Molecular Biology of the Cell - Volume 2 - Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. 1993

Cell Growth and Division
Regulation of Cell Division in Multicellular Organisms

In single-celled organisms such as Yeasts, Bacteria, or Protozoa, natural Selection favors Cells that grow and divide as rapidly as possible. Consequently, The rate of Cell Division is typically limited only by the speed at which nutrients can be absorbed from the environment and converted into cellular material. In contrast, the cells of Multicellular animals are specialized and form a complex community, where the primary objective is the survival of the Organism as a whole rather than the survival or reproduction of individual cells. For a multicellular organism to survive, some of its cells must refrain from dividing even when nutrients are abundant. However, when new cells are needed—for example, to repair damage—cells that were previously non-dividing must rapidly switch to a division cycle; conversely, in cases of continuous tissue "wear and tear," the rates of cell production and death must always be tightly balanced. Therefore, complex regulatory mechanisms of a much higher order than those operating in simple organisms like Yeast must exist. This section is devoted to such "social control" at the single-cell level. In Chapters 17 and 21, we will explore how this system Functions in multicellular systems to maintain and renew body Tissues, and how its failures lead to Cancer, while in Chapter 16 we will see how an even more sophisticated system governs cell division during embryonic development.

13.3.1. Differences in cell division rates are driven by variations in the duration of the post-mitotic pause [16]

Human body cells, numbering up to 10¹³, divide at vastly different rates. Neurons and Skeletal Muscle cells do not divide at all; others, such as Liver cells, typically divide only once every year or two, whereas certain intestinal epithelial cells divide more than twice a day to ensure the continuous renewal of the gut lining (Fig. 13-22). Most vertebrate cells fall somewhere between these extremes: they retain the capacity to divide, but usually do so infrequently. Almost all variations in cell division frequency are due to differences in the length of the gap between mitosis and the S phase; slowly dividing cells arrest after mitosis for weeks, months, or even years. Conversely, the time required for a cell to progress from THE START OF the S phase through the end of mitosis is very brief (typically 12 to 24 hours in mammals) and remarkably constant, regardless of the interval between successive divisions.

Class="center">

Fig. 13-22. Cell division and migration in the epithelial lining of the mouse Small Intestine. All cell divisions occur exclusively in the lower regions of tubular epithelial invaginations known as crypts. Newly formed cells migrate upward, forming the epithelium of the intestinal villi, where they carry out the Digestion AND ABSORPTION of nutrients from the gut lumen. Most epithelial cells have a short lifespan and are shed from the tip of the villus no later than five days after leaving the crypt. However, a ring of approximately 20 slowly dividing "immortal" cells (whose nuclei are highlighted in a darker tone) remains anchored at the Base of the crypt. These so-called stem cells yield two daughter cells upon division: on average, one remains in place and continues to function as an undifferentiated stem cell, while the other migrates upward, where it differentiates and becomes part of the villus epithelium. (Adapted from C.S. Potten, R. Schofield, L.G. Lajtha, Biochim. Biophys. Acta 560: 281-299, 1979.)

The time cells spend in a non-proliferating state (the so-called G₀ phase) varies depending not only on their cell type but also on physiological circumstances. Sex Hormones stimulate cells in the uterine wall to divide rapidly for a few days during each Menstrual cycle to replace tissue lost during menstruation; Blood loss triggers the proliferation of blood cell precursors; and liver damage prompts surviving hepatocytes to divide once or twice a day until the lost tissue mass is restored. Similarly, epithelial cells surrounding a wound begin intensive division to repair the damaged tissue (Fig. 13-23).

To regulate the proliferation of each cell type according to physiological demand, finely tuned and highly specific mechanisms exist. However, while Structure/19.html">The Importance of such regulation is obvious, its mechanisms are difficult to analyze within the complex context of a whole organism. Therefore, detailed studies of cell division control are typically performed using cell cultures, where external conditions can be easily manipulated and cells can be observed continuously over long periods.

Fig. 13-23. Epithelial cell proliferation in response to a wound. The lens epithelium was injured with a needle, and at specific time points ³H-thymidine was added to label cells in the S phase (highlighted); the tissues were then fixed and prepared for radioautography. In the diagrams on the left, S-phase cells are highlighted and M-phase cells are marked with crosses; the black spot in the center indicates the site of the wound. The stimulation of cell division gradually spreads outward from the wound, recruiting resting G₀ cells into the Cell Cycle, which results in an unusually robust response to a relatively minor injury. In the 40-hour preparation, cells located far from the wound are entering the S phase of their First Division cycle, whereas cells near the wound itself are entering the S phase of a Second Division cycle. The drawing on the right corresponds to the boxed area in the left diagram and is based on a photomicrograph of a 36-hour preparation stained to reveal cell nuclei. (After C. Harding, J.R. Reddan, N.J. Unakar, M. Bagchi, Int. Rev. Cytol. 31: 215-300, 1971.)

13.3.2. When growth conditions become unfavorable, animal cells, like yeast cells, arrest at a critical point in G₁—the restriction point [17]

Studies of The Cell cycle in vitro generally rely on established cell lines (Section 4.3.4) capable of indefinite proliferation. These are lines specifically selected for maintenance in culture; many of them—so-called untransformed cell lines—are widely used as models for normal somatic cell proliferation.

Fibroblasts (such as various mouse 3T3 cell lines) typically divide faster when plated at a low density in culture dishes using a nutrient-rich medium supplemented with serum—the fluid obtained by blood clotting after removal of insoluble clots and Blood Cells. Depriving the culture of essential nutrients, such as Amino Acids, or adding a Protein Synthesis inhibitor causes the cells to behave much like the nutrient-starved yeast cells described above: the average duration of the G₁ phase increases, while the rest of the cell cycle remains largely unaffected. Once a cell has passed through G₁, it inexorably proceeds through the S, G₂, and M phases without delay, regardless of environmental conditions. This late G₁ transition point is often called the restriction point (R) because it is the precise juncture where the cell cycle can still be halted if external conditions preclude further progression. The restriction point corresponds to the "start" point in the yeast cell cycle; much like in yeast, it may partly serve as a mechanism for regulating cell size. However, in higher eukaryotes, its function is more complex than in yeast, and the G₁ phase may feature several subtly distinct restriction points associated with different MECHANISMS OF CELL proliferation control.

Fig. 13-24. The Variability in cell cycle duration typically observed within a homogeneous cell population in vitro. Such data are obtained by tracking individual cells under a Microscope and directly recording the elapsed time between successive divisions.

13.3.3. The cycle duration of proliferating cells appears to be probabilistic in nature [18]

Individual dividing cells in culture can be continuously monitored using time-lapse cinemicrography. Such observations reveal that even genetically identical cells exhibit significant variability in cycle length (Fig. 13-24). Quantitative analysis demonstrates that the time interval from one division to the next contains a stochastically varying component, which fluctuates primarily due to variations in the G₁ phase. Apparently, as cells approach the restriction point in G₁ (Fig. 13-25), they must "wait" for a certain period before proceeding to the remainder of the cycle, with the probability per unit time of crossing the R point being roughly equal for all cells. Thus, cells behave analogously to atoms undergoing radioactive decay: if half the cells cross the R point in the first three hours, half of the remaining cells will cross it in the next three hours, half of those remaining in the next three, and so on. A plausible mechanism explaining this behavior was proposed earlier in the context of S-phase activator production (Section 13.1.5). However, these random fluctuations in cell cycle duration mean that an initially synchronized cell population loses its synchrony after several cycles. While inconvenient for researchers, this stochastic behavior may be advantageous for a multicellular organism: otherwise, large clones of cells might undergo mitosis simultaneously, and since cells typically round up and lose strong intercellular adhesions during mitosis, this would severely disrupt the integrity of the tissue.

13-15

13-16

13.3.4. Different cell types require distinct growth factors for proliferation [19, 20]

The conditions required for an animal cell to grow and divide are considerably more complex than those for yeast. If vertebrate cells in a standard artificial culture medium are completely deprived of blood serum, they will in most cases fail to pass the restriction point—even when all essential nutrients are present—and will cease to grow as well. Assays reveal that the indispensable components of serum are highly specific Proteins present at very low concentrations (on the order of 10^-9 to 10^-11 M). Different cell types require distinct combinations of these proteins. Some Serum proteins are directly and specifically involved in stimulating cell division and are known as growth factors. A prime example is platelet-derived growth factor (PDGF). The method used for its isolation was suggested by the observation that cultured fibroblasts divide in the presence of serum, but fail to divide in plasma—the fluid component of blood obtained by removing blood cells while preventing clotting. During blood clotting, Blood Platelets (Section 17.5) release the contents of their secretory vesicles, and PDGF is among the substances released (along with clotting factors). PDGF is largely responsible for enabling fibroblast division in culture. It presumably exerts a similar action in vivo, stimulating Connective Tissue and smooth muscle cells to divide during wound healing (Fig. 13-26). Cells that respond to PDGF possess specific Plasma Membrane Receptors for it (as well as for other growth factors). Other cell types express distinct sets of receptors that interact with different growth factors (Table 13-1); some of these factors are also present in serum.

Fig. 13-25. This diagram illustrates that rapidly dividing mammalian cells in culture arrest temporarily in late G₁ at a point that may correspond to the restriction point (R). If Protein synthesis is blocked, cells can remain at this checkpoint indefinitely. Continuous microscopic observation shows that genetically identical cells—including two daughter cells derived from the same division—frequently pause for varying durations before passing the R point. This suggests that the delay is governed by a stochastic process (see text).

Fig. 13-26. The putative role of platelet-derived growth factor (PDGF) in wound healing. PDGF is secreted at the injury site by blood platelets and macrophages, and potentially also by endothelial and smooth muscle cells of damaged Blood Vessels. It induces the proliferation of fibroblasts and smooth muscle cells, stimulates fibroblasts to synthesize additional Extracellular matrix, and acts as a chemoattractant for both fibroblasts and macrophages. Wound healing is a complex process in which many other factors participate alongside PDGF.

Because growth factors are secreted in minute quantities, they are difficult to isolate. This difficulty is compounded by The complexity of their action, as most cell types appear to respond to specific combinations of growth factors rather than to any single factor in isolation. Although relatively few distinct growth factors have been characterized to date (fewer than 30), many have been independently rediscovered under different conditions and assigned new names, only for researchers to later realize they were already known molecules. This suggests that there may be only a limited repertoire of growth factors which, acting in various combinations, selectively regulate the proliferation of each of the numerous cell types in higher animals. Furthermore, it is becoming clear that these same factors can act under specific conditions as regulators of other processes, particularly Cell Differentiation. Some growth factors circulate in the bloodstream, but most act as local chemical mediators. The class of local chemical mediators likely includes A large number of poorly understood factors that help regulate cell division and differentiation during embryonic development (Section 16.2.3). In addition to growth factors that stimulate cell division, there are opposing factors that inhibit it, although the latter are generally less well characterized.

Table 13.1. Certain growth factors and their actions

13.3.5. Neighboring cells compete for growth factors [20, 21]

Through growth factors such as PDGF, cells of one type can control the proliferation of cells of another type. However, it is also important that Cells of the same type within a tissue interact with one another and coordinate their division rates to maintain an appropriate population density. This sort of "social" control is clearly evident in responses to tissue injury. For example, when the epithelium is damaged, cells at the margins of the wound are stimulated to divide (see Fig. 13-23) and migrate over the exposed surface until it is covered once again; at that point, rapid cell proliferation and movement cease. A similar phenomenon can be observed with dissociated cells in culture. Epithelial cells or fibroblasts placed in a culture dish in the presence of serum will attach to the surface, spread, and divide until a confluent monolayer is formed in which neighboring cells make contact. Thereafter, normal cells stop dividing—a phenomenon known as density-dependent inhibition of proliferation. If such a monolayer is "wounded" with a needle to create a cell-free strip on the dish, the cells at the edges of this strip begin to migrate into the vacant space and divide (Fig. 13-27).

Fig. 13-27. Cells scattered across The surface of a culture dish proliferate normally until they merge into a continuous monolayer. The sequence of events following the scraping away of a strip of cells is shown. Cells at the edges of the "wound" flatten and resume growth and division, which continue until the gap between them is filled. When the monolayer once again becomes continuous, cell proliferation ceases almost entirely.

Initially, these phenomena were attributed to "contact inhibition" of cell division, but this apparently does not capture The Essence of the process. The cell population density at which cells in a confluent monolayer stop dividing increases with the concentration of growth factors in the medium. Furthermore, it has been shown that if culture medium is flowed across the surface of a dish containing patches of cells, cells bathed by the medium that has just passed over other cells divide more slowly than those bathed by medium that has passed over cell-free areas. This gives the impression that the medium flowing over the cells is depleted of certain essential nutrients or growth factors. Incidentally, this could have been predicted. Indeed, PDGF is typically present in the medium at a concentration of about 10^-10 M (roughly one molecule in the volume of a sphere 3 µm in diameter). A single fibroblast has about 10⁵ PDGF receptors, each of which exhibits a very high affinity for the growth factor. Thus, each cell has enough receptors to bind all the PDGF molecules within a spherical volume roughly 150 µm in diameter. Moreover, it is believed that a significant fraction of the PDGF bound to cell-surface receptors is rapidly internalized via endocytosis and degraded (sec. 6.5.12). It is clear from this that neighboring cells compete with one another for even trace amounts of growth factors. Such competition would be important both for cells within a tissue and for cultured cells, as it would prevent population growth beyond a certain density threshold.

13-14

13.3.6. Normal animal cells in culture stop dividing when detached from their substratum [22]

Competition for growth factors and nutrients is not the only factor influencing the division rate in cell culture. Cell shape during spreading and movement over the substratum into vacant spaces also strongly affects their ability to divide. When normal cells are cultured in suspension—where they are not attached to a solid surface and therefore remain rounded—they almost never divide (anchorage dependence of division). The Effect of cell spreading on proliferation can be demonstrated by growing cells on substrata with varying surface adhesiveness or on substrata possessing only tiny adhesive patches to which a cell can attach but upon which it cannot spread. The rate of cell division increases with the degree of cell spreading. It is possible that highly spread cells can capture more growth factor molecules and take up more nutrients simply due to their larger surface area. However, certain cell types (such as 3T3 cells), which are virtually incapable of proliferation in suspension, readily divide as soon as they manage to contact a patch of substratum, even if that patch is so small that the cell cannot spread on it (Fig. 13-28). Such "focal" contacts serve as sites of linkage (albeit indirect) between intracellular Actin filaments and extracellular matrix molecules (sec. 11.2.8). These and other observations strongly suggest that the control of cell division is somehow tied to the Organization OF THE Cytoskeleton. Although the mechanism and functions of this link remain unclear, it seems likely that the anchorage dependence of cell division helps preserve tissue integrity and prevents the proliferation of cells that have broken away from their normal environment.

Fig. 13-28. Dependence of cell division on Cell size and anchorage. In the experiment illustrated here, cells are grown either in suspension or attached to patches of adhesive material (palladium) on a non-adhesive surface; the size of these patches determines the extent of cell spreading and whether division can occur. ³H-thymidine was added to the culture medium, and 1–2 days later the cultures were fixed and autoradiographs were prepared to determine the percentage of cells that had entered S phase. A. Rounded 3T3 cells in suspension divide very rarely, but attachment to a very small patch—one that does not permit the cell to spread—allows them to divide much more frequently. B and C. Scanning electron micrographs comparing a cell spread on a large adhesive patch with a cell attached to a small patch. (Photographs from C. O'Neill, P. Jordan, G. Ireland, Cell 44: 489–496, 1986. Copyright Cell Press.)

13.3.7. Cell division is accompanied by changes in cell junctions

The relationship between cell anchorage and cell division has several facets. On the one hand, most normal vertebrate cells require anchorage to pass the restriction point; on the other hand, for cells that have already passed this point, anchorage is not strictly required to complete the division cycle—they typically lose their contacts and round up as they enter M phase. This cycle of attachment and detachment presumably allows adhesive contacts, both between cells and between cells and the matrix, to be rearranged so that newly formed daughter cells can be integrated into the tissue before embarking on the next division cycle.

The relaxation of contacts appears to be a crucial feature of the proliferative behavior of most cell types. For example, the early response of fibroblasts to PDGF is marked by the disassembly of their focal contacts (sec. 13.4.6). Strikingly, the loss of growth control in cancer cells is almost invariably associated with a permanent decrease in cell adhesiveness, which also manifests as a loss of focal contacts when such cells are grown in culture. The relationship between cell division and anchorage is a highly complex problem, as we will discuss later; it also accounts for a significant gap in our understanding of the Transformation of a normal cell into a cancer cell (sec. 13.4.7).

13.3.8. Cells that are not supposed to divide enter a resting state—G₀ [17, 23]

When conditions do not favor division, a healthy cell will almost invariably reside in the G₁ phase of the cell cycle. When circumstances become favorable for mitosis, the cell resumes its progression through the cycle. For example, a cell deprived of growth factors will resume the cycle upon The addition of serum to the medium. However, following serum addition, the onset of S phase is almost always preceded by a significant delay, typically several hours longer than the total duration of G₁ in normally proliferating cells. Growth factor deprivation plunges the cell into a non-proliferating, heavily altered state in which it cannot pass the restriction point. Emerging from this state is a complex, time-consuming process consisting of a series of stages that differ in their sensitivity to growth factors.

The Link Between serum deprivation and the cell division cycle was elucidated through studies of 3T3 cells in culture. The restriction point in the cell cycle can be detected 3.5 hours after the completion of mitosis. Depriving cells of serum (or treating them with a protein synthesis inhibitor) for as little as one hour prior to this point arrests the cell in G₁ and results in an 8-hour delay (operating on an all-or-none basis) upon subsequent serum readdition before the cycle resumes. The same serum deprivation after the restriction point causes no delay in the current division cycle, though such a delay appears during the passage through G₁ of the following cycle. These observations can be interpreted quite simply. In the proliferative state, a cell contains a set of molecules that allow it to pass the restriction point; once this point is passed, these molecules, while usually persisting, are no longer required for the S, G₂, and M phases that follow automatically. These "division-licensing" molecules are rapidly degraded during periods of serum deprivation and take considerably longer to resynthesize after serum is restored. Degradation can occur during any phase of the cycle, but its consequences do not manifest until the cell reaches the restriction point. If this is the case, the state of the cell is determined by two independent parameters: (1) the phase of the chromosomal cycle and (2) the presence or absence of division-licensing molecules—that is, the Factors Determining the cell's proliferative state. A cell lacking the "permission" to divide will be unable to pass the restriction point and will arrest there; it is said to be halted in a resting state, or the G₀ state.

Cell culture studies reveal The Role of external factors in reversibly governing the choice between proliferation and rest. In Multicellular Organisms, however, many cell types enter the G₀ state As a result of terminal differentiation and lose the capacity to divide regardless of external stimuli. The formation of such cells is typically regulated by specialized mechanisms involving stem cells, which will be discussed in Chapter 17.

13.3.9. The chromosomal cycle can become independent of cell growth [24]

While resting with respect to the chromosomal cycle, a cell in the G₀ state generally differs from a proliferating cell in the balance between protein Synthesis and degradation as well; whereas a dividing cell grows during G₁, a G₀ cell maintains a constant size. G₀ cells typically contain fewer Ribosomes and lower amounts of RNA than corresponding G₁ cells, and their Rate of protein synthesis is more than halved. When growth factors stimulate a G₀ cell to proliferate, The change in its protein synthesis rate generally correlates with its progression through the chromosomal cycle: much like in yeast, Cell Growth and Division are coordinated to maintain normal cell size.

Fig. 13-29. Comparison of the sizes of a mammalian retinal neuron and a lymphocyte; both cells contain the same amount of DNA. During development, the neuron grows continuously while remaining in the G₀ state. Over this period, the Cytoplasm-to-DNA ratio increases enormously (by a factor of 10⁵ for certain neurons). [W. B. Boycott, in Essays on The Nervous system (R. Bellairs and E. G. Gray, eds.), Oxford, U.K.: Clarendon Press, 1974]

However, the coupling between PROTEIN SYNTHESIS AND the chromosomal cycle is not rigid. With the appropriate combination of Protein Synthesis Inhibitors and growth factors, protein synthesis can be suppressed in cultured cells without delaying their progression through the chromosomal cycle, or conversely, protein synthesis can be stimulated without triggering cell division. Furthermore, specialized cells of various types differ widely in their nucleocytoplasmic ratio, and certain G₀ cells, such as neurons, can grow almost indefinitely without METABOLISM/36.html">DNA Replication (Fig. 13-29).

13-17

13.3.10. The transition probability into G₀ generally increases with the number of cell divisions: cell senescence [25]

In mammals and birds, most normal cells show a striking reluctance to divide indefinitely. This distinguishes them from stable cultured cell lines, such as 3T3, which appear to have undergone certain genetic changes that render them "immortal." For instance, fibroblasts taken from a human fetus undergo only about 50 population doublings when grown in a standard medium; toward the end of this period, proliferation slows down and eventually halts, and all cells, after spending some time in a quiescent state, die. Similar cells obtained from a 40-year-old individual cease dividing after roughly 40 doublings, and those from an 80-year-old after approximately 30 doublings. Fibroblasts from animals with a shorter lifespan stop dividing in culture after fewer cycles. By analogy with organismal Aging as a whole, this phenomenon has been termed cellular senescence. Cellular senescence is a puzzling phenomenon. Short programmed series of cell divisions that culminate in differentiation are a characteristic feature of embryonic development (Section 16.3.4), yet it is difficult to conceive how cells could keep a long-term count of their mitotic cycles and halt after, say, 50 divisions. According to one theory, cellular senescence is the result of a catastrophic accumulation of self-perpetuating errors in the cell's biosynthetic machinery; these errors are inconsequential under natural conditions, where most animals succumb to other causes long before a significant fraction of their cells undergoes senescence. From this perspective, cellular senescence simply reflects imperfections in cell physiology that are entirely natural given the very weak selection pressure directed toward their elimination. However, this view fails to explain how germ-line cells, "immortal" cultured cell lines, and even ordinary somatic cells under certain specialized conditions (described below) retain the capacity for endless proliferation. Another hypothesis is that cellular senescence results from a mechanism evolved to protect against cancer by limiting tumor growth. Yet such a defense would seem inefficient, as fifty division cycles are more than sufficient for a tumor of substantial size to develop. A further suggestion is that cell aging in the highly artificial environment of cell culture reflects the tendency of cell proliferation within the organism to slow down gradually with age, and that this cellular behavior evolved as a means of stabilizing adult body size.

Fig. 13-30. Demonstration of differences in the heritable capacity of cells for division. Individual cells within a clone, even when genetically identical, vary in the number of division cycles they can undergo. Shown here are different pairs of sister cells from a single analyzed clone, alongside histograms illustrating the number of cells undergoing a given number of divisions. If one sister cell fails to divide altogether, the other typically either fails to divide as well or undergoes few divisions (left); conversely, if one sister cell divides 8 or more times, the other usually undergoes 8 or more divisions as well (right). This example reveals heritable differences between genetically identical cells in the number of division cycles they are capable of executing. However, these distinct heritable states are not entirely stable, such that sister cells occasionally behave differently. Further studies have demonstrated that as a population ages, cells undergo stochastic transitions to a reduced capacity for division. (Based on J. R. Smith, R.G. Whitney, Science 207: 82–84, 1980.)

Whatever the function of cellular senescence, abundant evidence indicates that this process is strongly influenced by extracellular environmental factors. For example, epidermal cells from infant Skin senesce after approximately 50 division cycles in the absence of epidermal growth factor, and after about 150 cycles when this factor is present. "Immortal" 3T3 cells exhibit signs of senescence when growth factors are depleted. Cells from normal mouse embryos can continue to divide indefinitely without the slightest sign of senescence if placed in a chemically defined medium containing a mixture of purified growth factors instead of serum; conversely, the addition of serum leads to a cessation of proliferation. This suggests that senescence is partly attributable to serum components that inhibit cell proliferation by counteracting the action of growth factors.

Certain inherited human disorders, such as Werner syndrome, lead to premature aging. Fibroblasts obtained from patients who typically die before the age of 50 cease dividing in culture after an unusually small number of mitotic cycles. Interestingly, these fibroblasts are unresponsive to PDGF and fibroblast growth factor, yet they can proliferate vigorously in response to other growth factors.

Although cell senescence occurs at a predictable time within a cell population under specific conditions, it is not rigidly programmed at the level of the individual cell. Within a clone of apparently identical normal fibroblasts growing under standard conditions, some cells divide numerous times, whereas others divide only a few times. Individual cells apparently stop dividing as a result of a stochastic transition into an alternative state, the probability of which increases in each successive cell generation until a point is reached where no dividing cells remain in the population (Fig. 13-30).

Studies on cell clones indicate that cells identical in all other respects differ in their division capacity. Apparently, the transition of a senescing cell into a non-proliferating state is simply the end result of a series of temporally stochastic steps involving the deterioration of Gene Expression that regulates the readiness of cells to pass the restriction point. The molecular nature of these events is beginning to be elucidated through The Study of cancer cells, which, among other properties, are "immortal" and immune to senescence.

Summary

Cell division in multicellular animals depends on complex "social" regulatory mechanisms, and the proliferation of various cell types is controlled by diverse combinations of protein growth factors. These act at very low concentrations, and many serve as local chemical mediators that help regulate cell population density. Furthermore, most normal cells are incapable of division without attachment to the extracellular matrix. Upon a shortage of growth factors or an inability to attach to the matrix, cells arrest after mitosis, entering a specialized quiescent state—G₀—from which they can emerge only hours later upon the addition of growth factors. Once a cell has exited G₀ and passed the restriction point in G₁, it rapidly traverses the S, G₂, and M phases independently of attachment or growth factors. In a proliferating cell population, passage through the restriction point represents an all-or-none event characterized, much like radioactive decay, by a definite probability of occurrence. In addition to the immediate control of cell proliferation, there exist long-term mechanisms that lead to senescence and the cessation of division in normal mammalian somatic cells in culture after a limited number of division cycles.