Checkpoints and Cell Cycle Control

Cancer is caused by loss of normal controls of cell proliferation and differentiation, yet is commonly treated with radiation and genotoxic drugs which damage or inhibit the replication of DNA. Our research focuses on understanding the molecular mechanisms, which enable tumour cells to respond to and survive genotoxic stress. By understanding how these processes are controlled and how they interact with those which regulate normal cell growth and division we hope to find ways of exploiting the mutations which occur in cancer cells to make anti-cancer therapies more effective.

Eukaryotic cells respond to DNA damage or blockage of replication by triggering "checkpoint" responses, which delay cell cycle progression, promote repair, and protect genome integrity. Checkpoints are best understood in model organisms such as fission yeast, where the presence of aberrant DNA structures is relayed via signal transduction pathways, which culminate in the activation of the so-called checkpoint kinases, Chk1 and Cds1 (Figure 1). In fission yeast Chk1 is responsible for the mitotic delay that is induced by DNA damage, whereas Cds1 stabilises stalled replication forks and delays mitosis when DNA synthesis is inhibited. These responses are important for cell survival, since yeast mutants that are deficient in one or both kinases are sensitive to DNA damage and/ or replication stress.

Chk1 and Cds1 have counterparts in vertebrate cells (the vertebrate homologue of Cds1 is referred to as Chk2) raising the question of: 1) whether these kinases control analogous checkpoint functions, and 2) whether processes controlled by Chk1 and Chk2 help tumour cells to survive exposure to genotoxic stresses. This latter issue is of more than academic interest, since radiation and many other important anti-cancer chemotherapies act by damaging or interfering with DNA replication.

To dissect the molecular functions of Chk1 and Chk2 in tumour cells we have used gene targeting to "knockout" each kinase in the hyper-recombinogenic DT40 B-lymphoma cell line. By comparing how these isogenic mutant cell lines respond to agents that damage or interfere with DNA replication, we have begun to build up a picture of which checkpoints are controlled by Chk1 and Chk2. Collectively, these experiments reveal that whereas the basic mechanisms of some checkpoints have been conserved from fission yeast to vertebrates, others appear to operate in quite different ways. Our studies have also begun to provide clear evidence that processes controlled by checkpoint kinases do indeed influence tumour cell survival.

Remarkably, Chk1-deficient DT40 cells lack any measurable G2/ M arrest in response to ionising radiation, even at very high doses which induce massive amounts of DNA damage (Zachos et al., EMBO J 2003; 22: 713). Furthermore, as in fission yeast, Chk1-deficient cells are significantly radiosensitive compared to wild-type. We believe this is attributable to direct killing through division of cells with lethal DNA damage, since cells lacking Chk1 show relative resistance to radiation-induced apoptosis (Zachos et al., EMBO J 2003; 22: 713). In comparison, Chk2 seems to play only a very minor role in G2/ M arrest, although the biological significance of this contribution remains under investigation.

Even more surprisingly, it proves to be Chk1, rather than Chk2, that is responsible for replication fork stabilization when DNA synthesis is blocked. This was unexpected, since in fission yeast this is the responsibility of Cds1 (i.e. the Chk2 homologue) rather than Chk1 (Figure 1), yet as with DNA damage Chk2 appears to play little if any role in this process. In general, the molecular mechanisms and targets underlying these replication checkpoint responses are unknown; however they are clearly vital for cell survival since Chk1-deficient DT40 cells are also hypersensitive to killing by aphidicolin (Zachos et al., EMBO J, 2003; 22: 713).

Fig. 1: Checkpoint pathways in fission yeast and vertebrates. See text for details.

Chk1 is also required for mitotic delay triggered by inhibition of DNA synthesis. In fission yeast this delay is often considered to be mechanistically equivalent to that induced by DNA damage, even though they are implemented by distinct effector kinases (i.e., Chk1 or Cds1; Figure 1). When DNA synthesis is inhibited, Chk1-deficient DT40 cells enter mitosis with partially replicated DNA leading to a dramatic mitotic catastrophe phenotype (Figure 2). In marked contrast to the situation with DNA damage however, where no mitotic delay is detected, this premature mitosis occurs only after a pronounced initial delay. Remarkably, this delay is not due to compensation by Chk2. Instead, it seems to be determined by the persistence of viable replication forks, which are of course unstable in Chk1-/- cells (Zachos et al., Mol Cell Biol 2004; in press). These experiments lead us to believe that it is the process of DNA replication per se which inhibits mitosis during S-phase, and that the principal role of the replication checkpoint (and thus Chk1) is to ensure that replication forks are preserved until genome duplication is achieved. If correct, this means that the S-M checkpoint mechanism in vertebrates operates in a fundamentally different way to that which is currently accepted in fission yeast (Figure 1).

Fig 2: Mitotic catastrophe triggered by DNA synthesis inhibition in Chk1-deficient DT40 cells. Condensed chromosomes are visualised in red using an antibody specific for histone H3 phosphorylated on serine 10, whilst the mitotic spindle is visualised in green using an antibody specific for alpha-tubulin.

The picture which emerges is one in which Chk1 is both the lynchpin of the DNA damage and replication checkpoints and an important arbiter of survival in tumour cells exposed to genotoxic stress (Figure 1). Pharmacological inhibition of Chk1 therefore could in principle provide a rational strategy for enhancing the efficacy of genotoxic anti-cancer therapies, and in future it will be important to extend these studies to encompass clinically important agents. Much also remains to be learned about checkpoint mechanisms; the role of Chk2 remains to be clearly defined, we know little of the substrates through which Chk1 and Chk2 kinases exert their effects, whilst connections between checkpoints and other potential therapeutic determinants such as DNA repair and apoptosis are also under investigation.