INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Eukaryotic cells employ a set of surveillance mechanisms to coordinate the onset of one event and the completion of the preceding event during the cell cycle. The mechanisms that ensure the proper ordering of cell cycle events have been termed checkpoint controls in eukaryotes (11). When DNA is damaged or DNA replication is blocked, the activation of checkpoint pathways arrests the cell cycle and induces the transcription of genes that facilitate DNA repair and/or replication (5, 33).

Checkpoint pathways are an evolutionarily conserved feature of eukaryotic cells. This feature is typified in the ATM and ATR family genes which encode phosphatidylinositol 3-kinase-related proteins possessing protein kinase activity (33). In the budding yeast Saccharomyces cerevisiae, MEC1 encodes an ATR-related protein and plays a critical role in checkpoint controls (14, 17, 32). Mec1 physically interacts with Pie1 (also called Lcd1 or Ddc2), a protein that exhibits limited homology to the fission yeast Rad26 protein (17, 19, 32). Likewise, in fission yeast the ATR family protein Rad3 forms a complex with Rad26 (4). DNA damage responses have been well characterized in budding yeast and consist of the G₁-, S-, and G₂/M-phase damage checkpoints (14). Both Mec1 and Pie1 are essential for all three DNA damage checkpoints, as well as the DNA replication block checkpoint.

In addition to MEC1 and PIE1, a number of genes that control the checkpoints in budding yeast have been identified. These include DDC1, MEC3, RAD9, RAD17, RAD24, and RAD53 (5, 14, 33). RAD53 encodes a protein kinase and functions downstream of MEC1 in the checkpoint pathway. Like Mec1, Rad53 plays an essential role in both the replication block and DNA damage checkpoints. Following DNA damage and replication block, the Rad53 protein is hyperphosphorylated and activated by a mechanism dependent on Mec1 (20, 26). Thus, Mec1 and Rad53 constitute a central checkpoint pathway in budding yeast. RAD9, RAD17, MEC3, DDC1, and RAD24 are also required for DNA damage checkpoints. Rad9 is hyperphosphorylated following DNA damage, and the phosphorylated Rad9 protein binds to Rad53, possibly to modulate its activity (6, 27, 30). Genetic evidence has suggested that RAD17, RAD24, MEC3, and DDC1 operate in a common checkpoint pathway. Indeed, Ddc1, Mec3, and Rad17 interact physically with each other and function in a complex to control the DNA damage checkpoints (12). It has been shown that Ddc1, Mec3, and Rad17 are structurally related to PCNA (1, 28, 29). RAD24 encodes a protein structurally related to the subunits of replication factor C (RFC) which is required for DNA replication and repair. RFC consists of one large subunit, Rfc1, and four small subunits, Rfc2, Rfc3, Rfc4, and Rfc5 (3). Rad24 also interacts with the four small RFC subunits, Rfc2, Rfc3, Rfc4, and Rfc5, to form an RFC-related complex (9, 16). Genetic evidence has indicated that Rad24 functions upstream of the Ddc1-Mec3-Rad17 complex in the checkpoint pathway (12). RFC loads PCNA onto the primer terminus of DNA, and then DNA polymerases  and  bind to the resulting DNA-RFC-PCNA complex to form a processive replication complex (31). By analogy, the RFC-related RAD24 complex is proposed to recruit a complex consisting of Ddc1, Mec3, and Rad17, each of which is related to PCNA, to damaged DNA (9, 16, 33).

We have shown that rfc5-1 mutants are defective not only in the DNA damage checkpoint but also in the DNA replication block checkpoint (24, 25). The observation that the rad24 mutation enhances the replication block checkpoint defect in rfc5-1 mutants suggests that Rad24 plays a role in the DNA replication block checkpoint (22). However, the rad24 mutation alone causes no obvious defect in the DNA replication block checkpoint (15, 22). These results suggest that RAD24 functions redundantly with other genes in this checkpoint pathway.

The CHL12 (also called CTF18) gene encodes a protein homologous to the Rad24 and RFC proteins (3, 13). CHL12 was identified in a screen for mutants exhibiting increased rates of mitotic loss of chromosomes and has been suggested to play a critical role in DNA metabolism (13). In this paper, we show that CHL12 and RAD24 function redundantly in the DNA replication block checkpoint: this checkpoint operates normally in the single chl12 and rad24 mutants but is defective in chl12 rad24 double mutants. We also show that Chl12 interacts physically with the four small RFC subunits to form a complex that is related to, but distinct from, the RFC and RAD24 complexes. Thus, Chl12 and Rad24 are both required for the DNA replication block checkpoint.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains, media, and general methods. The yeast strains used in this study are isogenic and are listed in Table 1. Standard genetic techniques were used for manipulating yeast strains. Synthetic complete (SC) medium containing 0.5% Casamino Acids and the appropriate supplements was used to maintain selection of URA3 plasmids.

Plasmids and gene disruptions. To create the CHL12 disruption plasmid, the amino- and carboxyl-terminal regions were amplified by PCR with the amino-terminal primers KS158 (5'-ACAGAACTATCTGCAGAATGTGAATATTGTC-3') and KS162 (5'-CCAGAAGCTTGTTCTATATCACC-3') or the carboxyl-terminal primers KS163 (5'-ACTGAATTCAATCCAAAATCGGTCAGAA-3') and KS161 (5'-CGGAGCTGTGTCGACTACCGGCAGGATTGATTGCTAAT-3'). The amino- and carboxyl-terminal fragments were digested with the appropriate restriction enzymes (PstI-HindIII and EcoRI-SalI, respectively) and cloned into YIplac128 (8). The CHL12 disruption plasmid was cleaved by SalI and transformed into a diploid strain. The heterozygous diploid was sporulated, and the tetrads were dissected. Disruption of the CHL12 gene was confirmed by PCR. To construct a hemagglutinin (HA)-tagged version of CHL12, the carboxyl-terminal region of CHL12 was amplified by PCR with the primers KS160 (5'-CCATATCATGGTTTAAAATCGTGAACCAATT-3') and KS171 (5'-CTCGGATCCCTTCCCACAGGTTATTCCAAGTC-3'). The SnaBI-BamHI-treated carboxyl-terminal fragment of CHL12 and a BamHI-SalI fragment containing sequence encoding HA epitopes were cloned into the SmaI-SalI-linearized YIplac128, creating YIp-CHL12-HA. The RAD24-FLAG integration plasmid YIpU-RAD24-FLAG was constructed as follows. The carboxyl-terminal region of RAD24 was amplified by PCR with the primers KS551 (5'-TCTGAGCTCGGGCGACATCAAGTGGAAGTT-3') and KS552 (5'-CTCGGATCCGAGTATTTCCAGATCTGAATCTGAAAGGGA-3'). After treatment with SacI and BamHI, the fragment was cloned into SacI-BamHI-linearized pRS306FLAG (a gift from N. Lowndes). YCpRAD53-HA was described previously (24). To construct YIp-RAD24-myc, the EcoRI-HindIII fragment from YCpRAD24-myc (22) was cloned into YIplac204 (8). The CHL12-HA and RAD24-FLAG strains were generated by transforming YIp-CHL12-HA and YIp-RAD24-FLAG after treatment with XbaI. The RAD24-myc strains were obtained by transforming YIp-RAD24-myc after digestion with NcoI. The construction of strains carrying FLAG-tagged RFC1, RFC2, RFC3, RFC4, and RFC5 was described previously (9). Cells containing the tagged constructs expressed appropriate-sized proteins from their own promoters and showed no growth defect or increased sensitivity to DNA-damaging agents.

UV radiation and drug sensitivities. The hydroxyurea (HU), UV radiation, or methyl methanesulfonate (MMS) sensitivity assay was performed as described previously (32).

MMS synchrony experiments. To analyze cell cycle delay at the G₂/M transition, log-phase cultures at 30°C were arrested with 15 µg of nocodazole/ml for 150 min to synchronize cells in G₂/M. Cells arrested in G₂/M were incubated with 0.25% MMS for 30 min and then washed to remove the nocodazole and MMS and released into fresh yeast extract-peptone-dextrose (YEPD). At timed intervals, cells were withdrawn and stained with 4',6'-diamidino-2-phenylindole (DAPI) for microscopic examination. To examine regulation of the S-phase progression, cells were synchronized in G₁ and released into MMS at 30°C as described previously (16). Briefly, cells were grown in YEPD and treated with 6 µg of -factor/ml for 150 min. Cells synchronized with -factor were released into YEPD containing 0.05% MMS. An experiment to analyze cell cycle delay at the G₁/S transition following MMS treatment was carried out at 30°C (23). Cells were grown in YEPD and treated with 6 µg of -factor/ml for 150 min. Cells arrested in G₁/S were incubated with 0.25% MMS for 30 min and then washed to remove the -factor and MMS and released into fresh YEPD. At timed intervals, cells were withdrawn for microscopic examination.

Immunofluorescence microscopic analysis. To examine spindle elongation at 30°C, the culture was synchronized in the G₁ phase by addition of 6 µg of -factor/ml at 30°C for 150 min. The cells were then washed to remove the -factor and released into YEPD with or without 10 mg of HU/ml at 30°C. Aliquots of cells were removed and processed for DNA flow cytometry analysis and indirect immunofluorescence microscopy as described previously (32).

Immunoblotting. Protein extracts for immunoblotting were prepared and resolved by electrophoresis on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels as previously described (32). The proteins were then transferred to nylon membranes and subjected to immunoblotting analysis with the monoclonal anti-HA (3F10 or 16B12), anti-myc (9E10), and anti-FLAG (M2) antibodies, and antibody binding was detected using an ECL kit (Amersham Pharmacia Biotech).

Immunoprecipitation. Cells were grown in preculture, diluted in YEPD, and allowed to grow for 3 h at 30°C. The cells were next harvested, washed, and resuspended in lysis buffer (24). An equal volume of glass beads was added, and the cells were lysed by vortexing. Extracts were prepared as described previously (24) and incubated at 4°C for 2 h with protein G-Sepharose beads bound with anti-HA (3F10) or anti-FLAG (M2) antibody. Immunoprecipitates were washed with lysis buffer containing 300 mM NaCl and subsequently with a wash buffer and boiled immediately in 1× SDS-PAGE sample buffer. The proteins were detected after immunoblotting was performed with the antibodies described above.

Sucrose density gradient centrifugation. Extracts prepared in lysis buffer were separated by sucrose density gradient sedimentation in an SW60 rotor at 40,000 rpm for 16 h at 4°C as described previously (12). The gradients were fractionated from the top and subjected to immunoblotting with the antibodies described above.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

CHL12 encodes a protein structurally related to Rad24. Rad24 has an essential role in the DNA damage checkpoint and forms an RFC-related complex with Rfc2, Rfc3, Rfc4, and Rfc5 (9, 16). Rad24 has been suggested to participate in the DNA replication block checkpoint (22). However, rad24 single mutants are not defective in this checkpoint (15, 22). One possible explanation is that Rad24 functions redundantly with other proteins in this checkpoint pathway. The CHL12 gene encodes a 741-amino-acid protein which is structurally related to the Rad24 and RFC proteins in S. cerevisiae (13). All of the RFC proteins contain seven conserved domains, termed RFC boxes II to VIII (3). Chl12 contains RFC boxes II to V, VII, and VIII (3), whereas the Rad24 protein contains RFC boxes II, III, and VIII (15) (Fig. 1). CHL12 and RAD24 encode proteins of similar sizes, although Chl12 shows more significant homology to the RFC proteins than to Rad24. Because of its similarities to Rad24, we investigated the possible redundant roles of Chl12 with Rad24 in the DNA replication block checkpoint.

View larger version (16K): [in this window] [in a new window]   FIG. 1.   Structures of the S. cerevisiae Chl12, Rad24, and RFC proteins. There are eight RFC boxes numbered consecutively from the amino terminus to the carboxyl terminus. All of the RFC proteins possess the RFC boxes II to VIII. Box I is present only in the largest RFC subunits. The solid and shaded boxes indicate high and moderate degrees of homology, respectively. a.a., amino acids.

Chl12 and Rad24 play redundant roles in the DNA replication checkpoint. To address the potentially redundant roles of Chl12 and Rad24 in the DNA replication block checkpoint, we generated chl12 and rad24 single or double mutants and examined their sensitivities to HU treatment (Fig. 2). While neither chl12 nor rad24 single mutants were sensitive to HU, chl12 rad24 double mutants became sensitive. To test whether chl12 rad24 double mutants are defective in the replication block checkpoint, cells were synchronized with -factor at G₁ and released into medium with or without HU. Flow cytometric analysis showed that DNA replication was blocked by HU treatment in cells during the experiments (data not shown). In the absence of HU, half of the cells exhibited elongated spindles at 60 min after release from -factor, indicating that cells of all strains entered into mitosis in similar manners (Fig. 3). In the presence of HU, wild-type, chl12, and rad24 mutant cells were arrested as large budded cells with short spindles. Although chl12 rad24 double mutants delayed progression into mitosis, 20% of the cells displayed elongated spindles at 120 min after release into HU (Fig. 3). These results indicate that chl12 rad24 double mutants are partially defective in the DNA replication block checkpoint.

View larger version (12K): [in this window] [in a new window]   FIG. 2.   Viability of chl12 and chl12 rad24 mutants following exposure to HU, MMS, or UV light. Wild-type (KSC006), chl12 (KSC1148), rad24 (KSC1090), and chl12 rad24 (KSC1266) cells were grown in log phase at 30°C, treated with HU or MMS, and irradiated with UV light. The viability of the cells was estimated as described in Materials and Methods.

View larger version (17K): [in this window] [in a new window]   FIG. 3.   DNA replication block checkpoint in chl12 rad24 mutants. Cells were arrested at G1 with -factor and then released in YEPD with or without 10 mg of HU/ml at 30°C. Aliquots of cells were collected at the indicated times and stained with anti-tubulin antibodies. The percentage of cells with elongated spindles was scored as described in Materials and Methods. The strains used were the wild type (KSC006), chl12 (KSC1148), rad24 (KSC1090), and chl12 rad24 (KSC1266).

chl12 rad24

chl12

rad24

chl12 rad24

View larger version (17K): [in this window] [in a new window]   FIG. 4.   Effect of chl12 rad24 mutation on Rad53 modification following replication block. Cells carrying YCpRAD53-HA were grown at 30°C, arrested in G1 with -factor, and then released in YEPD containing 5 mg of HU/ml. Aliquots of cells were collected at the indicated times and subjected to immunoblotting analysis as described in Materials and Methods. The strains used were the wild-type (KSC006), chl12 (KSC1148), rad24 (KSC1090), and chl12 rad24 (KSC1266).

Effect of the chl12 mutation on response to DNA damage. While the rad24 mutation affects the DNA replication checkpoint in combination with other mutations as described above, the rad24 mutation alone causes a defect in the DNA damage checkpoint. We therefore characterized the possible role of CHL12 in the DNA damage checkpoint. We first examined the sensitivity of chl12 and rad24 single or double mutants to UV irradiation and MMS treatment (Fig. 2). Both chl12 and rad24 single mutants showed sensitivity to MMS, and chl12 rad24 double mutants became more sensitive to MMS than either single mutant. In contrast to rad24, the chl12 mutation had no apparent effect on sensitivity to UV irradiation. Moreover, chl12 rad24 double mutants were no more sensitive to UV irradiation than the single rad24 mutants. These results suggest that CHL12 is required for the proper response to DNA damage following MMS treatment but not following UV irradiation. It has been shown that RAD24 is required for the G₁-, S-, and G₂/M-phase DNA damage checkpoints following MMS treatment (7, 18). We therefore examined the checkpoint defect of chl12 single mutants and chl12 rad24 double mutants after MMS treatment. The G₂/M-phase DNA damage checkpoint was examined by monitoring mitotic division following DNA damage (Fig. 5A). When cell cultures were released from nocodazole arrest after MMS treatment, wild-type cells showed delayed nuclear division while rad24 cells proceeded through mitosis faster than wild-type cells. In contrast to rad24 mutants, chl12 mutants underwent mitosis at the same rate as wild-type cells. Moreover, chl12 rad24 double mutants did not proceed through mitosis any faster than rad24 single mutant cells. The S-phase DNA damage checkpoint was analyzed by monitoring the DNA content of cells experiencing DNA damage after release from G₁ block (Fig. 5B). When treated with MMS and released from -factor arrest, wild-type cells exhibited lower rates of DNA synthesis. Although rad24 mutants progressed through the S phase faster than wild-type cells, chl12 mutants went through the S phase at the same rate as wild-type cells. Again, there was no difference between chl12 rad24 double mutants and rad24 single mutants with respect to the rate of DNA synthesis. The G₁-phase DNA damage checkpoint was also examined (Fig. 5C). After MMS treatment, wild-type cells delayed the G₁/S transition as judged by budding whereas rad24 mutants went through this transition faster. Once again, the chl12 mutation had no effect on the G₁/S transition following DNA damage, either as a single mutation or in combination with the rad24 mutation. These results indicate that CHL12 is not required for the G₁-, S- and G₂/M-phase DNA damage checkpoints and does not have a role overlapping that of RAD24 in the DNA damage checkpoints.

View larger version (71K): [in this window] [in a new window]   FIG. 5.   DNA damage checkpoints in chl12 and chl12 rad24 mutants. (A) G2/M-phase DNA damage checkpoint in chl12 and chl12 rad24 mutants. Cells were grown at 30°C, arrested with nocodazole, and treated or not treated with MMS. At the indicated times after release of MMS-treated (+MMS) and untreated (MMS) cultures from nocodazole, the percentage of uninucleate large budded cells was scored by DAPI staining. (B) S-phase DNA damage checkpoint in chl12 and chl12 rad24 mutants. Cells were synchronized with -factor in G1 and released in either the presence or the absence of MMS at 30°C as described in Materials and Methods. Aliquots of cells were collected at the indicated times after release from -factor treatment and examined for DNA content by flow cytometry. The dotted lines indicate the DNA content of 1C and 2C cells. The top panels represent asynchronous (As) cells not treated with MMS at 30°C and are included as a reference. (C) G1-phase DNA damage checkpoint in chl12 and chl12 rad24 mutants. Cells were synchronized with -factor in G1 and treated with MMS (+MMS) or not treated (MMS). At the indicated times after release from -factor, the percentage of small budded cells was scored under microscopy. The strains used are the wild type (KSC006), chl12 (KSC1148), rad24 (KSC1090), and chl12 rad24 (KSC1266).

Chl12 forms a novel RFC-related complex with the RFC small subunits, Rfc2, Rfc3, Rfc4, and Rfc5. Rad24 forms an RFC-related complex with the four small RFC subunits, Rfc2, Rfc3, Rfc4, and Rfc5. Given the analogous functions of Chl12 and Rad24 in the replication block checkpoint, we examined whether Chl12 also interacts physically with Rfc2, Rfc3, Rfc4, and/or Rfc5 in vivo by immunoprecipitation experiments. We generated strains in which the corresponding genomic alleles were replaced with the HA-tagged CHL12 gene and/or one of each of the RFC genes tagged with FLAG. Extracts were prepared from cells and subjected to immunoprecipitation with anti-FLAG antibody. The immunoprecipitates were then analyzed by immunoblotting them with antibodies against the HA and FLAG epitopes. Chl12-HA was coprecipitated with Rfc2-FLAG, Rfc3-FLAG, Rfc4-FLAG, and Rfc5-FLAG in cells expressing both the tagged Chl12 and RFC proteins (Fig. 6A). In the reverse coimmunoprecipitation experiment, Chl12 was also found to interact physically with Rfc2, Rfc3, Rfc4, and Rfc5 (Fig. 6B). Rad24 and Rfc1 have been shown to form a complex with Rfc2, Rfc3, Rfc4, and Rfc5. We then tested whether Chl12 interacts physically with Rad24 or Rfc1. However, coprecipitation of Chl12 with Rfc1 or Rad24 was not detected (Fig. 6C). These observations demonstrate that Chl12 physically interacts in vivo with Rfc2, Rfc3, Rfc4, and Rfc5 but not with Rad24 or Rfc1.

View larger version (31K): [in this window] [in a new window]   FIG. 6.   Physical interaction of Chl12 with Rfc2, Rfc3, Rfc4, and Rfc5. (A and B) Interaction of Chl12 with Rfc2, Rfc3, Rfc4, and Rfc5. Extracts were prepared from CHL12-HA cells containing no FLAG construct or the indicated FLAG-tagged construct (A) or cells containing the indicated FLAG-tagged construct and no HA construct () or CHL12-HA (+) (B) and subjected to immunoprecipitation (IP) with anti-FLAG antibody (A) or anti-HA antibody (B). An F after a gene name indicates the addition of FLAG epitopes. The immunocomplexes were separated by SDS-PAGE and immunoblotted with anti-FLAG or anti-HA antibody. Whole extracts were immunoblotted with anti-HA antibody (A) or anti-FLAG (B) antibody. (C) Interaction of Chl12 with Rad24 and Rfc1. Extracts were prepared from RAD24-FLAG and RFC1-FLAG cells containing no HA construct () or CHL12-HA (+) and subjected to IP and immunoblotting analysis as for panel A. (D) Interaction of Chl12 with Rfc3 in RFC5 and rfc5-1 mutant cells. Extracts were prepared from RFC5 or rfc5-1 cells containing CHL12-HA and RFC3-FLAG and were subjected to IP and immunoblotting analysis as for panel B.

chl12

View larger version (24K): [in this window] [in a new window]   FIG. 7.   Sedimentation of Chl12, Rad24, and Rfc1 in sucrose density gradient centrifugation. Extracts were prepared from CHL12-HA RAD24-myc RFC1-FLAG (KSC1433) cells and separated by centrifugation in a 10 to 40% sucrose gradient for 16 h. The load on the gradient (L) and fractions (removed from the top of the gradient) were analyzed by immunoblotting using anti-FLAG, anti-HA, or anti-myc antibody. An F after a gene name indicates the addition of FLAG epitopes. Bovine serum albumin (4.5S) and thyroglobulin (16.5-19S) were separated simultaneously in an independent gradient as markers.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The RFC complex is required for DNA replication and repair and consists of one large and four small subunits (31). In S. cerevisiae, the large subunit of RFC is encoded by RFC1 (also called CDC44), and the four small subunits are encoded by RFC2, RFC3, RFC4, and RFC5. RFC is a structure-specific DNA-binding protein complex that recognizes a primer-template junction. Following its association with DNA at a primer end, RFC recruits PCNA onto DNA and then tethers DNA polymerase  or  to the primer junction (31). Rad24 shares a limited homology with the RFC subunits and forms an RFC-related complex with the four small RFC subunits (9, 16). RAD24 plays an essential role in the DNA damage checkpoint and functions upstream of Ddc1, Mec3, and Rad17. Ddc1, Mec3, and Rad17 possess PCNA-like structures and interact physically with each other. One model suggests that the RAD24 complex may recognize aberrant DNA structures and recruit the Ddc1-Mec3-Rad17 complex to damaged DNA.

It has been shown that the small RFC subunits Rfc2 and Rfc5 are involved in the DNA replication block checkpoint. This checkpoint response occurs normally in rad24 mutants, even though the rad24 mutation increases the severity of the checkpoint defect caused by the rfc5-1 mutation. This raises the possibility that the replication block checkpoint may be redundantly controlled by Rad24 and other RFC-related complexes. In this study, we have provided evidence indicating that Chl12, which is structurally related to the Rad24 and RFC proteins, forms an RFC-related complex and plays a role in the DNA replication block checkpoint. Although neither chl12 nor rad24 single mutants are defective, chl12 rad24 double mutants become defective in the DNA replication checkpoint. Consistent with this, phosphorylation of Rad53 following HU treatment is decreased in the chl12 rad24 double-mutant cells. Chl12 was found to interact physically with Rfc2, Rfc3, Rfc4, and Rfc5 but not with Rad24 or Rfc1. Moreover, Chl12 does not cosediment with Rad24 or Rfc1 in a sucrose density centrifugation. Thus, Chl12 forms a novel RFC-related complex and functions redundantly with Rad24 in the replication block checkpoint pathway. Because of their structural similarities to RFC, it is suggested that the Chl12- and Rad24-containing complexes function to recognize specific DNA structures and to recruit the proper apparatus to a stalled replication fork.

In rfc5-1 mutants, the interaction among the small RFC subunits is decreased, resulting in defective association between Rad24 and the small RFC subunits (16). We found that the association between Chl12 and the small RFC subunits was similarly affected by the rfc5-1 mutation. These observations suggest that the checkpoint defect observed in rfc5-1 mutants may be attributed, at least in part, to the decreased interaction of Chl12 and Rad24 with the small RFC subunits. However, these checkpoint defects could not be due entirely to the disruption of the CHL12 and RAD24 complexes, since the DNA replication block checkpoint is more defective in rfc5-1 single and rfc5-1 rad24 double mutants than in chl12 rad24 double mutants (data not shown). One possible explanation is that the remaining small RFC subunits are able to form a subcomplex retaining partial function. In fact, the human small RFC subunits form a subcomplex in vitro, although its in vivo function is not yet established (2). Alternatively, the replication block checkpoint might require the RFC complex containing Rfc1. Since disruption of RFC1 is lethal, this possibility would be best addressed if conditional rfc1 mutations were available. A nonlethal mutation allele of RFC1, cdc44-1, has no apparent effect on the DNA replication block checkpoint, either by itself or in combination with either chl12 or rad24 (data not shown). However, a more systematic analysis with rfc1 mutations would be required to argue against this possibility.

Although CHL12 plays a role in the DNA replication block checkpoint, CHL12 does not appear to be involved in the DNA damage checkpoints. In contrast, RAD24 plays a role in both the DNA replication block and damage checkpoints. If both the CHL12 and RAD24 complexes could bind to specific DNA structures, it is puzzling that CHL12 has no apparent role in the DNA damage checkpoints. One explanation could be that other checkpoint machineries might cooperate with the CHL12 or RAD24 complex to detect aberrant DNA structures and activate the checkpoint pathways. Following DNA damage, such checkpoint machineries might bind to the same DNA structures that the RAD24 complex could recognize but not ones that the CHL12 complex could. If this were the case, each of the CHL12 and RAD24 complexes would recognize distinct DNA structures resulting from DNA damage. Consistent with this model, chl12 and rad24 single mutants behave differently in response to DNA damage; rad24 mutants are sensitive to MMS and UV, whereas chl12 mutants are sensitive to MMS but not to UV. Moreover, an increase in the dosage of either complex failed to compensate for lack of the other complex in response to DNA damage; overexpression of CHL12 or RAD24 did not suppress MMS sensitivity of the other deletion mutation (data not shown). Finally, although our results indicate that Chl12 functions in a complex with the small RFC subunits, we cannot exclude the possibility that Chl12 could form a separate complex with other proteins. Such separate CHL12 complexes might be involved in repair of MMS-induced DNA lesions but not checkpoint response.

In fission yeast, the rad17⁺ gene, a homolog of RAD24, has been isolated and characterized (10), and the gene product has been shown to interact with the small RFC subunit (21). Thus, the fission yeast Rad17 protein might also form an RFC-related complex. Genetic evidence has suggested that the fission yeast rad17⁺ and budding yeast RAD24 genes might be functionally different; the rad24 mutation is defective only in the DNA damage checkpoint, whereas the rad17 mutation is defective in both the DNA damage and replication block checkpoints. However, given our observation that RAD24 plays a redundant role in the replication block checkpoint, the functions of these two related genes appear to be much more conserved than they initially appeared to be. Interestingly, a homolog of CHL12 has also been identified in the fission yeast genome database (data not shown), although it has not been characterized yet. It is possible that this Chl12 homolog plays an overlapping role with Rad17 in the fission yeast checkpoints. Homologs of the CHL12 and RAD24 genes have also been identified in humans (reference 33 and data not shown). The structural and functional conservation of the yeast genes suggests that these human homologs might control the DNA replication checkpoint in human cells.