The Molecular Basis of G2 Arrest

To understand how the cell cycle is regulated in response to DNA damage, mutants of budding and fission yeast were identified in which the cell cycle was no longer delayed in response to DNA damage (58). This approach led to the concept of the cell cycle checkpoint, and uncovered many genes that form an important part of what we know about the workings of the DNA-damage response. Cell cycle checkpoints are regulatory mechanisms that ensure that cell cycle processes occur at the right time and in the right order. Early studies in budding yeast with the rad9 mutant indicated that G2 delay was owing to intracellular signaling that blocked entry into mitosis in the presence of DNA

damage (58). A number of other "Rad" mutants that fail to arrest in the cell cycle in response to ionizing radiation have been uncovered, and orthologs in various organisms including mammals have been identified (reviewed in 5961). Studies on the components of the DNA-damage response have uncovered three main groups of proteins involved in initiating G2 arrest. The sensors recognize damaged DNA, and the transducers transmit the signal downstream to the effectors, whose activity is modulated to bring about arrest (61,62). Although these groups of proteins are sufficient to induce the arrest, in time, many checkpoints become spontaneously inactivated, and additional mechanisms are used to maintain the arrest.

4.1. Effectors

Studies with mammalian cells in tissue culture showed that Cdc2 is an important effector for the G2 checkpoint. Ionizing radiation and other forms of DNA damage blocked the dephosphorylation of Cdc2 at tyrosine 15 and threonine 14, causing it to remain inactive (63,64). Inhibition of Cdc2 activity occurs rapidly after inducing DNA damage. For example, the loss of activity can be detected within an hour of adding the DNA-damaging agent etoposide to Chinese hamster ovary cells (63). One of the main effects of DNA damage is to interfere with the dephosphorylation of Cdc2 by the Cdc25C phosphatase.

4.2. Transducers

The mechanism by which Cdc25C is inactivated in response to DNA damage involves its phosphorylation by Chkl and Chk2 (65-67) (Fig. 1). Chkl and Chk2 (also known as CDS1) were originally identified by genetic screens in S. pombe, where Chkl is required for cell cycle arrest in response to damaged DNA and Chk2-Cds1 is required for arrest in response to unreplicated DNA (68,69). Chkl and Chk2 are protein kinases whose activity increases in response to damaged or unreplicated DNA in yeast and mammals (reviewed in 70). Cdc25C is phosphorylated by either Chkl or Chk2, which creates a binding site on Cdc25C for proteins of the l4-3-3 family (65-67,71). Binding to l4-3-3 sequesters Cdc25C in the cytoplasm and blocks its ability to dephos-phorylate Cdc2 (Fig. 1).

Two kinases, Atm and Atr, are responsible for the activation of Chkl and Chk2 in response to stress (71-73) (Fig. 1) . In mammals, it appears that both Atm and Atr can phosphorylate either Chkl or Chk2 as well as other substrates (reviewed in 74). Parallels between mammals and yeast extend upstream of Chkl and Chk2-Cdsl. Budding yeast Mecl and fission yeast Rad3 are the orthologs of Atr and are required for the activation of Chkl and Chk2-Cdsl in response to DNA damage or unreplicated DNA (61). Tell in budding and fission yeast is an Atm ortholog.

Studies in mammalian cells on the phosphorylation of the p53 tumor suppressor by Atm and Atr have indicated that these kinases respond to different types of damage. For example, the phosphorylation of p53 on serine 15 in response to ionizing radiation is significantly, but not completely, reduced in cells lacking Atm, whereas there is no defect in the phosphorylation of p53 on serine 15 in response to ultraviolet radiation (75-77). Inactivation of Atr using a dominant negative version of the protein severely reduced p53 phosphorylation in response to ultraviolet radiation, and the initial phosphorylation of p53 was normal when cells were exposed to ionizing radiation (78,79). This suggests that Atr primarily mediates the response to ultraviolet radiation and Atm mediates the response to ionizing radiation. Although p53 was phosphorylated in response to ionizing radiation in cells expressing the dominant-negative Atr, this phosphorylation was lost much faster than in parental cells (79). This suggests that Atr may also provide a backup function in response to ionizing radiation. Atr has also been implicated in the response to unreplicated DNA caused by blocking DNA synthesis with hydroxyurea (78). Atm was originally identified as the gene mutated in the recessive autosomal disease ataxia telangiectasia (80). Among a number of symptoms, patients with ataxia telangiectasia are prone to cancer, which probably reflects the roles played by Atm in the cellular response to DNA damage.

There is evidence that Chk1 and Chk2 are not the only kinases that phospho-rylate Cdc25 in response to DNA damage. For example, the p38 stress-activated kinase can phosphorylate both Cdc25B and Cdc25C, leading to their increased binding to 14-3-3 proteins in vitro (81). Also, the immediate G2 arrest that normally occurs after ultraviolet radiation was attenuated in cells treated with SB202190, a chemical inhibitor of p38 (81). In vivo studies showed that the binding of Cdc25B to 14-3-3 in cells exposed to ultraviolet radiation was reduced by treatment with the p38 inhibitor. However, the inhibitor had no effect on the binding of Cdc25C to 14-3-3 proteins in vivo, suggesting that Cdc25B is the main target of p38 in the G2 arrest response (81).

There is also evidence that inhibiting Cdc25 is not the only way by which Atm and Atr block the activation of Cdc2. Division of fission yeast normally requires Cdc25, but cells with the hypermorphic cdc2-3w allele can survive if Cdc25 is deleted (82). Induction of DNA damage in a cdc2-3w, Acdc25c mutant still caused a mitotic delay, showing that Cdc25 is not the only determinant of this response (82). Further studies pinpointed the Mik1 kinase as an important target of this Cdc25-independent arrest. Deletion of either Chk1 or Mik1 abrogated the residual arrest that occurred in the cdc2-3w, Acdc25c mutant. Overexpressing Chk1 caused an arrest that was dependent on Mik1, suggesting that Mik1 acts downstream of Chk1. Also, upregulation of Mik1 proteins in response to DNA damage was found to depend on Chk1 and Rad3 (82).

Similarly to Weel, Mikl phosphorylates tyrosine 15 of Cdc2, suggesting that DNA damage not only turns off the Cdc25 phosphatase that targets tyrosine 15 of Cdc2, but also turns on a kinase that phosphorylates this residue. Upregulation of the rate of inhibitory phosphorylation of Cdc2 has not been implicated as a mechanism of G2 arrest in mammals.

An additional substrate of the Atm and Atr kinases has been uncovered that may contribute to G2 arrest. Plkl is inactivated in response to DNA damage by Atm/Atr-dependent phosphorylation (83). Because Plkl can phosphorylate cyclin B1 to block export of cyclin B1 from the nucleus, one interesting possibility is that inactivation of Plk1 leaves cyclin B1 stranded in the cytoplasm (40,41). Because Cdc2 can also phosphorylate cyclin B1 in its nuclear export signal, the regulation of cyclin B1 localization is likely to be more complicated (42) (Fig. 1).

4.3. Sensors

Evidence is accumulating regarding how Atm, Atr, Chk1, and Chk2 are activated in response to DNA damage. There is clearly an involvement of damage sensors in the activation of at least some of these kinases (Fig. 1). There are several major damage-sensing machines, one of which shows striking similarity to the PCNA and RFC complexes needed for processive DNA synthesis (reviewed in 59-61). PCNA forms a homotrimeric clamp around the double helix, which is loaded onto DNA by the heteropentameric RFC complex composed of the RFC1-5 subunits. The DNA damage response involves the Rad17 protein, which forms a complex with RFC2, 3, 4, and 5, and by homology has been suggested to form a clamp loader (84,85). Rad1, Hus1, and Rad9 proteins form a trimer with structural similarity to the PCNA sliding clamp complex (86). Interactions between Rad17 and components of the Rad1-Hus1-Rad9 trimer suggest that the Rad17 complex may load the Rad1-Hus1-Rad9 complex onto DNA at sites of DNA damage (87,88). Rad1, Rad9, Rad17, and Hus1 are all required for the cell cycle delay in response to DNA damage and for the activation of checkpoint proteins such as Chk1 (59,89). Thus, the Rad17-RFC complex may recognize damaged DNA and load the Rad1-Hus1-Rad9 complex. Because Rad1 is a 3'-5' exonuclease, the Rad1-Hus1-Rad9 complex may function to increase the amount of single-stranded DNA at sites of damage to facilitate signaling to the checkpoint transducer proteins (90).

Additional complexes act proximal to DNA damage and are important in allowing the signal transducers to become activated. Brca1, originally identified as a locus of susceptibility in human breast cancer, is localized to sites of DNA damage and may be needed to recruit other proteins, such as Atm, to these sites (91-93). Atm and Atr can be found in a large complex containing Brca1, called Brca1-associated genome surveillance complex (BASC) (94).

BASC also contains the proteins Mrell, Nbsl, and Rad50, which are essential for the recognition and repair of double-strand breaks in DNA (94). Mrell, Nbsl, and Rad50 form a complex that is recruited to sites of DNA damage and helps to repair the break through nonhomologous end-joining (95,96). Studies in yeast have shown that these three proteins are needed for proper checkpoint function, but the detailed mechanism of how these proteins signal to downstream signal transducers has not been uncovered (97).

Rad3, the Atr ortholog in budding yeast, is activated in response to DNA damage, and this process is independent of the Radl7-RFC and Radl-Husl-Rad9 complexes (98). This observation is important, because it argues against a simple model in which these proximal DNA damage-sensing complexes signal to the Atm/Atr proteins. Also important is the fact that the activation of Chkl does require the Radl7-RFC and Radl-Husl-Rad9 complexes (89). The phosphorylation of Chkl appears to be directly catalyzed by Rad3 (99). If Rad3 is active in cells lacking Radl7, why can't it phosphorylate Chkl? One possibility is that Radl7 provides a docking site for Chkl and activated Rad3, either directly or by allowing damaged DNA to be processed into foci where DNA damage-signaling proteins accumulate.

A similar situation may occur in mammalian cells. Atm exists as an inactive dimer or oligomer in unstressed cells (100). Atm is phosphorylated at an autophosphorylation site and dissociates into a monomeric active kinase very rapidly (within approx 30 s) after cells are exposed to ionizing radiation. This rapid phosphorylation occurs on approx 50% of the Atm in the cell and happens at doses of radiation that would create fewer than 20 double-strand breaks (100). These data suggest that it is unlikely that every Atm dimer must diffuse to the site of damage to be activated. This suggestion is consistent with literature on budding yeast showing that Rad3 does not need the damage-sensing Rad proteins to be activated. An alternative model has been proposed in which DNA strand breaks cause changes in the higher order topology of chromatin, which can act at a distance to signal the activation of Atm. This model is supported by the observation that chloroquine and trichostatin A, drugs that can alter higher order chromatin topology, can also induce the phosphorylation of Atm without causing detectable DNA damage (100). Once activated, Atm may then diffuse to foci in the nucleus that contain damage recognition and repair complexes like BASC (Fig. 1). At those sites, Atm could phosphorylate some of its downstream targets that mediate G2 arrest, such as Chkl and Brcal.

Additional proteins may act as adapters to bring substrate and enzyme together at damage-induced nuclear foci. For example, the 53BPl protein shows regions of homology to Brcal, and is relocalized to damage-induced foci (101). 53BPl binds to Chkl and Brcal in unstressed cells, and these associations are disrupted by ionizing radiation at the same time as Chkl and Brcal become phosphorylated (102). 53BP1 was identified because it binds to p53 (103). Reduction of 53BP1 levels with small interfering RNA molecules reduces the accumulation of p53 and the phosphorylation of Chk1 that occurs in response to ionizing radiation (102). One interpretation is that 53BP1 is an adapter that brings Atm substrates to damage-induced foci, where they are phosphorylated by monomeric active Atm that has diffused to those same sites.

Biochemical and genetic approaches have uncovered a large amount of information about how the cell responds to DNA damage. An important branch of this response ultimately leads to the inactivation of Cdc2 and G2 arrest. As described above, rapid events lead to the inactivation of Cdc2 through inactivation of Cdc25C. Recent experiments in mammalian cells suggest that the p53 tumor suppressor participates in pathways that help to maintain G2 arrest in response to DNA damage.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment