1. Introduction

27.16 Summary

The cell cycle consists of transitions from one regulatory state to another. The change in regulatory state is separated by a lag period from the subsequent changes in cell phenotype. The transitions take the form of activating or inactivating a kinase(s), which modifies substrates that determine the physical state of the cell. Checkpoints can retard a transition until some intrinsic or extrinsic condition has been satisfied.

The two key control points in the cell cycle are in G1 and at the end of G2. During G1, a commitment is made to enter a replication cycle; the decision is identified by the restriction point in animal cells, and by START in yeast cells. After this decision has been taken, cells are committed to beginning an S phase, although there is a lag period before DNA replication initiates. The end of G2 is marked by a decision that is executed immediately to enter mitosis.

A unifying feature in the cell cycles of yeasts and animals is the existence of an M phase kinase, consisting of two subunits: Cdc2, with serine/threonine protein kinase catalytic activity; and a mitotic cyclin of either the A or B class. Homologous subunits can be recognized in (probably) all eukaryotic cells. The genes that code for the catalytic subunit in yeasts are the eponymous cdc2 in S. pombe and CDC28 in S. cerevisiae. Animal cells usually contain multiple mitotic cyclins (A, B1, B2); in S. pombe, there is only a single cyclin at M phase, a B class coded by cdc13, although S. pombe has several CLB proteins.

The activity of the M phase kinase is controlled by the phosphorylation state of the catalytic subunit. The active form requires dephosphorylation on Tyr-15 (in yeasts) or Thr-14/Tyr-15 (in animal cells) and phosphorylation on Thr-161. The cyclins are also phosphorylated, but the significance of this modification is not known. In animal cells, the kinase is inactivated by degradation of the cyclin component, which occurs abruptly during mitosis. Cyclins of the A type are typically degraded before cyclins of the B type. Destruction of at least the B cyclins, and probably of both classes of cyclin, is required for cells to exit mitosis.

A comprehensive analysis of genes that affect the cell cycle has identified cdc mutants in both S. pombe and S. cerevisiae. The best characterized mutations are those that affect the components or activity of M phase kinase. Mutations cdc25 and wee1 in S. pombe have opposing effects in regulating M phase kinase in response to cell size (and other signals). wee1 is a kinase that acts on Tyr-15 and maintains cdc2 in an inactive state; cdc25 is a phosphatase that acts on Tyr-15 and activates cdc2. The existence of wee1 and cdc25 homologs in higher eukaryotes suggests that the apparatus for cell cycle control is widely conserved in evolution.

By phosphorylating appropriate substrates, the kinase provides MPF activity, which stimulates mitosis or meiosis (as originally defined in Xenopus oocytes). A prominent substrate is histone H1, and H1 kinase activity is now used as a routine assay for M phase kinase. Phosphorylation of H1 could be concerned with the need to condense chromatin at mitosis. Another class of substrates comprises the lamins, whose phosphorylation causes the dissolution of the nuclear lamina. A general principle governing these (and presumably other) events is that the state of the substrates is controlled reversibly in response to phosphorylation, so that the phosphorylated form of the protein is required for mitotic organization, while the dephosphorylated form is required for interphase organization. Phosphatases are required to reverse the modifications introduced by M phase kinase.

Transition from G1 into S phase requires a kinase related to the M phase kinase. In yeasts, the catalytic subunit is identical with that of the M phase kinase, but the cyclins are different (the combinations being CDC28-cig1,2 in S. pombe, cdc2-CLN1,2,3 in S. cerevisiae). Activity of the G1/S phase kinase and inactivity of the M phase kinase are both required to proceed through G1. Initiation of S phase in S. pombe requires rum1 to inactivate cdc2/cdc13 in order to allow the activation of cdc18, which may be the S phase activator.

In mammalian cells, a family of catalytic subunits is provided by the cdk genes, named because they code for the catalytic subunits of cyclin-dependent kinases. There are ~10 cdk genes in an animal genome. Aside from the classic cdc2, the best characterized product is cdk2 (which is well related to Cdc2). In a normal cell cycle, cdk2 is partnered by cyclin E during the G1/S transition and by cyclin A during the progression of S phase. cdk2, cdk4, and cdk5 all partner the D cyclins to form kinases that are involved with the transition from G0 to G1. These cdk-cyclin complexes phosphorylate RB, causing it to release the transcription factor E2F, which then activates genes whose products are required for S phase. A group of cki (inhibitor) proteins that are activated by treatments that inhibit growth can bind to cdk-cyclin complexes, and maintain them in an inactive form.

Checkpoints control progression of the cell cycle. One checkpoint responds to the presence of unreplicated or damaged DNA by blocking mitosis. Others control progress through mitosis, for example, detecting unpaired kinetochores.

Apoptosis is achieved by an active pathway that executes a program for cell death. The components of the pathway may be present in many or all higher eukaryotic cells. Apoptosis may be triggered by various stimuli. A common pathway involves activation of caspase-8 by oligomerization at an activated surface receptor. Caspase-8 cleaves Bid, which triggers release of cytochrome c from mitochondria. The cytochrome c causes Apaf-1 to oligomerize with caspase-9. The activated caspase-9 cleaves procaspase-3, whose two subunits then form the active protease. This cleaves various targets that lead to cell death. The pathway is inhibited by Bcl2 at the stage of release of cytochrome c. An alternative pathway for triggering apoptosis that does not pass through Apaf-1 and caspase-9, and which is not inhibited by Bcl2, involves the activation of JNK. Different cells use these pathways to differing extents. Apoptosis was first shown to be necessary for normal development in C. elegans, and knockout mutations in mice show that this is also true of vertebrates. Every cell may contain the components of the apoptotic pathway and be subject to regulation of the balance between activation and repression of cell death.