Faithful replication of chromosomes is crucial to genome integrity. In yeast, the ORC binds replication origins throughout the cell cycle. However, Cdc45 binds these before S-phase, and, during replication, it moves along the DNA with MCM helicase. When replication progression is inhibited, checkpoint regulation is believed to stabilize the replication fork; the detailed mechanism, however, remains unclear. To examine the relationship between replication initiation and elongation defects and the response to replication elongation block, we used fission yeast mutants of Orc1 and Cdc45—orp1-4 and sna41-928, respectively—at their respective semipermissive temperatures with regard to BrdU incorporation. Both orp1 and sna41 cells exhibited HU hypersensitivity in the absence of Chk1, a DNA damage checkpoint kinase, and were defective in full activation of Cds1, a replication checkpoint kinase, indicating that normal replication is required for Cds1 activation. Mrc1 is required to activate Cds1 and prevent the replication machinery from uncoupling from DNA synthesis. We observed that, while either the orp1 or the sna41 mutation partially suppressed HU sensitivity of cds1 cells, sna41 specifically suppressed that of mrc1 cells. Interestingly, sna41 alleviated the defect in recovery from HU arrest without increasing Cds1 activity. In addition to sna41, specific mutations of MCM suppressed the HU sensitivity of mrc1 cells. Thus, during elongation, Mrc1 may negatively regulate Cdc45 and MCM helicase to render stalled forks capable of resuming replication.
DURING DNA replication, various kinds of intermediate DNA structures are formed, including single-stranded, nicked, and Y-shaped DNA. These replication intermediates appear to be unstable, especially when the progression of replication is hampered by dNTP starvation, nonhistone DNA-binding proteins, or DNA lesions (Ivessa et al. 2003; Branzei and Foiani 2005). The collapsed DNA structures resulting from stalled replication forks are a great threat to genome integrity. Therefore, stalled replication forks must be processed faithfully to prevent chromosomal aberrations that can cause cell death or cancer in multicellular organisms.
DNA replication is initiated at specific chromosomal loci known as replication origins. Origin recognition complexes (ORCs) bind the origin and serve as landing pads for other replication factors (for reviews, see Tye 1999; Lei and Tye 2001; Bell and Dutta 2002). Prior to entry into the S-phase, the minichromosome maintenance (MCM) complex is loaded onto the origin via the function of Cdc18/Cdc6 and Cdt1. Several lines of evidence indicate that the MCM complex is likely to be a replicative DNA helicase that functions with the aid of other factors and protein modifications (Aparicio et al. 1997; Ishimi 1997; Kelman et al. 1999; You et al. 1999; Chong et al. 2000; Labib et al. 2000). Cdc45 appears to be an essential accessory factor for the MCM helicase since Cdc45 is associated with the MCM helicase only during the S-phase and since DNA unwinding by the MCM helicase is stimulated by the presence of Cdc45 (Zou and Stillman 1998; Tercero et al. 2000; Walter and Newport 2000; Masuda et al. 2003). In addition, Cdc45 as well as MCM moves along the DNA, and the destruction of these factors prevents further replication following replication initiation (Aparicio et al. 1997; Labib et al. 2000; Tercero et al. 2000). Therefore, in contrast to the ORC, Cdc45 participates in both the initiation and the elongation phases of replication.
The progression of DNA replication is monitored by the checkpoint mechanism to ensure that stalled replication forks are stabilized and mitosis occurs only after all the chromosomes have completely replicated (for reviews, see Hartwell and Weinert 1989; Carr 2002; Nyberg et al. 2002; Zhou and Bartek 2004). Most factors involved in the checkpoint have been conserved from yeast to humans. In the fission yeast Schizosaccharomyces pombe, Rad3 is the central player in the checkpoint mechanism and is required for the phosphorylation and activation of the downstream kinases, Cds1 and Chk1 (Murakami and Okayama 1995; Walworth and Bernards 1996; Lindsay et al. 1998). The phosphorylation of Cds1 and Chk1 results in increased kinase activity (Lindsay et al. 1998; Lopez-Girona et al. 2001; Capasso et al. 2002). The activated kinases inhibit cell cycle-dependent kinase (CDK) by phosphorylating CDK regulators (e.g., Wee1, Mik1, and Cdc25), thus causing cell cycle arrest (Rhind et al. 1997; Furnari et al. 1999; Raleigh and O'Connell 2000). In addition, Cds1 and Chk1 may contribute to DNA metabolism by regulating the expression, localization, and activity of a set of proteins that is involved in the repair of DNA damage and/or in the processing of stalled replication forks (Huang et al. 1998; Boddy et al. 2000, 2003; Caspari et al. 2002; Sogo et al. 2002; Kai et al. 2005). Although Cds1 and Chk1 share many activation factors, including the Rad3–Rad26 complex (homolog of the ATR–ATRIP complex in humans), they are activated in different situations (Lindsay et al. 1998; Martinho et al. 1998; Harris et al. 2003). Cds1 is activated when replication is blocked during the S-phase, while Chk1 is activated when DNA damage occurs either within or outside the S-phase. The activation of Cds1 specifically depends on Mrc1, while that of Chk1 depends on Crb2/Rhp9 (Saka et al. 1997; Alcasabas et al. 2001; Tanaka and Russell 2001). In addition to its function in Cds1 activation, Mrc1 has been shown to prevent the extensive uncoupling of the Cdc45-containing replication machinery from DNA synthesis when replication is hindered by dNTP depletion (Katou et al. 2003). Mrc1 associates with the replication fork and preferentially binds branched DNA structures in vitro (Katou et al. 2003; Zhao and Russell 2004; Calzada et al. 2005; Nedelcheva et al. 2005). However, its mechanism of action in preventing the uncoupling of the replication machinery from DNA synthesis remains unclear.
In this study, we used the fission yeast mutants of Orc1 (a subunit of the ORC) and Cdc45—orp1-4 and sna41-928, respectively—and demonstrated that these mutants are hypersensitive to hydroxyurea (HU) in the absence of Chk1 and partially defective in Cds1 activation at their respective semipermissive temperatures with regard to the ability to incorporate a nucleoside analog, 5-bromodeoxyuridine (BrdU). These results are consistent with the concept that normal replication is required for full activation of the replication checkpoint. Interestingly, either the orp1 or the sna41 mutation suppresses the HU sensitivity of a cds1Δ mutant; however, only the sna41 mutation suppresses the HU sensitivity of an mrc1Δ mutant. The sna41 mutation suppresses the defect in the recovery from HU arrest but does not increase Cds1 activity in mrc1Δ cells. In addition to the sna41 mutation of Cdc45, mutations of the MCM protein suppress the HU sensitivity of an mrc1Δ mutant in an allele-dependent manner. These results suggest that Mrc1 negatively regulates Cdc45 and MCM helicase to render the stalled replication forks capable of resuming replication.
MATERIALS AND METHODS
Fission yeast strains and media:
The yeast strains used in this study are listed in Table 1. Yeast media were prepared and standard genetic procedures were carried out as described elsewhere (Alfa et al. 1993). HU (Sigma, St. Louis) was dissolved in water to 1 m, sterilized by filtration, stored at −20°, and used at the indicated final concentrations. mrc1∷kanMX6 and cds1∷kanMX6 strains were generated by transforming yeast cells with the PCR product obtained using the pFA6a-kanMX6 plasmid (Bahler et al. 1998) as a template. The primer pair used for mrc1∷kanMX6 comprised 5′-CTAAGGAGGACTAAGAGATGTATCGCGGCAAAGCAACTACCATTACTCGTTCAATAAGAGCTTTGTGGTGCTTAAATCTCGGATCCCCGGGTTAATTAA and 5′-GTTATGTAAATTATCAATACCTCATTCAAAAAAAACAAGTTTGACAAGTCCAGCTCGTCAAATCCCCTTTCTTAGCCACGAATTCGAGCTCGTTTAAAC (the sequence complementary to the mrc1+ flanking region is underlined), and that used for cds1∷kanMX6 comprised 5′-TTGATCACTCATTTGCACGTTTATTTGTGTTTACTGATATACATGGTTAAAGAATTCATCCAGTTTTTCTGTTTTTAAGAATTCGAGCTCGTTTAAAC and 5′-CTATTTACAATATTATAAATTTGACGGTCTAAGTATAAAAATTAATTAATTATCATTTAGAATACTAAATATTAATAATCGGATCCCCGGGTTAATTAA (the sequence complementary to the cds1+ flanking region is underlined). Yeast transformants were selected on yeast extract (YE) plates containing 100 μg/ml of G418 disulfate (Nacalai Tesque), and correct integration was confirmed by PCR using the primer pairs that complemented the flanking regions of the integrated DNA (the primer sequences used are available upon request). To obtain the adh1 promoter, a 0.8-kbp fragment was amplified from yeast genomic DNA using the following primer pair: 5′-GCTCTAGATCGATGACATTCGAATGGCATGCCC and 5′-GGGGTACCATATGTATGTGGTTAGAAAAAAGAAAAGAC (the sequence complementary to the adh1 promoter region is underlined). An ade6+:(adh1)p-hENT construct was created to locate a 0.8-kbp XbaI-KpnI fragment in the adh1 promoter that is followed by a 1.4-kbp KpnI-XbaI fragment containing the human equilibrative nucleoside transporter (hENT) gene from pKS007 (Katou et al. 2003), which is 70 bp downstream of the ade6+ gene. A ura4+:(adh1)p-TK construct was created to locate a 0.8-kbp ClaI-NdeI fragment in the adh1 promoter that is followed by a 1.3-kbp BamHI fragment containing the thymidine kinase (TK) gene from pYK001 (Katou et al. 2003), which is 440 bp downstream of the ura4+ gene.
Incorporation of 5-bromodeoxyuridine:
G2 cells from log-phase cultures in YE medium were collected by elutriation with a Beckman J6-MC centrifuge and resuspended in fresh medium at a concentration of 5 × 106 cells/ml. Next, BrdU and HU were added to a final concentration of 100 μm and 10 mm, respectively. After a 3-hr incubation, ∼4 × 108 cells were collected and washed with washing buffer (5 mm EDTA, 50 mm NaF), and DNA was prepared as described (Bahler et al. 1998). The DNA was fragmented to ∼0.5 kbp by sonication with Sonifier 250 (Branson, Plainview, NY) three times at a tune level of 2 for 10 sec. Following ethanol precipitation, the DNA was suspended in 1.7 ml of TEN buffer [10 mm Tris-HCl (pH 7.4), 1 mm EDTA, 150 mm NaCl] supplemented with cesium chloride (CsCl) at a final concentration of 1 g/ml, and the DNA solution was centrifuged at 80,000 rpm for 14 hr in a Hitachi CS120 with a RP120VT rotor (Hitachi). The DNA samples that were recovered from 14 fractions at the top of the centrifuge tube were dialyzed against 10:1 TE [10 mm Tris-HCl (pH 7.4), 1 mm EDTA] using the GIBCO BRL microdialysis apparatus (Bethesda Research Laboratories, Gaithersburg, MD). Following heat denaturation (95° for 5 min), an equal volume of 20× SSC [0.3 m Na-citrate (pH 7.0), 3 m NaCl] was added to the DNA samples, and they were transferred to a Nytran nylon membrane (Schleicher & Schuell, Keene, NH) using an SHM-48 Slot Blot hybridization manifold (Scie-Plas). For Southern hybridization, a 3.2-kbp NotI-XbaI fragment that was prepared from pXN289 (Okuno et al. 1997) and a 1.0-kbp XbaI fragment from p2BN052-1 (Okuno et al. 1997) were labeled with 32P using the Megaprime DNA labeling system (Amersham, Arlington Heights, IL) and employed as the autonomous replication sequence (ARS) and non-ARS fragment probe, respectively. Southern blot signals were detected using a Fuji BAS2500 phosphorimager and were measured using the Image Gauge software (Fujifilm).
Preparation of cell extracts:
Cells were washed with one-half culture volume of ice-cold washing buffer (50 mm NaF, 5 mm EDTA), frozen in liquid nitrogen, and stored at −80° before use. They were suspended in lysis buffer [50 mm HEPES (pH 7.4), 100 mm K-acetate, 2 mm EDTA, 1 mm DTT, 0.1% NP-40, 10% glycerol, 50 mm NaF, 60 mm β-glycerophosphate, 1 mm Na-orthovanadate, 1 mm PMSF, 1 mm benzamidine, 1 μg/ml leupeptin, 1 μg/ml pepstatin A] and disrupted using Bead-Beater (Biospec Products, Bartlesville, OK) after the addition of 5 μl of protease inhibitor cocktail (Sigma) and glass beads; the beads were then removed by centrifugation. After the addition of 20% Triton X-100 to a final concentration of 1%, the cell lysate was incubated at 4° for 30 min with rotation of the tube. The protein extract was cleared by centrifugation at 15,000 rpm for 10 min at 4°, and the protein concentration was determined by the Bradford assay (Protein Assay; Bio-Rad, Hercules, CA).
Cds1 kinase assays:
Cds1 kinase assays were performed essentially as described elsewhere (Boddy et al. 1998). In a total volume of 0.5 ml, 1.5 mg of cell extracts were mixed with glutathione-Sepharose-bound GST-Wee170 in lysis buffer supplemented with 1% Triton X-100. After rotation for 1.5 hr at 4°, the Sepharose beads were washed with 0.3 ml of the lysis buffer supplemented with 1% Triton X-100 and then with 0.3 ml of kinase buffer [50 mm Tris-HCl (pH 7.6), 10 mm MgCl2] three times each. The mixture of beads was then incubated with 50 μl of kinase buffer containing 100 μm ATP (Pharmacia, Piscataway, NJ) and 0.25 μl of [γ-32P]ATP (>4000 Ci/mmol, ICN Biomedicals) at 30° for 30 min. The reaction was terminated by the addition of 20 μl of Laemmli sample buffer (Bio-Rad), followed by heat denaturation. The Fuji BAS2500 phosphorimager was used to detect 32P-labeled proteins that were then measured using Image Gauge software (Fujifilm).
Flow cytometry analysis:
The cells were fixed in 70% ethanol and stored at 4°. After washing in 50 mm Na-citrate, they were suspended in 50 mm Na-citrate solution containing 100 μg/ml RNase and 500 ng/ml propidium iodide, incubated at 37° for 2 hr, and then subjected to sonication to resolve cell aggregation (Sonifier 250, Branson). The DNA content of the cells was measured using the FACScan system with CellQuest software (Becton Dickinson, Franklin Lakes, NJ).
At their respective semipermissive temperatures, the orp1-4 and sna41-928 mutants exhibit HU hypersensitivity in the absence of Chk1, a DNA damage checkpoint kinase:
Orc1 is required solely for replication initiation, while Cdc45 is required for both initiation and elongation. To examine the relationship between replication initiation and elongation and the response to replication elongation block, we used the orc1/orp1-4 and cdc45/sna41-928 mutants. To demonstrate that DNA synthesis is partially defective in these mutants at their respective semipermissive temperatures, the TK and hENT genes were integrated into the yeast genome, allowing the incorporation of a heavy thymidine analog, BrdU, into the synthesized DNA. TK converts exogenous thymidine to thymidine monophosphate, which then enters the yeast pathway of thymidine triphosphate formation (McNeil and Friesen 1981; Hodson et al. 2003). hENT is a membrane protein that facilitates the diffusion of nucleosides down their concentration gradients (Griffiths et al. 1997). Synchronous cultures in the G2 phase were obtained by centrifugal elutriation and divided into two equal aliquots. To ensure that DNA synthesis occurred only around the origin, HU, which is a specific inhibitor of ribonucleotide reductase, was added to both aliquots. BrdU was added to one aliquot to a final concentration of 100 μm. After a 3-hr incubation in the presence of HU, DNA was extracted and separated on CsCl density gradients. The fragment of interest in the gradients was determined by slot blot hybridization using the ARS and non-ARS fragments as probes (Figure 1A). BrdU incorporation into the ARS region was evident by its shift from the light to the heavy position (Figure 1B, top). As expected, the non-ARS region showed essentially the same profile in the presence or absence of BrdU (Figure 1B, bottom). At 28°, 47 and 29% of the ARS region shifted to the heavy position in the wild-type and orp1 mutant strains, respectively (Figure 1C). At 30°, the efficiency of BrdU incorporation was reduced in the sna41 mutant compared to that of the wild-type strain (Figure 1D). These results demonstrate that DNA replication is partially defective in the orp1-4 and sna41-928 mutants at 28° and 30°, respectively.
To study the effect of defective replication on the response to replication elongation block, we examined the HU sensitivity of the orp1 and sna41 mutants by a serial dilution test with logarithmically growing cells (Figure 2). Since the DNA damage checkpoint pathway is activated and functions as a backup system to maintain cell viability when the replication fork that is stalled by HU treatment is collapsed (Boddy et al. 1998; Lindsay et al. 1998), we carried out the HU sensitivity test in the presence or absence of Chk1. Although the orp1 mutant exhibited similar sensitivity as the wild-type strain, the orp1 chk1Δ mutant exhibited higher sensitivity than the chk1Δ mutant at 25° and 28° (Figure 2A). The sna41 mutant also exhibited HU hypersensitivity at 30° in the absence of Chk1 (Figure 2B). The enhancement of HU sensitivity by the elimination of Chk1 was observed in the cds1Δ mutant (Figure 2B), suggesting that the orp1 and sna41 mutants have defects in the Cds1 pathway. The temperature sensitivity of the orp1 and sna41 mutants was also enhanced by a chk1 deletion, and this underlines the importance of Chk1 in replication-deficient cells (Figure 2, data not shown).
Activation of Cds1, a replication checkpoint kinase, is partially defective in the orp1-4 and sna41-928 mutants:
To determine whether the orp1 and sna41 mutations, at their respective semipermissive temperatures, affect the activation of the Cds1 pathway, we examined the Cds1 kinase activity that is induced by HU treatment. Cell extracts of the wild-type, orp1, and cds1Δ mutant strains were prepared before and after HU treatment at 28°, and an in vitro kinase assay was performed in the presence of [γ-32P]ATP and bacteria-expressed GST-Wee170 as the substrate (Figure 3, A and B). In the wild-type strain, ∼10-fold Wee1 phosphorylation induction was observed after HU treatment. In the cds1Δ mutant, almost no Wee1 phosphorylation was observed (Figure 3A), indicating that in this assay, Wee1 phosphorylation depends on Cds1. Wee1 phosphorylation in the presence of HU was lower in the orp1 mutant than in the wild-type strain. Quantification of the phosphorimager signal revealed that the Cds1 kinase activity was slightly but significantly decreased in the orp1 mutant (Figure 3B). Further, Wee1 phosphorylation at 30° was also lower in the sna41 mutant than in the wild-type strain (Figure 3, C and D). These results are consistent with the notion that normal replication is required for full activation of the replication checkpoint kinase (see discussion).
The HU sensitivity of the mrc1 mutant is suppressed by the sna41-928 mutation but not by the orp1-4 mutation:
To explore the genetic relationship between the orp1-4 and sna41-928 mutations and the replication checkpoint mutations, a series of double mutants were constructed and examined for their sensitivity to HU (Figure 4). We observed that both the orp1 and the sna41 mutations suppressed the HU sensitivity of cds1Δ cells (Figure 4, A and B). It is possible that a reduction in the number of replication forks can partially alleviate the problem that occurs in the absence of Cds1 (see discussion).
In addition to its role in Cds1 activation, Mrc1 is required to prevent the uncoupling of the replication machinery from DNA synthesis. Contrary to the case of cds1Δ cells, mrc1Δ cells did not exhibit suppressed HU sensitivity due to the orp1 mutation (Figure 4A). However, the sna41 mutation specifically suppressed the HU sensitivity of mrc1Δ cells (Figure 4B). To confirm that the sna41 mutation is a bona fide cause of the HU sensitivity suppression in mrc1Δ cells, we introduced an empty vector (p.vector) and a plasmid containing the sna41+ or the mrc1+ gene (p.sna41 and p.mrc1, respectively) into yeast cells and examined the HU sensitivity. Figure 4C shows that sna41 mrc1Δ cells harboring p.sna41 exhibit higher sensitivity than those harboring p.vector or p.mrc1. Importantly, p.sna41 did not affect the HU sensitivity of mrc1Δ cells as well as the wild-type and sna41 cells (Figure 4C, data not shown). These results imply that a mutation of Cdc45 specifically suppresses the HU sensitivity of the mrc1Δ mutant.
A time-course analysis using an acute dose of HU revealed the efficient suppression of the HU sensitivity of mrc1Δ cells by the sna41 mutation at 30°. Figure 5A shows that mrc1Δ cells are less sensitive than cds1Δ cells and that within the time course, the sensitivity of mrc1Δ cells is almost entirely suppressed by the sna41 mutation. At the 6-hr time point, the relative viability of mrc1Δ cells was 2.6 ± 0.9%, while that of sna41 mrc1Δ cells was 118 ± 33%. Given the observation that the HU sensitivity of mrc1Δ cells was efficiently suppressed by the sna41 mutation, it is possible that the sna41 mutation increases Cds1 activity in mrc1Δ cells. To test this, we performed an in vitro kinase assay to measure the Cds1 activity in the mrc1Δ, sna41 mrc1Δ, and wild-type strains grown at 30° (Figure 5, B and C). We observed that the Cds1 kinase activity was equally low in mrc1Δ and sna41 mrc1Δ cells. Thus, it appears that the sna41 mutation suppresses the HU sensitivity but not the defective Cds1 activation of mrc1Δ cells. It is also noted that there is a residual activation of Cds1 kinase even in the absence of Mrc1, consistent with the different HU sensitivity in mrc1Δ and cds1Δ cells (Figure 5A). These results suggest that there are Mrc1-dependent and Mrc1-independent pathways for Cds1 activation.
The sna41-928 mutation alleviates the defect in recovery from replication block in the mrc1 mutant:
Mrc1 as well as Cds1 is required for recovery from the replication block that is induced by HU treatment. To examine the recovery, logarithmically growing cells of the wild-type, mrc1Δ, sna41, and sna41 mrc1Δ strains were treated with an acute dose of HU for 3 hr to block replication, washed with distilled water, and released into HU-free medium to allow them to resume cell cycle progression; the DNA content of the cells was then monitored by FACS analysis (Figure 6A). The wild-type and sna41 cells showed essentially the same phenotype; the DNA content reached 2C from 1C by 60 min after release. In contrast, the DNA content of mrc1Δ cells increased very slowly, indicating that Mrc1 is required for recovery from replication block. However, the FACS profile of sna41 mrc1Δ cells was more similar to those of the wild-type and sna41 cells than to that of mrc1Δ cells (Figure 6A, see 40 and 60 min). Recovery from the replication block was also examined by staining the cells with DAPI and counting the number of binucleate cells (Figure 6B). With the wild-type and sna41 strains, binucleate cells started accumulating at 80 min after release and reached a peak at 100 min. With the mrc1Δ strain, no accumulation of binucleate cells was apparent. However, with the sna41 mrc1Δ strain, binucleate cells accumulated although later than in the wild-type and sna41 strains. In summary, these results show that the sna41 mutation alleviates the defect in recovery from replication block in the mrc1Δ mutant.
Mutations in the MCM helicase can also suppress the HU sensitivity of the mrc1 mutant:
Given that the MCM helicase as well as Cdc45 is essential for both initiation and elongation phases of replication and that there are genetic and physical interactions between them (Miyake and Yamashita 1998; Zou and Stillman 1998; Masuda et al. 2003; Yamada et al. 2004), mutations in MCM might also suppress the HU sensitivity of the mrc1Δ mutant. To test this possibility, we constructed a series of double mutants and examined their sensitivity to HU (Figure 7). We first examined mutant alleles of the Mcm2 subunit of the Mcm2–7 complex. A temperature-sensitive mutation of Mcm2, cdc19-P1 (Forsburg and Nurse 1994), suppressed the HU sensitivity of both mrc1Δ and cds1Δ cells at 25° (Figure 7A). However, a cold-sensitive mutation of Mcm2, nda1-367 (Miyake et al. 1993), suppressed the HU sensitivity of cds1Δ but not mrc1Δ cells at 33° (Figure 7B). These results demonstrate that the HU sensitivity of mrc1Δ cells is suppressed by a mutation of Mcm2 in an allele-dependent manner.
Another subunit of the MCM complex, Mcm5, was examined. It was observed that a cold-sensitive mutation of Mcm5, nda4-108 (Miyake et al. 1993), suppressed the HU sensitivity of mrc1Δ as well as cds1Δ cells at 35° (Figure 7C). It was confirmed that the nda4 mutation is a bona fide cause of the suppression in mrc1Δ cells, by comparison of the HU sensitivity of nda4 mrc1Δ transformants harboring an empty vector or the plasmid containing the nda4+ gene (p.vector and p.nda4, respectively) (Figure 7D). As was seen in the case of sna41-928 (Figure 5A), the nda4 mutation efficiently suppressed the HU sensitivity of mrc1Δ cells in a time-course analysis using an acute dose of HU (Figure 7E). Thus, it appears that the HU sensitivity of mrc1Δ cells can be suppressed not only by a mutation of Cdc45 but also by specific mutations of the MCM helicase.
Throughout the cell cycle, the ORC binds replication origins and is required exclusively for replication initiation. In contrast, the Cdc45 protein is essential to both the initiation and the elongation phases of replication. To examine the relationship between initiation and elongation defects and the response to replication elongation block, we used the temperature-sensitive fission yeast mutants of Orc1 and Cdc45, i.e., orp1-4 and sna41-928, respectively. We observed that at their respective semipermissive temperatures, the extent of BrdU incorporation was lower in orp1 and sna41 cells than in the wild-type cells. At the same temperatures, orp1 or sna41 cells exhibited HU hypersensitivity in the absence of Chk1 kinase and were partially defective in the HU-induced activation of Cds1 kinase. While the HU sensitivity of the cds1Δ mutant was partially suppressed by either the orp1 or the sna41 mutation, the sensitivity of the mrc1Δ mutant was specifically suppressed by the sna41 mutation.
Roles of Orc1 and Cdc45 in the activation of the replication checkpoint:
To monitor the replication efficiency of orp1-4 and sna41-928 cells at their respective semipermissive temperatures, we examined the incorporation of the thymine analog BrdU into genomic DNA and observed that it was lower in the orp1 and sna41 cells than in the wild-type cells. It is likely that the initiation of replication is defective in the orp1 mutant, while both initiation and elongation are defective in the sna41 mutant. At the same temperatures, the orp1 and sna41 cells exhibited HU hypersensitivity in the absence of Chk1. A synergetic enhancement of HU sensitivity is observed between the cds1 and chk1 mutants (Boddy et al. 1998; this study). Cds1 and Chk1 checkpoint kinases are involved in the replication and DNA damage checkpoint pathways, respectively. In the absence of Cds1, at least a fraction of the stalled forks is converted into aberrant DNA structures (e.g., double-strand breaks) that in turn induce the Chk1 pathway. Thus, the enhancement of HU sensitivity by a chk1 deletion that is observed in the orp1 and sna41 mutants suggests that these mutants have a defect in the Cds1 pathway. In fact, the HU-induced activation of Cds1 kinase is partially impaired in the orp1 and sna41 mutants. Since the orp1 and sna41 mutants show HU hypersensitivity only in the absence of Chk1, the reduced activity of Cds1 kinase might be slightly insufficient for all the replication forks and/or it could not arrest cell-cycle progression sufficiently without the aid of Chk1 kinase. These results are consistent with the notion that normal replication is required for full activation of the replication checkpoint mechanism. It is possible that the number of replication forks is reduced in the replication mutant, and this accounts for the defect in checkpoint activation, as was proposed in budding yeasts (Shimada et al. 2002; Lee et al. 2003; Tercero et al. 2003). Recently, using Xenopus egg extracts, it has been shown that even after DNA synthesis is inhibited, the MCM helicase together with Cdc45 continues a certain extent of DNA unwinding to generate single-stranded DNA, which is important for checkpoint activation (Zou and Elledge 2003; Byun et al. 2005). Thus, it is also possible that at its semipermissive temperature, the replication elongation mutant is partially defective in the formation of a single-stranded region that is sufficiently long for checkpoint activation. These two possibilities are not mutually exclusive, and we consider that both hold true for the sna41 mutant, as discussed below.
A mutation of either Orc1 or Cdc45 suppresses the HU sensitivity of the cds1 mutant:
The serial dilution test revealed that the sna41-928 mutation partially suppresses the HU sensitivity of the cds1Δ mutant. Since the orp1-4 mutation also suppresses the HU sensitivity of cds1Δ, it appears that a decrease in the number of replication forks can alleviate the defect in cds1Δ. It is possible that the protein factor that is required for the stability of stalled forks is limited in a cell, and its activation by Cds1 is essential for a number of stalled forks in wild-type cells. Alternatively, Cds1 might be required for the coordinated processing of neighboring forks, since a decrease in the number of replication forks can lead to an increase in the distance between the forks. Further studies are required to address these possibilities.
A mutation of Cdc45 but not Orc1 suppresses the HU sensitivity of the mrc1 mutant:
The HU sensitivity of the mrc1Δ mutant was suppressed by the sna41-928 mutation but not by the orp1-4 mutation. This specific suppression by the sna41 mutation suggests that a defect in the elongation phase of replication accounts for the suppression of HU sensitivity in mrc1Δ cells. A time-course experiment using an acute dose of HU revealed that mrc1Δ cells are less sensitive than cds1Δ cells and that the sensitivity of mrc1Δ cells is almost entirely suppressed by sna41. When replication is blocked by HU, Mrc1 is required to activate Cds1 and prevent the uncoupling of the replication machinery from DNA synthesis (Tanaka and Russell 2001; Katou et al. 2003). The Cds1 activation defect in mrc1Δ cells does not appear to be suppressed by sna41 since the in vitro activity of Cds1 kinase was similarly low in the mrc1Δ and sna41 mrc1Δ cells. However, the resumption of DNA synthesis and cell cycle progression in mrc1Δ cells after release from HU arrest was facilitated by the sna41 mutation. Thus, it is likely that the role of Mrc1 in the prevention of uncoupling is related to Cdc45.
Mutations in the MCM helicase suppress the HU sensitivity of the mrc1 mutant in an allele-specific manner:
We found that the HU sensitivity of mrc1Δ cells can be suppressed not only by a mutation of Cdc45 but also by mutations of the MCM helicase. Mutations of the Mcm2 subunit of MCM suppress the HU sensitivity of mrc1Δ cells in an allele-dependent manner; the cdc19-P1 but not the nda1-376 mutation of Mcm2 suppresses the HU sensitivity of mrc1Δ cells. However, at the same temperatures, the HU sensitivity of cds1Δ cells is suppressed by either the cdc19 or the nda1 mutation, indicating that both mutations cause some defect under the experimental conditions. It has been shown that the nda1 mutation causes an accumulation of cells with a 1C DNA content at a restrictive temperature (Forsburg and Nurse 1994). Chromatin binding of MCM is impaired in nda1 as well as orc1/orp1-4 cells at their restrictive temperatures (Yamada et al. 2004), indicating that replication initiation is defective in nda1 cells. The initiation defect in nda1 cells may account for the HU-sensitivity suppression of cds1Δ cells, as was discussed for the orc1/orp1-4 case (see above). Inability of the nda1 mutation to suppress the HU sensitivity of mrc1Δ cells suggests that some specific defect in the elongation may be required for the suppression. Contrary to the nda1 mutation, the cdc19 mutation does not accumulate cells with a 1C DNA content at a restrictive temperature (Forsburg and Nurse 1994), suggesting that cdc19 causes a defect in an elongation phase of replication. The cdc19 mutation decreases expression levels of Mcm2 and impairs interaction between MCM subunits even at a nonrestrictive temperature of 25° (Sherman et al. 1998), the temperature at which the HU sensitivity of mrc1Δ cells is suppressed by cdc19 (this study). Thus, the unstable replication machinery formed in cdc19 cells, which may cause defects in both initiation and elongation phases of replication, could account for the suppression of both cds1Δ and mrc1Δ cells.
In addition to a mutation of Mcm2, we found that a mutation of Mcm5, nda4-108, also suppresses the HU sensitivity of mrc1Δ and cds1Δ cells. Amino acid residues conserved among the orthologs from different organisms are altered by sna41-928 (A410T), nda4-108 (R50C), and cdc19-P1 (P257L and T272I) mutations (Forsburg et al. 1997; Yamada et al. 2004), suggesting that conserved functions of Cdc45 and MCM are affected by these mutations. Interestingly, there is an intimate relationship between the nda4-108 and the sna41-928 mutations. sna41-928 was originally identified as one of the Cdc45 mutations that can suppress the cold-sensitive growth defect of nda4-108 cells (Miyake and Yamashita 1998; Yamada et al. 2004). In both nda4 and sna41 cells at their restrictive temperatures, association of Cdc45 to replication origins is severely impaired although MCM as well as another replication factor, Sld3, is loaded and accumulated on the origin (Yamada et al. 2004). In addition, in nda4 cells at a restrictive temperature, physical interaction between MCM and Cdc45 proteins detected in wild-type cells is disrupted while the MCM complex appears to be intact, and the cold-sensitivity of nda4 is partially alleviated by overexpression of the wild-type sna41+ gene. Given these observations, it seems plausible that, even at the semipermissive temperature, the association between Mcm5 and Cdc45 may be partially impaired in nda4 and sna41 cells. Collectively, the MCM complex integrity and/or its association to Cdc45 appear to be impaired in the mutants that can suppress the HU sensitivity of mrc1Δ cells. Our findings provide genetic evidence that Mrc1 negatively regulates the replication machinery containing Cdc45 and the MCM helicase for the recovery from replication block.
We are grateful to Mitsuhiro Yanagida, Paul Russell, Katsunori Tanaka, Paul Nurse, Teresa Wang, Ayumu Yamamoto, Kunihiro Ohta, and Katsuhiko Shirahige for providing strains, plasmids, and antibodies. We also thank Sanae Miyake and Shigeru Yamashita for sharing their unpublished results and the members of our laboratory for helpful discussions. This work was supported by a grant-in-aid for cancer research from the Ministry of Education, Science, Technology, Sports, and Culture of Japan and by funding from the Sumitomo Foundation and the Naito Foundation awarded to T.N.
Communicating editor: P. Russell
- Received April 30, 2006.
- Accepted July 6, 2006.
- Copyright © 2006 by the Genetics Society of America