In biology, apoptosis (from the Greek words apo = from and ptosis = falling, pronounced ap-a-tow'-sis) is one of the main types of programmed cell death (PCD). As such, it is a process of deliberate suicide by an unwanted cell in a multicellular organism. In contrast to necrosis, which is a form of cell death that results from acute tissue injury, apoptosis is carried out in an ordered process that generally confers advantages during an organism's life cycle. For example, the differentiation of human fingers in the developing embryo requires the cells in between the fingers to initiate apoptosis so that the fingers can separate. As will be explained further on, the way the apoptotic process is executed facilitates the safe disposal of cell corpses and fragments.
Since the beginning of the 1990s, research on apoptosis has grown spectacularly. In addition to its importance as a biological phenomenon, defective apoptotic processes have been implicated in a very wide variety of diseases. As will also be explained in this article, too much apoptosis causes cell-loss disorders, while too little results in uncontrolled cell proliferation, as in cancerous tumors.
Not all forms of PCD share the characteristic shapes (the morphology) and sequences of apoptosis, but all types of PCD are highly regulated processes.
Apoptosis can occur, for instance, when a cell is damaged beyond repair, or infected with a virus. The "decision" for apoptosis can come from the cell itself, from its surrounding tissue or from a cell that is part of the immune system.
If a cell's capability of apoptosis is damaged (for example, by mutation), or if the initiation of apoptosis is blocked (by a virus), a damaged cell can continue dividing without restrictions, developing into cancer. For example, as part of the hijacking of the cell's genetic system carried out by human papillomaviruses (HPV), a gene called E6 is expressed in a product that degrades p53 protein, which is a vital piece of the apoptotic pathway. This severe interference in the apoptotic capability of cells plays a critical role in the fact that persistent infection by oncogenic HPVs can result in the development of cervical cancer.
Stress conditions (such as starvation) as well as damage to the cell's DNA resulting from toxicity or exposure to ionizing radiation, such as ultraviolet or X-rays, can induce a cell to begin an apoptotic process. A fascinating example, resulting from damage to the genome in the cell nucleus, is cell suicide triggered by the nuclear enzyme poli(ADP-ribose) polymerase-1, or PARP-1. This enzyme plays a crucial role in maintaining genomic integrity, and massive activation of PARP-1 can deplete the cell of energy-providing molecules, an event that sends signals from the nucleus for the mitochondrion to start the apoptotic process.
In the adult organism, the number of cells within an organ or tissue has to be constant within a certain range. Blood and skin cells, for instance, are constantly renewed by their respective progenitor cells; but proliferation has to be compensated by cell death. This balancing process is part of the homeostasis required by living organisms to maintain their internal states within certain limits. Some authors and researchers like Steven Rose and Antonio Damasio have suggested homeodynamics as a more accurate and eloquent term (Damasio 1999, p. 141).
From 50 to 70 billion cells die each day due to apoptosis in the average human adult. In a year, this amounts to the proliferation and subsequent destruction of a mass of cells equal to an individual's body weight.
Homeostasis is achieved when the rate of mitosis (cell proliferation) in the tissue is balanced by cell death. If this equilibrium is disturbed, either of two things happen:
Both states can be fatal or highly damaging.
For instance, misregulation of Hedgehog (Hgg) protein signalling (see Development, below) has been implicated in several forms of cancer. Hgg, which conveys an anti-apoptotic signal, has been found to be overexpressed in pancreatic adenocarcinoma tissues.
Programmed cell death is an integral part of both plant and metazoan (multicellular animals) tissue development. It does not resemble the sort of reaction that comes as a result of tissue damage due to accident or pathogenic infection (cell death by necrosis). Instead of swelling and bursting - hence spilling their possibly damaging internal contents into extracellular space - apoptotic cells and their nuclei shrink, and often fragment. In this way, they can be efficiently phagocytosed (and, as a consequence of this, their components reused) by macrophages or by neighboring cells.
Research on chick embryos - specifically on chick neural tube development - has suggested how selective cell proliferation, combined with selective apoptosis, sculpts developing tissues in vertebrates. During vertebrate embryo development, structures called the notochord and the floor plate secrete a gradient of the signaling molecule Sonic hedgehog (Shh), and it is this gradient that directs cells to form patterns in the embryonic neural tube: cells that receive Shh in a receptor in their membranes called Patched1 (Ptc1) survive and proliferate; but, in the absence of Shh, one of the ends of this same Ptc1 receptor (the carboxyl-terminal, inside the membrane) is cleaved by caspase-3, an action that exposes an apoptosis-producing domain (see the Perspective by Isabel Guerrero and Ariel Ruiz i Altaba and the research report by Chantal Thibert et al.)
Research like the one carried out by Thibert and her colleagues has begun to clarify some of the fundamental aspects of morphogenesis, or the development of organisms from fertilized eggs to fully-developed animals and plants. It has also suggested specific answers to why normal cells carry out apopotosis when they end up outside the places they should be in body tissues.
B cells and T cells are sophisticated –and very effective– front-line players in the body's defenses against infectious agents, as well as against local cells that have acquired or developed a malignancy. In order to carry out their job, B and T cells must have the ability to discriminate "self" from "nonself", and "healthy" from "unhealthy" antigens (protein segments that make a good fit, like a key and a lock, with specialized receptors in B and T cell membranes). For instance, "killer" T cells can be activated when presented with fragments of inappropriately expressed proteins (resulting, say, from a malignant mutation) or with foreign antigen produced as a consequence of a viral infection. After becoming activated, they migrate out of the lymph nodes in which they reside, proliferate, recognize the affected cells and commit them to programmed cell death.
The receptors in immature B and T cell membranes are not tailored precisely to coincide with "known" antigen. Rather, they are generated through a highly variable process that results in an immense variety, capable of making a good fit with an astounding number of precise molecular shapes. This means that most of these immature cells can be either ineffective (because the almost random shapes of their receptors do not engage any antigen of significance), or dangerous to their own organism, because their receptors could make a good molecular fit with healthy self antigen. If they would be let loose without any further processing, many could become autoreactive and attack healthy body cells. The way the immune system regulates this process is by "deleting" both the ineffective and the potentially damaging immature cells via apoptosis.
As has just been described in the previous section on development, all tissue in multicellular animals depends on continuous receipt of survival signals. In the case of T cells, as they develop and mature in the thymus, the survival signal depends on their capability to engage foreign antigen. Those that fail in this test, amounting to about 97% of the freshly produced T cells, are committed to programmed cell death. The survivors are tested as well for potentially damaging autoimmune reactions, and those that show high affinity to healthy self antigen are killed via apoptosis.
Be aware that the above paragraphs present a highly simplified picture: the actual process in which B and T cells are driven to proliferation, differentiation or apoptosis comprises a complex interplay between positive and negative regulators
A cell undergoing apoptosis shows a characteristic morphology that can be seen under a microscope:
The dying cells that have just been described display "eat me" signals, like phosphatidylserine (PS, a phospholipid from the inner cell-membrane). Phagocytic scavengers, such as macrophages, have specialized receptors that recognize PS and carry out their disposal job in an orderly manner without eliciting an inflammatory response , .
In studies on mouse embryos lacking PS receptors conducted by Ming O. Li and colleagues , un-ingested cells undergoing apoptosis accumulated in the brain and lungs, leading to neonatal lethality. These studies show how critical is the role of PS receptor (PSR) in the development of complex organisms such as mammals.
Apoptotic messages from outside the cell (called extrinsic inducers) will be described in the next section, on biochemical execution of apoptosis.
Apoptotic messages from inside the cell (intrinsic inducers) are a response to stress, such as nutrient deprivation or DNA damage, as explained by Chiarugi and Moskowitz in their previously mentioned article on PARP-1.
Both extrinsic and intrinsic pathways have in common the activation of central effectors of apoptosis, a group of cysteine proteases called caspases, which carry out the cleaving of both structural and functional elements of the cell, resulting in the previously described morphological changes.
Caspases are normally suppressed by IAP (inhibitor of apoptosis) proteins . When a cell receives an apoptotic stimulus, IAP activity is relieved after SMAC (Second Mitochondria-derived Activator of Caspases, or its mouse homolog, called DIABLO), a mitochondrial protein, is released into the cytosol. SMAC binds to IAPs, and in doing so "inhibits the inhibitors", effectively preventing them from arresting the apoptotic process.
But before we go on to a short description of how SMAC is released, lets take a look at two well-studied extrinsically induced apoptotic processes: the TNF and the Fas pathways. Keep in mind, however, that both activating and inhibiting factors are present at each step of these pathways.
Tumor necrosis factor (TNF), a 157 amino acid inter-cellular signaling molecule (cytokine) produced mainly by activated macrophages, and is the major extrinsic mediator of apoptosis. The cell membrane has two specialized receptors for TNF: TNF-R1 and TNF-R2. The binding of TNF to TNF-R1 has been shown to fire-off the pathway that leads to activating the caspases .
Fas (a.k.a. Apo-1 or CD95), is another receptor of extrinsic apoptotic signals in the cell membrane, and belongs to the TNF receptor superfamily . The Fas ligand (FasL, the protein that binds to Fas and activates the Fas pathway) is a transmembrane protein, and is part of the TNF family. The interaction between Fas and FasL results in the formation of the death-inducing signaling complex (DISC), which contains the Fas-associated death domain protein (FADD) and caspases 8 and 10. In some types of cells (type I), processed caspase-8 directly activates other members of the caspase family, and triggers the execution of apoptosis; while in other types of cells (type II), the Fas DISC starts a feed-back loop that spirals into increasing release of pro-apoptotic factors from mitochondria (see below), and the amplified activation of caspase-8.
Downstream from TNF-R1 and Fas activation - at least in mammalian cells - a balance between pro- (like BAX, BID, or BAD) and anti-apoptotic (Bcl-Xl and Bcl-2) members of the Bcl-2 family is compromised. This balance is the proportion of pro-apoptotic homodimers that form in the outer-membrane of the mitochondrion. This homodimers (of molecules like BAK and BAX) are required in order to make the mitchondrial membrane permeable for the release of caspase activators. Just how BAX and BAK are controlled under the normal conditions of cells that are not undergoing apoptosis, is incompletely understood. But it has been found that a mitochondrial outer-membrane protein, VDAC2, interacts with BAK to keep this potentially lethal apoptotic effector under control. When the death signal is received, products of the activation cascade - such as tBID, BIM or BAD - displace VDAC2: BAK and BAX are activated, and the mitochondrial outer-membrane becomes permeable, aparently these members of the Bcl-2 family have a pore-forming domain. This results in the release of caspase activators, namely cytochrome c , ., but other molecules like SMAC or AIF are also released.
Once the cytochrome c is released, it binds with Apaf-1 and ATP which then binds to pro-caspase-9, creating a multi-protein complex known as apoptosome. The apoptosome cleaves this pro-caspase rendering the active form of caspase-9, which in turn activates effector caspase-3. (See also the articles on caspases and the Bcl-2 protein family).
The whole process requires energy and a cell machinery not too damaged. If the cell damage is between certain levels, the cell can start the earliest events of apoptosis and then continue with a necrosis.
Readers should be aware, however, that the apoptotic pathways that have been summarily described are subject to regulatory mechanisms, and that there is not a 1-to-1 relationship between the reception of TNF or FasL and the complete execution of an apoptotic pathway. Fas, for instance, has been implicated - in a seemingly ironic way - in cell proliferation, through pathways that are not yet well understood; and both Fas and TNF-R1 trigger events that activate the transcription factor nuclear factor kappa B (NF-κB), which induces the expression of genes that play an important role in diverse biological processes, including cell growth and death, development, and immune responses.
The link between TNF and apoptosis shows why an abnormal production of TNF plays a fundamental role in several human diseases, especially (but not only) in autoimmune diseases, such as diabetes and multiple sclerosis.
In their Nature article on the "Integration of interferon-alpha/beta signaling to p53 responses..." (see previous section on Cell damage or infection), Takaoka and co-workers have described their research on how interferon alpha and beta (IFN-alpha/beta)induce transcription of the p53 gene, resulting in the increase of p53 protein level and enhancement of cancer cell-apoptosis. p53 is a tumor suppressor, and is considered as a negative-growth and anti-oncogenic factor.
Work carried out by Takaoka and colleagues has contributed to clarify the role played by interferon in the treatment of some forms of human cancer, and has provided knowledge on the link between p53 and IFN alpha/beta. The p53 response not only contributes to tumor suppression, but is also important in eliciting an apoptotic response to viral infection and consequent damage to the cell's reproductive cycle.
Liling Yang et al. reported in the Feb. 15, 2003, issue of Cancer Research  the results of their work in the role played by a defective death signal in a type of lung cancer cells called NCI-H460 (human non-small cell lung cancer cells). They found that the X-linked inhibitor of apoptosis protein (XIAP) is overexpressed in H460 cells. XIAPs bind to the processed form of caspase-9, and suppress the activity of apoptotic activator cytochrome c (see previous section on biochemical execution).
The apoptotic pathway was found to be dramatically restored in H460 cells with a Smac peptide (SmacN7) that targets IAPs. Yang and her team successfully developed a SmacN7 peptide that selectively reversed apoptosis resistance - and hence tumor growth - in H460 cells in mice.
An interesting case of re-use and feed-back of apoptotic products was presented by Matthew L. Albert in a research article that won him an Amersham Biosciences & Science Prize for Young Scientists in Molecular Biology, and published in Science Online in December, 2001. Albert described how dendritic Cells, a type of antigen-presenting cells, phagocytose (that is, engulf) apoptotic tumor cells. Upon maturation, these dendritic cells present antigen (derived from the apoptotic corpses) to killer T cells, which are then primed for the eradication of cells undergoing malignant transformation. This apoptosis-dependent pathway for T cell activation is not present during necrosis, and has opened exciting possibilities in tumor immunity research.
Sydney Brenner's studies on animal development began in the late 1950s in what was to become the Laboratory of Molecular Biology (LMB) in Cambridge, UK. It was at this lab that during the 1970s and 80s, a team led by John Sulston succeeded in tracing the nematode C. elegan's entire embryonic cell lineage. In other words, Sulston and his team had traced where each and every cell in the roundworm's embryo came from during the division process, and where they ended up.
H. Robert Horvitz arrived from the US at the LMB in 1974, where he collaborated with Sulston. Both would share the 2002 Nobel Prize in Physiology or Medicine with Brenner, and Horvitz would go back to the US in 1978 to establish his own lab at the Massachusetts Institute of Technology.
Brenner's original interests were centered in genetics and in the development of the nervous system, but cell lineage and differentiation inevitably led to the study of cell fate: "One aspect of the cell lineage particularly caught my attention: in addition to the 959 cells generated during worm development and found in the adult, another 131 cells are generated but are not present in the adult. These cells are absent because they undergo programmed cell death", as Horvitz narrated in his Nobel Lecture "Worms, Life and Death", delivered on 8 Dec. 2002 .
Programmed cell death had been known long before "the worm people" began to publish their celebrated findings. In 1964 Richard A. Lockshin and Carroll Williams published their contribution on "Endocrine potentiation of the breakdown of the intersegmental muscles of silkmoths" in the Journal of insect physiology (10 p. 643), where they used the concept of "programmed cell death". Unfortulately, though, not much research was being carried out on this topic. John W. Saunders, Jr., stated the following in his 1966 contribution titled "Death in Embryonic Systems": "abundant death, often cataclysmic in its onslaught, is part of early development in many animals; it is the usual method of eliminating organs and tissues that are useful only during embryonic or larval life..." . A little further on, this author lamented that too little had been done to analyze the significance of this process. Saunders, it should be noted, recognized that he was building on earlier work by A. Glcksmann, and others.
Saunders and Lockshin reciprocally acknowledged that they benefitted from each other's work, and both pointed out the possibility that cell death might be regulated. Their observations helped to lead later work toward the genetic pathways of programmed cell death.
In a signal article published in 1972, John F. Kerr, Andrew H. Wyllie and A. R. Currie, coined the term "apoptosis" in order to differentiate naturally occurring developmental cell death, from the necrosis that results from acute tissue injury . They also noted that the structural changes characteristic of apoptosis (see the section on Morphology, above) were present in cells that died in order to maintain an equilibrium between cell proliferation and death in a particular tissue (see Homeostasis, above).
In 1991, Ron Ellis, Junying Yuan and Horvitz released a rounded and up-to-date account of research on programmed cell death in their "Mechanisms and Functions of Cell Death" . Among other important work at Horvitz's laboratory, graduate students Hilary Ellis and Chand Desai had made the first discovery of genes that encode apoptosis-inducing proteins: ced-3 and ced-4.
Ron Ellis also identified a gene with an opposite effect: ced-9. The product of this gene, CED-9, protects cells from programmed cell death, so its expression (or lack of) conveys a life-or-death decision on individual cells. As part of the same research, and not long afterwards, on February 1992, Michael Hengartner found that ced-9 had a human homolog: bcl-2 (which is not, actually, a single gene but a whole family of mammalian genes). Indeed, around four years before, in landmark research by David L. Vaux and colleagues, the anti-apoptotic and tumorigenic (tumor-causing) role of bcl-2 had been identified . Researchers had been hot in the track of oncogenes (genes that played a prominent role in causing cancer), and now more and more of the pieces were falling into place.
Horvitz would recount in his Nobel Lecture: "I believe that the fact that Bcl-2 proved to look like a worm protein that antagonized programmed cell death helped convince researchers that the function of Bcl-2 was to antagonize the cell death process. I also believe that this similarity made the worm cell-death pathway suddenly a topic of major interest in the biomedical community, as this pathway was no longer simply an abstract formalism derived from complicated genetic studies of a microscopic soil dwelling roundworm but rather a framework for a process fundamental to human biology and human disease."
In 1992, two independent teams working at pharmaceutical companies had identified and purified interleukin-1-beta converting enzyme (ICE) in human cells, and succeeded in cloning the DNA sequence of this cysteine protease , . That same year, graduate student Shai Shaham working in Horvitz's laboratory identified ICE as the mammalian counterpart of CED-3 (that is, the product of the ced-3 gene in C. elegans).
In 1997, a protein similar to CED-4 was identified, as well, at the laboratory of Xiaodong Wang (Department of Biochemistry, University of Texas Southwestern Medical Center at Dallas), which they called Apaf-1 (apoptotic protease activating factor). The team published their results in an article titled "Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3 .
Wang and his team identified and reconstituted the mitochondrial pathway to apoptosis (see Biochemical execution, above). Their published results illuminated whole new avenues of research on inflammatory diseases, cancer, and apoptosis in general.
By 1998, research on the topic had already picked a good deal of wind in its sails, as attested in the editorial "Cell Death in Us and Others", written by an important contributor to apoptosis research, Pierre Golstein, in the 28 Aug. 1998 issue of Science: "Although there have been scattered reports on the topic of cell death for more than a century, the 20,000 publications on this topic within the past 5 years reflect a shift from historically mild interest to contemporary fascination." 
Kerr, Wyllie, and Currie who coined the term apoptosis (falling leaves in Greek) meant, among other features, to remark on the de-adhesiveness of apoptotic cells from their natural surroundings, following programmed cell death. Anoikis ("homeless" in Greek) is chronologically an inverse process: de-adhesiveness of viable cells from their surroundings inducing programmed cell death. Integrins are essential adhesive molecules in this process but additional factors probably play a role. Beyond the physiological importance, understanding these patterns will be relevant to maintain the vitality of cells used for cell therapy. Abnormal apoptosis and clearance of apoptotic cells is a fundamental factor in the pathogenesis of numerous diseases including cancer, neuro-degenerative and ischemic diseases, AIDS and autoimmunity. In systemic lupus erythematosus (SLE) the antigen responsible for most anti-DNA antibodies, exclusively generated in this disease, are derived from nucleosomes. As nucleosomes are mainly generated during programmed cell death, excess of apoptotic material and altered clearance may induce autoreactive immune responses. On the other side of the spectrum, failure to die, as exemplified in MRL/1pr mice and human lymphoproliferative disorder, may allow persistence of autoreactive cells and prevent the resolution of inflammation. When combined, we may conclude that dying properly is essential for living properly.