Apoptotic cell death in heart failure

Time for primary review 29 days. Progressive deterioration of left ventricular (LV) function is a characteristic feature of the failing heart. This hemodynamic deterioration, often accompanied by worsening of clinical symptoms, occurs despite the absence of clinically apparent intercurrent adverse events [1,2] and, invariably, culminates in the syndrome of congestive heart failure. The exact mechanisms that drive this process are not known. For years, a so-called ‘vicious circle’ postulate was adopted to explain this phenomenon whereby compensatory mechanisms, elicited to maintain homeostasis, themselves become factors that exacerbate the heart failure state. Such compensatory mechanisms included LV hypertrophy, LV chamber dilation, and enhanced and sustained activity of the sympathetic nervous system and renin–angiotensin system [3–6]. In recent years, we and others have put forward a working hypothesis that progressive LV dysfunction in heart failure may result, in part, from ongoing loss of cardiomyocytes. The notion that progression of heart failure may be due, in part, to ongoing loss of functional cardiac units was based on the presence of ultrastructural degenerative changes of cardiomyocytes in the failed human heart as well as in hearts of animals with experimentally-induced heart failure [7–10]. Structural abnormalities included myofibrillar disruption and disarray [8], abnormalities of mitochondria characterized by disruption of the internal and external membranes, hyperplasia and reduced organelle size [7], and abnormalities of the cardiac interstitium characterized by accumulation of collagen [9]. These observations provided some support, albeit indirect, to the concept that ongoing myocyte degeneration and loss may occur in the failing heart. Objective evidence in support of this concept, however, was only recently put forward [11]. Studies in dogs with heart failure produced by intracoronary microembolizations showed, for the first time, that progressive LV dysfunction was associated with increased volume fraction of replacement fibrosis [11] or, in other words, increased proportion of scarred to viable myocardium suggesting ongoing myocyte loss. While these studies provided support to the concept of ongoing loss of viable myocardium in the failing heart, the means by which myocytes were lost was not identified. Even before these studies where completed, investigations in dogs with heart failure [12] and in explanted failed human hearts [13,14] showed that cardiomyocyte death through apoptosis occurs in advanced heart failure, a finding that supports the original working hypothesis. In this review, results of research in experimental animals as well as in humans on the identification of cardiomyocyte apoptosis in heart failure will be discussed. The discussion will also address potential molecular and pathophysiological triggers that may drive this process of cell death in the failing heart. Finally, an attempt will be made to address a central theme of whether cardiomyocyte apoptosis plays an important role in the progression of LV dysfunction that is characteristic of the heart failure state. Unlike necrosis, apoptosis is an active, precisely regulated, energy requiring process which appears to be orchestrated by a genetic program [15] and hence the interchangeable use of the terms ‘apoptosis’ and ‘programmed cell death’. Apoptosis plays a crucial role in the regulation of proliferating cell populations in adult tissues and in normal tissue development [16,17]. Cells such as neurons and cardiac myocytes, even though terminally differentiated contain the genes and signal transduction pathways necessary for programmed cell death and thus retain the ability to die by apoptosis [18]. In humans and other mammals, adult cardiac myocytes are thought to have, at best, a very limited capacity for self-renewal [19], and are intended to survive and actively function for the entire life of the organism. Viewed from this perspective, death of a significant number of adult cardiac muscle cells can have lasting adverse consequences on overall cardiac performance. Myocardial ischemia and infarction represent the major etiologies that underscore the development of congestive heart failure. Cardiomyocyte loss secondary to prolonged ischemia has long been thought to result from overt necrosis. While this form of cell death remains a primary cause of tissue injury, recent studies have suggested that cardiomyocyte loss after acute myocardial infarction can also be caused by apoptosis [20–25]. In studies in rats with myocardial infarction, internucleosomal DNA fragmentation, evidenced by DNA laddering on agarose gel electrophoresis, was detected as early as 3 h after coronary artery occlusion and was also present for up to 1 month after coronary ligation primarily in regions adjacent to infarcted tissue and to a lesser extent in myocardial regions remote from the infarction [22]. Cardiomyocyte apoptosis has also been observed in humans following acute myocardial infarction [23,24]. In hearts obtained from patients with acute myocardial infarction who died within 10 days of the onset of symptoms, Olivetti et al. [24] reported DNA strand breaks suggestive of apoptosis in 12% of myocytes in the infarct border zone and in 1% of constituent myocytes of myocardial regions remote from the infarction. In a study by Saraste et al. [23], hearts from patients who died of an acute myocardial infarction and had patent infarct-related arteries at autopsy, also showed extensive DNA strand breaks in cardiomyocytes. In this same study, apoptotic cardiomyocytes were observed primarily in myocardial regions that bordered the infarction [24]. As will be discussed in subsequent parts of this review, the observation of a high prevalence of cardiomyocyte apoptosis in the peri-infarct border region in comparison to myocardial regions remote from the infarction is also evident in myocardium of both humans with chronic heart failure secondary to ischemic cardiomyopathy as well as in animal models of chronic heart failure produced by intracoronary microembolizations [12,26]. Heart failure can result from sustained pressure overload as in long-standing hypertension or aortic valvular stenosis. Ventricular hypertrophy is associated with loss of cardiac myocytes that result in focal sites of replacement fibrosis traditionally attributed to necrosis [27]. Recent studies, however, have shown that experimentally-induced LV hypertrophy is associated with myocyte apoptosis [28–30]. In rats with LV hypertrophy produced by aortic banding, Teiger et al. [28] identified myocyte apoptosis during the first 7 days after instituting aortic banding. Studies from other laboratories suggested that cardiomyocyte apoptosis may be important in the transition from compensated hypertrophy to heart failure [29]. In spontaneously hypertensive rats (SHR) with symptoms of heart failure, Li et al. [29] showed a near five-fold increase in the number of cardiac myocytes undergoing apoptosis compared to nonfailing SHR rats. In this rat model, the transition to heart failure was accompanied by features characteristic of the heart failure state including cardiac pump dysfunction [31], myocardial fibrosis [32], and reduction in the volume fraction of cardiac myocytes [33]. The incidence of apoptotic myocyte nuclei in failed SHR was ∼40 cells per 100 000 nuclei compared to ∼8 per 100 000 nuclei in non-failed SHR rats. In age-matched WKY rats the incidence of apoptotic nuclei of myocyte origin was ∼2 per cells per 100 000 nuclei [29]. These results, while interesting, do not establish a cause and effect relationship between apoptosis and transition to heart failure. Left ventricular chamber enlargement is a characteristic adaptation of the failing heart regardless of etiology. Chronic ventricular enlargement and failure can result from long-standing volume overload as in aortic or mitral valve insufficiency or the development of large conduit vessel arterio-venous fistulas. As with ventricular hypertrophy, LV chamber dilation is associated with loss of cardiac myocytes that result in focal sites of fibrosis. Recent studies have shown that passive myocardial stretch is also associated with cardiomyocyte apoptosis [34]. In vitro studies by Cheng et al. [34] showed a 21-fold higher incidence of myocyte DNA strand breaks in rat posterior papillary muscles exposed to high tension levels as a result of overstretch compared to papillary muscles exposed to lower tension levels. DNA laddering studies using extracts from muscles exposed to high stretch also revealed the presence of DNA fragments consistent with apoptosis while degradation of DNA was not observed in non-overstretched papillary muscles [34]. These data suggest that myocardial stretch alone, as can occur under conditions of acute or chronic volume overload, may be associated with cardiomyocyte loss through apoptosis. In inflammatory heart muscle disease, autoimmunity is considered to play a role in the pathogenesis of impaired cardiac performance [35]. Marked depression of cardiac function occurs in patients with dilated cardiomyopathy in the absence of extensive loss of viable myocardium. In a subset of this population, the etiology and pathogenesis points to an inflammatory origin as in myocarditis. Under such circumstances, the immune response to the invading pathogen is the major determinant of the severity, course and progression of the disease. Secretory products of immune cells such as macrophages and other infiltrating cells could well contribute to abnormalities of contraction and relaxation that are seen, for instance, in myocarditis [35]. Proinflammatory cytokines such as tissue necrosis factor-α (TNF-α), interleukin (IL)-1, IL-2 and IL-6 are antigen-nonspecific glycoproteins that are synthesized rapidly and released locally by immune cells in response to injury [36]. Cytokines have been shown to reduce the positive inotropic response of isolated cardiac myocytes to adrenergic agonists [37]. TNF-α and IL-1 have also been shown to uncouple agonist-occupied receptors from adenylate cyclase in isolated cardiac myocytes [37]. Of particular importance to the present discussion is the potential role for proinflammatory cytokines, and in particular TNF-α, in the induction of cardiomyocyte apoptosis [38]. TNF-α is overexpressed in patients with heart failure regardless of etiology [39]. Plasma levels of TNF-α were shown to be elevated in patients with NYHA class I–III heart failure compared with age-matched controls and were progressively elevated in relation to decreasing functional status [40]. Transgenic mice with cardiac-specific overexpression of TNF-α have been shown to manifest lymphohistocytic myocarditis, cardiomegaly and congestive heart failure [41]. In this transgenic mouse, overexpression of TNF-α was associated with increased incidence of cardiomyocyte apoptosis shown by both DNA laddering and terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL) assay. The increased incidence of myocyte apoptosis in this model was also associated with activation of multiple members of the apoptosis pathway that included Fas and Bax [42]. The fact that TNF-α is a potent inducer of apoptosis is not surprising. Cardiomyocytes express functional TNF type-1 receptors and have been shown to undergo apoptosis after stimulation with TNF-α in vitro [38]. Clinical trials are currently underway to determine whether chronic inhibition of TNF-α is beneficial in the treatment of patients with heart failure. Even though the exact mechanisms that underly the process of progressive LV dysfunction in heart failure is not known, the possibility that this process results, in part, from ongoing loss of viable cardiomyocytes is no longer relegated to the realm of pure speculation. Studies in myocardium of animal models of experimentally-induced heart failure as well as studies in end-stage explanted failed human hearts have provided strong evidence for the existence of cardiac myocyte apoptosis. The very fact that myocyte apoptosis occurs in chronic heart failure, albeit at a pace yet to be defined and with a magnitude yet uncertain, supports the concept of ongoing loss of functional cardiac units. Cardiomyocyte apoptosis was identified in dogs with heart failure produced by multiple sequential intracoronary microembolization [12]; a model that manifests many of the classic sequelae of heart failure seen in humans, including marked and sustained depression of LV systolic function, LV hypertrophy and dilation and enhanced activity of the sympathetic nervous system [43]. In dogs with heart failure, the existence of cardiomyocyte apoptosis was established by uncovering ultrastructural features of this form of cell death as well as by immunohistochemical staining for nuclear DNA strand breaks using the TUNEL assay. Electron microscopic evidence of cardiomyocytes at various stages of apoptosis were identified in LV tissue obtained from dogs with heart failure but not in LV tissue obtained from normal dogs [12]. Features of early stages of apoptosis included compaction of nuclear chromatin that on the nuclear In these the were and the was Features of cardiomyocytes in advanced stages of apoptosis included an in the presence of with only of and that the of these as cardiomyocytes in in advanced stages of apoptosis also or with cell cardiomyocytes were primarily to LV regions were by large collagen and had an with absence of associated the a feature of apoptosis compared to necrosis [12]. In the same model, the number of cardiomyocytes undergoing apoptosis was based on an of the number of myocytes for nuclear DNA strand breaks per cardiomyocytes [11]. were made in LV regions or tissue as well as in LV regions remote from As in studies the number of cardiomyocytes undergoing apoptosis was higher in LV regions compared to LV regions remote from nuclear DNA events per This observation that in the failing heart, the peri-infarct border region may be a primary of myocyte loss through a finding consistent with the observation of ultrastructural abnormalities of cardiomyocyte in these regions Cardiomyocyte apoptosis was also in dogs with heart failure produced by ventricular DNA with of DNA to and was detected in myocardium of dogs with heart failure but not in dogs with this finding an of myocyte nuclei per was detected as undergoing apoptosis based on positive consistent with observations in the microembolization model of heart failure, of apoptotic myocytes were in of of replacement fibrosis is an for the development of congestive heart failure. the process in humans and in animals is also associated with a significant loss of cardiac myocytes is that the process may contribute to loss of of of LV myocytes This ongoing cell loss may be for the increased of development of ventricular dysfunction and failure in the studies have apoptosis as a process to the overall loss of cardiomyocyte in the heart et al. rats of from 3 to to cardiomyocyte necrosis and apoptosis. showed that in the LV the extent of both myocyte necrosis and apoptosis increased with necrosis, based on in labeling with in at 3 of and increased to at of Cardiomyocyte apoptosis was evident in 10 at 3 and increased to at of In this rat model, the progressive increase in apoptotic and cell death was associated with the development of ventricular dysfunction and failure which clinically apparent between and of finding of this study was the observation in to cell necrosis, which to early and between and of apoptosis to increase between and of life that triggers of both of cell death may be This between necrosis and apoptosis can also be on the of of of is that cell necrosis, which may have to sites of replacement fibrosis and apoptosis in viable myocardial regions that border the as seen in models of heart failure as well as in explanted failed human hearts The exact or mechanisms for apoptosis in peri-infarct regions is that myocardial regions that border are to a to apoptosis. Studies in tissues obtained from explanted hearts of patients with end-stage heart failure have the presence of cardiomyocyte apoptosis et al. tissue from explanted failed human hearts for evidence of cardiomyocyte apoptosis. of patients in heart failure was to dilated cardiomyopathy had immunohistochemical evidence of cardiomyocyte nuclear DNA by TUNEL and DNA laddering consistent with apoptosis. In of patients with ischemic cardiomyopathy had evidence of cardiomyocyte nuclear DNA by TUNEL and were positive for apoptosis based on DNA This and study, was by a and study by Olivetti et al. In tissue from normal Olivetti and reported TUNEL positive cardiomyocyte nuclei in only 10 In failed this increased from a of to a of was no in the number of apoptotic myocyte nuclei in patients with compared to patients with heart failure to In to the results of the study by et al. showed DNA laddering in myocardium of patients with and In a recent study, a higher incidence of cardiomyocyte apoptosis was reported in failed hearts of patients with compared to patients with apoptotic The higher number of apoptotic myocytes in was to a near higher incidence of apoptosis in regions compared to myocardial regions remote from apoptotic In patients with et al. reported a near increase in apoptosis of cardiomyocytes compared to that observed in myocardial tissue obtained from papillary muscle of patients undergoing mitral valve apoptosis is characterized by internucleosomal of DNA by and activity during apoptosis et al. showed that which is from is increased in myocardium of patients with end-stage heart failure compared to myocardium of The studies in human considered in suggest that apoptosis of cardiomyocytes occurs in heart failure regardless of the The of some of which such as apoptosis and others such as Bax which apoptosis is of the of the apoptotic process The of to the so-called is often as an of apoptosis. increase in this is to of the apoptotic process a in the is to of the apoptotic In viable myocardium of rats with myocardial infarction and ventricular failure, Cheng et al. reported a in the of and an increase in the of an that apoptosis. In myocardium of SHR with heart failure, of was compared to SHR rats [29]. In explanted failed human Olivetti et al. reported a near of the of in cardiac tissue changes in the of a that from apoptosis. in apoptosis is the in cell through of a The is to apoptosis in response to DNA and other such as increased of in a of cell In myocardium of SHR rats with heart failure, Li et al. [29] showed a significant increase levels in comparison to levels in myocardium of rats. of as recently to as have recently a and as primary of apoptosis. Studies in rats with acute myocardial infarction have suggested that apoptosis based on the ability of such as a to the apoptosis In a recent study, the of was in LV tissue obtained from failing human hearts In this study, was in myocytes recent and to a lesser extent in failing hearts to dilated of with apoptotic cardiomyocytes has been reported in rats following myocardial ischemia and The of or the of from mitochondria may be an important pathway for the activation of with apoptosis in the failing heart from mitochondria has been shown to activation in apoptotic cardiomyocytes during ischemia in the rat of appears to activation of the by of abnormalities have been in patients and in dogs with heart failure that disruption of the and reduced organelle size In myocardium of dogs with chronic heart failure we have also shown a marked in compared to normal dogs While requiring studies, the role of mitochondria in myocyte apoptosis in the failing heart could be central crucial in apoptosis of other cell cell a of the TNF is also in the regulation of apoptosis by as a for the which a that has become as the Recent studies have shown that levels of a that the of Fas and, is increased in patients with congestive heart failure In to these studies, other studies have reported increased levels of Fas in patients with congestive heart failure is a that between Fas and and apoptosis. role of mitochondria in apoptosis from can in apoptosis in of and the other of apoptosis to to which can and 3 can also to of various death that apoptosis. can with and can also the of and, in from apoptosis. cell cell progression in the of DNA and cell in terminally differentiated cells are potent of apoptosis hypertrophy and failure are associated with DNA in myocytes and with of molecular of cell progression increase in proliferating cell nuclear a nuclear necessary for DNA and cell progression was reported in myocardium of dogs with heart failure by ventricular of events that occur during the cell is also on a of as to be important for the progression from to through also of genes including and and the that the activity of the In a recent study, et al. showed a in the of in myocardium of patients with end-stage heart failure compared to normal a finding consistent with of cell Studies by and suggested that adult cardiac myocytes are to and that this capacity during cardiac including heart failure. the overall of such cell remains very and in the state at the present time, possibility is that cardiomyocytes to are apoptosis. for this can be in studies in which DNA in cardiomyocytes with in apoptosis cell The discussion only signal transduction pathways that are to cardiomyocyte apoptosis in the failing heart. treatment of the apoptotic pathways can be in many on the is evidence to suggest that pathophysiological to the heart failure are important of cardiac myocyte apoptosis is often suggested that apoptosis may be by the same that necrosis with the of cell death on the of the The observation that increased can to apoptosis is consistent with this factors as triggers of cardiomyocyte apoptosis the of to levels of levels of and increased levels of cytokines such as the role of and limited of the myocardium have in recent and for and sustained activity of the renin–angiotensin system and the sympathetic nervous system as well as or even are in many characteristic features of the failing heart and have long been in the progression of the disease. of isolated adult rat cardiomyocytes to was shown to cause a near five-fold increase in apoptosis cardiomyocyte were exposed to in the presence of a apoptosis was with this we observed an of cardiomyocyte apoptosis in dogs with heart failure with the inhibition has also been shown to apoptosis in rats with heart failure [29]. of isolated adult rat cardiomyocyte to caused a near increase in apoptosis myocytes were exposed to in the presence of the and the effect was with these in isolated rat myocytes, we observed a marked reduction of cardiomyocyte apoptosis in dogs with the The of myocyte apoptosis with was associated with an increase in the to Bax a finding that cell studies support for the role of as a potential inducer of cardiomyocyte apoptosis. of rat cardiac myocytes to was shown to apoptosis as evidenced by positive labeling for nuclear DNA In the same study, enhanced of Fas was also in response to have also been shown to be in the apoptosis in cardiac myocytes has also been suggested to increase the and nuclear accumulation of such as and that also have been in the induction of cell progression and apoptosis factors in cardiomyocyte apoptosis of the failing heart. renin–angiotensin tissue necrosis appears to be evidence based on studies in the end-stage failed human heart as well as studies in animal models of heart failure to support the concept that cardiomyocyte apoptosis occurs in heart failure. While the of this finding in the of the overall of this state and, in in relation to the progressive deterioration of LV function, remains is as yet cardiomyocyte apoptosis play an important role in the progression of heart While to this are at is to address this are that the human LV myocytes, a cardiomyocyte apoptosis remote from of myocyte for for nuclear DNA fragmentation, a of h for myocyte death to occur from and no cardiomyocyte can a near loss of LV cardiomyocytes per from apoptosis. some of the are and the proportion of cardiomyocyte loss to apoptosis the course of 1 is even in the of such a magnitude of loss of viable myocardium can have a significant adverse on LV performance. of the at which myocytes are lost in the failing heart as a result of apoptosis will many of the While studies be to this or even chronic studies in animal models of heart failure can be to determine inhibition of apoptosis with progressive LV after these have been will the importance of cardiomyocyte apoptosis in the of heart failure be in part, by a from the and