Cardiac anti-remodelling effects of phosphodiesterase type 5 inhibitors: afterload-(in)dependent?

Intracellular cyclic GMP (cGMP) is an important second messenger regulating many different cellular functions. In the cardiovascular system, cGMP and its primary effector cGMP-dependent protein kinase I (cGKI) play a pivotal role in diminishing smooth muscle vascular tone by cGKI-mediated phosphorylation of regulatory proteins involved in calcium handling and sarcomere shortening.1 cGMP is thereby critically involved in the regulation of systemic and pulmonary arterial blood pressure.1,2 In the heart, cGMP/cGKI signalling appears to exert more subtle effects. Increases in cGMP can counteract the trophic actions of growth hormones on cultured neonatal cardiomyocytes and cardiac fibroblasts.2 In genetic mouse models, cardiomyocyte-specific activation of cGMP synthesis suppressed cardiac remodelling in response to pressure overload,3 whereas inhibition exacerbated the response.4 These local, blood pressure-independent, cardiac anti-remodelling effects of cGMP were mild, but significant. cGMP levels in vascular smooth muscle and cardiac myocytes are regulated by the activities of three different guanylyl cyclases (GCs): soluble GC (sGC), the intracellular receptor for endothelial nitric oxide; particulate GC-A, a membrane-bound receptor for the cardiac natriuretic peptides atrial and B-type natriuretic peptides; and GC-B, a specific receptor for the endothelial C-type natriuretic peptide.2 cGMP formed is hydrolysed by members of the phosphodiesterase (PDE) family, mainly PDE5 and possibly PDE9, as well as by the dual-substrate, cGMP/cAMP-hydrolysing PDEs 1 and 2.5 PDE5 remains best characterized, and its inhibition by drugs such as sildenafil (Viagra®) is clinically used to treat erectile dysfunction and pulmonary hypertension.5 In particular, many small studies and a large randomized controlled (Super-1) trial have demonstrated the safety and efficacy of sildenafil in improving mean pulmonary artery pressure, pulmonary vascular resistance, cardiac index, and exercise tolerance in patients with pulmonary arterial hypertension.6 Despite the fact that it is expressed in cardiomyocytes, the role of PDE5 within the heart has long been considered minimal. PDE5 expression and activity in the heart are two orders of magnitude lower than in the lungs. However, more recent experimental studies indicated potent, direct cardiac anti-remodelling effects of selective PDE5 inhibitors, ameliorating ischaemia/reperfusion7 and effectively preventing and even reversing left ventricular (LV) hypertrophy, fibrosis, and dilatation in mice subjected to constant LV pressure overload induced by surgical aortic constriction.8 Based on these and other data, the National Institutes of Health initiated a multicentre trial (RELAX; http://clinicaltrials.gov/ct2/show/NCT00763867) to test the utility of sildenafil for treating heart failure with a preserved ejection fraction (diastolic heart failure). These patients commonly display systemic arterial hypertension and substantial ventricular hypertrophy. Schäfer et al.9 elegantly address the question whether PDE5 inhibition by sildenafil also exerts direct anti-remodelling effects on the right ventricular (RV) myocardium working against increased afterload. PDE5 expression seems to be low in normal human and rat RV, but is markedly upregulated in hypertrophied RV.10. It was previously shown that sildenafil prevents RV hypertrophy in rats with pulmonary hypertension induced by monocrotaline11 and reduces RV hypertrophy in patients with pulmonary arterial hypertension.12 However, in these studies, the relative importance of decreases in afterload provoked by the pulmonary artery vasodilating actions of sildenafil vs. a direct anti-hypertrophic effect of PDE5 inhibition was difficult to determine. Schäfer et al.9 directly compared the effects of sildenafil on RV remodelling in rats subjected either to monocrotaline-induced pulmonary hypertension or to surgical pulmonary artery banding. Importantly, both methods increased RV afterload to a similar extent and provoked similar severe RV hypertrophy and dysfunction. In the monocrotaline model, sildenafil decreased pulmonary pressure, reduced RV hypertrophy, and prevented fibrosis marker gene upregulation. In contrast, after pulmonary banding (where RV load is kept constant), sildenafil could not reverse, but actually increased RV hypertrophy and gene markers of myocardial remodelling. Similar observations in pulmonary artery-banded rats have been recently made by Andersen et al.13 Taken together, both studies indicate that sildenafil prevents RV myocardial remodelling in pulmonary hypertension through an indirect action via RV unloading and not through direct actions on RV myocytes. Even more, both studies indicated that sildenafil aggravates RV remodelling when afterload remains increased. The discrepant effects of sildenafil on pressure overload-induced LV vs. RV hypertrophy in various rodent models could be simply related to drug pharmacokinetics and different treatment conditions. Of note, the catabolism of sildenafil is ∼100-fold higher in rodents when compared with humans. However, Schäfer et al.9 carefully adjusted the dose to achieve sildenafil plasma concentrations above the in vitro IC50 for rat PDE5 inhibition. This treatment was accompanied by increases in extracellular (plasma) and RV cGMP levels, indicating inhibition of PDE5-mediated cGMP hydrolysis. Rather surprisingly, a recent paper of Nagayama et al.14 proposed that PDE5 inhibition to suppress LV hypertrophic responses is more effective when cardiac stresses are of a higher magnitude. In particular, in this study with mice, sildenafil failed to prevent a modest, well-compensated LV hypertrophy induced by ‘moderate’ transverse aortic constriction. However, sildenafil became highly effective when the hypertrophic stimulus was severe and was accompanied by pronounced activation of stress kinases (ERK1/2, calcineurin). The authors concluded that cGMP/cGKI is not a general inhibitor of cardiac remodelling, but suppresses selective cascades involved in severe, maladaptive LV hypertrophy. For instance, cascades coupled to the activation of Gαq-coupled receptors, which can specifically be inhibited by increased cardiac cGMP (and therefore by PDE5 inhibitors) via cGKI-mediated activation of regulator of G-protein signalling-2 (RGS2).15 However, this pressure overload magnitude dependence of the anti-hypertrophic efficacy of PDE5A inhibition is unlikely to explain the lack of effect of sildenafil on RV hypertrophy in pulmonary banded rats. In the studies by Schäfer et al.9 and Andersen et al.,13 the extent of RV pressure overload and hypertrophy was obviously severe. If differences in animal species, dosage, and extent of cardiac stress do not account for the apparent differences in the actions of sildenafil on LV and RV hypertrophy, then what does? One insufficiently explored possibility is that cardiomyocytes in the RV and LV respond to stress with activation of different signalling cascades and that the regulators and targets of cGMP/cGKI (and therefore the impact of PDE5 inhibition) are less prominent in the RV when compared with the LV. Of course, PDE5 inhibition can only enhance myocardial cGMP/cGKI activity if there is sufficient cGMP synthesis produced by the relevant cyclase, and it can only ameliorate ventricular hypertrophy if cGMP-targeted pathways are involved. Thus, the lack of efficacy of PDE5 inhibition to reverse RV hypertrophy in the presence of fixed pressure overload could reflect insufficient expression and/or activation of GCs (sGC, GC-A, GC-B) and insufficient cGMP production in cardiomyocytes of the RV, insufficient involvement of PDE5 in cGMP degradation, or inadequate involvement of cGMP-modulated hypertrophic pathways. The mechanisms involved in the development of RV hypertrophy, remodelling and failing of the RV, and the potential of pharmacological interventions are only sparsely understood. Much more is known about the cellular mechanisms involved in the development of hypertrophy and remodelling in the LV. Here, the cGMP cascade is one of the pathways playing a modulatory role.3,4,8 There are important differences in the gene expression, embryology, and physiology between the LV and RV.16,17 In animal models, inhibition of PDE5 raised the cGMP content in the LV15 and RV.9 Whereas this apparently directly protected the LV from severe pressure overload-induced hypertrophy,8,14,15 the RV indirectly benefited from cGMP-provoked pulmonary vasodilatation and pressure unloading.9 Are there differences in cGMP formation and signalling in myocytes from the left and right ventricles? Do PDE5 inhibitors exert different actions on both heart chambers? Fortunately, many questions are left for our future research! Conflict of interest: none declared.