CHAPTER 11 APOPTOSIS
CHAPTER 11 APOPTOSIS
ROBERTA A. GOTTLIEB
Features of Programmed Cell Death
Endogenous Prevention of Apoptosis
Apoptosis in Human Disease
Apoptosis is a physiologic form of cell death that has evolved in multicellular organisms as a mechanism of eliminating unwanted cells. Apoptosis is a cell-autonomous process that may be triggered through a receptor or through the detection of cellular damage. It involves a coordinated series of enzymatic steps orchestrated by activation of a special class of proteases (caspases) and is controlled by inhibitors at each step, conferring tight control over this lethal process. The cell destruction process is accompanied by alterations in most organelles—particularly mitochondria—as well as changes to the cytoskeleton, plasma membrane, and ion transport systems, and culminates in the degradation of nuclear DNA through the action of endonucleases.
Acronyms and abbreviations that appear in this chapter include: AIF, apoptosis-inducing factor; CAD, caspase-activated DNase; FADD, Fas-associated death domain; GM-CSF, granulocyte-macrophage colony-stimulating factor; ICAD, CAD inhibitor; MTP, mitochondrial permeability transition pore (MPT); TNF-a, tumor necrosis factor a; TRAIL, TNF-related apoptosis-inducing ligand.
Apoptosis is a term originally coined by Kerr, Wyllie, and Currie1 to describe a form of cell death characterized by cell shrinkage and nuclear condensation, and is derived from the Greek term for the shedding of leaves or petals. This physiologic, tightly regulated process is initiated by eukaryotic cells in response to internal or external cues. Apoptosis occurs in all multicellular organisms as the means to balance cell proliferation in continuously renewing tissues in order to maintain a constant organ size.
In the hematopoietic system, cell production is delicately balanced against cell death and removal through the monocyte-macrophage system.
A panoply of cytokines and growth factors regulate cell survival, proliferation, and apoptosis.2 Stem cell factor, Flt ligand, thrombopoietin, and IL-3 suppress apoptosis, while IL-6 and IL-11 stimulate proliferation of early progenitors. Granulocyte-macrophage colony-stimulating factor (GM-CSF) can both suppress apoptosis and trigger proliferation. Tumor necrosis factor a (TNF-a), Fas ligand, TNF-related apoptosis-inducing ligand (TRAIL), and interferon-gamma promote apoptosis of cells expressing the appropriate receptors.
Apoptosis occurs at defined times and locations during development, thus earning it the name programmed cell death. It is a critical process during embryogenesis, where remodeling requires highly regulated cell death. Programmed cell death is the process by which tadpoles lose their tails, and its occurrence has been studied in detail in the nematode Caenorhabditis elegans. Three of the most important genes that control apoptosis have been identified through studies of development in C. elegans. Two of them—designated ced-3 and ced-4 (C. elegans death gene)—were shown to be essential for programmed cell death to occur, and one gene, ced-9, was shown to be essential for opposing cell death.3,4 These genes are conserved throughout evolution and are represented by large families of mammalian homologs. Ced-3 is a cysteine protease with the unusual characteristic of cleaving peptides after aspartic acid residues. The first mammalian homolog of Ced-3 to be identified was interleukin-1b–converting enzyme (ICE). Subsequently, a family of more than 10 related cysteine proteases (“death proteases”) has been identified; they have been designated caspases, for cysteine aspartases.5,6 The nematode death gene, ced-4, encodes a protein that controls the activation of the caspase, Ced-3. Ced-4 is in turn regulated through interaction with Ced-9. Apaf-1 has been shown to be the mammalian homolog of Ced-4. Bcl-2 is the mammalian homolog of the anti-apoptosis gene, ced-9, and was first identified as an oncogene created by a chromosomal 8;14 translocation in B-cell lymphoma.7 Studies of the mammalian homologs of the C. elegans death genes have led to an understanding of the critical elements of the “death machinery” of apoptosis.
FEATURES OF PROGRAMMED CELL DEATH
Mitochondrial alterations, caspase activation, and chromatin fragmentation are among the key events that characterize apoptosis and are discussed in greater detail below. Upon initiation of the death program, cells undergo dramatic volume loss, membrane blebbing, cytoplasmic acidification, rearrangement of the cytoskeleton, and loss of contact with adjacent cells and extracellular matrix. Cells exhibit disordered ion homeostasis characterized by diminished proton elimination (and/or increased proton production), and volume loss accomplished largely through potassium and chloride efflux, which is accompanied by water loss. Calcium homeostasis is also disturbed, since mitochondrial sequestration of calcium is impaired. Membrane blebbing and phosphatidylserine externalization have been attributed to proteolytic cleavage of the membrane cytoskeletal protein fodrin (a spectrin homolog) and to activation of the phospholipid scramblase, which is activated by low pH and elevated calcium levels. Cytoskeletal alterations are due in part to proteolytic cleavage of actin, as well as to changes in the activity of kinases and G proteins that regulate the state of assembly of cytoskeletal components. A variety of signaling pathways that participate in survival signaling are proteolytically inactivated.8
The cell is marked for ingestion by neighboring cells or professional phagocytes through the upregulation of certain adhesion markers and through the externalization of phosphatidylserine. In the intact organism, apoptotic cells are removed before membrane integrity is lost, thereby preventing spillage of cellular contents. The magnitude of this clearance process is demonstrated by the clearance of inflammatory cells during the resolution of pneumonia.9
The mitochondria are complex organelles consisting of an outer membrane, an inner membrane, an intermembrane space, and the matrix, which is enclosed by the inner membrane. The outer membrane is permeable to small molecules, while the inner membrane is highly impermeable and able to maintain a proton gradient equivalent to 1 pH unit. The electron transport system is embedded in the inner membrane and oxidizes substrates in order to move protons across the inner membrane. This proton gradient is the driving force for ATP synthesis. The inner membrane, which has extensive infoldings (cristae) to increase the available sites for ATP synthesis on the matrix face of the inner membrane, has a much greater surface area than does the outer membrane. Cytochrome c, which serves as an electron carrier, is located in the intermembrane space and is electrostatically associated with the inner membrane.
In addition to their role in ATP production, the mitochondria play a key role in the regulation of apoptosis. Sequestered in the space between the inner and outer mitochondrial membranes are cytochrome c, which is a cofactor for caspase activation; some caspases; and apoptosis-inducing factor (AIF), which promotes DNA fragmentation and chromatin condensation.
Two major mitochondrial alterations occur in apoptosis: loss of cytochrome c and depolarization of the inner membrane. The earliest event is the dissociation of cytochrome c from the electron transport chain, followed somewhat later by its release from the mitochondria to the cytosol, where it may participate in caspase activation. Loss of cytochrome c from its normal association with the electron transport chain prevents delivery of electrons to complex IV and thence to oxygen. However, it may be possible for electrons to “bleed off” at the upstream site of ubiquinone, leading to superoxide production. This may explain the production of free radicals frequently observed in apoptosis. Effective mitochondrial respiration is shut down, and dissipation of the proton gradient may result in loss of membrane potential across the inner membrane. In some cases, the F0F1 ATPase may run in reverse, hydrolyzing any available ATP so as to maintain mitochondrial inner membrane potential through proton pumping. The mitochondria also become engaged in futile calcium cycling (uptake and release); the combination of the two processes may lead to a crisis state for the mitochondria, at which point the inner mitochondrial membrane becomes freely permeable to solutes, resulting in loss of mitochondrial membrane potential. This is often referred to as opening of the mitochondrial permeability transition (MPT) pore, although the event may be less specific than the term pore implies. Loss of ion homeostasis across the inner mitochondrial membrane may lead to swelling of the mitochondrial matrix compartment and eventual rupture of the outer mitochondrial membrane, leading to release of cytochrome c, caspases, and AIF.
The participation of the mitochondria in these apoptotic events is regulated by members of the Bcl-2 family, some of which oppose apoptosis while others promote apoptosis. Bcl-2, some of which opposes apoptosis, is able to prevent the release of cytochrome c from mitochondria, while the proapoptotic Bax promotes cytochrome c release.10,11 Proteolytic processing of Bid, or dephosphorylation of Bad, results in their translocation to the mitochondria, where they cause cytochrome c release. Bax has also been reported to translocate from cytosol to mitochondria during apoptosis. Since these molecules bear structural similarity to the pore-forming colicins, much attention has been directed toward their potential role as pore formers in the outer mitochondrial membrane. They may interact with the voltage-dependent anion channel, a large-conductance porin in the outer mitochondrial membrane, as well as with the adenine nucleotide transporter, an inner-membrane protein believed to be a component of the MTP pore in addition to its primary role in ATP-ADP exchange.12 Bcl-2 family members are extremely important in the regulation of cell death. Bcl-2 family members and their function (pro- or antiapoptotic) are shown in Fig. 11-1 and reviewed in ref. 13.
FIGURE 11-1 Bcl-2 family members and their role in apoptosis regulation. Bcl-2 homology (BH) domains represent conserved sequences among family members. TM designates the transmembrane domain. The Bcl-2 subfamily promotes cell survival. Proapoptotic members are grouped into the Bax subfamily and the BH3 subfamily, which has high sequence divergence outside the BH3 domain. (Adapted from JM Adams,29 with permission.)
The caspases can now be grouped into three functional categories. The first group includes ICE (also designated caspase-1) and two related caspases: caspase-4 and caspase-5. Although ICE primarily participates in cytokine processing, the function of the other members of this group is not clear. The second group consists of the effector caspases, such as caspase-3, which possess a short prodomain (<3 kDa). The effector caspases are responsible for cleaving many of the important intracellular protein substrates that are degraded during apoptosis.14 The third, and perhaps most interesting, group includes the signaling caspases, which possess a large prodomain. Most cells express multiple caspases, probably related to the observed redundancy in pathways that initiate cell death; it also appears that the caspases may work in a cascade fashion, perhaps analogous to the amplification seen in the clotting system. The caspase family is diagrammed in Fig. 11-2.
FIGURE 11-2 Phylogenetic tree for the caspase family. Caspase designations and their aliases are shown. (Adapted from ES Alnemri et al5 and EW Humke et al,30 with permission.)
Caspases are synthesized as proenzymes with an amino-terminal prodomain that is removed by proteolytic processing. The enzyme is further processed into large (»20 kDa) and small (»10 kDa) fragments, which form a heterodimer. Two heterodimers assemble to form the active tetrameric protease (Fig. 11-3). Caspases are capable of autoprocessing under certain circumstances. Recently it has been suggested that effector caspases are dependent on proteolytic activation, while the signaling caspases, such as caspase-9, depend on interaction with a cofactor, such as Apaf-1.15,16
FIGURE 11-3 Diagram of caspase processing and assembly. Members of the cysteine aspartate protease (caspase) family are characterized by an amino-terminal prodomain and a catalytic domain, which is processed by cleavage after Asp residues to release large (»20 kDa) and small (»10 kDa) subunits. The two fragments assemble into a heterodimer. Two heterodimers form a tetramer in the active form of the caspase.
Activation of caspases may be accomplished through multiple pathways, two of which have been worked out in detail (Fig. 11-4). The receptor-mediated pathway involves a cell surface receptor, such as Fas or the receptor for TNF-a. Occupancy of the receptor with its ligand causes recruitment of a cytosolic adapter molecule containing a protein-protein interaction region termed the death domain. The adapter molecule [e.g., Fas-associated death domain (FADD) or TNFR–associated death domain (TRADD)] then recruits a signaling procaspase such as caspase-8 or caspase-10, which docks with the adapter molecule and undergoes proximity-induced processing. Caspase-8 or caspase-10 may directly activate caspase-3 and related effector caspases.17
FIGURE 11-4 Two main signaling cascades lead to caspase activation. Extracellular signaling from the death receptors TNFR, Fas, and TRAIL lead to activation of caspase-8 and/or caspase-10, which leads to activation of caspase-3 and other end effectors of apoptosis. Caspase-8 and caspase-10 are inhibited by CrmA. Cell stressors and related signals lead to mitochondrial alterations resulting in caspase-3 activation through the interaction of Apaf-1, cytochrome c, and caspase-9. Bcl-2 opposes the pathway that involves mitochondria.
An alternative pathway exists involving activation of caspase-9 through interaction of Apaf-1, which must bind cytochrome c and dATP or ATP (Fig. 11-5).18 Apaf-1, which has homology to Ced-4, is present in the cytosol and inactive until cytochrome c is available for interaction. It possesses a caspase activation and recruitment domain (CARD) that is essential for its function. Additional Apaf-1-like molecules are now being identified that presumably will be activated through different pathways.
FIGURE 11-5 Model for caspase activation. Caspase-9 is activated through an interaction with Apaf-1, cytochrome c, and ATP or dATP. The reaction is normally held in check by sequestration of cytochrome c within the mitochondria. However, in response to apoptotic stimuli, cytochrome c becomes accessible to the other components. Activated caspase-9 cleaves the downstream effector caspase-3, resulting in widespread cleavage of cellular proteins.
Cytotoxic T lymphocytes inject granzyme B into cells to trigger apoptosis. Granzyme B is a serine protease that cleaves caspases after Asp residues to generate active caspases. The introduction of granzyme B into the cytosol of target cells results in the rapid activation of caspase-3 and subsequent cell death.
It is generally believed that activation of proteases constitutes an irreversible event and that the caspases are the ultimate effectors of apoptotic cell destruction. There are exceptions however. For example, caspase-3 is activated in a subset of T cells and participates in processing of IL16 without leading to apoptosis. Conversely, inhibition of caspases may not always prevent cell death. Since multiple enzymatic pathways are activated in apoptosis, it is reasonable to think that completion of any subset of processes will result in the death of the cell or, at the very least, in loss of its ability to divide.
The classic histologic manifestations of apoptosis are the condensation of nuclear chromatin and fragmentation of the nucleus. DNA condensation and classic apoptotic body formation depend on proteolysis of lamin by one or more caspases.19 DNA digestion is accomplished by several distinct endonucleases, eventually resulting in fragments representing multiples of the approximately 200 base pair nucleosome (the so-called nucleosomal ladder). The search for the responsible endonucleases has yielded several candidates, including DNase I and DNase II. DNA fragmentation factor (DFF) consists of a 40-kDa caspase-activated DNase (CAD) bound to its 45-kDa inhibitor (ICAD). Cleavage of CAD releases it from ICAD, thereby enabling it to function as an endonuclease.20 Although DNA fragmentation is commonly observed in apoptosis, it is not an essential feature.
In addition to the characteristic morphologic changes in the nucleus, it is possible to detect DNA fragmentation using a histologic method known as terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL), in which labeled deoxynucleotides are incorporated into nuclear DNA at sites of nicking. The incorporated nucleotides are then detected using conventional histochemical staining or fluorescence detection. This method has become widely utilized to detect DNA fragmentation in tissue sections and in cell suspensions evaluated by flow cytometry.
ENDOGENOUS PREVENTION OF APOPTOSIS
Cells have evolved a variety of safeguards to prevent inappropriate apoptosis. Viruses have also exploited these safeguards to prevent the cell from undergoing apoptosis in response to the presence of the virus. Bcl-2, which opposes apoptosis, has corresponding viral homologs, including E1B-19K. The cell also has inhibitors of apoptosis (IAPs), which bind to caspases and inhibit their activity. The cowpox response modifier protein, CrmA, is a viral gene product that performs the same function. Transcriptional regulation of antiapoptotic genes is mediated in part through NFkB, which is mimicked by the viral transcription factor v-Rel.
APOPTOSIS IN HUMAN DISEASE
The occurrence of apoptosis in pathophysiologic settings or its absence in physiologic settings result in human disease. More simply, diseases can be grouped according to whether there is too much apoptosis or too little.
Since apoptosis must occur at defined times during development, a failure of apoptosis to occur in the appropriate settings would be expected to give rise to developmental defects. However, genetically defined abnormalities in known elements of the process of apoptosis have not been identified in human developmental disorders. Some insights have been derived from gene knockout studies in mice. Deletion of the gene encoding caspase-3, arguably the most important death protease, results in mice that die in utero or soon after birth with an excess of brain tissue, owing to a failure of normal programmed cell death during neuronal development.
In the immune system, deletion of self-reactive T cells is essential to prevent autoimmune disorders.21 Signaling for lymphocyte deletion is accomplished through engagement of one or more cell surface receptors, including a molecule known as Fas/APO-1/CD95. Engagement of Fas by its ligand results in aggregation of proteins (FADD and FLICE) through self-association regions known as death domains, culminating in caspase activation and death of the cell. Fas is widely expressed on lymphocytes and is one means by which unwanted T cells are eliminated. Mutation of Fas or its ligand in mice results in a disease that strongly resembles systemic lupus erythematosus. In humans, mutations of Fas occur in the heritable autoimmune lymphoproliferative syndrome, in which CD3+CD4–CD8– T cells fail to undergo apoptosis and contribute to autoimmune disease.22
In areas of immune sanctuary, such as the testis, Sertoli cells express high levels of Fas ligand to prevent invading T cells from surviving long enough to mount an immune response against sperm cells (which are recognized as foreign by the body). A similar mechanism of protection is involved in limiting inflammatory responses in viral infections in sensitive organs such as the eye. Immune-mediated rejection of transplanted organs rests in part on the induction of apoptosis in the foreign cells. This mechanism has been exploited by genetic manipulation to express Fas ligand on pancreatic islet cells to prevent induction of apoptosis in the transplanted cells, with resulting prolonged survival of the allograft.
A number of viral proteins block apoptosis signaling or effector pathways. Baculovirus p35 and cowpox viral protein CrmA directly inhibit caspases; adenovirus E1B inhibits caspase-3 activation, and herpesvirus-poxvirus FLICE inhibitor proteins block downstream death domain signaling. The function of these proteins may be critical to viral virulence by blocking host defense against viral replication; infected cells engage the apoptotic machinery and mark themselves for phagocytic ingestion, thus limiting the extent of viral infection.
Polycythemia vera is characterized by an abnormal clone of erythroid progenitors that proliferates independently of erythropoietin. These clonal cells overexpress Bcl-xL, which prevents apoptosis and may contribute to the survival of erythroid progenitors in the absence of erythropoietin.23
Tumor growth rate is determined by the imbalance between apoptosis and mitosis. For example, the function of the Bcl-2 gene product was discovered because its overexpression prevents the normal death of B cells, leading to a lymphoma associated with a normal rate of proliferation but reduced apoptosis.24 A malignant cell may arise when a cell fails to undergo apoptosis when it should have. Loss of a necessary growth factor or removal from the normal extracellular matrix should trigger a cell to commit suicide. If, however, the cell fails to die, it may survive and proliferate sufficiently for its progeny to acquire other mutations, including loss of p53 and activation of other oncogenes. The p53 gene product is a transcription factor activated by DNA damage to induce a family of p53-dependent genes that regulate the cell cycle and induce apoptosis.25 Mutations of p53 have been found in many malignant tumors and in some families with hereditary cancer syndromes. Thus, the first step in oncogenesis may be a failure of apoptosis. Chronic myelogenous leukemia represents another example of the blockage of normal apoptosis due to the effects of the BCR-ABL oncogene. Modulation of apoptosis is widely considered a key target for cancer therapy.26 Although malignant cells are generally considered more resistant to the induction of apoptosis, they still possess the necessary cellular machinery and, when exposed to appropriate chemotherapeutic agents (or radiation), usually die by apoptosis, not necrosis.27 Efforts to decrease their resistance to apoptosis are directed at targets such as Bcl-2. Evaluation of apoptosis in response to chemotherapeutic agents may correlate with prognosis and might eventually direct the selection of agents on an individualized basis.
One hypothesis about mechanisms of aging is that too little apoptosis occurs, permitting the survival of cells that have sustained DNA damage. Such damaged cells would function inefficiently at best, owing to the accumulation of mutations in essential genes, and could undergo malignant transformation. Eventually, such marginally functioning and precancerous cells would predominate, with more generalized cellular dysfunction as time went on.
Excessive cell death is of particular concern in organs that are populated by terminally differentiated, nondividing cells. Any cells lost, whether by apoptosis or necrosis, are irreplaceable. In settings where cell death is inevitable, inhibiting the enzymatic processes of apoptosis may not salvage the cell but may merely convert its demise to a necrotic form. However, if a cell is damaged beyond repair, a tidy, noninflammatory apoptotic death may still be preferable, avoiding collateral damage from inflammation.
Excessive apoptosis is now being recognized in a variety of hematopoietic disorders. In some cases, this may be due to unavailability of a necessary growth factor, an inability to respond to the growth factor, or alterations in the balance of proapoptotic and antiapoptotic Bcl-2 family members. The myelodysplastic syndrome, which is characterized by peripheral cytopenias and (at least in the early stages) marrow hyperplasia, is now recognized to be associated with excessive apoptosis in the later stages of myeloid differentiation, resulting in ineffective myelopoiesis. Eventually, apoptosis-resistant clones often emerge, with concomitant progression to acute myelogenous leukemia. Fanconi’s anemia is accompanied by increased susceptibility to apoptosis mediated by Fas and TNF-a. It seems likely that abnormalities in apoptosis will be recognized as features of additional myeloid disorders.28
Wyllie AH, Kerr JFR, Currie AR: Cell death: the significance of apoptosis. Int Rev Cytol 68:251, 1980.
Wickremasinghe RG, Hoffbrand AV: Biochemical and genetic control of apoptosis: relevance to normal hematopoiesis and hematological malignancies. Blood 93:3587, 1999.
Ellis HM, Horvitz HR: Genetic control of programmed cell death in the nematode C. elegans. Cell 44:817, 1986.
Metzstein MM, Stanfield GM, Horvitz HR: Genetics of programmed cell death in C. elegans: past, present and future. Trends Genet 14:410, 1998.
Alnemri ES, Livingston DJ, Nicholson DW, et al: Human ICE/CED-3 protease nomenclature [letter to the editor]. Cell 87:171, 1996.
Li H, Yuan J: Deciphering the pathways of life and death. Curr Opin Cell Biol 11:261, 1999.
Korsmeyer SJ: Chromosomal translocations in lymphoid malignancies reveal novel proto-oncogenes. Annu Rev Immunol 10:785, 1992.
Wolf BB, Green DR: Suicidal tendencies: Apoptotic cell death by caspase family proteinases. J Biol Chem 274:20049, 1999.
Savill J, Haslett C: Granulocyte clearance by apoptosis in the resolution of inflammation. Semin Cell Biol 6:385, 1995.
Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD: The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science 275:1132, 1997.
Yang J, Liu X, Bhalla K, et al: Prevention of apoptosis by Bcl-2: release of cytochrome c from mitochondria blocked. Science 275:1129, 1997.
Tsujimoto Y: Role of Bcl-2 family proteins in apoptosis: apoptosomes or mitochondria? Genes Cells 3:697, 1998.
Reed JC: Bcl-2 family proteins. Oncogene 17:3225, 1998.
Tewari M, Quan LT, O’Rourke K, et al: Yama/CPP32, a mammalian homolog of CED-3, is a Crm-A-inhibitable protease that cleaves the death substrate poly(ADP-ribose) polymerase. Cell 81:801, 1995.
Salvesen GS, Dixit VM: Caspases: intracellular signaling by proteolysis. Cell 91:443, 1997.
Salvesen GS: Programmed cell death and the caspases. APMIS 107:73, 1999.
Ashkenazi A, Dixit VM: Death receptors: signaling and modulation. Science 281:1305, 1998.
Li P, Nijhawan D, Budihardjo I, et al: Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479, 1997.
Lazebnik YA, Takahashi A, Moir RD, et al: Studies of the lamin proteinase reveal multiple parallel biochemical pathways during apoptotic execution. Proc Natl Acad Sci USA 92:9042, 1995.
Sakahira H, Enari M, Nagata S: Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis. Nature 391:96, 1998.
Los M, Wesselborg S, Schulze-Osthoff K: The role of caspases in development, immunity, and apoptotic signal transduction: lessons from knockout mice. Immunity 10:629, 1999.
Straus SE, Sneller M, Lenardo MJ, Puck JM, Strober W: An inherited disorder of lymphocyte apoptosis: the autoimmune lymphoproliferative syndrome. Ann Intern Med 130:591, 1999.
Fernandez-Luna JL: Apoptosis and polycythemia vera. Curr Opin Hematol 6:94, 1999.
Hockenberry DG, Nunez C, Milliman RB, Schreiber RD, Korsmeyer SJ: Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature 348:334, 1990.
el-Deiry WS: Regulation of p53 downstream genes. Semin Cancer Biol 8:345, 1998.
Hannun YA: Apoptosis and the dilemma of cancer chemotherapy. Blood 89:1845, 1997.
Brown JM, Wouters BG: Apoptosis, p53, and tumor cell sensitivity to anticancer agents. Cancer Res 59:1391, 1999.
Haurie C, Dale DC, Mackey MC: Cyclical neutropenia and other periodic hematological disorders: a review of mechanisms and mathematical models. Blood 92: 2629, 1998.
Adams JM, Cory S: The Bcl-2 protein family: arbiters of cell survival. Science 281:1322, 1998.
Humke EW, Ni J, Dixit VM: ERICE, a novel FLICE-activatable caspase. J Biol Chem 273:15702, 1998.
Copyright © 2001 McGraw-Hill
Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, and Uri Seligsohn