Apoptosis can be thought of as a signalling cascade that results in the death of the cell. Properly executed apoptosis is critically important for both development and homoeostasis of most animals. Accordingly, defects in apoptosis can contribute to the development of autoimmune disorders, neurological diseases and cancer. Broadly speaking, there are two main pathways by which a cell can engage apoptosis: the extrinsic apoptotic pathway and the intrinsic apoptotic pathway. At the centre of the intrinsic apoptotic signalling pathway lies the mitochondrion, which, in addition to its role as the bioenergetic centre of the cell, is also the cell’s reservoir of pro-death factors which reside in the mitochondrial IMS (intermembrane space). During intrinsic apoptosis, pores are formed in the OMM (outer mitochondrial membrane) of the mitochondria in a process termed MOMP (mitochondrial outer membrane permeabilization). This allows for the release of IMS proteins; once released during MOMP, some IMS proteins, notably cytochrome c and Smac/DIABLO (Second mitochondria-derived activator of caspase/direct inhibitor of apoptosis-binding protein with low pI), promote caspase activation and subsequent cleavage of structural and regulatory proteins in the cytoplasm and the nucleus, leading to the demise of the cell. MOMP is achieved through the co-ordinated actions of pro-apoptotic members and inhibited by anti-apoptotic members of the Bcl-2 family of proteins. Other aspects of mitochondrial physiology, such as mitochondrial bioenergetics and dynamics, are also involved in processes of cell death that proceed through the mitochondria. Proper regulation of these mitochondrial functions is vitally important for the life and death of the cell and for the organism as a whole.
There are multiple types of cell death, and each type has distinct morphological characteristics, physiological roles and mechanisms of execution . Apoptosis (known as ‘Type I cell death’) is one form of cell death, and plays a very important role in development and tissue homoeostasis (i.e. the balance between cell proliferation and cell death). Disrupting tissue homoeostasis by altering apoptosis can contribute to many human diseases. For example, aberrant apoptosis can lead to autoimmune and neurodegenerative diseases, while blocks in apoptosis can contribute to carcinogenesis by allowing cells to grow uncontrollably and ignore normal anti-growth signals . Contrary to other forms of cell death, such as necrosis, apoptosis involves the activation of a signalling cascade that causes cells to lose viability before they lose membrane integrity . The molecular steps to apoptotic cell death are outlined below and in Figure 1.
Apoptosis is a signalling cascade carried out by caspases
The molecules responsible for propagating the apoptotic signalling cascade are enzymes known as caspases, which are cysteine-dependent aspartate-directed proteases . To date, 17 mammalian caspase proteins have been identified, however, only seven of these are thought to function primarily in apoptosis (others have roles in immune responses or have uncharacterized functions) . Apoptotic caspases are classified as either initiator caspases or effector caspases. In healthy cells, caspases exist as zymogens (inactive enzymes); following an apoptotic stimulus, initiator caspases are activated via dimerization and subsequent autocleavage in a process termed ‘induced proximity’ . Once activated, initiator caspases activate downstream effector caspases via cleavage at specific aspartate residues . Active effector caspases then proteolytically cleave numerous regulatory and structural proteins within the cell . Cleavage of these proteins by active effector caspases results in the physical and morphological manifestations of apoptosis (i.e. membrane blebbing, phosphatidylserine externalization and chromatin condensation). The orderliness of this signalling cascade is advantageous to the organism, as the cell is neatly dismantled from the inside out and packaged for removal by macrophages, thus avoiding potential damage to neighbouring cells and an immune response due to leakage of cellular contents (as is seen in other forms of cell death) .
Apoptosis can be broadly defined by two main pathways
There are two principal apoptotic pathways, the extrinsic, or death receptor-mediated pathway, and the intrinsic, or mitochondrial-mediated pathway (Figure 1). The extrinsic pathway is engaged when ‘death receptors’ on the exterior plasma membrane [i.e. TNFα (tumour necrosis factor α), TRAIL (TNF-related apoptosis-inducing ligand) and Fas receptors] are engaged by their cognate ligands (which can be soluble or membrane-bound, and in general are produced by members of the innate immune system) , leading to ligation of the receptors and subsequent downstream signalling through the initiator caspase 8. Although both pathways are physiologically important, the focus of the present chapter will be on the mitochondrial, or intrinsic, apoptotic signalling pathway.
Intrinsic apoptotic signalling
Unlike the extrinsic pathway, the intrinsic pathway is activated by stress such as DNA damage, growth-factor withdrawal and exposure to certain chemotherapeutic agents. During intrinsic apoptotic signalling, permeabilization of the OMM (outer mitochondrial membrane) occurs. MOMP (mitochondrial outer membrane permeabilization) leads to the release of proteins from the IMS (intermembrane space) into the cytosol. Some of these proteins are considered ‘innocent bystanders’ and do not elicit any particular cellular response following their release from the IMS. However, others, such as cytochrome c and Smac/DIABLO (Second mitochondria-derived activator of caspase/direct inhibitor of apoptosis-binding protein with low pI), promote cell death by promoting the activation of caspases. For example, following its release from the IMS, cytochrome c forms a complex with dATP, Apaf-1 and the initiator caspase 9. This multimeric complex (known as the ‘apoptosome’)  functions as an activating platform for caspase 9. Following its activation in the apoptosome through induced proximity, active caspase 9 cleaves and activates effector caspases in the cytosol (Figure 1 and ). Cytochrome c is absolutely required for formation of the apoptosome and activation of caspase 9, thus, without MOMP and the release of cytochrome c from the IMS, caspase 9 activation and subsequent activation of downstream effector caspases does not occur . The IMS protein Smac/DIABLO also functions to promote caspase activation upon its release from the mitochondria, although via a different mechanism from cytochrome c. Smac/DIABLO promotes caspase activation by neutralizing IAPs (inhibitor of apoptosis proteins), which bind to and prevent the activities of caspase 9 and effector caspases in the cytosol; following its release from the IMS, Smac/DIABLO neutralizes the IAPs, thus promoting caspase 9 and effector caspase activation . Although MOMP-induced caspase activation is absolutely necessary for mitochondrial apoptotic cell death, inhibition of caspase activity in the presence of MOMP can still lead to non-apoptotic forms of cell death, due to loss of mitochondrial function (reviewed in ). Thus the factors that control the integrity of the OMM literally control the life and death of the cell. In addition, factors such as mitochondrial dynamics, mitochondrial bioenergetics and mitochondrial cristae remodelling also contribute to MOMP and the overall health of the mitochondria, as discussed at the end of this chapter.
The Bcl-2 (B-cell lymphoma-2) family controls MOMP
The integrity of the OMM and MOMP is controlled by the Bcl-2 family of proteins. The founding member of this family, Bcl-2, is a potent anti-apoptotic molecule that blocks MOMP and was first identified in follicular B-cell lymphoma as a translocation to the immunoglobulin heavy chain locus t(14:18), which renders the gene constitutively hyperexpressed . Following the discovery of Bcl-2, numerous proteins with sequence and structural similarity were identified that either negatively or positively influence MOMP; these proteins comprise the ‘Bcl-2 family’ (Figure 2). Family members are classified into one of three functional groups: anti-apoptotic proteins, pro-apoptotic effectors and BH3-only proteins, which are pro-apoptotic proteins which regulate the other two classes of Bcl-2 proteins. The Bcl-2 proteins share sequence similarity in the form of BH (Bcl-2 homology) domains. The anti-apoptotic proteins and the pro-apoptotic effector proteins contain BH domains 1–4 , and the BH3-only proteins, as their name suggests, only contain the third BH domain. The anti-apoptotic members of the Bcl-2 family (e.g. Bcl-2, Bcl-xL, Bcl-w, A1 and Mcl-1) serve to inhibit MOMP and cytochrome c release by binding to and sequestering pro-apoptotic family members  and in general are found tethered to the OMM. The two other functional groups are pro-apoptotic and exert their pro-death functions in different ways (described in detail below). The majority of these pro-apoptotic proteins are cytosolic, with the exception of the effector molecule Bak, which is located on the OMM.
Following an apoptotic stimulus, the effector molecules Bak and Bax undergo conformational changes that lead to the formation of homo-oligomers in the OMM. Bak and Bax homo-oligomers form pores in the OMM, and affect its permeabilization . These openings serve as portals through which IMS proteins move from the IMS to the cytosol, thus engaging caspases in the cytosol and propagating the apoptotic signal. It should be noted, however, that the exact nature and structure of these pores is unknown. Thus a critical step in the control of mitochondrial apoptosis is the activation of Bak and Bax. Indeed, genetic studies have proven that Bak and Bax are functionally redundant and completely necessary for MOMP and mitochondrial apoptosis. While bak−/− or bax−/− mice appear developmentally and phenotypically normal (although bax−/− mice do have minor defects relating to the neurological and haematopoietic systems and spermatogenesis), the vast majority of bak−/− bax−/− mice (approx. 90%) die embryonically due to an inability to complete the apoptotic processes that are required to remove excess cells during development [15–17]. Furthermore, cells from these mice are resistant to a variety of apoptotic stimuli known to rely on the intrinsic pathway. Although another protein, Bok, shares similar domain structure with Bak and Bax, there is little evidence to suggest that it functions as an effector molecule, and its relationships with the Bcl-2 family and the intrinsic apoptotic pathway remain unclear .
The activity of effector molecules is tightly regulated by the other members of the Bcl-2 family (Figure 3). As mentioned above, the anti-apoptotic proteins bind to and inhibit the pro-apoptotic members of the Bcl-2 family. The BH3-only proteins do not permeabilize the OMM like Bak and Bax, but rather are thought to promote the activation of Bak and Bax through either directly activating them and/or neutralizing the anti-apoptotic proteins (the two models for how BH3-only proteins function are discussed below). The BH3-only proteins are regulated by different stimuli during apoptosis (reviewed in ) in order to engage the mitochondrial pathway of apoptosis. For example, the BH3-only protein Bid is a much more potent pro-apoptotic molecule following proteolytic cleavage of its inhibitory N-terminal region to its truncated form (termed ‘tBid’) by active caspase 8; caspase 8 is activated by the extrinsic apoptotic pathway (see above and Figure 1), thus caspase-8-mediated cleavage of Bid links the extrinsic and intrinsic apoptotic pathways.
The majority of signals to the mitochondrial pathway of apoptosis, however, come from intracellular stress. For example, basal levels of Bim are very low in healthy cells due to sustained phosphorylation by ERK (extracellular-signal-regulated kinase) signalling on Ser69 of Bim , which leads to subsequent proteasomal degradation of Bim. Upon cessation of ERK signalling following detachment of cells from the extracellular matrix or growth-factor withdrawal, cellular levels of Bim increase, thereby promoting apoptosis. Similar to Bim, Bad is also regulated by post-translational modifications. In healthy cells, growth-factor signalling leads to phosphorylation of Bad on multiple residues, which results in its sequestration by 14-3-3 proteins [20–23], thereby preventing its pro-apoptotic function. Upon growth-factor withdrawal, this phosphorylation is relieved, and Bad is free to promote MOMP by interacting with and inhibiting anti-apoptotic Bcl-2 proteins (see below). Other BH3-only proteins respond to cellular stress and engage the mitochondrial pathway of apoptosis following transcriptional up-regulation. For example, E2F1 up-regulates the BH3-only proteins Bim, Puma and Noxa , while DNA damage up-regulates Puma and Noxa via p53. These are merely a few examples of the differential regulation of BH3 proteins and how they are engaged by intrinsic apoptotic stress, which emphasizes the complexity of Bcl-2 protein regulation. The complexity of Bcl-2 family protein regulation underscores the importance of inducing MOMP in a tightly controlled manner. The regulation of Bcl-2 proteins is further co-ordinated by the specific binding patterns displayed by family members, in that Bcl-2 proteins display differential preference for other Bcl-2 proteins. Experiments using the BH3 domains of the Bcl-2 proteins have illustrated that BH3-only proteins preferentially interact with particular subsets of anti-apoptotic proteins (Figure 4) [25–27]. For example, Noxa only interacts with Mcl-1 and A1. The BH3-only protein Bad, on the other hand, interacts with Bcl-2, Bcl-xL and Bcl-w, but not Mcl-1 and A1. Bid, Bim and Puma are not restricted in their binding patterns and can interact with all of the anti-apoptotic Bcl-2 proteins. The interactions between Bcl-2 proteins have in large part shaped the two models for Bak and Bax activation (discussed below).
There are two models to explain the induction of MOMP
There are two models to explain the activation of Bak and Bax, the direct activator model and the neutralization model (Figure 5) . It should be noted that the two models are probably not mutually exclusive. The direct activator model suggests that Bak and Bax are inactive until they are directly engaged by the BH3-only proteins tBid, Bim and possibly Puma (the role of Puma as a direct activator is controversial). This direct binding induces a conformational change that allows for oligomerization of Bak and Bax and insertion into the OMM, thus leading to MOMP. Until recently, the main evidence for this model came from studies with peptides and purified proteins, which illustrated that Bid and Bim, but not the other BH3-only proteins, can induce conformational changes in Bak and Bax that cause Bak and Bax to form pores in liposomes (which recapitulate the lipid environment of an OMM) in the absence of other mitochondrial proteins [26,27]. These and other studies have also demonstrated that the other BH3-only proteins (i.e. Bad, Bmf, Noxa, etc.) are apparently not capable of directly activating Bak or Bax, but rather function to promote MOMP by binding to and neutralizing anti-apoptotic Bcl-2 proteins that have sequestered pro-apoptotic proteins [25–27]. Therefore in the direct activator model, Bid and Bim are responsible for directly activating Bak and/or Bax, while the other BH3-only proteins function to promote Bak and Bax activation by neutralizing the anti-apoptotic family members [13,25–27].
Technically, it is difficult to detect interactions between Bak and Bax and the BH3-only proteins by co-immunoprecipitation, which is the main experimental method employed to detect protein–protein interactions in cells. However, recent progress has defined interactions between Bak and Bax and the BH3-only proteins, which has led us to consider new aspects of Bcl-2 protein function, the most surprising of which is that the mitochondrial membrane itself plays an important role in modulating the activities of Bcl-2 proteins and aiding in Bax oligomerization [28,29]. For example, experiments with recombinant proteins have demonstrated that neither Bax nor Bcl-xL bind to membranes until tBid is inserted , indicating that the first step in tBid-induced Bax activation is not tBid–Bax interaction, as might have been predicted from the direct activation model, but rather is the insertion of tBid into the membrane. In addition, using techniques such as FRET (Förster energy resonance transfer) and NMR, structural, spatial and temporal interactions in vitro between tBid and Bax and Bim and Bax (respectively) have been observed [29,30]. That highly technical methods were needed to detect these interactions and that, at least in the case of tBid, binding to Bax requires membranes, potentially explains why co-immunoprecipitation experiments from cell lysates failed to reveal interactions between BH3-only proteins and pro-apoptotic effector molecules.
While the direct activator model states that engagement of Bak and Bax by a direct activator BH3 is required for their activation, the neutralization model does not consider this to be an obligate step to MOMP. The neutralization model suggests that Bak and Bax are always in an active conformation, but remain sequestered by anti-apoptotic proteins that prevent their oligomerization; therefore Bak and Bax do not require a physical engagement by a BH3-only protein in order to become active. Following an apoptotic stimulus, BH3-only proteins become ‘activated’ (either by being released from their subcellular compartments or by being up-regulated) and neutralize the anti-apoptotic proteins, thus releasing active Bak and Bax, allowing for Bak and Bax oligomerization and MOMP. In this model, Bid and Bim are the most potent inducers of MOMP, because they have the capacity to interact with all of the anti-apoptotic Bcl-2 proteins (see Figure 4). One argument for this model and against the direct activation model surrounds the phenotype of the bak−/− bax−/− and the bid−/− bim−/− mice. As discussed above, it is known that the bak−/− bax−/− phenotype is embryonically lethal, and that murine embryonic fibroblasts from these animals are resistant to inducers of the intrinsic apoptotic pathway. However, the bid−/− bim−/− phenotype is developmentally normal, and cells from these mice undergo apoptosis in response to some intrinsic apoptotic stimuli . The direct activator model would argue that if direct activation of Bak and Bax was necessary, these two mice would share the same phenotype. Proponents of the neutralization model thus cite this finding as evidence that direct activation is not necessary. However, an alternative interpretation is that proteins other than Bid and Bim can directly activate Bak and/or Bax. Indeed, there has been extensive work on the role of cytosolic p53 and Bak and Bax activation, which concludes that p53 in the cytoplasm can activate Bak and Bax (reviewed in ). In addition to p53, other non-Bcl-2 proteins have also been proposed to influence Bak or Bax activation, including ATG5, MAP-1 and others. Thus we have much to learn about how MOMP is regulated by the Bcl-2 (and possibly non-Bcl-2) proteins, and it is very likely that aspects of both models are required for optimal cell death to occur.
Mitochondrial alterations upon MOMP
During apoptosis, the mitochondrial transmembrane potential (ΔΨm) dissipates, and this event occurs at about the time of MOMP or shortly thereafter. However, this event is not causative; the loss of ΔΨm is delayed by inhibition or disruption of caspase activation, while MOMP proceeds unimpeded ; therefore, the loss of ΔΨm following MOMP is a result of active caspases gaining access to the intermembrane space through the now permeable OMM. Other mitochondrial effects observed upon MOMP, including ROS (reactive oxygen species) generation and gross changes in mitochondrial morphology, are also attributable to the action of caspases at the IMM (inner mitochondrial membrane) . One caspase substrate is particularly important for the rapid loss of ΔΨm upon MOMP: NDUFS1, which is a component of complex I of the electron transport chain. In cells expressing a non-cleavable mutant of NDUFS1, MOMP occurs but is not accompanied by ROS production, loss of ΔΨm, gross changes in mitochondrial morphology or rapid loss of ATP .
Nevertheless, even in the absence of caspase activation, a delayed loss of ΔΨm and electron transport function occurs that cannot be attributed to only the dilution of cytochrome c . As a consequence, cells that undergo MOMP can submit to a CIDC (caspase-independent cell death), which is an effect that can be observed in some cells in wild-type animals undergoing developmental cell death (reviewed in ), presumably due to inefficient engagement of caspases. CIDC is not inevitable if MOMP occurs in the absence of caspase activation; a combination of elevated glycolysis and effective mitophagy can protect some cells from CIDC if, for example, elevated GAPDH (glyceraldehyde-3-phosphate dehydrogenase) levels are enforced .
Other mitochondrial events upon MOMP may be more central to the process. The IMM is folded to form cristae, and much of the cytochrome c in the IMS appears to be sequestered in these. The entrances to the cristae are occluded by the protein OPA1 (optic atrophy 1), and changes in OPA1 distribution during MOMP may contribute to subtle alterations in cristae morphology, facilitating the release of cytochrome c and other proteins upon MOMP . Rhomboid proteases, present in the mitochondrial membranes, are likely to be involved in OPA1 cleavage and cristae remodelling (reviewed in ).
Upon MOMP, mitochondria often fragment, and this is probably due to an inhibition of mitochondrial fusion. How this occurs is not clear. Anti-apoptotic Bcl-2 proteins such as Bcl-xL have been shown to bind to the fusion protein Mfn-2, and overexpression of Bcl-xL can promote mitochondrial fusion . In contrast, cells lacking Bak and Bax display decreased mitochondrial fusion . This leads to the idea that changes in Bcl-2 family interactions during MOMP affect their interactions with the fusion machinery.
The situation is made more complex by observations that components of the mitochondrial fission mechanisms may have roles in MOMP. In particular, alterations in the functions of the fission protein Drp1 through the use of siRNA (small interfering RNA) , dominant-negative Drp1  or pharmacological inhibitors  block or delay MOMP in some settings. However, MOMP can occur irrespective of changes in mitochondrial dynamics, and the detailed relationship between fission, fusion and MOMP remains tantalizing, but obscure.
Mitochondria and necrosis
Because mitochondrial function is essential for cell survival, changes in mitochondrial function can cause cell death that is due to energy collapse, leading to necrosis. This is particularly relevant to cell death as a consequence of ischaemia/reperfusion injury.
One mechanism that leads to necrosis following various forms of severe cellular stress is the MPT (mitochondrial permeability transition). A well-described, but incompletely understood, event in mitochondrial physiology is the MPT. This MPT involves the opening of a channel in the IMM, allowing solutes of less than 1500 Da to freely diffuse. As a result of MPT, the ΔΨm dissipates, and the matrix swells, sometimes rupturing the OMM. MPT can be readily produced in isolated mitochondria by exposure to high levels of calcium, and a number of factors, including ROS, ADP levels and many toxins, increase sensitivity for this effect.
The MPT is often described to involve the ANT (adenine nucleotide translocator) in the IMM, and the voltage-dependent anion channel in the OMM, but a number of studies have cast doubt on the involvement of these in the formation of the MPT pore (reviewed in ). In contrast, a matrix protein, CypD (cyclophilin D), clearly plays a prominent role in MPT. Mitochondria from mice lacking CypD are profoundly deficient in MPT . Strikingly, these animals are relatively resistant to ischaemia/reperfusion injury of the heart or brain. Importantly, developmental cell death in such mice proceeds normally and cells from these animals display normal MOMP and apoptosis in response to stress. Such studies underscore the involvement of MPT in some forms of necrotic death, but argue strongly against a major role for MPT in apoptosis or MOMP. That said, an extensive literature on MPT implicates this event in a variety of forms of cell death, mostly (but not all) necrotic, and this has been explored in more detail elsewhere [45,46].
Other roles for mitochondria in cell death have been described. As a major source of ROS in the cell, mitochondrial damage and/or defects in ROS scavenging can engage pathways leading to apoptosis or necrosis. The former involves enlisting the Bcl-2 family proteins to cause MOMP; that is, mitochondrial damage itself is not sufficient to release sufficient cytochrome c to activate caspases.
A consequence of ROS production is DNA damage that can lead to necrotic death, apparently due to depletion of NADH by the repair enzyme PARP (poly-ADP ribose polymerase). Intriguingly, cells from animals deficient in PARP are resistant to cell death induced by ROS .
As we discussed above, MOMP is followed by cytochrome c-induced caspase activation and apoptosis. However, when such caspase activation is blocked or disrupted, another MOMP-dependent form of cell death generally ensues [10,13]. This CICD has mainly been demonstrated with caspase inhibitors or genetic deletion of key components downstream of mitochondria, but may also occur naturally .
Two models of CICD have been proposed . In one, IMS proteins released upon MOMP affect CICD. These include AIF (apoptosis-inducing factor), endonuclease G and Omi/HtrA2. Difficulties in formally proving this model include the essential role for AIF in development and for mitochondrial function, and concerns that cultured cells adapted to its absence have altered requirements for mitochondrial function. Meanwhile, genetic ablation of endonuclease G or Omi/HtrA2 have not supported important roles in CICD. The alternative model suggests that a decline in mitochondrial function following MOMP produces an energetic catastrophe that results in necrotic cell death .
Mitochondria have also been implicated in the death of cells targeted by cytotoxic lymphocytes. Such death is mediated in large part by the release of a pore-forming protein, perforin, and several serine proteases called granzymes. Of these, Granzyme B has been shown to be capable of engaging the mitochondrial pathway of apoptosis via the cleavage and activation of Bid (see above). Another granzyme, Granzyme A, produces a necrotic death that appears to involve mitochondria as well. Granzyme A can be imported into mitochondria, where it cleaves a component of complex I, NDUFS3, whereupon ROS are produced and contribute to the death of the cell .
Finally, mitochondria may be involved in a form of programmed necrosis triggered by death receptors in a caspase-independent manner (and probably by other signals as well). This cell death involves the activities of two kinases, RIP (receptor-interacting protein)-1 and RIP-3 [50–52], and results in mitochondrial perturbations , although the mechanism of cell death remains obscure.
Mitochondria are highly dynamic organelles that are integral to the life and death of the cell. Specifically, permeabilization of the OMM is induced in response to, and is necessary for, apoptotic cell death induced by intrinsic stimuli. The Bcl-2 family of proteins, as well as non-Bcl-2 proteins on the inside and the outside of the mitochondria, all play a role in promoting or inhibiting MOMP. Shortly following MOMP, mitochondria display numerous bioenergetic and morphological changes, including a reduction in ΔΨm and an increase in ROS generation, among other changes, which are due to the action of active caspases. Other mitochondrial events, such as cristae remodelling via OPA1 redistribution, as well as fission and fusion of mitochondrial networks, may also impact MOMP. Also, although MPT is tightly linked to mitochondrial function and morphology, a role for MPT in MOMP or apoptosis remains obscure, although it is probably involved in some forms of necrosis. Thus multiple aspects of mitochondrial function, some well understood and some still elusive, co-ordinate to provide multicellular organisms with energy, and also hold the key to executing cell death at the appropriate time during development and homoeostasis.
• Apoptosis is a signalling cascade leading to cell death, which is characterized by distinct morphological features and occurs in response to specific physiological cues.
• The intrinsic, or mitochondrial, apoptotic signalling cascade is one of two main apoptotic pathways.
• MOMP is the key step in apoptosis that involves the mitochondria.
• Bcl-2 family proteins are both pro- and anti-apoptotic in nature, and are the main regulators of MOMP.
• Bak or Bax is necessary for MOMP.
• MOMP leads to the release of pro-death factors from the IMS, which engage caspases to carry out the apoptotic signalling cascade.
• There are two models to explain the activation of MOMP, the direct activation model and the neutralization model.
• Mitochondrial bioenergetics become altered upon MOMP due to the activation of caspases
• Mitochondrial networks undergo fission and fusion; the Bcl-2 family interacts with the fission and fusion machinery during MOMP, although the consequences of these interactions remain unknown.
• MPT is an important aspect of mitochondrial physiology, but probably does not play a major role in MOMP.
- © The Authors Journal compilation © 2010 Biochemical Society