Cell death signalling mechanisms in heart failure

In 2003, cardiovascular disease was the most costly disease in Canada, and it is still on the rise. The loss of properly functioning cardiomyocytes leads to cardiac impairment, which is a consequence of heart failure. Therefore, understanding the pathways of cell death (necrosis and apoptosis) has potential implications for the development of therapeutic strategies. In addition, the role of B-cell lymphoma-2 family members is discussed and the importance of mitochondria in directing cell death mechanisms.

Abstract

Cardiac disease is a global epidemic that is on the rise, despite the recent advances in cardiovascular research. Once the myocardium is injured, it has a limited capacity to activate reparative mechanisms to restore proper cardiac function, leading to the development of systemic heart failure. Autophagy, under certain conditions, may result in cell death, further emphasizing the controversial issues regarding the autophagic process as an adaptive or maladaptive biological response. Although significant progress in understanding the signalling mechanisms of cell death in myocytes has been made, the role of apoptotic cell death and programmed necrosis during heart failure is not completely understood. Insight to how myocytes determine whether to activate apoptotic or programmed necrosis signalling machinery remains under current investigation because it is a major problem for both scientists and clinicians in treating heart failure patients. Herein, the different modes of cell death implicated in heart failure are highlighted, as well as the role of B-cell lymphoma-2 family members and how mitochondria act as central organelles in directing such cell death mechanisms.

Cardiovascular disease is on the rise, and is a major concern to public health in terms of treatment costs and management. In 2003, cardiovascular disease was the most costly disease in Canada, responsible for 33% of all deaths (1). The loss of functioning cardiomyocytes results in cardiac impairment, and is a detrimental consequence of heart failure. Understanding how cardiac cells die and the underlying molecular mechanisms thereof have great potential to bring about the development of therapeutic strategies to treat this pervasive epidemic. According to the American Heart Association, cardiovascular disease is expected to increase in prevalence over the next 10 years due to an aging population (2), emphasizing the importance of improving cardiovascular health through research and the potential benefits preventive therapy may have for the growing number of heart failure patients (3,4).

Cardiac cell fate can be influenced by at least two recognized pathways of cell death that are ‘programmed’: necrosis and apoptosis – both of which result in the death of cardiomyocytes. Autophagy is a cellular process that provides a mechanism for cells to survive under conditions of extreme stress, but contributes to cell death if overactivated. However, no pathway that underlies autophagic cell death is known to exist. Because these processes are defined by morphological characteristics and signalling pathways, there is evidence of overlap between the two modes of programmed cell death, which contribute to the loss of myocytes. Identifying the underlying mechanisms that determine how a cell may commit to a specific cell death pathway is ongoing.

AUTOPHAGY

Autophagy is defined as a process in which intracellular organelles and proteins are degraded by lysosomal enzymatic action, and is the cell’s natural way of recycling and removing unwanted material to maintain homeostasis. Cells experience basal autophagy, but nutrient shortage during cellular stress further activates autophagy to sustain increased metabolic requirements. Excess or deregulated activation of autophagy has been implicated as a trigger of cell death (5–7) and in heart failure (8). Irregular autophagic activity has been documented as a contributor to many diseases, including neurodegenerative disorders, aging, cancer, diabetes and cardiovascular disease, as reviewed by Essick and Sam (9), and Beau et al (10).

Over the past 30 years, the role of defective lysosomal machinery has been documented in many diseases (11). This is most evident in patients with Danon disease, who develop cardiomyopathy due to a deficiency in lysosomal-associated membrane protein-2 (12,13). Mice deficient in lysosomal-associated membrane protein-2 displayed large aggregates of autophagosomes in isolated cardiac cells (14) and developed contractile impairment (15). Maladaptive autophagy in the myocardium has been shown to be detrimental in various disease pathologies including pressure overload-induced cardiac hypertrophy (16,17), oxidative stress (18) and myocardial ischemia/reperfusion (19,20) – all of which contribute to the progression of heart failure.

The process of autophagy is regulated by autophagy-related genes (Atg) (21) that have been evolutionarily conserved from yeast to mammals (22) and identified as important regulators during the four stages of autophagy: induction, elongation and autophagosome formation, docking/fusion and lysosomal degradation (23,24). The initial stage consists of Atg proteins that are responsible for phagophore elongation and recruitment of microtubule light chain-3 (LC3). Proteolytic cleavage of LC3 produces LC3-I and a lipid conjugation system converts LC3-I to an autophagic membrane-associated form, LC3-II (25). This forms an invaginated structure called an autophagosome, a double-membrane structure consisting of engulfed material. Disassembly of Atg proteins allows fusion of the autophagosome with a lysosome, resulting in lysosomal hydrolyase degradation (26).

The process of autophagy is under tight regulation, and involves molecular repressors and activators. When the activated form of the mammalian target of rapamycin, a serine/threonine kinase, interferes with autophagosome formation and inhibits autophagy (27,28), induction of autophagy occurs by activation of a protein complex comprised of class III phophatidylinositol-3 kinase and Beclin-1 (19,29). This complex induces autophagy by recruiting autophagic proteins. However, the activity of Beclin-1 is known to be suppressed when bound to Bcl-2, which is an antiapoptotic factor that reduces the ability of Beclin-1 to induce autophagy (30,31). Disruption of the Beclin-1/Bcl-2 complex restores induction of autophagy (32). Beclin-1, in contrast, cannot inhibit the antiapoptotic function of Bcl-2 (33). Recent studies have identified inactivation of autophagy by caspase-mediated cleavage of Beclin-1 (34,35), in which caspases are key players in apoptotic cell death (which will be discussed in a later section). Such interaction involving Beclin-1 with key regulators of apoptosis promotes crosstalk between autophagy and cell death. Although the mechanisms underlying cardiac autophagy are not well understood, specific cellular conditions resulting in deregulated autophagy that cannot maintain metabolic requirements of the cell may lead to activation of apoptotic cell death.

APOPTOSIS

In a healthy organism, programmed cell death is an integral part of the tissue renewal process. In humans, the outer layer of epidermal cells are lost, and is replaced by a new layer daily. When the body suffers physical injury, it activates a program to specifically choose and eliminate damaged cells so they can be replaced with new, fully functional cells. This process is made possible by apoptosis, which is the body’s way of deciding which cells need to be destroyed and/or replaced, independent of accidental death.

Physiological apoptosis plays an essential homeostatic role. Apoptosis during embryogenesis removes excess cells, as in the case of Mullerian structures in the male fetus for male organ development (36), and it is also important for developmental stages during tadpole tail regression (37). Apoptotic inhibition during embryonic limb development leads to the emergence of a ‘webbed’ phenotype, due to excess tissue formation between fingers and toes – the tissue that would have otherwise been targeted for apoptosis (38). Significant work conducted in Caenorhabditis elegans and other animal models (39) has identified many evolutionarily conserved genes, regulatory mechanisms and the specificity of apoptosis during certain biological processes.

Defined as a highly regulated form of cell death during the 1970s (40), apoptosis is considered to be a process that responds to external and internal stimuli, activating separate signalling cascades resulting in cell death (41). A defect in the apoptotic machinery, resulting in deregulated or abnormal cell death, disturbs cellular homeostasis, and contributes to human pathologies including cancer, neurodegenerative disorders and cardiovascular disease. It is well documented that apoptosis occurs in the myocardium during hypoxia (42), ischemia/reperfusion (43), myocardial infarction (44) and a failing heart (45). Although apoptosis plays a homeostatic role during development and a detrimental role in disease, studies suggest that there is a possibility that mechanisms involved in apoptotic execution may be tissue specific and regulated differently during development and disease.

Apoptosis is characterized by several factors: the presence of membrane blebbing without the loss of membrane integrity, formation of apoptotic bodies, chromatin condensation, cell shrinkage, DNA fragmentation and a lack of inflammation (46). After cells are genetically manipulated or exposed to conditions such as hypoxia, they can be stained with Hoechst 33258 to distinguish nuclei (42): healthy cells display well-rounded nuclei with definite shape, while cells undergoing apoptosis exhibit condensed and abnormally shaped shrunken nuclei (47). DNA laddering is another biochemical technique used to detect fragmented DNA, which is created through the cleaving action of nucleases during apoptosis, creating DNA fragments of distinct sizes (48). Analysis of isolated DNA by electrophoresis reveals apoptotic DNA running in specific sizes, as opposed to necrotic DNA, which runs as a smear (49). Caspase activity is another key hallmark unique to apoptosis; caspase signalling cascades are activated by apoptotic triggers not otherwise observed in other forms of cellular death (50).

Caspases must be activated to functionally cleave molecular targets and activate other caspases. With a key role in apoptosis, caspases are believed to work in a hierarchical manner in which upstream initiator caspases, including caspase-8, caspase-9 and caspase-10, activate downstream executioner caspases including caspase-3, caspase-6 and caspase-7 (51). Interaction involving the death effector domain of procaspase-8 with the death-inducing signalling complex leads to auto-activation of caspase-8, which activates the caspase signalling cascade ending in apoptotic cellular demise (52). External and internal stimuli activate receptor-mediated or mitochondrial-mediated pathways, resulting in apoptosis.

Extrinsic apoptotic death pathway

External death ligands activate the extrinsic pathway mediated by transmembrane receptors at the surface of the cell membrane. Extracellular death signals, such as Fas ligand (53), tumour necrosis factor-α (TNF-α) (54) and TNF-related apoptosis-inducing ligand (55), are received by death receptors belonging to the TNF receptor (TNFR) super family such as TNFR1 (56) and death receptor-4 (57). Ligand receptor binding induces receptor trimerization, which recruits cytosolic adaptor molecules, such as Fas-associated death domain (58) and TNFR-associated death domain (59), which assemble the death-inducing signalling complex. This complex is responsible for the recruitment and activation of initiator caspases that proteolytically cleave downstream executioner caspases, committing the cell to an apoptotic fate.

Intrinsic apoptotic death pathway

Compared with ligand-induced death, the intrinsic pathway of apoptosis becomes activated by various signals, both of external and internal sources including ultraviolet radiation, treatment with drugs or toxic agents, hypoxic conditions and genetic damage. Because this arm of apoptosis is also termed the mitochondrial pathway, mechanisms are mediated through the ‘powerhouse’ of the cell, in which the mitochondria undergo functional changes and release proapoptotic factors into the cytoplasm. The cardiac mitochondria are responsible for providing cellular energy to myocytes. Managing cellular metabolic demands by generating ATP is accomplished through the mitochondrial respiratory chain and oxidative phosphorylation (60). Healthy mitochondria consist of an impermeable inner membrane that creates an electrochemical gradient between the matrix and the intermembrane space. ATPase, an enzyme localized in the inner mitochondrial membrane, pumps H⁺ protons and is responsible for generating the mitochondrial membrane potential (ΔΨm) (60). This is essential for generating ATP that is used as fuel to drive cellular processes.

Although mitochondria are the site of energy production, the organelle is also the source of reactive oxygen species generation that contributes to oxidative stress, triggering mitochondrial dysfunction. Under certain circumstances, mitochondrial disturbances involve opening of the mitochondrial permeability transition pore, but selective permeabilization of the outer mitochondrial membrane is the major mitochondrial event in apoptotic cell death, resulting in impairment of oxidative phosphorylation machinery (60). Permeabilization of the outer mitochondrial membrane leads to a lethal dissipation of the ΔΨm, contributing to the initial stages of apoptotic cell death, and causing the outer membrane to rupture and release proapoptotic factors (61). The mitochondria release intermembrane proteins such as cytochrome c (62), Smac/DIABLO (second mitochondria-derived activator of caspase/direct inhibitor of apoptosis protein [IAP]-binding protein with low isoelectric point) (63), apoptosis-inducing factor (64) and endonuclease G (65). The IAPs (66) repress apoptosis by preventing the activation of mature caspases. Following a death signal, the release of Smac/DIABLO in the cytoplasm prevents the inhibitory function of IAPs, thus promoting apoptotic death (67). Endonuclease G translocates from the cytoplasm to the nucleus, which activates DNA fragmentation, while the apoptosis-inducing factor promotes progression of apoptotic damage (65). Cytochrome c is recruited to an apoptosome (consisting of apoptotic protease activating factor-1 [68], deoxyadenosine triphosphate (69) and procaspase-9 [70]) that is responsible for activating downstream caspases. It is believed that the cell’s ‘choice’ in signalling to activate apoptosis may be determined by the balance between proapoptotic and antiapoptotic proteins within a cell.

This relative ratio involves specific proteins belonging to the B-cell lymphoma-2 (Bcl-2) family, in which members are both pro-apoptotic and antiapoptotic. Discovered in the 1980s, Bcl-2 was first identified in human follicular lymphoma (71) and found to be resistant to apoptosis (72). On characterization, Bcl-2 family members must share one or more of the four conserved Bcl-2 homology domains, but not all (73,74). In relation to cardiac apoptosis, prosurvival members include Bcl-2 itself, Bcl-w and Bcl-X_L, while proapoptotic members consist of Bax, Bad, Bak, Bid, Bim, Puma, Nix and Bnip3 (75,76). An abundance of proapoptotic Bcl-2 family members will inhibit life-promoting Bcl-2 family members, and result in mitochondrial damage by disruption of the mitochondrial membrane followed by the release of apoptotic factors.

As a hypoxia-inducible member of the Bcl-2 family, Bcl-2/adenovirus E1B 19 kDa protein-interacting protein-3 (Bnip3) has been identified as a proapoptotic protein in tissue cultures including isolated cardiomyocytes sharing only the BH3 domain with Bcl-2 (77). Recent work has reported abnormal Bnip3 activation associated with neurodegenerative disorders such as Huntington’s disease (78) and Parkinson’s disease (79), neuroexcitotoxicity (80,81), rheumatoid arthritis (82), endometrial cancer (83) and colorectal cancer (84). Studies in the Kirshenbaum laboratory have documented Bnip3 as a unique protein that becomes activated under hypoxic injury, mediates mitochondrial defects consistent with mitochondrial permeability transition pore opening with loss of ΔΨm in cardiomyocytes (77) and is upregulated in ischemic cardiac models (77,85). Loss of the transmembrane domain of Bnip3 prevented hypoxia-induced mitochondrial dysfunction and reduced cell death mediated by apoptosis in neonatal myocytes (86), demonstrating that Bnip3 integration into the mitochondrial membrane is critical in triggering mitochondrial disruption. Along with hypoxic injury, Bnip3 has been recognized as an important molecular player during myocardial ischemia/reperfusion (87). Therefore, proapoptotic members of the Bcl-2 family, such as Bnip3, play a critical role in human pathologies and contribute to activation of mitochondrial-mediated apoptotic cell death in cardiomyocytes.

Endoplasmic reticulum stress in apoptosis

Although mitochondrial defects are implicated in cardiac dysfunction, other organelles have been documented to play a functional role. Until the past decade, advances in research have emphasized the role of the endoplasmic reticulum (ER) in cardiac cell death. The ER is an organelle involved in performing diverse functions in mammalian cells including lipid production, regulation of Ca²⁺ handling, protein synthesis and conducting post-translational modification before protein export (88). Accumulation of unfolded proteins or disruption of ER function activates the unfolded protein response (UPR) (89,90) to restore ER functionality and remove irregular proteins through ER-associated degradation (ERAD) (91,92). The unique ER lumen consists of an oxidizing environment essential for the formation of disulfide bonds that contributes to oxidative stress, resulting in apoptosis (93) and myocarditis (94).

Furthermore, chronic or severe ER stress during ischemia/reperfusion and hypoxia can activate apoptotic cell death (95–97) – it is implicated in neurodegenerative (98,99) and metabolic disorders (100), cystic fibrosis (60,101) and cardiovascular disease (102–104). Tissue cultures can be stimulated by various substances, and result in altering the oxidative environment, calcium levels and protein folding. These substances include dithiothreitol, thapsigargin and tunicamycin, which can be used as experimental tools in understanding the UPR (105,106).

UPR activation involves an ER resident molecule, binding immunoglobin protein/glucose-regulated protein 78 (BiP/GRP78) that senses ER stress (107). Under normal conditions, BiP/GRP78 is tethered to ER transmembrane proteins and UPR activation releases BiP/GRP78 into the lumen where it corrects the conformation of unfolded proteins (108). The three integral ER proteins that activate the UPR are protein kinase-like ER kinase (PERK), inositol-requiring kinase-1 (IRE1) and activating transcription factor-6 (ATF6) (90). Dissociation of the transmembrane proteins from BiP/GRP78 (109) induces cytoplasmic and nuclear signalling pathways (110). Auto-phosphorylation and dimerization of PERK, a serine/threonine kinase, leads to phosphorylation of eukaryotic translation initiation factor-2, resulting in inhibition of protein translation (109). Increased PERK activation leads to expression of ATF4, which regulates CCAAT/enhancer-binding protein homologous protein (CHOP), a proapoptotic transcription factor (111). Similar to PERK, phosphorylation and dimerization of IRE1 activates the endoribonuclease domain (110), which is critical for IRE1 function in mediating the messenger RNA splicing event of X-box binding protein-1 (XBP1) messenger RNA that yields a transcriptionally active form of XBP1 involved in the UPR and ERAD (112). Cleavage-mediated activation of the third transmembrane protein, ATF6, also functions in regulating CHOP and XBP1 messenger RNA (113). In failing hearts, ER stress activation of XBP1 was found to regulate cardiac brain natruretic peptide, an important molecule used in diagnosing heart failure (114).

Ischemic injury in cardiomyocytes has been shown to activate ERAD (115), upregulate BiP/GRP78 expression (116,117) and caspase-12 (118), indicating that ER stress contributes to cardiac apoptosis. In a heart failure animal model, activation of BiP/GRP78 and CHOP were significantly increased in induced chronic myocardial ischemic hearts (119). Recent work by Miyazaki et al (120) found ER stress induced by myocardial ischemia/reperfusion in CHOP-deficient mice inhibited myocyte apoptosis and repressed cardiac injury, while the expression of inflammatory cytokines induced by thapsigargin was also suppressed in CHOP-deficient myocytes, identifying CHOP as a molecular player contributing to myocardial injury. Earlier it was reported that the p65 subunit of nuclear factor kappa-B repressed CHOP promoter activity, inhibited tunicamycin-induced ER stress-mediated cell death and increased cell survival in breast cancer cells (121). Research suggests that CHOP may mediate apoptosis by disrupting the balance between proapoptotic and antiapoptotic Bcl-2 members, including Bcl-2, itself (122), Bim (123), Bid and Bnip3l/Nix (124). Furthermore, blocking the activation of the CHOP/puma signalling pathway inhibited ER stress-induced apoptosis in cardiomyocytes (104), supporting the work of another independent study (125). Such reports suggest that CHOP plays a critical role in mediating ER stress-induced apoptotic death in cardiomyocytes. ER-induced stress mediated by IRE1 involves interaction between the α-isoform of IRE1 with proapoptotic Bcl-2 members, Bak and Bax, in mediating mitochondrial apoptosis (126). Furthermore, IRE1 recruits TNF receptor-associated factor 2, which activates a death mechanism by recruiting apoptosis signal-regulating kinase-1 (102,127,128) and caspase-12 (129), contributing to cardiac apoptosis (130).

Activation of the different arms of the UPR in response to severe ER stress results in irreversible cardiac damage by activating the apoptotic signalling machinery. These findings suggest that ER stress in non-UPR pathways may be mediated by other mechanisms because interest in ER and mitochondria communication has greatly increased. Along with inducing UPR signalling molecules, the ER lumen has the highest concentration of Ca²⁺ in the cell. Mitochondrial Ca²⁺ is used in energy production, and fluctuations in ER Ca²⁺ release by proapoptotic Bcl-2 family members have been shown to promote apoptosis (131). Therefore, disruption of ER-mitochondrial coupling contributes to ER stress-induced apoptotic cell death (132,133); understanding such a close organelle relationship can identify potential therapeutic targets for preventing cardiac disease.

PROGRAMMED NECROSIS

Recognized as a form of cell death for more than a century, the necrotic death response is triggered by cells experiencing traumatic or severe damage caused by nutrient shortage, heat shock or exposure to harsh environments (134). In such conditions, cells die uncontrollably, resulting in the failure of entire regions of tissue (135,136). It is understood that the accidental death of cells via necrosis was a nonapoptotic mode of cell death, in which necrotic cells exhibited unique morphological changes. During the initial stages of necrosis, a cell experiences organelle and cytoplasmic swelling (137), as well as disrupted cellular functions, eventually leading to severe ATP depletion and plasma membrane rupture, which is also a morphological feature unique to necrosis. Bursting of the membrane releases cellular contents at the site of necrotic injury; the release of cellular contents negatively affects neighbouring cells, inducing inflammation (138). Necrotic cells do not undergo formation of apoptotic bodies and autophagic vacuoles; therefore, activation of necrotic death is morphologically distinct from apoptosis and autophagy.

Recent studies have shown that the chemical compound necrostatin-1 (Nec-1) inhibits necrosis without affecting apoptosis (139), suggesting that programmed necrosis is a caspase-independent death pathway, potentially occuring when apoptotic machinery is deregulated. Programmed necrotic cell death has been associated with various pathological conditions including viral infections (140), retinal disorders (141,142) and Huntington’s disease (143). Inhibition of programmed necrosis was found to reduce ischemic neuronal tissue damage, improve motor function and neurological impairment (144), and reduce oxidative damage and hypoxic injury in mouse models after administering Nec-1 (145).

Smith et al (146) showed that cardiac cells exhibited peroxide-induced necrosis and, in an isolated perfused mouse heart and open chest mouse model, reduced cell death and infarct size after ischemic/reperfusion was observed following administration of Nec-1. The report was among the first to conclude that Nec-1 could inhibit myocardial injury, identifying Nec-1 as having a cardioprotective role during cardiovascular disease (146).

CONCLUSION

Significant advances in research in ventricular myocytes involving autophagy, apoptosis and programmed necrosis have deepened our understanding of how these cell death pathways contribute to the decline in cardiac performance and contractile impairment after myocardial injury. Recognizing and identifying many key molecular players, such as prodeath proteins, propels a curiosity to further gain insight into how apoptotic cell death is regulated. The mitochondrial disturbances and irregular ER-mitochondrial interaction that regulate the signalling mechanisms involved in apoptosis demonstrates a high level of complexity that requires further investigation for developing potential therapies for treating injured hearts. Although research in basic sciences has provided an in-depth knowledge of the mechanisms underlying cell death, clinical studies based on utilizing pharmacological agents selective for targeting key molecules and repressing mitochondrial dysfunction are crucial in developing treatments that provide cardioprotection in the hopes of treating heart failure patients.