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

Cell Growth and Division
Phases of the cell cycle and their causal interrelationships

Under the Microscope, Introduction/5.html">Eukaryotic Cell division looks like a breathtaking spectacle. During mitosis, the nuclear contents condense to form visible Chromosomes that, through a series of remarkably coordinated movements, segregate into two daughter sets; then, during cytokinesis, The Cell itself splits into two daughter Cells, each receiving one of the two chromosome sets. Because of how easily they can be observed, mitosis and cytokinesis were the primary focus of early researchers. However, these two events together comprise only a brief phase of the entire Cell Cycle, known as the M phase (from "mitosis"). The much longer period between two consecutive M phases is known as interphase. Under the microscope, interphase deceptively resembles an intermission with no visible action, as the cell merely grows slowly in size. More sophisticated Methods reveal that interphase is actually a period during which complex preparations for mitosis take place in a strictly defined sequence. In this section, we will discuss how The sequence of events in interphase can be studied and what causal relationships exist between the phases of the cell cycle.

13-3

13-4

13.1.1. Nuclear METABOLISM/36.html">DNA Replication occurs during a specific period that forms part of interphase [1]

In most cells, nuclear DNA Synthesis occupies only a fraction of interphase—a period called the S phase of the cell cycle. Typically, between the end of the M phase and THE START OF DNA synthesis, there is an interval known as the G₁ phase (from "gap"); another interval, called the G₂ phase, separates the end of DNA synthesis from the beginning of the next M phase. Thus, interphase consists of the sequential G₁, S, and G₂ phases and usually accounts for at least 90% of the total cell cycle time. For example, in rapidly dividing cells of higher eukaryotes, the M phase typically recurs every 16–24 hours, whereas it lasts only 1 to 2 hours itself. A typical cell cycle with its four successive phases is shown in Fig. 13-1; some details are provided in the legend to this figure.

Class="center">

Fig. 13-1. The four successive phases of the cell cycle in a typical eukaryotic cell. Following the M phase, which consists of nuclear division (mitosis) and cytoplasmic division (cytokinesis), the daughter cells enter the interphase of a new cycle. Interphase begins with the G₁ phase, during which intense biosynthetic processes—sharply suppressed during mitosis—resume. The S phase is the period of DNA synthesis; it ends when the nuclear DNA content has doubled and the chromosomes are fully replicated (each chromosome now consists of two identical "sister chromatids"). The cell then enters the G₂ phase, which continues until the onset of mitosis, i.e., the M phase. During the M phase, the duplicated chromosomes condense and become clearly visible under a Light Microscope. The nuclear envelope breaks down (with the exception of single-celled eukaryotes, such as Yeast, where it remains intact); the sister chromatids separate to form two new nuclei, and the Cytoplasm divides to yield two daughter cells, each with a single nucleus. The process of cytokinesis completes the M phase, and the interphase of the next cell cycle begins. The figure illustrates a typical 24-hour cycle, though the duration of the cell cycle in higher eukaryotes varies widely: it can be shorter than 8 hours, and in adult animals, more than a year, with these differences depending primarily on the duration of the G₁ phase.

Fig. 13-2. Autoradiograph obtained after brief incubation of cells with ³H-thymidine. This technique is described in Section 4.5.2. The presence of silver grains in the photoemulsion over the nuclei (dark areas) indicates that the cells incorporated ³H-thymidine into DNA and, consequently, spent some time in the S phase during the exposure period. (Courtesy of James Cleaver.)

The timing of DNA synthesis within the cell cycle was first uncovered in the early 1950s using autoradiography. The standard method employs ³H-thymidine, a radioactive precursor of a compound that the cell uses exclusively for DNA synthesis. ³H-thymidine can either be injected into an animal to study Cell Division cycles in Tissues or added to an in vitro culture medium (Fig. 13-2). In the former case, at specific times following ³H-thymidine administration, tissue is harvested from the animal and autoradiographs are prepared. Those cells that were synthesizing DNA during the preceding period (and therefore were in the S phase) can be identified by the silver grains over their nuclei. By counting the fraction of cells in the M phase at various durations of ³H-thymidine incorporation, one can demonstrate that the cell cycle comprises the four phases described above and measure the duration of each.

Suppose a single injection is given and, shortly thereafter—say, half an hour later—the tissue is fixed for autoradiography. In a typical cell population where all cells are dividing rapidly but asynchronously, about 30% of the cells will become radiolabeled. These will be the cells that were synthesizing DNA during the brief exposure period in the presence of ³H-thymidine, and their fraction in the cell population reflects the proportion of the cell cycle occupied by the S phase (Fig. 13-3). Only about 5% of the cells will be in mitosis at the time of fixation (the small magnitude of this mitotic index means that mitosis occupies only a tiny fraction of the cell cycle), and none of them will be radiolabeled. This indicates that the M and S phases are distinct PARTS OF THE cell cycle. On the other hand, if preparations are fixed several hours after ³H-thymidine administration, some cells undergoing mitosis will become radiolabeled; presumably, these cells were still synthesizing DNA at the time of injection. The minimum interval between the injection and the appearance of labeled mitotic cells equals the duration of the G₂ phase. Studies of this kind make it possible to determine the duration of all four phases of the cycle. An example of this method's application is shown in Fig. 13-4.

The duration of cell cycles in different tissues, across different species, and at various stages varies over a very wide range—from less than an hour (e.g., in early frog embryos) to more than a year (e.g., in the adult mammalian Liver). Although all phases of the cell cycle can vary in length to some extent, this is particularly true of the G₁ phase, the duration of which can range from virtually zero (in the early frog embryo) to values so large that the cells appear to have stopped dividing altogether (in the mature human liver). Cells in such a quiescent G₁ phase are often said to be in the G₀ state (see Section 13.3.8).

13-5

13.1.2. The cell cycle is easiest to study in in vitro cultures [1, 2]

The mechanisms underlying the cell cycle are difficult to study in the complex and inaccessible tissues of an intact animal. Working with cell cultures is much easier. For example, using time-lapse cinematography (Section 4.1.5), one can observe an individual cell undergo mitosis, grow, and then enter mitosis again; this enables direct measurement of the duration of the M phase and the entire cell cycle. Cells synthesizing DNA in culture can be detected in the same manner as in the intact Organism—by autoradiography using ³H-thymidine. The course of the cell cycle can also be tracked by directly measuring the cellular DNA content; this task is greatly facilitated by The Use of a flow cytometer (Fig. 13-5).

Fig. 13-3. The length of each cell cycle phase is roughly equal to the fraction of cells in that phase at any given moment, multiplied by the total cycle duration (assuming the cell population grows uniformly and all cells divide at the same rate). However, an accurate calculation of each phase's duration requires a "correction factor" that ranges from 0.7 for early G₁ cells to 1.4 for mitotic cells. This coefficient is necessary because a uniformly growing population always contains more young (recently divided) cells than old ones.

Fig. 13-4. Method commonly used to measure the duration of cell cycle phases. An asynchronously proliferating cell population is briefly maintained in a medium containing ³H-thymidine and then washed; the cells then continue progressing through their cycle. At various times following the exposure, cell samples are taken and autoradiographs are prepared. To interpret the results, it is helpful to represent the cells as if they were distributed on a uniformly rotating wheel, with each cell's position corresponding to the cycle stage it occupies. Initially, cells in the S phase become radiolabeled (shaded area), whereas cells in the G₂, M, and G₁ phases do not. After a time interval equal to the duration of G₂, labeled cells begin to enter the M phase; when a time equal to G₂ + M has elapsed, all cells currently in the M phase become labeled, and so forth. By recording the times at which labeled cells enter the M phase and subsequent phases until they return to M, one can in principle determine the average durations of the G₂, M, and S phases and the total cycle duration, and thereby (by subtraction) the duration of G₁. In the example shown, these values are 3, 7, 22, and 11 h, respectively.

Fig. 13-5. Typical data obtained by analyzing the DNA content of individual cells in a growing population using a flow cytometer—an electronic instrument based on the same principle as fluorescence-activated cell sorting (see Section 4.3.1). The cells are stained with a dye that becomes fluorescent upon binding to DNA. Therefore, the fluorescence intensity is directly proportional to the DNA content of each cell. Cells are divided into three categories: those with unreplicated DNA (1 arbitrary unit), i.e., in the G₁ phase; those with fully replicated DNA (2 arbitrary units), i.e., in the G₂ or M phase; and those with an intermediate DNA content, in the S phase. The cell distribution in the example shown indicates that the number of cells in the G₁ phase is greater than in G₂ and M combined. This means that in this population, the G₁ phase is longer than G₂ + M.

A further simplification in cell cycle analysis involves using a large population of cultured cells that pass through the same cell cycle phases simultaneously. Such synchronized cell populations can be obtained in various ways. The earliest method consisted of maintaining cells in a solution of a substance that disrupts a specific stage of the cell cycle; prolonged exposure of the culture to such a solution causes all cells to arrest at that stage, and upon release of the block, they resume the cycle and progress synchronously. However, many different processes occur in parallel during the cell cycle, and it is unlikely that all of them would be blocked simultaneously. Therefore, it would be preferable whenever possible to use methods for obtaining synchronized populations that do not disrupt the normal progression of the cell cycle. For most mammalian cells, the simplest and best method takes advantage of cytoskeletal Changes in the M phase that lead to cell "rounding up." Cells that have rounded up in the M phase attach so loosely to the bottom of the culture dish that they can be dislodged by gentle shaking (Fig. 13-6). Mitotic cells harvested in this way constitute a synchronized population that enters the G₁ phase of the cell cycle almost immediately. Another method is also used: because cells increase in size as they progress through the cycle, centrifugation can be employed to isolate subpopulations of cells at different Stages of the cycle.

Fig. 13-6. A standard method for obtaining a synchronized population of animal cells in culture. Mitotic cells are collected by shaking them off The surface of the dish in which they are growing. Upon transfer to a new dish, these cells undergo subsequent cycles synchronously. Due to random variations in individual division rates, synchrony is lost after a few division cycles.

Fig. 13-7. Graph of cell mass increase over the course of the cell cycle. Most cellular components are synthesized more or less evenly throughout interphase, with their rate of formation typically increasing as Cell size and biosynthetic activity grow (with a brief pause during the M phase). For the average cell size to remain constant in a proliferating population, The amount of each component must precisely double during the cycle.

13.1.3. Critical events of the cell cycle occur abruptly against the backdrop of continuous cell growth [3]

A synchronized population of dividing cells allows for a more detailed Study of the chemical changes occurring during the cell cycle. Under growth-favorable conditions, the total protein content of a typical cell increases more or less continuously throughout the entire cycle (Fig. 13-7). RNA Synthesis also occurs at a constant rate, except during the M phase, where chromosome Condensation apparently hinders Transcription, bringing RNA synthesis nearly to a halt while protein production decreases. Analysis of individual Protein Synthesis (Fig. 13-8) shows that the vast majority of them are synthesized throughout the entire cycle. Thus, during cell growth, most of its components are formed gradually and continuously—their synthesis briefly pauses only when the cell divides into two.

Against the backdrop of this continuous growth, a series of sharp changes occurs, associated with critical moments of the cell cycle. Some of these, such as the onset of DNA synthesis, are easily detected, whereas others are harder to spot. For example, it turns out that for most cells, There is a critical point in the G₁ phase where their cell cycle pauses if environmental conditions are unfavorable for growth. Upon passing this point, known as the restriction point, internal changes take place within the cell, after which it must proceed through all subsequent stages of the cell cycle according to a strict time schedule.

As one might expect given the presence of several critical points in the cycle, certain Proteins can be isolated (though very few) whose synthesis accelerates sharply at specific stages of the cycle (see Fig. 13-8). For instance, Histones required for the assembly of new Chromatin are synthesized at a high rate only during the S phase; this apparently also applies to certain Proteins of the DNA replication machinery.

But what determines the timing of such critical events and how are they coordinated with one another and with the continuous process of cell growth? To answer these questions, we will first examine what serves as the trigger signal for DNA synthesis, i.e., the transition to the S phase of the cell cycle.

13.1.4. DNA synthesis is triggered by A change in the cytoplasm—the appearance of an S-phase activator [4]

By adding an appropriate agent to the culture medium (Section 4.3.5), one can induce the fusion of cells from two synchronized cell populations residing in different phases of the cell cycle. The results of such an experiment are extremely informative. When an S-phase cell fuses with a cell at an earlier G₁ stage, the Nucleus of the G₁ cell immediately commences DNA synthesis (Fig. 13-9, A). Clearly, this nucleus is already primed for DNA replication, but normal G₁ cells still lack some signal (or set of signals) required to activate the DNA synthesis machinery. Evidently, such a signaling factor is present in Abundance in the cytoplasm of cells residing in the S phase. The appearance of this S-phase activator obviously marks the boundary between the G₁ and S phases in a normal cell.

Fig. 13-8. Analysis of proteins synthesized in the G₁ and S phases using two-dimensional Polyacrylamide gel Electrophoresis. Synchronized mouse lymphoma cells were briefly labeled with a mixture of radioactive Amino Acids during early G₁ or late S phase. The presented autoradiographs reveal about 1,000 Newly synthesized proteins, but only two of them (marked with a circle and a square) are produced at substantially different rates in these two phases. Because histones carry a strong positive charge, they migrate out of the gel and are not visible here. (P. Coffino, V. E. Groppi, Adv. Cyclic Nucleotide Res. 14: 399-410, 1981.)

Does the S-phase activator disappear after DNA synthesis is complete and the cell transitions to the G₂ phase? The answer can again be obtained from Cell Fusion experiments. When a G₂ phase cell fuses with a G₁ phase cell (Fig. 13-9, B), the G₁ Cell Nucleus does not prematurely begin synthesizing DNA; however, if the same G₂ cell fuses with an S-phase cell, DNA replication continues in The Nucleus of the latter (Fig. 13-9, A). Clearly, the S-phase activator (or some vital component thereof) disappears shortly after this phase ends, and the cytoplasm of G₂ phase cells no longer contains either a diffusing activator or a diffusing inhibitor of DNA synthesis.

13.1.5. In each cycle, the entire genome is replicated only once [4, 5]

As discussed in Chapter 9 (Section 9.3.6), different parts of The Genome replicate at different moments of the S phase and cannot replicate again thereafter, as this is prevented by a chemical modification that occurs in every segment of each chromosome following its replication. Thanks to this block on rereplication, when a G₂ phase cell fuses with an S-phase cell (Fig. 13-9, A), the G₂ cell nucleus proves insensitive to the S-phase activator and does not resume DNA synthesis. The block is lifted when cells progress through mitosis into the onset of a new G₁ phase.

Fig. 13-9. Summary diagram of various outcomes from the fusion of two mammalian cells at different phases of the cell cycle.

However, another mechanism is required to ensure that the S-phase activator remains present until the replication of all DNA is complete. As we have seen, cell fusion experiments show that the cytoplasmic signal activating the DNA replication machinery at the start of the S phase (the S-phase activator) disappears by its end. Nonetheless, if a cell is artificially blocked in the S phase by DNA synthesis inhibitors, the DNA replication machinery remains functional even past the normal termination time of the S phase, such that upon removal of the inhibitor, DNA replication resumes and is brought to completion. Apparently, a chromosome that has not finished replication somehow maintains the replication machinery in an active state.

It is possible that replication forks themselves are responsible for this effect. As we saw in Chapter 9 (Section 9.3.1), these forks exist in pairs: the two forks of a single pair move in opposite directions from a common origin, and each ceases to exist only when it reaches the end of the chromosome or collides with a fork moving toward it. Thus, once a chromosome has initiated replication, at least one Replication fork will exist until the entire chromosome is fully duplicated. It is possible that in some unclear manner, such a fork provides additional production of the S-phase activator, effectively catalyzing The formation of new forks in other DNA regions. Indeed, the initiation of the first pair of replication forks could serve as a trigger mechanism for the onset of the S phase, operating on an all-or-none principle. Such a single initiation event would depend on a rare random collision between a DNA start sequence and an initiator molecule present at a low concentration. Indeed, the variation in the timing of the G1→S transition is stochastic in nature, which is consistent with this assumption (Section 13.3.3).

13-6

13.1.6. A cytoplasmic signal delays preparation for mitosis until DNA replication is complete [4, 6]

A nucleus that has completed the S phase and entered G₂ normally condenses its chromosomes and enters mitosis after a specific time interval. However, if DNA synthesis is artificially blocked, mitosis is delayed until the block is lifted and DNA synthesis is completed. Similarly, following the fusion of an S-phase cell with a G₂ phase cell, the nucleus of the latter is detained at this stage until the other nucleus "catches up," and eventually both nuclei enter mitosis together. The simplest assumption is that the mitotic delay is caused by a cytoplasmic signal generated when DNA replication is incomplete. This signal may or may not be identical to the S-phase activator; in either case, some clue to its possible mechanism is contained in the fact that if a G₂ phase cell is damaged (e.g., by X-irradiation), mitosis is delayed until repair is achieved. Both during repair and during DNA replication, single-stranded DNA must be present in the cell;

meanwhile, it is known that an excess of such DNA in Bacteria triggers The production of a cytoplasmic signal that delays cell division (the SOS Response—see Section 5.2.9). It is possible that single-stranded DNA also generates an M-phase delay signal in Eukaryotic cells. This does not bring us much closer to understanding The Nature of such a signal, save for one very interesting observation: if DNA synthesis is artificially blocked in cultured mammalian cells (via inhibitors or DNA damage), The addition of caffeine to the medium can cause them to prematurely enter mitosis before their DNA has fully replicated. Exactly what caffeine acts upon in this process remains unknown.

13-7

13.1.7. Mitosis is triggered by the "M-phase-promoting factor" (MPF) [4, 7]

The disappearance of signals delaying the M phase is not in itself sufficient to trigger mitosis—another cytoplasmic factor is required. A normal G₂ phase can be viewed as a period of preparation for the production of this decisive factor, which activates the mitotic mechanism after the delay factors have dissipated. Data regarding this have also been obtained from cell fusion experiments.

When an M-phase cell is fused with a cell at any stage of interphase (G₁, S, or G₂), the interphase nucleus rapidly enters the M phase, undergoing chromosome condensation and preparing for division even if this threatens (as in the case of G₁ or S phase nuclei) to disrupt the entire subsequent course of division (Fig. 13-9, B and 13-10). Evidently, M-phase cytoplasm contains a potent M-phase-promoting factor (MPF) to which the nucleus responds regardless of the cell cycle phase. Presumably, the mitotic delay factors mentioned earlier inhibit the production of MPF, but cannot block its action once it has already formed.

Fig. 13-10. Premature condensation of interphase chromosomes following the fusion of interphase Cells of the marsupial PtK with human mitotic cells. A. The PtK cell was in phase C₁, so its prematurely condensed chromosomes appear as single chromatids. B. The PtK cell was in phase S, and its chromatin now assumes a "pulverized" appearance. C. The PtK cell was in phase G₂, and now the chromatids, although very long compared to normal human metaphase chromosomes, are duplicated. (K. Sperling, P. Rao, Humangenetik 23: 235-258, 1974.)

13.1.8. Chromosomal cycle events: interconnected links of a single chain [8]

The experiments described above can serve as a basis for the Structure/81.html">Functional Classification of certain molecules that apparently govern the events of the chromosomal cycle. Three diffusible controlling factors have been described; for the sake of brevity, it is convenient to assume that each of them represents a single molecule, although in reality they may be more complex. These are: 1) the S-phase activator, which is normally present in the cytoplasm of cells only during the S-phase and turns on DNA synthesis; 2) the M-promoting factor (MPF), which is contained in the cytoplasm only during the M-phase and triggers chromosome condensation; 3) the DNA-dependent M-phase-delaying factor (possibly identical to the S-phase activator), which is present in the cytoplasm during the S-phase and inhibits the processes leading to MPF production.

The moments of rapid appearance and disappearance of these diffusing factors in the cytoplasm delineate a series of cell cycle events, and the time intervals between them determine the duration of the entire cycle.

Causal relationships among the three factors (and possibly others not yet known) ensure that the events of the chromosomal cycle always proceed in a specific sequence, preventing disastrous mishaps such as chromosome condensation in the middle of the DNA synthesis phase. Each subsequent step depends on the preceding one. Therefore, a cell cannot enter mitosis until the M-promoting factor appears; and it cannot appear until the M-delaying factor disappears; the M-delaying factor and the S-phase activator cannot disappear until DNA synthesis is complete; DNA synthesis will not cease until all DNA is replicated; the next round of DNA replication cannot begin until the block on rereplication is lifted during the transition to G₁. Later we will encounter another example: a cell cannot transition from mitosis to G₁ until chromosomes are segregated by the mitotic spindle (Section 13.5.7). All these observations, as well as those to be discussed later (Sections 18.2.1, 19.8.2), indicate that most events and processes of the chromosomal cycle are interrelated, forming a dependent sequence.

13.1.9. During early Cleavage divisions, when cells do not grow, the cell cycle is shortened [9]

Experiments with eggs and early embryos of the clawed frog Xenopus have proven particularly useful for studying the molecules that control the chromosomal cycle. The Xenopus egg, like that of many other species, is an unusually large spherical cell. Its diameter is just over a millimeter, and it contains a supply of virtually all the substances (except DNA) required to build an early embryo. All these substances are accumulated during a long growth period of the immature egg, called the oocyte. This prolonged stage is best defined as the G₂ phase of the first meiotic division cycle (although it is usually called the prophase of the first meiotic division, in many respects it resembles a normal G₂ phase—see Section 15.2.7). During ovulation, hormone action triggers egg maturation, so that by the time of laying, it has already passed through the subsequent stages of Meiosis and arrests in the M-phase of the second meiotic division (Section 15.3.3). Fertilization then triggers an exceptionally rapid sequence of cell divisions: one giant cell cleaves to form an embryo consisting of thousands of smaller cells (Fig. 13-11). Practically no growth occurs during this process—only the DNA required to form the requisite number of nuclei and a small amount of protein are synthesized from macromolecules. After the First Division, which lasts about 90 minutes, the subsequent 11 divisions occur more or less synchronously at 30-minute intervals, producing 4096 (2¹²) cells in about 7 hours. The preliminary accumulation of substances in the egg makes such rapid cell cycles possible by eliminating the time required for cell growth during each cycle. DNA Replication and division cycles are shortened because the S and M phases are accelerated, while G₁ and G₂ become so short as to be virtually indistinguishable.

Fig. 13-11. In about 7 hours, the Xenopus egg undergoes 12 very rapid synchronous division cycles consisting of alternating S and M phases without discernible G₁ and G₂ phases. These divisions cleave the egg into 4096 (2¹²) smaller cells. One cell is highlighted by color at each stage.

13.1.10. M-promoting factor (MPF) induces mitosis in A wide variety of cells [10]

Because Xenopus oocytes and eggs are very large, it is easy to inject various substances into their cytoplasm. Furthermore, the oocyte, egg, and early embryo serve as abundant sources of cytoplasm strictly specific to particular stages of the cell cycle. This is especially important when studying the M-promoting factor (MPF), mentioned above. This factor was first discovered in mature unfertilized Xenopus eggs, which are in the M-phase. If cytoplasm from such an egg is injected into an oocyte, it drives the oocyte out of the G₂ stage and forces it to enter the M-phase. This initiates oocyte maturation (originally, the abbreviation MPF stood for maturation-promoting factor; see Section 15.3.6). Active MPF also appears in the cleaving egg (embryo) during each M-phase (Fig. 13-12). Thus, the Xenopus egg and oocyte can serve both as a source of material in attempts to purify MPF and as an assay system to determine its activity (Fig. 13-13).

Fig. 13-12. Levels of M-promoting factor (MPF) activity in the Xenopus oocyte, egg, and early embryo. The oocyte arrests in the G₂ phase of meiosis with a low level of MPF; the mature laid egg arrests in the M phase of meiosis with a high level of MPF; following fertilization, the early embryo undergoes alternating S and M phases with correspondingly fluctuating levels of MPF activity.

Fig. 13-13. MPF assays by injection into a Xenopus oocyte. The presence of MPF is detected by its ability to drive the oocyte into the M phase. The large nucleus (the "germinal vesicle") of the oocyte breaks down during the Formation of the mitotic spindle.

MPF is of universal significance for eukaryotic cells and is highly conserved in evolution: extracts prepared from mitotic cells of very diverse organisms, such as mammals, sea urchins, Mollusks, and yeast, drive Xenopus oocytes into the M-phase when injected. A purified preparation with MPF activity has been obtained from mature Xenopus eggs. It behaves as a large protein composed of Two Types of subunits; one such subunit is a protein kinase, and it appears capable of phosphorylating the other. Therefore, MPF is likely capable of self-activation: if a small amount of preparation with MPF activity is injected into a Xenopus oocyte, the cell responds by producing a much larger quantity of MPF from its own inactive reserves (Section 15.3.6). These and other data suggest that the appearance and disappearance of MPF activity throughout the normal cell cycle depend on protein modification—specifically phosphorylation and dephosphorylation—rather than on Synthesis and degradation. However, the normal "triggering" of MPF activity requires the synthesis of another protein called cyclin (see below); therefore, all cell types are unable to transition from interphase to M-phase when Protein synthesis is blocked.

It appears that many of the molecular changes occurring in mitosis are brought about by phosphorylation; MPF kinase directly phosphorylates certain substrates, such as histone H1, which may facilitate chromosome condensation (Section 9.2.2); MPF may trigger the entire complex of mitosis-related events through a cascade of phosphorylation reactions.

13.1.11. MPF is generated by a cytoplasmic oscillator [8, 10, 11]

The sharp increase in MPF levels that occurs every 30 minutes in the Xenopus embryo during cleavage is caused by a cytoplasmic oscillator that operates even in the absence of a nucleus. By constricting an activated egg with a fine Hair before the first division, it can be divided into two roughly equal parts, one containing a nucleus and the other not (Fig. 13-14). The nucleated part will continue normal cleavage. Remarkably, a series of oscillations—manifested as repeated cycles of mild periodic contraction and increased rigidity of the cortical cytoplasm—will also occur in the anucleated portion. These periodic "spasms" occur almost exactly synchronously with the cleavage Divisions of the other, nucleated half of the egg. By sampling cytoplasm from the oscillating anucleated cell and testing its activity by injection into oocytes, it can be shown that the visible oscillations are accompanied (and possibly caused) by fluctuations in the concentration of active MPF.

Fig. 13-14. Method for detecting the oscillatory process in the cytoplasm associated with the cell division cycle in a cleaving Xenopus egg. A freshly fertilized egg is constricted into two parts with a loop of fine human hair; one half contains the nucleus and continues to divide, while the other, devoid of a nucleus, does not divide. Photographs show that the anucleated half periodically changes its size due to alterations in the stiffness of the cell cortex. These oscillations occur strictly synchronously with the divisions in the other half. (K. Hara, P. Tydeman, M. Kirschner, Proc. Natl. Acad. Sci. USA 77: 462-466, 1980.)

These and other experiments suggest that cleavage divisions in the early Xenopus embryo involve two parallel cyclic processes—the Chromosome replication cycle and the cytoplasmic MPF cycle—which are normally coordinated because each new chromosomal cycle can only begin when the block to DNA replication is lifted by the next pulse of MPF in the M-phase. Such an interaction between the two cycles prevents the chromosomal cycle from "getting ahead" and maintains the coordinated progression of both cycles as long as there is no risk that a too-slow chromosomal cycle will fail to ensure complete DNA replication before the rise in MPF levels. In the Xenopus egg, with its unusually rapid S phases and regular division cycles, this danger appears small, and simple interaction between cycles seems sufficient. However, in mammalian cells discussed earlier (and presumably in most eukaryotic cells excluding cleaving eggs), there is an additional mechanism: as we saw, unreplicated DNA generates an M-phase delay signal that prevents the cytoplasmic MPF cycle from running ahead of the chromosomal cycle. Experiments blocking DNA replication with inhibitors show that this additional control does not operate in the cell cycles of the early Xenopus embryo. Furthermore, judging by the reduction of the G₁ phase, the S-phase activator appears to be present continuously. Thus, in the early Xenopus embryo, the cell cycle is simplified and compressed in time.

The phenomena just described imply that a cytoplasmic oscillator may be present in all cells, but they reveal nothing about its mechanism. The clue might be related to another protein, cyclin, which has been discovered in cleaving eggs of Xenopus, sea urchins, and bivalve mollusks. Cyclin, much like MPF, belongs to a small group of proteins whose activity depends heavily on the phase of the cell cycle. Although cyclin is synthesized at a roughly constant rate throughout the cycle, it abruptly degrades in the middle of the M-phase. Consequently, in each cycle its concentration gradually rises from zero and then drops sharply back to zero. Cyclin genes have been cloned, making it possible to prepare pure mRNA for this protein. When such RNA is injected into a Xenopus oocyte, it has the same effect as an MPF injection, driving the oocyte from G₂ to the M-phase. Such findings have led to the hypothesis that the rise of MPF in the M-phase is caused by an increase in cyclin concentration to a certain threshold level, while cyclin destruction is linked to some event in the M-phase; the subsequent disappearance of MPF may be a consequence of cyclin destruction (Fig. 13-15). In this case, the interval between two mitoses would be determined mainly by the time required for the cyclin concentration to rise from zero to the threshold value; under these conditions, the cell cycle should arrest in interphase upon Treatment with Protein Synthesis Inhibitors, which is indeed observed.

Fig. 13-15. Coordinated rises and falls in MPF and cyclin levels associated with cell cycles. Measurements of cyclin concentration were performed primarily in marine invertebrate eggs, where cyclin accounts for 5% of the proteins synthesized during a brief incubation with radioactive amino acids.

Conclusion

The reproductive cycle of a typical eukaryotic cell can be divided into four phases designated as G_i (from the end of mitosis to the onset of DNA synthesis), S (DNA synthesis), G₂ (from the end of DNA synthesis to the onset of mitosis), and M (mitosis). Each S and M phase is initiated by a soluble cytoplasmic factor [the S-phase activator and M-phase promoting factor (MPF), respectively]. The S-phase activator is produced throughout the S phase and may also act as a factor that delays preparation for the M phase until DNA replication is complete. The M-phase promoting factor can be detected in M-phase cells across a wide range of organisms—from yeast to mammals—and its activity is presumably regulated by phosphorylation. In rapidly cleaving eggs, such as those of Xenopus, the cell cycle is shortened and simplified. In this case, the cycle appears to be regulated by interrelated fluctuations in MPF activity and cyclin concentration.