Signalling in cardiac disease: the molecular deficit at the heart of the problem

See article by Yoshida et al. [3] (pages 34–45) in this issue. Events leading to heart failure are characterized by a prolonged action potential and a deficit in cardiac performance which is paralleled by ionic remodelling and a loss of contractile function at the level of the single cardiomyocyte (see Ref. [1] for review). Increased sympathetic stimulation as well as a number of paracrine and autocrine factors lead to cardiac hypertrophy and an eventual decompensated failing phenotype [2]. However, several issues remain to be resolved in terms of a molecular mechanism for these events. First, is there a deficit in the contractile apparatus of the failing myocyte per se? Alternatively, are signalling pathways which modulate contractility and are known to be altered in the failing heart the trigger for these events? Further, the focus of work on whole cardiac tissue to date has not resolved the role of non-myocyte cells in the disease process. Finally, the subcellular localization and compartmentalization of signalling pathways and the proteins they modulate may be of critical importance as well. The study by Yoshida et al. [3] in this issue of Cardiovascular Research begins to address some of these important issues. The authors demonstrate that although contractile function is clearly maintained in isolated cardiomyocytes from animals with coronary artery ligation-induced congestive heart failure (CHF), their regulation by the β-adrenergic system is significantly altered resolving a dilemma in the literature about the actual deficit in failing hearts. Further, they show alterations in levels of Gsα but not Giα subunits in cardiomyocytes, while in non-myocytes there are significant alterations in both subtypes. These results are important as they address the actual site of the molecular deficit in cardiac tissue. The regulation of [Ca2+]i plays a central role in cardiac performance. Ca2+-dependent signalling is critically important in the development of cardiac hypertrophy (see Refs. [4,5] for reviews). Changes in the expression or activities of Ca2+ handling proteins have been observed in both human dilated cardiomyopathy and experimental models of heart failure [6,7]. The sarcoplasmic reticulum (SR) contains important components of the Ca2+ handling machinery, which may be functionally altered in the disease state. In the rat infarct model, SR Ca2+ ATPase (SERCA) 2 activity is increased in the hyperfunctioning right ventricle during a compensatory stage, but depressed at the heart failure stage [8], and the activities of SERCA2 and of phospholamban, which modulates SERCA2 function, correlate with the steady state levels of their respective mRNAs [6]. The levels of the SR Ca2+ storage proteins, calreticulin and calsequestrin, are also changed in some models. Overexpression of calreticulin results in severe hypertrophy in young mice, indicating that calsequestrin is both a storage and a regulatory protein in myocardial Ca2+ homeostasis [9]. The density of the cardiac ryanodine receptor isoform (RyR2) is decreased in response to either increases in hemodynamic load [10] or phospholamban ablation [11]. A number of sarcolemmal Ca2+ handling proteins are also implicated in hypertrophy and heart failure, among them L-type Ca2+ channels and the Na+/Ca2+ exchanger. However, at the present time it remains unclear whether or not the levels of any or all of these proteins are altered in heart failure and to some extent depends on the actual model used [1]. Regulatory alterations in the function of various SR proteins may indeed be responsible for these functional changes. In the study of Yoshida et al. [3] it is clearly demonstrated that contractile function as measured by cell shortening remains intact in animals with coronary artery ligation-induced heart failure although there is a significant reduction in cardiac performance in the working heart. In isolated cardiomyocytes from sham-operated or infarcted animals, basal contractility, calcium transients and receptor-independent stimulation by direct activation of adenylyl cyclase or via an indirect ouabain-stimulated pathway were similar between the two groups while β-adrenergic receptor-stimulated responses (as measured with isoproterenol) were blunted in the failing hearts. These results argue that signalling pathways which control the cardiac contractility are altered in disease and that these alterations are responsible for reduced cardiac function, rather than a direct deficit in the contractile apparatus itself. In the failing heart it is known that both β1AR and β2AR responses are significantly desensitized owing to higher sympathetic drive [12,13]. β1AR density is reduced in animal models of heart disease and in tissue from human patients while β2AR density tends to remain intact [14,15]. These observations are supported in the study of Yoshida et al. [3] where a significant reduction in βAR-stimulated contractility in isolated myocytes was demonstrated. The role of various G proteins in heart disease has also been studied. Increases in relative levels of Gi proteins in ventricular tissue have been noted by a number of studies (see Refs. [12,16] for reviews). However, are these alterations in cardiomyocytes? This key question has been addressed by Yoshida et al. [3]. They observed no changes in levels (as assessed by semi-quantitative Western blotting) of Giα1, Giα2 or Giα3 proteins in cardiomyocytes isolated from viable right and left ventricle (RV, LV) or in the septum but significant increases in the general tissue levels of these proteins in the infarct scar and in the different regions of the heart. A concomitant increase in collagen levels in left ventricle and the infarct scar suggests a proliferation of fibroblasts in these regions. Alterations in Gi levels in non-myocytes may lead to alterations in release of paracrine or autocrine mediators which may have significant effects on cardiomyocyte contractility and/or hypertrophy. Levels of Gsα were slightly increased in the infarct scar, unchanged in viable LV but decreased in tissue RV and septum. Levels of Gsα were decreased in cardiomyocytes isolated from all three regions of viable cardiac tissue suggesting that they may play a role in the reduced contractile response to isoproterenol. However, it was not clear from this study whether or not modulation of contractility or calcium handling by direct G protein activation with GTP analogs was modified in the infarcted animals nor was it determined if levels of βAR differed from control animals. Thus it is difficult to conclude whether the alterations in βAR or Gs levels in cardiomyocytes are responsible for the observed contractile deficit. Basal levels of cAMP were decreased in isolated LV cardiomyocytes in the infarcted animals. However, no alterations in cAMP production by direct activation of adenylyl cyclase (AC) with colforsin daropate were seen as compared to control animals suggesting that adenylyl cyclase itself in the infarcted animals was intact. Other studies have demonstrated that forskolin-stimulated AC levels in tissue are reduced in the failing heart [17–19]. It remains to be seen whether AC activity in non-myocytes is altered. Also the specific alterations in signalling pathways in cardiomyocytes versus non-myocytes which play roles in ionic remodelling leading to action potential prolongation and ultimately myocyte hypertrophy and failure, need to be more clearly delineated. The paper by Yoshida et al. [3] sets the stage for further dissection of endogenous cardiomyocyte signalling pathways and paracrine factors released by other cell types in the heart. This work opens several lines of inquiry that will need to be pursued in the coming years. First, knowing that several autocrine and paracrine regulators can be released by cardiomyocytes and non-myocytes alike, it will be of critical importance to determine if release of these mediators (including endothelin, epidermal growth factor, cardiotrophin, etc.) is altered in heart failure and to what extent alterations in G protein-coupled signalling pathways are responsible. It is not clear what the specific roles of β1AR and β2AR and possibly other βAR subtypes play in this process. There is evidence that the β1AR and β2AR are coupled to distinct signalling pathways and much work remains to characterize their separate roles in the disease process [13,15]. Further studies are also needed to resolve the relative roles of receptor versus G protein in development of CHF. Are the responses desensitized in the failing heart because of a combination of receptor- and G protein-specific alterations? Are other signalling pathways coupled to G proteins including αAR, M2 and M3 muscarinic receptors (among others) also altered in a differential fashion between cardiomyocytes and other cell types in the heart under normal and disease conditions? Finally, as our appreciation of differences in the ultrastructural organization and subcellular distribution of signalling pathways increases, a number of issues relevant to heart failure are evolving. We are beginning to perceive the players in these pathways less like ships in the night who encounter each other transiently after receptor activation but rather as parts of more stable signalling complexes which are targetted to specific regions of the cell. Caveolae and lipid rafts may be sites for staging and assembly of these signalling complexes [12,20,21]. Scaffolding proteins and adaptors such as PDZ proteins and SH2/SH3 domain-containing proteins are also involved in targetting of signalling molecules [22]. Stable interactions between receptors and G proteins (and possibly the downstream effectors they modulate) may represent a molecular mechanism for signalling specificity in vivo [23–26]. Is there a coordinate downregulation of these signalling complexes in heart disease? Development of more efficacious and specific therapies for heart failure may depend on resolution of the organization of signalling pathways. Knowing which cells contain them and how they are altered in cardiac disease is an excellent place to start. T.E.H. is a MacDonald Scholar of the Heart and Stroke Foundation of Canada. This work was supported by grants from the Heart and Stroke Foundation of Québec and the Canadian Institutes for Health Research.