO E

cGMP affects cAMP signaling via cross-talk regulation by lar, calcineurin, plays a major role in this process.50,51 Fiedler cGMP-regulated PDEs (PDE2 or PDE3), and modulates car- et al first demonstrated in cultured myocytes that PKG deacti- diac function. Using a real-time cyclic nucleotide imaging tech- vates calcineurin via inhibition of the L-type Ca2+ channel, nique, Mongillo et al reported that catecholamine stimulates blunting the cellular hypertrophy response.52 Recently, studies cGMP via β3AR coupled to eNOS, and activates PDE2-cAMP have implicated a role for transient receptor potential canonical hydrolysis, attenuating localized accumulation of cAMP.19 A (TRPC) 6, a receptor-operated Ca2+ channel, in this regula- more recent study has suggested that PDE2 activity is con- tion.53 PKG phosphorylates TRPC6 at Thr 69, 70 and Ser 322, fined to the PKA RII isoforms residing compartment that is and suppresses its conductance, resulting in inhibition of cal- functionally coupled to phospholamban or troponin I phos- cineurin-NFAT signaling. However, calcineurin inhibition is phorylation.42 cGMP-cAMP cross-talk involving PDE3 was not the sole mechanism, as mice lacking calcineurin Aβ can implicated in pathological right hearts. In a rat model of right still develop moderate cardiac hypertrophy in response to ventricular hypertrophy, acute PDE5 inhibition increased pressure overload that is inhibited by PDE5 inhibitor.54 Stud- myocardial cAMP (as well as cGMP) and cardiac contraction, ies have demonstrated a central role for Regulator of G protein associated with reduced PDE3 activity.22 A similar positive Signaling (RGS2 and RGS4) in the cGMP-mediated anti-hy- inotropic effect by PDE5 inhibition was observed in human pertrophy/remodeling mechanism.55,56 RGS proteins de-acti- failing right hearts.24 vate G protein-coupled signaling via GTPase activating prop- erties. PKGIα phosphorylates (activates) RGS2 and RGS4, and cGMP Regulation of Cardiac Hypertrophy and Remodeling terminates the Gq signal, inhibiting hypertrophy. We reported The heart develops hypertrophy in response to pathological that PKG activation by PDE5 inhibitor fails to de-activate stress.43 Although cardiac hypertrophy provides the initial func- multiple hypertrophy signaling cascades in hearts lacking tional compensation response to stress, it evolves into mal- RGS2, resulting in a lack of the antihypertrophic/remodeling adaptive remodeling/dilation with functional de-compensation, effect,55 and Tokudome et al reported that RGS4 overexpres- and can result in overt failure. cGMP signaling plays a “brake”- sion ameliorates cardiac spontaneous hypertrophy in NPRA like role (negative regulator) in the cardiac hypertrophy re- (GCA) null animals.56 Recent study suggested that PKG- modeling process. Gq-agonist-induced hypertrophy in cultured RGS2 is also involved in the ANP-GCA antihypertrophy.57 cardiac myocytes is suppressed by ANP, NO, cGMP or PKG.1 Lastly, it is worthwhile mentioning the differential response Although an early study using conventional PKGI null ani- between the right and left heart to PDE5 inhibitor in hypertro- mals with vascular re-introduction of PKGIβ failed to demon- phy/remodeling. Interestingly, PDE5 inhibitor fails to exert a strate a significant role for PKGI in cardiac hypertrophy or direct antihypertrophic effect on right heart hypertrophy in- remodeling, this could be partly because their pressure-load duced by surgically induced pressure overload.58 This differ- protocol was not severe enough to induce the pathological ence might be attributable to fundamental anatomical, bio- signal activation that cGMP-PKG targets.44 The 2 most recent logical and physiological differences between the right and studies clearly demonstrated the role of PKGI as a key nega- left sides of the heart. tive regulator of cardiac hypertrophy and remodeling.45,46 Mice with reduced myocyte PKG activity by cardiac myocyte Cardiac Protection by cGMP Signaling: Ischemic Injury and PDE5 overexpression develop exacerbated hypertrophy and Cardiomyopathy remodeling in response to pressure overload, and normalizing The cGMP-PKG pathway has been implicated in cardiac pro- PKG activity by turning off PDE5 overexpression ameliorates tection against cell death-inducing stress, such as ischemic the once established remodeling.45 Mice with cardiac-specific injury and doxorubicin toxicity.59,60 NO, BNP and 8Br-cGMP deletion of PKGI revealed worse remodeling to angiotensin II all ameliorate cell death in cultured cardiac myocytes exposed or pressure overload.46 Studies of animals with a modified to ischemia-reoxygenation,61 and mice lacking PKGI develop NP-cGMP pathway have shown consistent results.10,47 Mice a larger infarct than controls after ischemia-reperfusion.62 In lacking NPRA throughout the body show arterial hypertension vivo and ex vivo studies have revealed that NO donors limit with a disproportionate degree of cardiac hypertrophy, and infarct size and improve post-ischemic functional recovery. mice with cardiac-specific NPRA deletion develop exagger- Such beneficial effects of NO, however, were not consistently ated hypertrophy in response to pressure overload.10 Converse- observed, which might be attributable to impaired bioavail- ly, cardiac overexpression of constitutive active GC inhibits ability of NO because of reactive oxygen species generated pressure-overload hypertrophy.47 during early reperfusion.59 Inhibition of NOS during ischemic Most importantly, small molecule cGMP-PDE inhibitors preconditioning or postconditioning abolishes its cardiac pro- ameliorate hypertrophy and remodeling. We first reported that tective effects, suggesting the essential role of NOS-cGMP- the PDE5 inhibitor, sildenafil, activates myocardial PKG and PKG in the cardiac protection conferred by preconditioning or ameliorates cardiac hypertrophy/remodeling in a mouse model post-conditioning.59 On the other hand, NPs also limit the in- of pressure overload.25 Sildenafil efficacy in volume overload farct size from ischemia-reperfusion injury in animal models remodeling was recently demonstrated in a rat model of and in humans. Kitakaze et al reported that patients with acute chronic mitral valve regurgitation.48 Miller et al reported that myocardial infarction who were given carperitide (recombi- a PDE1 inhibitor (IC86340) has antihypertrophic effects in nant human ANP)63 as an adjunct to reperfusion therapy had cultured cardiac myocytes exposed to Gq-agonist and in chron- smaller infarcts, fewer reperfusion injuries and better clinical ic isoproterenol-induced hypertrophy in mice.49 PDE1 and outcomes than controls.64 Recent studies have revealed potent PDE5 inhibitors show additive antihypertrophic effects, sug- cardioprotection by PDE5 inhibitors. The infarct-limiting ef- gesting that these enzymes do not hydrolyze identical intracel- fect of PDE5 inhibitor in myocardial ischemia-reperfusion lular cGMP pools.49 injury has been reported in mouse, rat and rabbit hearts.60 Several mechanisms have been suggested for the antihyper- The underlying mechanisms are complex, involving mito- trophic/remodeling effects of the cGMP signals. The cardiac chondrial KATP channel (mitoKATP) regulation and stress respon- hypertrophic response involves activation of various intracel- sive (survival) signaling cascades.59,60 Ischemia-reperfusion lular signaling cascades. Gq-coupled signaling, and in particu- induces mitochondrial permeability transition (MPT) pore

Circulation Journal Vol.76, August 2012 cGMP-Dependent Signaling in Cardiac Myocytes 1823 formation that induces a permeability change of mitochondria, treatment improved cardiac function and exercise performance, leading to cell death. Ischemic preconditioning potently inhib- associated with reverse remodeling of left atrial volume index its this process, which is mediated by transient opening of mi- and LV mass index, in heart failure patients (New York Heart toKATP.65 Opening the mitoKATP partially compensates the mem- Association class II–III).75 Giannetta et al reported anti-remod- brane potential (ΔΨm), which enables additional protons to be eling effects and improved cardiac kinetics from 3 months silde- pumped out to form an H+ gradient for both ATP synthesis and nafil treatment in diabetic cardiomyopathy patients.76 These Ca2+ transport.65 Costa et al reported the mechanism that links results suggest that PDE5 inhibition is a promising new ap- PKG to mitoKATP and MPT: PKG increases opening of mito- proach to treating various human heart diseases. A large pla- KATP by PKCε via indirect activation.66 Rapid influx of +K cebo-controlled multicenter trial (RELAX, NCT00763867) increases the matrix pH and increases H2O2 production from led by NIH is ongoing to test sildenafil efficacy in heart failure complex I, to further activate PKCε coupled to MPT, resulting patients with preserved ejection fraction. A single center study in MPT inhibition. In addition to direct PKG mitochondrial is currently testing sildenafil efficacy in cardiomyopathy with regulation, other mechanisms might be at play in cGMP-PKG Duchenne and Becker muscular dystrophies (REVERS-DBMD, mediated cardioprotection. Phosphorylation of ERK and gly- NCT01168908). These studies will soon provide more infor- cogen synthase kinase 3β (GSK3β) and upregulation of anti- mation on the clinical benefit of PDE5 inhibitor treatment in apoptotic bcl2 were reported in ischemia re-perfused hearts human heart disease. On the other hand, BNP (nesiritide) was treated with PDE5 inhibitor.67 Although ERK activation lies introduced into clinical practice in 2001 for early relief of upstream of this regulation,68 mitochondrial effects might be dyspnea in acute heart failure patients, because of its vasodila- directly mediated by GSK3β inactivation (phosphorylation) and tory properties. However, recent evaluation in a large random- upregulated bcl2, the former inhibiting MPT pore formation69 ized trial showed minimal benefit and an increased risk of and the latter inhibiting mitochondrial apoptosis. Fiedler et al hypotension.77 The efficacy of combined use of tadalafil and reported PKG inhibition of pro-apoptotic p38 MAPK.62 Isch- nesiritide is being tested in subclinical heart failure patients, emia-reperfusion activates p38 via TAB1 (TAK1 binding pro- in whom the expected benefit is renal function improvement tein), independently of upstream kinases MKK3 or 6, inducing rather than a direct cardiac effect (NCT01544998). Consider- apoptosis. PKGIα physically associates with p38 MAPK and ing that the enzymes for cGMP synthesis, including NOS, sCG prevents TAB1 interaction, resulting in reduced p38 activation and GC-A, could be “uncoupled” from cGMP synthesis under and ischemia-reperfusion injury.62 Another protective mecha- chronic stress conditions, stimulation of sGC with a redox- nism involves sarcolemmal Na+/K+ ATPase regulation. Mad- insensitive activator and/or re-coupling of NOS with tetrahy- hani et al reported that sildenafil or PKG phosphorylates phos- drobiopterin (BH4) might have intriguing potential as a strat- pholemman and activates Na+/K+ ATPase, to limit Na+ overload egy to enhance cGMP signaling other than inhibiting cGMP and thus Ca2+ overload during reperfusion.70 degradation. The cardioprotective effects of cGMP signaling has been In summary, the cardiac cGMP system plays a key role in demonstrated in some types of cardiomyopathy. Doxorubicin protection against various cardiac pathologies, including hy- (Dox), an anticancer drug, has cardiac toxicity that ultimately pertrophy, heart failure, ischemic injury and cardiomyopathy. leads to cardiomyopathy. PDE5 inhibitor (sildenafil or tadala- PDE5 inhibitors, in particular, are a promising therapeutic fil) improved cardiac function and survival in a rodent model modality to enhance this signal, and efficacy has been demon- of Dox-cardiomyopathy by attenuating oxidative stress and strated in human heart failure and diabetic cardiomyopathy. apoptosis, which might be attributable to upregulation of bcl2 Ongoing and future large-scale clinical trials will determine and mitochondrial superoxide dismutase (MnSOD).60 A ge- whether the therapy will have long-term beneficial effects on netic mouse model of Duchenne muscular dystrophy lacking cardiac remodeling and a reduction in major cardiac events. (mdx) develops cardiomyopathy as do human mus- NP (nesiritide) failed to show benefit in acute decompensated cular dystrophy patients. Enhancing cGMP signal by cardiac heart failure, but its cardioprotective effect, particularly against overexpression of sGC or by PDE5 inhibitor, sildenafil, ame- ischemic injury, warrants further investigation.78 liorates the cardiac phenotype of mdx, including cardiac con- tractile performance, metabolic status and sarcolemmal integ- Acknowledgments 71 rity, via mechanisms that might involve inhibition of MPT This work was supported by National Institute of Health (RO1 093432) formation.72 PDE5 inhibitor is effective also in diabetic car- and American Heart Association (11GRNT7700071). diomyopathy.60 Tadalafil improves cardiac function in leptin- deficient mice, an animal model for type II diabetes, which is References associated with an altered proteomic profile of cytoskeletal 1. Tsai EJ, Kass DA. Cyclic GMP signaling in cardiovascular patho- rearrangement and redox regulation. physiology and therapeutics. Pharmacol Ther 2009; 122: 216 – 238. 2. Francis SH, Busch JL, Corbin JD, Sibley D. cGMP-dependent pro- tein kinases and cGMP phosphodiesterases in nitric oxide and cGMP Conclusion action. Pharmacol Rev 2010; 62: 525 – 563. 3. Fischmeister R, Castro LR, bi-Gerges A, Rochais F, Jurevicius J, cGMP Signal Modification as Therapy for Human Heart Leroy J, et al. Compartmentation of cyclic nucleotide signaling in the Disease heart: The role of cyclic nucleotide phosphodiesterases. Circ Res Accumulating evidence suggests that cGMP-PKG signaling 2006; 99: 816 – 828. plays a cardioprotective role against pathological stress. Among 4. Lima B, Forrester MT, Hess DT, Stamler JS. S-nitrosylation in car- diovascular signaling. Circ Res 2010; 106: 633 – 646. several strategies to enhance this signaling, PDE5 inhibition 5. Nakahira A, Minamiyama Y, Takemura S, Hirai H, Sasaki Y, Okada has been gaining great attention, with 3 inhibitors (sildenafil, S, et al. Coadministration of carvedilol attenuates tolerance vardenafil and tadalafil) already in clinical use to treat erectile by preventing cytochrome p450 depletion. Circ J 2010; 74: 1711 – dysfunction and pulmonary hypertension. Recent small human 1717. 6. Ziolo MT, Kohr MJ, Wang H. Nitric oxide signaling and the regula- studies revealed beneficial cardioprotective effects from chron- tion of myocardial function. J Mol Cell Cardiol 2008; 45: 625 – ic sildenafil treatment, while it also provides acute hemody- 632. namic benefits.73,74 Guazzi et al reported that 1-year sildenafil 7. Friebe A, Koesling D. The function of NO-sensitive guanylyl cy-

Circulation Journal Vol.76, August 2012 1824 TAKIMOTO E

clase: What we can learn from genetic mouse models. Nitric Oxide protein kinase G eliminates oxidant sensing to cause hypertension. 2009; 21: 149 – 156. Nat Med 2012; 18: 286 – 290. 8. Heo KS, Fujiwara K, Abe J. Disturbed-flow-mediated vascular reac- 31. Wegener JW, Nawrath H, Wolfsgruber W, Kuhbandner S, Werner C, tive oxygen species induce endothelial dysfunction. Circ J 2011; 75: Hofmann F, et al. cGMP-dependent protein kinase I mediates the 2722 – 2730. negative inotropic effect of cGMP in the murine myocardium. Circ 9. Takimoto E, Kass DA. Role of oxidative stress in cardiac hypertro- Res 2002; 90: 18 – 20. phy and remodeling. Hypertension 2007; 49: 241 – 248. 32. Layland J, Li JM, Shah AM. Role of cyclic GMP-dependent protein 10. Kishimoto I, Tokudome T, Nakao K, Kangawa K. Natriuretic pep- kinase in the contractile response to exogenous nitric oxide in rat tide system: An overview of studies using genetically engineered cardiac myocytes. J Physiol 2002; 540: 457 – 467. animal models. FEBS J 2011; 278: 1830 – 1841. 33. Wang H, Kohr MJ, Traynham CJ, Ziolo MT. Phosphodiesterase 5 11. Kuhn M. Structure, regulation, and function of mammalian mem- restricts NOS3/Soluble signaling to L-type Ca2+ brane guanylyl cyclase receptors, with a focus on guanylyl cyclase- current in cardiac myocytes. J MolCell Cardiol 2009; 47: 304 – 314. A. Circ Res 2003; 93: 700 – 709. 34. Wang H, Viatchenko-Karpinski S, Sun J, Gyorke I, Benkusky NA, 12. Kalra PR, Clague JR, Bolger AP, Anker SD, Poole-Wilson PA, Kohr MJ, et al. Regulation of myocyte contraction via neuronal nitric Struthers AD, et al. Myocardial production of C-type natriuretic pep- oxide synthase: Role of ryanodine receptor S-nitrosylation. J Physi- tide in chronic heart failure. Circulation 2003; 107: 571 – 573. ol 2010; 588: 2905 – 2917. 13. Wollert KC, Yurukova S, Kilic A, Begrow F, Fiedler B, Gambaryan 35. Lee DI, Vahebi S, Tocchetti CG, Barouch LA, Solaro RJ, Takimoto E, S, et al. Increased effects of C-type natriuretic peptide on contractil- et al. PDE5A suppression of acute beta-adrenergic activation requires ity and calcium regulation in murine hearts overexpressing cyclic modulation of myocyte beta-3 signaling coupled to PKG-mediated GMP-dependent protein kinase I. Br J Pharmacol 2003; 140: 1227 – troponin I phosphorylation. Basic Res Cardiol 2010; 105: 337 – 347. 1236. 36. Yang L, Liu G, Zakharov SI, Bellinger AM, Mongillo M, Marx SO. 14. Schlossmann J, Ammendola A, Ashman K, Zong X, Huber A, Protein kinase G phosphorylates Cav1.2 alpha1c and beta2 subunits. Neubauer G, et al. Regulation of intracellular calcium by a signalling Circ Res 2007; 101: 465 – 474. complex of IRAG, IP3 receptor and cGMP kinase Ibeta. Nature 37. Borlaug BA, Melenovsky V, Marhin T, Fitzgerald P, Kass DA. Silde- 2000; 404: 197 – 201. nafil inhibits beta-adrenergic-stimulated cardiac contractility in hu- 15. Francis SH, Blount MA, Corbin JD. Mammalian cyclic nucleotide mans. Circulation 2005; 112: 2642 – 2649. phosphodiesterases: Molecular mechanisms and physiological func- 38. Takimoto E, Belardi D, Tocchetti CG, Vahebi S, Cormaci G, Ketner tions. Physiol Rev 2011; 91: 651 – 690. EA, et al. Compartmentalization of cardiac beta-adrenergic inotropy 16. Johnson WB, Katugampola S, Able S, Napier C, Harding SE. Profil- modulation by phosphodiesterase type 5. Circulation 2007; 115: ing of cAMP and cGMP phosphodiesterases in isolated ventricular 2159 – 2167. cardiomyocytes from human hearts: Comparison with rat and guinea 39. Lainchbury JG, Burnett JC Jr, Meyer D, Redfield MM. Effects of pig. Life Sci 2012; 90: 328 – 336. natriuretic peptides on load and myocardial function in normal and 17. Vandeput F, Wolda SL, Krall J, Hambleton R, Uher L, McCaw KN, heart failure dogs. Am J Physiol Heart Circ Physiol 2000; 278: H33 – et al. Cyclic nucleotide phosphodiesterase PDE1C1 in human car- H40. diac myocytes. J Biol Chem 2007; 282: 32749 – 32757. 40. Bishu K, Hamdani N, Mohammed SF, Kruger M, Ohtani T, Ogut O, 18. Vandeput F, Krall J, Ockaili R, Salloum FN, Florio V, Corbin JD, et et al. Sildenafil and B-type natriuretic peptide acutely phosphorylate al. cGMP-hydrolytic activity and its inhibition by sildenafil in normal titin and improve diastolic distensibility in vivo. Circulation 2011; and failing human and mouse myocardium. J Pharmacol Exp Ther 124: 2882 – 2891. 2009; 330: 884 – 891. 41. Kruger M, Kotter S, Grutzner A, Lang P, Andresen C, Redfield MM, 19. Mongillo M, Tocchetti CG, Terrin A, Lissandron V, Cheung YF, et al. Protein kinase G modulates human myocardial passive stiffness Dostmann WR, et al. Compartmentalized phosphodiesterase-2 activ- by phosphorylation of the titin springs. Circ Res 2009; 104: 87 – 94. ity blunts beta-adrenergic cardiac inotropy via an NO/cGMP-depen- 42. Stangherlin A, Gesellchen F, Zoccarato A, Terrin A, Fields LA, dent pathway. Circ Res 2006; 98: 226 – 234. Berrera M, et al. cGMP signals modulate cAMP levels in a compart- 20. Zaccolo M, Movsesian MA. cAMP and cGMP signaling cross-talk: ment-specific manner to regulate catecholamine-dependent signaling Role of phosphodiesterases and implications for cardiac pathophys- in cardiac myocytes. Circ Res 2011; 108: 929 – 939. iology. Circ Res 2007; 100: 1569 – 1578. 43. Frey N, Olson EN. Cardiac hypertrophy: The good, the bad, and the 21. Sun B, Li H, Shakur Y, Hensley J, Hockman S, Kambayashi J, et al. ugly. Annu Rev Physiol 2003; 65: 45 – 79. Role of phosphodiesterase type 3A and 3B in regulating platelet and 44. Kass DA, Takimoto E. Regulation and role of myocyte cyclic GMP- cardiac function using subtype-selective knockout mice. Cell Signal dependent protein kinase-1. Proc Natl Acad Sci USA 2010; 107: 2007; 19: 1765 – 1771. E98; author reply E99. 22. Nagendran J, Archer SL, Soliman D, Gurtu V, Moudgil R, Haromy A, 45. Zhang M, Takimoto E, Hsu S, Lee DI, Nagayama T, Danner T, et al. et al. Phosphodiesterase type 5 is highly expressed in the hypertro- Myocardial remodeling is controlled by myocyte-targeted gene reg- phied human right ventricle, and acute inhibition of phosphodiester- ulation of phosphodiesterase type 5. J Am Coll Cardiol 2010; 56: ase type 5 improves contractility. Circulation 2007; 116: 238 – 248. 2021 – 2030. 23. Pokreisz P, Vandenwijngaert S, Bito V, Van den BA, Lenaerts I, 46. Frantz S, Klaiber M, Baba HA, Oberwinkler H, Volker K, Gabetaner Busch C, et al. Ventricular phosphodiesterase-5 expression is increased B, et al. Stress-dependent dilated cardiomyopathy in mice with car- in patients with advanced heart failure and contributes to adverse diomyocyte-restricted inactivation of cyclic GMP-dependent protein ventricular remodeling after myocardial infarction in mice. Circula- kinase I. Eur Heart J 2011 December 23 [Epub ahead of print]. tion 2009; 119: 408 – 416. 47. Zahabi A, Picard S, Fortin N, Reudelhuber TL, Deschepper CF. Ex- 24. Shan X, Quaile MP, Monk JK, French B, Cappola TP, Margulies KB. pression of constitutively active guanylate cyclase in cardiomyocytes Differential expression of PDE5 in failing and nonfailing human inhibits the hypertrophic effects of isoproterenol and aortic constric- myocardium. Circ Heart Fail 2012; 5: 79 – 86. tion on mouse hearts. J Biol Chem 2003; 278: 47694 – 47699. 25. Takimoto E, Champion HC, Li M, Belardi D, Ren S, Rodriguez ER, 48. Kim KH, Kim YJ, Ohn JH, Yang J, Lee SE, Lee SW, et al. Long- et al. Chronic inhibition of cyclic GMP phosphodiesterase 5A prevents term effects of sildenafil in a rat model of chronic mitral regurgita- and reverses cardiac hypertrophy. Nat Med 2005; 11: 214 – 222. tion: Benefits of ventricular remodeling and exercise capacity. Cir- 26. Lu Z, Xu X, Hu X, Lee S, Traverse JH, Zhu G, et al. Oxidative stress culation 2012; 125: 1390 – 1401. regulates left ventricular PDE5 expression in the failing heart. Cir- 49. Miller CL, Oikawa M, Cai Y, Wojtovich AP, Nagel DJ, Xu X, et al. culation 2010; 121: 1474 – 1483. Role of Ca2+/calmodulin-stimulated cyclic nucleotide phosphodies- 27. Senzaki H, Smith CJ, Juang GJ, Isoda T, Mayer SP, Ohler A, et al. terase 1 in mediating cardiomyocyte hypertrophy. Circ Res 2009; 105: Cardiac phosphodiesterase 5 (cGMP-specific) modulates beta-adren- 956 – 964. ergic signaling in vivo and is down-regulated in heart failure. FASEB 50. Dorn GW 2nd, Force T. Protein kinase cascades in the regulation of J 2001; 15: 1718 – 1726. cardiac hypertrophy. J Clin Invest 2005; 115: 527 – 537. 28. Takimoto E, Champion HC, Belardi D, Moslehi J, Mongillo M, Mergia 51. Fiedler B, Wollert KC. Interference of antihypertrophic molecules E, et al. cGMP catabolism by phosphodiesterase 5A regulates car- and signaling pathways with the Ca2+-calcineurin-NFAT cascade in diac adrenergic stimulation by NOS3-dependent mechanism. Circ cardiac myocytes. Cardiovasc Res 2004; 63: 450 – 457. Res 2005; 96: 100 – 109. 52. Fiedler B, Lohmann SM, Smolenski A, Linnemuller S, Pieske B, 29. Burgoyne JR, Madhani M, Cuello F, Charles RL, Brennan JP, Schroder Schroder F, et al. Inhibition of calcineurin-NFAT hypertrophy sig- E, et al. Cysteine redox sensor in PKGIa enables oxidant-induced naling by cGMP-dependent protein kinase type I in cardiac myo- activation. Science 2007; 317: 1393 – 1397. cytes. Proc Natl Acad Sci USA 2002; 99: 11363 – 11368. 30. Prysyazhna O, Rudyk O, Eaton P. Single atom substitution in mouse 53. Koitabashi N, Aiba T, Hesketh GG, Rowell J, Zhang M, Takimoto E,

Circulation Journal Vol.76, August 2012 cGMP-Dependent Signaling in Cardiac Myocytes 1825

et al. Cyclic GMP/PKG-dependent inhibition of TRPC6 channel ac- tosol to mitochondria. Circ Res 2005; 97: 329 – 336. tivity and expression negatively regulates cardiomyocyte NFAT ac- 67. Das A, Xi L, Kukreja RC. Protein kinase G-dependent cardioprotec- tivation Novel mechanism of cardiac stress modulation by PDE5 in- tive mechanism of phosphodiesterase-5 inhibition involves phos- hibition. J Mol Cell Cardiol 2010; 48: 713 – 724. phorylation of ERK and GSK3beta. J Biol Chem 2008; 283: 29572 – 54. Hsu S, Nagayama T, Koitabashi N, Zhang M, Zhou L, Bedja D, et al. 29585. Phosphodiesterase 5 inhibition blocks pressure overload-induced car- 68. Das A, Salloum FN, Xi L, Rao YJ, Kukreja RC. ERK phosphoryla- diac hypertrophy independent of the calcineurin pathway. Cardio- tion mediates sildenafil-induced myocardial protection against isch- vasc Res 2009; 81: 301 – 309. emia-reperfusion injury in mice. Am.J Physiol Heart Circ Physiol 55. Takimoto E, Koitabashi N, Hsu S, Ketner EA, Zhang M, Nagayama 2009; 296: H1236 – H1243. T, et al. Regulator of G protein signaling 2 mediates cardiac compen- 69. Sanada S, Komuro I, Kitakaze M. Pathophysiology of myocardial sation to pressure overload and antihypertrophic effects of PDE5 in- reperfusion injury: Preconditioning, postconditioning, and transla- hibition in mice. J Clin Invest 2009; 119: 408 – 420. tional aspects of protective measures. Am J Physiol Heart Circ Physi- 56. Tokudome T, Kishimoto I, Horio T, Arai Y, Schwenke DO, Hino J, ol 2011; 301: H1723 – H1741. et al. Regulator of G-protein signaling subtype 4 mediates antihyper- 70. Madhani M, Hall AR, Cuello F, Charles RL, Burgoyne JR, Fuller W, trophic effect of locally secreted natriuretic peptides in the heart. et al. Phospholemman Ser69 phosphorylation contributes to sildena- Circulation 2008; 117: 2329 – 2339. fil-induced cardioprotection against reperfusion injury.Am J Physiol 57. Klaiber M, Kruse M, Volker K, Schroter J, Feil R, Freichel M, et al. Heart Circ Physiol 2010; 299: H827 – H836. Novel insights into the mechanisms mediating the local antihyper- 71. Khairallah M, Khairallah RJ, Young ME, Allen BG, Gillis MA, trophic effects of cardiac atrial natriuretic peptide: Role of cGMP- Danialou G, et al. Sildenafil and cardiomyocyte-specific cGMP sig- dependent protein kinase and RGS2. Basic Res Cardiol 2010; 105: naling prevent cardiomyopathic changes associated with dystrophin 583 – 595. deficiency.Proc Natl Acad Sci USA 2008; 105: 7028 – 7033. 58. Schafer S, Ellinghaus P, Janssen W, Kramer F, Lustig K, Milting H, 72. Ascah A, Khairallah M, Daussin F, Bourcier-Lucas C, Godin R, Allen et al. Chronic inhibition of phosphodiesterase 5 does not prevent pres- BG, et al. Stress-induced opening of the permeability transition pore sure-overload-induced right-ventricular remodelling. Cardiovasc Res in the dystrophin-deficient heart is attenuated by acute treatment with 2009; 82: 30 – 39. sildenafil.Am J Physiol Heart Circ Physiol 2011; 300: H144 – H153. 59. Burley DS, Ferdinandy P, Baxter GF. Cyclic GMP and protein ki- 73. Lewis GD, Lachmann J, Camuso J, Lepore JJ, Shin J, Martinovic nase-G in myocardial ischaemia-reperfusion: Opportunities and ob- ME, et al. Sildenafil improves exercise hemodynamics and oxygen stacles for survival signaling. Br J Pharmacol 2007; 152: 855 – 869. uptake in patients with systolic heart failure. Circulation 2007; 115: 60. Kukreja RC, Salloum FN, Das A, Koka S, Ockaili RA, Xi L. Emerg- 59 – 66. ing new uses of phosphodiesterase-5 inhibitors in cardiovascular dis- 74. Lindman BR, Zajarias A, Madrazo JA, Shah J, Gage BF, Novak E, eases. Exp Clin Cardiol 2011; 16: e30 – e35. et al. Effects of phosphodiesterase type 5 inhibition on systemic and 61. Gorbe A, Giricz Z, Szunyog A, Csont T, Burley DS, Baxter GF, et al. pulmonary hemodynamics and ventricular function in patients with Role of cGMP-PKG signaling in the protection of neonatal rat car- severe symptomatic aortic stenosis. Circulation 2012; 125: 2353 – diac myocytes subjected to simulated ischemia/reoxygenation. Basic 2362. Res Cardiol 2010; 105: 643 – 650. 75. Guazzi M, Vicenzi M, Arena R, Guazzi MD. PDE5 inhibition with 62. Fiedler B, Feil R, Hofmann F, Willenbockel C, Drexler H, Smolenski sildenafil improves left ventricular diastolic function, cardiac geom- A, et al. cGMP-dependent protein kinase type I inhibits TAB1-p38 etry, and clinical status in patients with stable systolic heart failure: mitogen-activated protein kinase apoptosis signaling in cardiac myo- Results of a 1-year, prospective, randomized, placebo-controlled cytes. J Biol Chem 2006; 281: 32831 – 32840. study. Circ Heart Fail 2011; 4: 8 – 17. 63. Takahashi R, Uchiyama A, Iguchi N, Mashimo T, Fujino Y. Effects 76. Giannetta E, Isidori AM, Galea N, Carbone I, Mandosi E, Vizza CD, of continuous venovenous hemofiltration on the pharmacology of et al. Chronic inhibition of cyclic GMP phosphodiesterase 5a im- carperitide, a recombinant human atrial natriuretic peptide. Circ J proves diabetic cardiomyopathy: A randomized, controlled clinical 2010; 74: 1888 – 1894. trial using magnetic resonance imaging with myocardial tagging. 64. Kitakaze M, Asakura M, Kim J, Shintani Y, Asanuma H, Hamasaki T, Circulation 2012; 125: 2323 – 2333. et al. Human atrial natriuretic peptide and as adjuncts to 77. O’Connor CM, Starling RC, Hernandez AF, Armstrong PW, Dickstein reperfusion treatment for acute myocardial infarction (J-WIND): K, Hasselblad V, et al. Effect of nesiritide in patients with acute de- Two randomised trials. Lancet 2007; 370: 1483 – 1493. compensated heart failure. N Engl J Med 2011; 365: 32 – 43. 65. Ardehali H, O’Rourke B. Mitochondrial K(ATP) channels in cell 78. Sezai A, Wakui S, Akiyama K, Hata M, Yoshitake I, Unosawa S, et al. survival and death. J Mol Cell Cardiol 2005; 39: 7 – 16. Myocardial protective effect of human atrial natriuretic peptide in 66. Costa AD, Garlid KD, West IC, Lincoln TM, Downey JM, Cohen MV, cardiac surgery: “hANP shot” in clinical safety trial. Circ J 2011; 75: et al. Protein kinase G transmits the cardioprotective signal from cy- 2144 – 2150.

Circulation Journal Vol.76, August 2012