Programmed cell death, including apoptosis, is gene-directed. “The word comes from two Greek words, apo- and ptosis-, and the p is silent,” declares Jonathan C. Busser, a researcher in the department of neurology and neurosurgery at Case Western Reserve University School of Medicine. “Apo” means “separate from” and “ptosis” means “fall from”–a description of cells that naturally, and without any inflammatory fanfare, die as part of normal development, he explains.
The steps of apoptosis are distinctive. The cell forms a tight sphere and its membrane undulates, resulting in bulges called blebs. The nuclear membrane breaks, and endonucleases clip chromosomes where the DNA peeks out from protective proteins. This occurs at 180-base intervals, so the DNA pieces are all the same size. Then the cell fragments, with enough membrane sequestering toxic cell contents to prevent inflammation at the site. Finally, nearby cells consume the remains. (In contrast to this process is necrosis, a nonprogrammed form of cell death that is a response to injury, in which the cell swells and bursts, causing inflammation.)
The pieces of the cellular death machinery are present in the cytoplasm, proven by the fact that cells whose nuclei are removed can still undergo apoptosis. A hypothesized “death signal” activates the process. Apoptosis is so fast that researchers often can’t detect it, let alone sort out the sequence of events. “Once it starts, apoptosis probably takes from a few minutes to an hour,” says Douglas Green, head of the division of cellular immunology at the La Jolla Institute for Allergy and Immunology in California.
On a cellular level, apoptotic cells are visualized with vital dyes and electron microscopy. On a molecular level, electrophoresis gels are used to display the telltale same-sized DNA pieces, which resemble ladders. Alternatively, the DNA pieces can be detected by labeling their 3_ ends with a biotinylated thymine analog. “Cells that stain brightly are the ones with large numbers of 3_ ends. If there’s no bright stain, there’s no apoptosis,” comments Busser.
Apoptosis as part of normal development is a strategy to select certain cells for survival, sculpting a tissue’s specificity. In a vertebrate embryo’s limb, apoptosis carves fingers from webbing. In the developing brain it leaves behind only certain neural connections, and in the fetal thymus allows only T cells with “self” surfaces to complete development.
Later in life, apoptosis protects. Consider sunburn. A cell whose DNA is damaged by ultraviolet radiation in sunlight is either repaired or jettisoned via apoptosis–peeling (A. Ziegler et al., Nature, 372:773-6, 1994). “Such controls ensure that any one mutated cell cannot proliferate. Without this, tumors would be incredibly common,” Green explains.
Developmental biologists have long been familiar with cell death in carving a vertebrate’s digits and in insect metamorphosis. But today’s cell-death community credits a paper by University of Edinburgh researcher Andrew Wyllie and his colleagues as the seminal work in the field (J.F.R. Kerr, A.H. Wyllie, A.R. Currie, British Journal of Cancer, 26:239-57, 1972). They coined the term apoptosis, writing that it plays “a complementary but opposite role to mitosis in the regulation of animal cell populations.”
The paper created little excitement initially. “It was just one of those things in the literature that stayed dormant for 10 to 15 years. Then it was gradually rediscovered and gained recognition as a generally important mechanism,” reports L. Maximilian Buja, chairman of the department of pathology and laboratory medicine at the University of Texas Medical School at Houston.
What catapulted apoptosis into “hot topic” status was its meticulous demonstration in a tiny worm, followed by identification of death genes in other organisms (J. Sulston, H.R. Horvitz, Developmental Biology, 56:110-56, 1977). In the 1980s, the term “programmed cell death” was almost synonymous with Caenorhabditis elegans, the tiny, transparent nematode worm whose cell-death program removes precisely 131 of 1,090 cells to form the adult.
“What has pushed the field forward is Bob Horvitz’s work, which allowed us to look at the process in the worm,” says Osborne, who spent a sabbatical year in 1992 in the lab of Horvitz, a Howard Hughes Medical Institute (HHMI) investigator and a professor of biology at the Massachusetts Institute of Technology. “The fact that you have a certain number of cells and can trace their developmental fates and see what happens, and watch under the microscope and predict which cells will die, then isolate genes, has made the field blossom and flourish.”
Meanwhile, little was known about cell death in other types of animals. When David Hockenbery, Stanley Korsmeyer, and their HHMI group at the Washington University School of Medicine in St. Louis discovered that the proto-oncogene bcl-2 blocks programmed cell death (D. Hockenbery et al., Nature, 348:334-6, 1990), this and other work on bcl-2 refocused attention on apoptosis, contributing to the second blip of interest in the early 1990s.
Soon, researchers using worm genes with mutations called “ced” (for cell death abnormal) as probes identified death genes in other animals. “It was a great advance to realize that some apoptosis genes in the nematode are similar to genes in mammals,” says Hermann Steller, an associate professor of neurobiology and an HHMI investigator at MIT who recently discovered an apoptosis gene in Drosophila melanogaster (K. White et al., Science, 264:677-82, 1994).
Little is known about the many-tiered genetic control of apoptosis. Most apoptosis genes under investigation turn the process on or off.
“Apoptosis on” genes include ced-3 and ced-4 in C. elegans, and ICE and p53 in mammals. Expression of ced-3 and ced-4 is necessary for the cell death of normal worm development. The gene ced-4 encodes a novel protein, but ced-3 is a homolog of ICE (interleukin-1b converting enzyme).
Experiments demonstrate the link between the gene ICE and apoptosis. Rat fibroblasts genetically engineered to overproduce ICE die by apoptosis, and phagocytes gain ICE after gobbling apoptotic cells. Recently, Junying Yuan–an associate professor in the department of medicine at Harvard Medical School–and her colleagues, working with a chicken neuron cell culture, found that inhibiting ICE activity prevents the cells from dying when their supply of nerve growth factor is blocked (V. Gagliardini et al., Science, 263:826-8, 1994).
Another “apoptosis on” gene receiving much attention is p53. The gene, which encodes a transcription factor and is common in many human cancers, mediates cellular responses to some environmental damage. The p53 protein either temporarily halts cell division so the cell can repair altered DNA, or sends the cell to an apoptotic death.
“How p53 makes that choice is the $64,000 question. Maybe there is a threshold. If damage is minor, the cell takes the time to repair it. But if damage is above threshold, it bails out, choosing apoptosis. The threshold may be different in different cell types,” says Alexander Kamb, director of research at Myriad Genetics Inc. in Salt Lake City, Utah. Adds Steven Schreiber, an assistant professor of neurology at the University of Southern California School of Medicine: “Maybe the choice of cell death or arrest of the cell cycle depends on the proliferative capacity of the cell–if it has a certain number of cycles to go. Or, it may depend upon the stage of the cell cycle when the damage occurs.”
Work on p53 reveals that there are different means to an apoptotic end. “Our work with Tyler Jacks a professor of biology at MIT shows that p53 only kills thymocytes when the inducer to death is radiation,” says Osborne. Other thymic inducers include glucocorticoids and cross-linking T-cell receptors. In nerve tissue, two cell-surface proteins, TNF R1 and APO-1/Fas, induce apoptosis when they bind their ligands.
The bcl-2 gene is an “apoptosis off” control. In 1992, Korsmeyer discovered that bcl-2 is the mammalian equivalent of ced-9, a worm anti-death gene (S.J. Korsmeyer, Blood, 80:879-86). The proto-oncogene is an apoptosis “brake” in the skin. The gene is expressed in the basal layers, where stem cells must divide to supply more cells. In the upper layers, lack of bcl-2 protein permits apoptosis, preventing tumor formation. The bcl-2 protein binds a protein called bax. The bcl-2/bax ratio is critical to a cell’s fate. If bcl-2 is in excess, all available bax is bound, apoptosis is blocked, and the cell lives. If bax is in excess, all bcl-2 is bound, the brake is released, and the cell dies (Z. Oltvai, C. Milliman, S. Korsmeyer, Cell, 74:609-14, 1993).
Certain viruses also have apoptosis brake genes. “This keeps the cell it infects from committing suicide,” says Steller. Such genes are found in adenovirus, Epstein-Barr virus, African swine flu virus, and vaccinia.
Apoptosis’ ties to the number of cells in certain organs and tissues suggest applications in correcting medical problems stemming from particular cellular excess or deficiency. “We’ve always thought of cancer as a proliferative process. Now there’s a whole new way of thinking–the absence of cell death sets the stage for proliferation,” says Buja.
Kamb foresees applying knowledge about apoptosis to monitoring cancer treatment: “If you give chemotherapy that works through the apoptotic pathway, and can show that the tumor is not apoptosis-competent, you’d know that you’re just poisoning the patient. Paying attention to apoptosis competency in tumors, diagnostically and prognostically, may provide a way to tailor therapy.”
Ricki Lewis is a textbook author and science writer based in Scotia, N.Y.
The Scientist, Vol 9 #3 p. 15 February 6, 1995; Copyright The Scientist, Inc.
The Scientist, Vol 9 #3 p. 15 February 6, 1995; Copyright The Scientist, Inc.