Apoptosis, derived from the Greek word for the natural process of leaves falling from trees, is a distinct form of programmed cell death (Kerr et al., 1972; Kerr, 2002). Although such programmed deaths were described many decades ago, the significance of apoptosis was largely overlooked, in particular, its relevance to disease. During the past 25 years, the improved understanding of apoptotic signalling pathways, with the cloning and characterization of pro- or antiapoptotic genes, has attracted great interest to this process and raised the possibility that therapeutic strategies altering apoptotic pathways may be useful for the treatment of cancer, infectious diseases, degenerative syndromes and other pathological conditions.
Features of Apoptosis
Cell death may occur via at least two broadly defined mechanisms: necrosis or apoptosis. Necrosis is a cell death process generally occurring in response to trauma generated by external factors or overwhelming cellular injury. Necrosis is characterized by swelling and rupture of the plasma cell membrane (cell lyse), with the release of the cellular contents into the immediate extracellular space, which may cause inflammation or harm other neighbouring cells.
In contrast to necrosis (a dirty form of cell death), apoptosis is a genetically encoded and evolutionarily conserved form of cell death, in which any harm done to the organism by this process is minimized (a clean form of cell death). Apoptosis is characterized by several morphological and biochemical aspects: the condensation of the nucleus and cytoplasm, the activation of caspases and nucleases (which degrade cellular proteins and deoxyribonucleic acid (DNA), respectively), membrane blebbing and the fragmentation of cells into multiple small membrane-bound apoptotic bodies, which are rapidly phagocytosed by neighbouring cells. As a result, apoptotic cells are removed from tissues without leaking their cytoplasmic contents into the intercellular space, minimizing tissue inflammation and avoiding damage to neighbouring cells. Apoptosis may occur during normal physiological conditions, for example, during embryonic development, where unnecessary cells may die by apoptosis and deregulation of apoptosis may cause pathological conditions such as cancer or neurodegenerative diseases (Jacobson et al., 1997; Vaux and Korsmeyer, 1999). See also Apoptosis: Morphological Criteria and Other Assays, and Apoptosis: Regulatory Genes and Disease
Executioners of Apoptosis: Caspases
Although there might be some discussion of when death actually occurs in a cell triggered for apoptosis, there is little disagreement that, in most cases, activation of caspases represents a point of irreversibility in the cell death process. The pioneering work of H Robert Horvitz and colleagues with the nematode Caenorhabditis elegans established that the apoptotic gene ced-3 is a caspase, belonging to a family of proteases containing cysteine at their active site and being related to the mammalian interleukin 1-converting enzyme (Yuan et al., 1993). Since the discovery of the first caspase, many more members of this family have been identified in humans and in other species. All these enzymes are constitutively expressed in virtually all cells as inactive zymogens (procaspases) and, on activation, cleave substrates just C-terminal to aspartic acid residues. The activation of caspases is itself dependent on the proteolytic cleavage of the zymogens, which leads to the removal of an inhibitory N-terminal domain and the production of two subunits (one large and one small). These subunits associate as a heterotetramer (two large and two small) to form the active protease that cleaves many cellular proteins. Among the targets of caspases are other molecules of caspase zymogen, potentially amplifying the apoptotic cascade once the initial caspase activation has occurred. Therefore, much effort has been made in addressing the important mechanistic questions in the regulation of caspase activation and understanding the precise pathway(s) leading to the activation of apoptosis. See also Apoptosis: Morphological Criteria and Other Assays, Interleukins.
Regulation of caspase activation
The regulation of caspases occurs by two distinct molecular signalling pathways, depending on whether the cell signals activating apoptosis originate extracellularly, thereby activating the extrinsic pathway (more on this in the sections ahead), or intracellularly, thereby activating the intrinsic pathway (Figure 1).
Most of the current knowledge about the molecular mechanisms regulating the intrinsic pathway derives from studies using the nematode C. elegans or mammalian systems (Figure 1). In the classic pathway, the activation of caspases results from the formation of a multiprotein complex, termed the apoptosome, consisting of CED-4/Apaf-1, procaspase-9 and cytochrome c (in mammals only), which is required for the initial activation of procaspase-9, with the activation of other caspases occurring subsequent to caspase-9 activation (Shi, 2006). For this reason, caspase-9 homologues are also known as initiator caspases, whereas caspases that are directly or indirectly activated downstream of caspase-9 to participate in the degradation of the cellular components, such as caspase-3, are known as effector caspases. See also The Apoptosome: The Executioner of Mitochondria-
Members of the B-cell lymphoma protein 2 (Bcl-2) family of proteins are important regulators of caspase-9 activation, since they can facilitate or prevent the release of cytochrome c from mitochondria into the cytoplasm and regulate the formation of the apoptosome (Danial and Korsmeyer, 2004). A second branch of the intrinsic pathway is an inhibitory one, where caspase activation is blocked by the inhibitor of apoptosis proteins (IAPs) and is only unleashed on expression of IAP antagonists, such as reaper, grim and head involution defective (hid) in the fruitfly, Drosophila melanogaster, or Smac/Diablo in mammalians (Song and Steller, 1999). It appears that both these pathways are used in coordination to control the activation of caspases, in a manner similar to way the gas and the brake are used in the process of driving an automobile (Steller, 2008), but the contribution of each branch may vary according to the cell type and signalling paradigm. See also Inhibitor of Apoptosis (IAP) and BIR-
The decision to undergo apoptosis may also be determined by the balance between proapoptotic and antiapoptotic signalling events triggered by environmental (extracellular) factors, such as Fas ligand (FasL/CD95L), tumour necrosis factor (TNF), transforming growth factor (TGF) and cytokines. Most growth factors and cytokines promote cell survival, growth and differentiation, by triggering antiapoptotic signalling on their target cells. In fact, the loss of certain cell types due to mutations of critical growth factors can be rescued by targeted overexpression of the generic antiapoptotic factor Bcl-2. Results such as this suggest that, in certain situations, suppression of apoptosis is a major function of growth factors and cytokines, and that cell differentiation may represent the default pathway for precursor cells that survive. Among the intracellular (noncytokine) factors that have been shown to potently suppress apoptosis are the CD40 ligand, viral genes such as E1B from adenovirus or p35 from baculovirus and antiapoptotic members of the Bcl-2 family. A large number of DNA viruses have been demonstrated to encode factors which function to curtail the cellular apoptotic response, presumably a prerequisite for successful viral infection and propagation. See also Cytokines, Death Receptors, Oncogenes, and Transforming Growth Factor Beta: Role in Cell Growth and Differentiation
In mammals, the extrinsic pathway mediates apoptosis in response to the activation of cell-surface death receptors, such as Fas/CD95 and TNF receptor. Death receptors can induce apoptosis directly, through the activation of caspases or indirectly, by amplifying the death signal through the activation of the intrinsic/mitochondrial pathway. All Fas receptors contain a conserved extracellular death receptor domain (DR), where FasL binds, inducing the oligomerization of death receptors and initiating a cascade of events that leads to the activation of apoptosis in the target cells. Activated death receptors bind the adaptor molecule Fas-associated death domain (FADD) via the death domain (DD) and FADD recruits the initiator procaspase-8 and procaspase-10 into a complex, the death-inducing signalling complex (DISC), through the death effector domain (DED), which is present both in FADD and in the procaspase. The recruitment of procaspase-8 and procaspase-10 into the DISC complex leads to the autoproteolytic cleavage and activation of these caspases, with subsequent activation of the effector caspases (Figure 2). See also Mitochondrial Outer Membrane Permeabilization
Fas plays a critical role in T-cell mediated toxicity, being abundantly expressed in activated mature lymphocytes and in lymphocytes transformed with human immunodeficiency virus (HIV) or human T-cell leukaemia virus (HTLV-I). It is believed that the Fas apoptotic pathway is implicated in eliminating unwanted activated lymphocytes or virus-infected cells.
Contrary to the evolutionary conserved intrinsic pathway, the components of the extrinsic pathway are not conserved in all metazoan organisms. In C.elegans, no cell-death receptors of the TNF family have been found so far. In D. melanogaster, only a single TNF ligand (Eiger) and its associated receptor (Wengen) are present and it is believed that Eiger/Wengen activate the c-Jun N-terminal kinase (JNK) pathway via the adaptor proteins DTRAF1/2, inducing apoptosis indirectly. The Drosophila homologues of caspase-8 (Dredd) and FADD do not seem to be involved in the activation of apoptosis, but are instead required for the production of antimicrobial peptides in response to Gram negative () bacteria (Kuranaga and Miura, 2007).
Disregulation of the Fas-mediated apoptosis can lead to several pathological disorders, with reduced Fas-mediated apoptosis being associated with excessive cell proliferation in hepatocellular carcinoma and excessive Fas-mediated apoptosis being associated with viral and alcoholic hepatitis, Wilson disease and hepatic fibrosis (Guicciardi and Gores, 2009).
Bcl-2 family members
The members of the Bcl-2 family of proteins are important components of the intrinsic pathway, regulating mitochondrial outer membrane permeabilization (MOMP) and the release of proapoptotic factors, such as cytochrome c, from mitochondria. In addition, Bcl-2 family members are key regulators of apoptosis because they connect the extrinsic and intrinsic pathways (Figure 2). Bcl-2 was first identified as being the cause of human follicular lymphoma due to a chromosome translocation affecting Bcl-2 function. Subsequently, it was shown to promote tumuorigenesis by inhibiting apoptosis, instead of promoting cell proliferation (Chao and Korsmeyer, 1998).
In humans, there are presently 23 known members of the Bcl-2 family (Hardwick and Youle, 2009), all of which contain at least 1 of the 4 possible protein domains of homology (BH Bcl-2 homology). Bcl-2 family members can be classified into three categories according to their functions, as well as the number and type of BH protein domains: antiapoptotic proteins with multiple BH domains (BH1BH4), proapoptotic proteins with multiple domains (BH1BH3) and proapoptotic proteins with BH3-only domain (Table 1).
In the prevailing model for regulation of apoptosis by Bcl-2 family members (Figure 2), the antiapoptotic multidomain members, such as Bcl-2, bind to and neutralize the proapoptotic members (Bax) in nonapoptotic cells. On receiving death-inducing signals, BH3-only domain proteins inactivate the antiapoptotic multidomain proteins, releasing the proapoptotic proteins from the inhibitory interaction with the antiapoptotic proteins. The proapoptotic Bcl-2 family proteins then oligomerize, creating pores in the mitochondrial outer membrane and allowing the release of cytochrome c into the cytoplasm, which leads to caspase activation and cell death. In an alternative model, the antiapoptotic members directly bind and inhibit the BH3-only domain proteins, which otherwise directly induce the oligomerization of the proapoptotic multidomain proteins.
Recently, much more attention has been paid to the additional, nonapoptotic roles of Bcl-2 family proteins, including those in biological processes like the regulation of mitochondrial dynamics (fusion and fission of mitochondria) or autophagy (the process in which cells undergoing nutrient starvation degrade cellular components to survive and maintain a minimal metabolic activity). See also Autophagy in Nonmammalian Systems, and Mitochondria Fusion and Fission
In C. elegans, there are three Bcl-2-related proteins: CED-9, EGL-1 and CED-13. CED-9 is an antiapoptotic multidomain protein, which inhibits CED-4 (Apaf-1)-mediated activation of CED-3 (caspase). During apoptosis, CED-9 is inhibited by the BH3-only proteins EGL-1 and CED-13 to allow the activation of CED-3 (Figure 1). Two Bcl-2 family proteins exist in D. melanogaster, Buffy and Debcl, both containing multiple BH domains. However, Bcl-2 family members seem to have a limited role in the regulation of apoptosis in Drosophila, whereas the Diap pathway seems more preponderant in this biological system (Steller, 2008). See also The BCL-
Role of the c-myc oncogene in the control of apoptosis
Additional factors have been shown to directly or indirectly control apoptosis, including the proto-oncogene c-myc, identified 25 years ago, which has a central role in the regulation of growth control, cell differentiation and apoptosis, and is among the genes that most frequently contribute to the development of human tumours. c-MYC is a transcription factor which recognizes the CA[C/T]GTG element (E box) and also has the ability to repress transcription through a pyrimidine-rich cis element termed the initiator (Inr). The target genes of c-MYC are many, being involved in a variety of physiological processes, such as cell cycle regulation, metabolism, protein synthesis and cell adhesion. See also Transcriptional Gene Regulation in Eukaryotes
Overexpression or inappropriate expression in time of c-MYC has been found to promote apoptosis. In the early 1990s, it was observed that ectopic expression of c-MYC protein accelerated apoptosis in cells deprived of survival factors (Hoffman and Liebermann, 2008). Still, other observations have suggested that in B cells MYC may regulate apoptosis in precisely the opposite fashion: inhibition of c-MYC resulted in dramatic apoptosis, whereas overexpression of c-MYC protected B cells from apoptosis. Further, addition of antisense c-MYC oligonucleotides to immature T cells and to some T-cell hybridomas inhibited c-MYC expression and prevented T-cell receptor-mediated apoptosis. The precise molecular mechanisms of how MYC induces apoptosis remain unclear. However, it appears that multiple pathways are regulated by MYC, including one requiring the p53 tumour suppressor, where MYC-induced apoptosis is preceded by stabilization of p53. In mammalian cells, p53 is a major regulator of cell cycle arrest and apoptosis (see following sections). Deregulated MYC upregulates ARF (acute renal failure), which in turn activates p53 to regulate a group of target genes that activate apoptosis and cell cycle arrest (Figure 3). The loss of ARF and p53 caused an increase in tumourigenesis in mouse models, confirming the importance of the ARFMDM2 (mouse double minute 2)p53 pathway in MYC-induced apoptosis. Other reports have suggested, however, that MYC-dependent apoptosis is, in some contexts, independent of functional p53. This dependence on p53 seems to be determined by the cell type and the apoptotic signal triggered (Hoffman and Liebermann, 2008).
|Figure 3. Regulation of apoptosis by p53. Several stress conditions (DNA damage, hypoxia, oncogene activation, among others) lead to the transcriptional activation and an increase of the p53 protein levels. Activation of p53 promotes the transcriptional activation of many target genes involved in several cellular responses: apoptosis, cell cycle arrest, DNA repair and senescence. Under normal physiological conditions, the cytoplasmic p53 protein is maintained at low levels by MDM2 (an E3 ubiquitin ligase) that targets the p53 protein to ubiquitylation (Ub) and subsequent degradation by the proteosome. MDM2 is also a target of the nuclear p53 protein and MDM2 activity is counteracted by ARF in response to activation of oncogenes, such as c-MYC. Inhibition of MDM2 allows p53 to accumulate, both in the nucleus and the cytoplasm. In the cytoplasm, p53 can interact with proteins of the Bcl-2 family, being sequestered by Bcl-x|
In Drosophila, only one Myc protein (dMyc) has been identified, with similar biochemical and molecular characteristics to its mammalian orthologue. dMyc is also involved in the regulation of cell proliferation and apoptosis, and the functional conservation between the Drosophila and the mammalian orthologues has been shown by complementation studies in both organisms. However, dMyc differs from its vertebrate orthologue by the fact that its overexpression does not lead to the formation of tumours. Also, the apoptotic pathway induced in Drosophila is different from the mammalian one. Despite the existence of a p53 orthologue in Drosophila, neither ARF nor MDM2 orthologues are present in the fly genome. Recently, it was shown that the apoptosis caused by dMyc in Drosophila imaginal disc cells occurs by the process of cell competition: in neighbouring cells expressing different levels of dMyc, either due to dMyc loss-of-function mutations or overexpression of dMyc, the cells with higher levels of dMyc out-compete the cells with lower levels of dMyc, which undergo apoptosis (Moreno and Basler, 2004; de la Cova et al., 2004). One explanation for this phenomenon is that cells with higher levels of dMyc proliferate faster and sequester growth factors, which normally are produced in limited amounts, causing apoptosis in cells with lower levels of dMyc due to the lack of growth factors. Other studies have shown that dMyc induced apoptosis is accompanied by the induction of Drosophila p53 messenger ribonucleic acid (mRNA), but that dp53 activity is not essential for dMyc's ability to induce apoptosis.
Tumour suppressor gene p53 and apoptosis
The p53 protein was originally identified as a nuclear phosphoprotein that binds the large transforming antigen of the SV40 DNA virus (T antigen). Since then, p53-dependent signalling pathways have been the subjects of intense study (Fuster et al., 2007) (Vousden and Prives, 2009). p53 is now recognized to act as a transcription factor regulating the expression of genes involved in at least two major processes in response to DNA damage or cellular stress: regulation of cell cycle progression and apoptosis.
The first evidence of the involvement of p53 in apoptosis was obtained by introducing a temperature-sensitive mutant of p53 in a myeloid leukaemia cell line. At the permissive temperature, the p53 mutant triggered massive apoptosis of the cells. In genetically defined cell lines, loss of p53, due to mutation or knockout, confers resistance to apoptosis triggered by radiation and by chemotherapeutic agents. In fibroblasts, p53 triggers cell cycle arrest in primary cells, but apoptosis in oncogene-transformed cells. In humans, the loss of p53 comprise the most frequent genetic abnormality found in human tumours and is associated to the worst prognosis. Up to 50% of human cancers contain deleted or mutated p53 genes, including 80% of colon cancers, 50% of lung cancers and 40% of breast cancers. Many human p53 mutants have been described; most mutants loose the ability to bind DNA and accordingly fail to activate the transcription of target genes. Complexes of wild-type and mutant p53 protein are unable to bind the p53 site or transcriptionally activate p53 reporter constructs, suggesting that mutant p53 proteins may act in a dominant negative manner. See also Chronic Lymphocytic Leukaemia, EpsteinBarr Virus, and Tumour Suppressor Genes
The kind of cellular response following p53 activation is dependent on the type of cell and nature of the cellular stress. Recently, a novel transcription-independent proapoptotic function, mediated by the cytoplasmic form of p53, revealed that p53 can participate directly in the regulation of the apoptotic intrinsic pathway, by interacting with members of the Bcl-2 family to induce mitochondrial outer membrane permeabilization (Green and Kroemer, 2009).
One of the important target genes transcriptionally activated by p53 is the cyclin-dependent kinase (CDK) inhibitor p21. The p53 protein elicits an increase in p21 levels on cellular damage inflicted by radiation or other external toxic agents, leading to CDK inhibition and cell cycle arrest. The arrest of the cell cycle following DNA damage allows time for DNA repair to occur, leading to the concept that p53 acts as a guardian for genomic fidelity. Induction of p21 expression is very sensitive to even low levels of p53. This fine regulation permits a transient arrest in the G1 phase of the cell cycle under mild DNA damage or stress and the survival of cells until optimal cellular conditions are restored. However, in the case of pre-cancerous lesions, tumour progression can be avoided by an irreversible cell cycle arrest: p53 mediated senescence. The p53 tumour suppressor pathway acts therefore as a surveillance mechanism that is activated in response to cellular stress. Despite its involvement in cell cycle arrest, the p53 gene is largely dispensable for cell growth or differentiation.
Modulation of mammalian p53 activation is regulated at the levels of transcription and translation, as well at the posttranslational level (Lavin and Gueven, 2006). Activation of p53 can occur in consequence of several posttranslational modifications, such as phosphorilation, ubiquitination, acetilation and other proteinprotein interactions, that can either stabilize p53 protein or convert p53 into its active phosphorilated form, changing its subcellular localization (nuclear export and/or mitochondrial association). Modulators of p53 include MDM4, that inhibits p53 transcriptional activity, a variety of kinases, such as ATM (product of the ataxia telangiectasia disease gene) and the ubiquitin ligase MDM2. MDM2 stimulates nuclear export of p53 by monoubiquilation, leading to p53 proteasomal degradation in the cytoplasm. On DNA damage sensor proteins (RAD proteins, HUS1) induce ATM/ATR which phosphorylate the effector kinases (CHK1/CHK2), which modulate phosphorylation of p53. In unstressed cells, p53 protein is maintained at low levels due to its interaction with the E3 ubiquitin ligase MDM2. Different cellular stresses can lead to the stabilization of the p53 protein, and an important mechanism for p53 protein stabilization is via the ARF-dependent inactivation of MDM2 (Figure 3).
In C. elegans, the p53-like protein (CEP-1) only shares homology with the DNA-binding domain of its mammalian homologue, and is required for the radiation-induced apoptosis in the germ cells. CEP-1 mediates several stress-induced responses but not a cell cycle arrest upon DNA damage. CEP-1 is required for the induction of EGL-1 and CED-13 (proapoptotic BH3-only proteins) after DNA damage. In Drosophila p53 (dmp53) is required for transcriptional induction of the proapoptotic proteins Reaper, Hid and Sikcle. Activation of dmp53 induced by Drosophila CHK2 phosphorylation, but both in C. elegans and Drosophila regulation of p53 homologues by MDM2 is absent (Lu and Abrams, 2006).
The role of NFB in apoptosis
The nuclear factor--light-chain-enhancer of activated B cells (NFB) is a central mediator of immune and inflammatory responses in mammals. NFB originally discovered as a protein that bound specifically to an enhancer element of the immunoglobulin kappa light chain gene in activated B cells. Major members of the NFB protein family include RelA (p65), Rel B, c-Rel, NFB1(p105/p50), and NFkB2(p100/p52), which have been implicated in the regulation of a variety of important genes in the immune response (IL-2 and IL-2), in inflammatory and acute-phase response (interleukin 1 (IL-1), IL-6, TNF, and serum amyloid A protein) and in response to certain viruses (e.g. HIV-LTR, SV40, cytomegalovirus and adenovirus).
In resting cells, cytosolic NFB dimers are bound to IkB inhibitory proteins that prevent the nuclear translocation of NFB and the transcriptional activation of NFB target genes. Two distinct NFB activation cascades, that respond to different stimuli, can be triggered: the canonical (or classical) NFB pathway and the noncanonical (or alternative) NFB signalling cascade (Dutta et al., 2006).
The NFB canonical pathway is activated in response to injury, inflammation, infection and other stress conditions. Some of the target genes activated by the NFB pathway include antiapoptotic factors of the Bcl-2 family (Bcl-x
NFB has been shown to protect cells from apoptosis in a number of scenarios, including TNF signalling and antiimmunoglobulin M (anti-IgM)- or TGF-mediated apoptosis. Apoptosis induced by TNF signalling can be greatly enhanced when a dominant inhibitor of NFB, IkB-, is introduced into cells, preventing activation of endogenous NFB; furthermore, introduction of wild-type NFB impedes TNF-induced apoptosis. However under specific circumstances proapoptotic activity of NFB can also occur. NFB can induce the transcription of proapoptotic target genes, including the Fas death receptor and respective death-inducing ligand, p53 and the proapoptotic Bcl-2 family members Bax and Bcl-x
Apoptosis involves a cascade of complex events, from external apoptotic signals activating specific receptor complexes to the execution of apoptosis by activation of proteases and endonucleases. The commitment to apoptosis depends on the balance between proapoptotic and antiapoptotic signalling components within cells. The understanding of these signalling pathways and molecular regulators opens enormous opportunities that may lead to specific therapies for diseases caused by deregulation of the normal cell death processes.
Based in part on the previous version of this Encyclopedia of Life Sciences (ELS) article, Apoptosis: Molecular Mechanisms by Min Wu, Han-Fei Ding and David E Fisher.
- A metabolic, suicide pathway to cell death which often maintains homeostasis in multicellular organisms. Apoptosis is characterized by membrane blebbing, highly condensed chromatin and activation of an endonucleolytic process which sequentially cleaves chromosomal DNA to small fragments around 200 bp. As a result, cells shrink and condense into separate multiple membrane-bound apoptotic bodies that are eventually engulfed by surrounding cells.
- Cysteine-aspartic proteinases are a family of cysteine proteases that are related to mammalian interleukin 1-converting enzyme (ICE/caspase-1) and to the nematode apoptotic gene product Ced-3. Caspases are the executioners of apoptotic cell death via cleavage of numerous target proteins.
- A passive, catabolic and pathological cell death process, which generally occurs in response to external toxic factors such as inflammation, ischaemic or toxic injury. It is characterized by the swelling of mitochondria, early rupture of the plasma membrane and the total destruction of the intact structure of the cells. Necrosis is biochemically distinguishable from apoptosis by its lack of caspase activation or extensive endonucleolytic DNA cleavage.
- Programmed cell death
- A highly orchestrated mechanism which eliminates cells or tissues in a predictable fashion suggestive of an underlying developmental or genetic programme.
- Digestive enzymes that cause the breakdown of proteins. Often they recognize specific amino acid target sequences and selectively target them for cleavage.