Hypertension, cardiotrophin-1 and gp130: three points to heart failure?

Heart failure is a chronic condition that is a leading cause of death in the developed western world and results from any structural or functional disorder that impairs contractile performance. The transition of the ventricular myocardium from a healthy, vital organ to a decompensated, failing one is complex and is frequently the result of multiple factors such as underlying hypertension, coronary arterial disease, acute injury (as with myocardial infarction), fibrosis, biomechanical stress, hypertrophy and cardiomyocyte apoptosis [1]. Efforts to understand the complex biological signaling underlying the development and evolution of heart failure have resulted in the identification of a multiplicity of circulating and tissue-based neurohumoral factors such as the renin–angiotensin system, aldosterone, endothelin, tumor necrosis factor-α, the natriuretic peptides and adrenomedullin [2]. All of these factors activate cellular receptors, which in turn stimulate intracellular cardiomyocyte kinase pathways such as Jak-STAT, p42/p44 mitogen-activated protein (MAPK) and phosphatidylinositol-3/protein kinase B (PI3K-Akt), to signal, in broad terms, either survival or destruction of the myocyte [3]. A relatively new player in this system is the gp130/gp190 [leukemia inhibitory factor (LIF)] receptor system, which is stimulated by the interleukin (IL)-6 family of cytokines [4]. The IL-6 family of circulating cytokines includes IL-6, IL-11, LIF, ciliary neurotrophic factor, oncostatin M and cardiotrophin-1. They have pleiotropic actions in hematopoiesis, inflammation, liver and neuronal regeneration, embryonic development and cardiovascular physiology [5]. Of these factors, cardiotrophin-1 is emerging as a potential pivotal player in linking hypertension, left ventricular hypertrophy (LVH) and heart failure. Plasma levels of cardiotrophin-1 are thus elevated in patients with hypertension [6,7] and heart failure [8], and cardiotrophin-1 has been shown to have important in-vitro hypertrophic and cardioprotective actions [9]. In vivo, the ventricular expression of cardiotrophin-1 in spontaneously hypertensive stroke-prone rats is specifically elevated immediately prior to the development of LVH, and it has been suggested that cardiotrophin-1 is the natural cardiac ligand for the gp130/LIF receptor [10]. Furthermore, myocardial cardiotrophin-1 gene expression is increased in experimental congestive heart failure [11], and both cardiotrophin-1 and gp130 gene expressions are increased in the left and right ventricles of rats after myocardial infarction [12]. The biological effects of cardiotrophin-1 are signaled through a tripartite receptor complex that is made up of the proteins gp130, gp190 (also known as LIF-R) and a third component approximately 80 kDa in size [13]. Of this complex, gp130 appears to be a critical component in protecting the myocardium from hypertrophic overload and modulating the transition from compensatory hypertrophy to heart failure. In mice with global genetic deletion of gp130, therefore, the heart is hypoplastic and abnormally developed [14] – whereas the ventricles in mice with cardiac-specific deletion of gp130 have a greater proportion of apoptotic myocytes and have a much more rapid onset to pressure overload-induced cardiomyopathy, without hypertrophic compensation, compared with control littermates [15]. Conversely, when gp130 is continuously activated by transgenic overexpression of IL-6 and its receptor, the result is compensatory ventricular hypertrophy [16], which one presumes may contribute to physiological and pathophysiological regulation of the myocardium [16]. These effects of gp130 are thought to be resolved intracellularly by the Jak2-STAT3 pathway, with interplay between it and the p42/44 MAPK and suppressor of cytokine signaling kinases [17]. The clinical importance of gp130 receptor signaling has been suggested by data from patients with end-stage heart failure due to ischemic and dilated cardiomyopathy [18]. In this study, the authors demonstrated that heart failure patients had 90% higher levels of ventricular gp130 gene expression, compared with control samples, but the gp130 protein abundance was actually 30% lower. In contrast, neither gp190/LIF receptor component, suppressor of cytokine signaling-3 protein nor phosphorylated STAT3/total STAT3 ratios were different from control samples, but ventricular cardiotrophin-1 gene expression and protein levels were markedly increased. These data suggested that differential changes in ventricular gp130, suppressor of cytokine signaling-3 and cardiotrophin-1 protein ratios are the result of a balancing act to prevent excessive gp130 signaling in heart failure. From this combined experimental and clinical data, key questions emerge: (1) how might clinical hypertension/LVH, a common substrate for heart failure, influence myocardial gp130?, (2) is cardiotrophin-1 signaling through gp130 likely to be beneficial or deleterious to the heart in patients with hypertension and LVH?, and (3) how are intracellular signaling factors such as p42/44 MAPK, Jak2-STAT3 and PI3K-Akt affected in concert with changes in gp130 in the above patient group? In the present issue of the journal, González and colleagues [19] report their findings from hypertensive patients with and without heart failure and provide important new information regarding the possible timing of changes in gp130 protein abundance and downstream intracellular signaling. González et al. took three endomyocardial biopsies from each of 28 nonheart failure patients with primary (essential) hypertension and 24 hypertensive patients who had previous diagnosis of heart failure based on the presence of one major and two minor Framingham criteria (objective evidence of impaired systolic and/or diastolic left ventricular function and elevated left ventricular end-diastolic and pulmonary capillary wedge pressures). All patients had LVH assessed by echocardiography. These biopsies were then interrogated by gene expression and western blot studies to determine the synthesis and activation states of gp130, gp190/LIF, cardiotrophin-1, p42/44 MAPK, STAT3 and apoptotic indices. These data were then assessed against clinical, hemodynamic and biochemical indices to determine statistical associations. In the heart failure group, gp130 protein expression and activation (phosphorylation status) of p42/44 MAPK and PI3K/Akt were all decreased compared with the nonheart failure group, and there was a significant positive correlation between gp130 protein levels and left ventricular ejection fraction in all hypertensive patients. Significantly increased myocyte apoptosis was observed in the heart failure group of patients, compared with the nonheart failure group, and this was inversely correlated with p42/44 and PI3K-Akt phosphorylation states and with left ventricular ejection fraction in all hypertensive patients. There was no difference in apparent gp130 gene expression between the groups, nor were there any differences in phosphorylated STAT3 abundance. At the same time, there was no difference in gp190/LIF gene expression/protein abundance between the two groups, suggesting a preferential degradation of gp130 in the heart failure group. In this respect, the data from González et al.[19] differ slightly from those of Zolk et al.[18], in that the latter group reported a 91% increase in gp130 mRNA abundance in their group of patients with end-stage heart failure. Overall, however, the common theme from both studies is that, in the myocardium of patients with heart failure: (1) gp130 protein abundance does not appear to keep pace with gene expression, suggesting either impaired protein translation or increased turnover/degradation; (2) ventricular gp190 gene expression and protein abundance appears to be unaffected; and (3) reductions in the phosphorylation (or activated) states of protective cellular proteins such as PI3k-Akt are significantly correlated with apoptotic indices. What role might cardiotrophin-1 have in this environment? González et al.[19] report that clinical plasma and myocardial concentrations of cardiotrophin-1 are inversely correlated with ventricular gp130 protein levels and that, in vitro, cardiotrophin-1 stimulates phosphorylation of gp130 at Ser782 and increases the ubiquitination of gp130 in HL-1 cardiomyocytes (a lineage derived from an AT-1 mouse atrial cardiomyocyte tumor that maintains cardiac phenotype through repeated passages). Given that plasma concentrations of cardiotrophin-1 are known to be elevated in patients with hypertension [6,7] and heart failure [8], and that cardiotrophin-1 can be released by ventricular wall stretch [7], it is conceivable that excessive cardiotrophin-1 stimulation of the ventricular myocardium drives gp130 ubiquitination, and is partially responsible for the decreases in total gp130 observed by Zolk et al.[18] and by Gonzalez et al.[19]. The findings of Gonzalez et al.[19] suggest new and important directions for future research. First, they bring forward the results from the earlier work of Zolk et al.[18] and suggest that clinically important degradation of gp130 protein may be present in hypertensive patients with heart failure, prior to end-stage disease. Second, STAT3 activation in hypertensive heart failure does not appear to be modulated by changes in gp130 status. Indeed, the hypertrophic effect of cardiotrophin-1 has been postulated to be mediated not by Jak2-STAT3, but by the MEK5-ERK5 pathway [20]. Coupled with the demonstration by Gonzalez et al.[19] that p42/44 MAPK and PI3K-Akt activation is downregulated in heart failure, a picture is emerging that perhaps the MAPK pathway, as opposed to Jak-STAT signaling, may be more important in allowing gp130-mediated compensatory ventricular hypertrophy to proceed in the overloaded heart. Third, future work should identify clearly whether cardiotrophin-1 is a causal agent or an innocent bystander in the transition from underlying hypertension-based ventricular hypertrophy to decompensated heart failure. The work of Gonzalez et al.[19] has several limitations: (1) The number of patients in the study was small, and there is the potential confounding factor of long-term medication use – both points acknowledged by the authors. (2) Gonzalez et al.[19] documented changes in activated/total gp130 status and increases in ubiquitination in mouse HL-1 cells only, but could not provide similar data from their clinical groupings. (3) biopsies were taken only from hypertensive patients with or without heart failure, there being no normotensive control group to provide ‘control’ data – while experimental data from animal models suggest that gp130 signaling and activation is downregulated in heart failure compared with control individuals [14,15], without data from healthy, normotensive control participants in the study of Gonzalez et al. the full clinical picture regarding ventricular gp130 activation remains elusive; and (4) the role of the 80 kDa ‘third component’ of the cardiotrophin-1 receptor complex [13] was not studied at all; indeed, practically nothing is known regarding the role this component may play in the pathophysiology of cardiotrophin-1 signaling in the cardiovascular system, despite the fact that it renders the gp130-gp190/LIF complex sensitive and specific to cardiotrophin-1 – clearly future studies may need to take this into account, especially given the dual hypertrophic and cardioprotective actions that cardiotrophin-1 appears to possess [9]. Despite these limitations, the work of Gonzalez et al.[19] provides important new clinical information regarding gp130 signaling, cardiotrophin-1 and hypertension in heart failure, and adds strong support for further investigations into how these factors may combine along the path to ventricular hypertrophy and heart failure. Acknowledgements The author would like to thank Prof. Mark Richards, Prof. Gary Nicholls and Prof. Tim Yandle for their helpful advice and comments during the preparation of this commentary.