Contrasting effects of Elg1–RFC and Ctf18–RFC inactivation in the abse
http://www.100md.com
《核酸研究医学期刊》
1Wellcome Trust Centre for Cell Biology, University of Edinburgh Michael Swann Building, Mayfield Road, Edinburgh EH9 3JR, UK 2Institute of Molecular Biology and Physiology, University of Copenhagen S?lvgade 83H, DK-1307 Copenhagen K, Denmark
*To whom correspondence should be addressed. Tel: +45 35 32 20 05; Fax: +45 35 32 20 40; Email: s.a.macneill@mermaid.molbio.ku.dk
ABSTRACT
Proliferating cell nuclear antigen loading onto DNA by replication factor C (RFC) is a key step in eukaryotic DNA replication and repair processes. In this study, the C-terminal domain (CTD) of the large subunit of fission yeast RFC is shown to be essential for its function in vivo. Cells carrying a temperature-sensitive mutation in the CTD, rfc1-44, arrest with incompletely replicated chromosomes, are sensitive to DNA damaging agents, are synthetically lethal with other DNA replication mutants, and can be suppressed by mutations in rfc5. To assess the contribution of the RFC-like complexes Elg1–RFC and Ctf18–RFC to the viability of rfc1-44, genes encoding the large subunits of these complexes have been deleted and overexpressed. Inactivation of Ctf18–RFC by the deletion of ctf18+, dcc1+ or ctf8+ is lethal in an rfc1-44 background showing that full Ctf18–RFC function is required in the absence of fully functional RFC. In contrast, rfc1-44 elg1 cells are viable and overproduction of Elg1 in rfc1-44 is lethal, suggesting that Elg1–RFC plays a negative role when RFC function is inhibited. Consistent with this, the deletion of elg1+ is shown to restore viability to rfc1-44 ctf18 cells.
INTRODUCTION
The ring-shaped homotrimer proliferating cell nuclear antigen (PCNA) plays key roles in many aspects of DNA metabolism in eukaryotic cells. PCNA encircles DNA to form a sliding clamp that can move freely along the duplex and acts to tether a host of interacting proteins implicated in DNA replication, DNA repair and DNA modification, such as DNA ligase I, the nucleases Fen1 and XP-G, uracil-N-glycosylase and cytosine-5-methyltransferase. However, the central function of PCNA is to act as a processivity factor for the replicative DNA polymerases. In the absence of polymerase tethering by PCNA, chromosomal DNA replication cannot be completed efficiently, if at all .
The loading of PCNA onto DNA requires that the PCNA ring be opened and then closed around the duplex. This reaction is catalysed in an ATP-dependent manner by a clamp loader complex. Clamp loaders have been identified in all three kingdoms of life. In eukaryotic cells, the first clamp loader to be identified was replication factor C (RFC), discovered in the studies of SV40 viral DNA replication in vitro (3,4). RFC is a pentameric complex comprising a large subunit (Rfc1) and four smaller subunits (Rfc2–Rfc5). All the five subunits are members of the AAA+ protein superfamily of ATPase and ATPase-related proteins, and are essential for chromosomal DNA replication and cell survival in yeast. Recently, the structure of the yeast RFC complex was solved, revealing a complex with spiral geometry that closely matches to that of the DNA duplex (5).
In addition to RFC, three RFC-like complexes have been identified in recent years, two of which have been demonstrated to possess clamp loading activity, albeit with different sliding clamps . In this study, the fission yeast protein nomenclature is used when describing these complexes, although it should be noted that much of the information that has been gathered on their functions has come from the analysis of budding yeast (see Table 1 for nomenclature summary).
Table 1 Yeast RFC/RFC-like complex and sliding clamp nomenclature
The first RFC-like complex to be identified (Rad17–RFC in fission yeast) performs its function in DNA structure checkpoints . The Rad17–RFC comprises the same four small subunits as RFC, Rfc2–Rfc5, but the large subunit Rfc1 is replaced by the related Rad17 protein (8–10). The effect of substituting Rad17 for Rfc1 is to alter the biochemical properties of the complex, notably its DNA substrate specificity (11–13). However, the Rad17–RFC also functions with a different sliding clamp. Instead of PCNA, Rad17–RFC loads the 9-1-1 complex onto DNA. This alternative sliding clamp comprises the Rad9, Rad1 and Hus1 DNA checkpoint proteins (Table 1) (11–13). It is thought that the Rad17–RFC and 9-1-1 complex act together as sensors of DNA structure perturbations.
The second RFC-like complex to be identified was Ctf18–RFC (14,15). In this complex, the large subunit of RFC is substituted with Ctf18. Two additional factors, Dcc1 and Ctf8, also associate with Ctf18, resulting in a seven subunit complex. The Ctf18–RFC is required for the efficient establishment of chromosome cohesion during S-phase, the process by which the newly replicated sister chromatids are held together until anaphase. Although all the three genes are non-essential, Saccharomyces cerevisiae cells carrying ctf18, dcc1 or ctf8 mutations display genome instability and chromosome loss phenotypes (14,16). Biochemical evidence indicates that Ctf18–RFC functions as an ATP-dependent PCNA loader, loading PCNA onto primed templates 10-fold less efficiently than RFC (17,18).
The third RFC-like complex sees Rfc1 substituted by Elg1 . Mutants in the budding yeast ELG1 gene were identified in a number of independent genetic screens for factors affecting genome stability in diverse ways (20–27): elg1 mutants display increased levels of mitotic recombination, chromosome loss, transposition and gross chromosomal rearrangement. The elg1 mutants also display telomere lengthening and are synthetically lethal or sick when combined with mutations in various DNA processing pathways, such as homologous recombination (synthetic interactions with rad50, rad51, rad52, rad54, rad55, rad57, mre11 and xrs2), replication fork re-start (mus81 and mms4) and DNA replication (pol32, dna2 and rad27). The elg1 cells are also mildly sensitive to the alkylating agent methylmethanesulphonate (MMS) but not to hydroxyurea (HU) or UV treatment. Despite this analysis, the precise function of the Elg1–RFC is unknown, although it has been shown that the complex co-immunoprecipitates with PCNA (23). The complete subunit composition of the Elg1–RFC is also unknown.
Although the small and large subunits of the clamp loader complexes are related to one another, the large subunits have extended N- and C-terminal regions. In the case of Rfc1, the function of the N-terminal extended region has been analysed previously (28–30). This region includes both a conserved PCNA-binding motif with a potential role in targeting RFC to replication factories (31) and a BRCT domain (32), and possesses DNA-binding activity (28,33) but is not required either for PCNA loading in vitro or for RFC function in vivo. The function of the C-terminal domain (CTD) has been less well characterized, although in vitro reconstitution experiments point to an important role for this domain in RFC complex assembly (28). Recent structural studies suggest a role for this domain in stabilizing the RFC complex by bridging the ends of the spiral assembly of RFC subunits but the functional consequences of this are not known (5).
In this study, the CTD of Schizosaccharomyces pombe Rfc1 is shown to be essential for RFC function in vivo, and the isolation and the analysis of novel mutant rfc1 alleles generated by random and targeted mutagenesis of the CTD, including a temperature-sensitive mutant, rfc1-44, is described. Analysis of rfc1-44 shows the cells to be incapable of completing DNA replication at their restrictive temperature, sensitive to DNA damaging agents, and synthetically lethal with a number of DNA replication mutants. To investigate this further, genes encoding S.pombe orthologues of Elg1, Ctf18, Dcc1 and Ctf8 are identified, elg1, ctf18, dcc1 and ctf8 strains constructed and analysed, and rfc1-44 ctf18, rfc1-44 dcc1 and rfc1-44 ctf8 double mutants shown to be inviable, implying that Ctf18–RFC function is essential for cell viability in the absence of fully functional RFC. The purification of the Elg1–RFC from fission yeast is also described, demonstrating for the first time in any organism that this complex has a five subunit structure. In support of a model in which Elg1–RFC plays a negative role when RFC function is impaired, elevated levels of Elg1–RFC are shown to be lethal in rfc1-44 and the deletion of elg1+ shown to restore viability to rfc1-44 ctf18 cells. Finally, the identification of suppressors of rfc1-44 mapping within the rfc5 gene is reported, suggesting that the functional defect in RFC in rfc1-44 cells may lie in the Rfc1–Rfc5 interaction revealed by the crystal structure.
MATERIALS AND METHODS
Yeast strains, media and methods
Haploid S.pombe strains used for strain construction were as follows: Sp236 (leu1-32 ura4-D18 h+), Sp292 (leu1-32 ura4-D18 ade6-M210 h–), Sp293 (leu1-32 ura4-D18 ade6-M216 h+), Sp322 (leu1-32 ura4-D18 h–), Sp347 (972 h–) and Sp348 (h+S). PCR-mediated gene targeting was used to construct the following strains: Sp323 (elg1::natMX6 leu1-32 ura4-D18 h–), Sp325 (elg1::kanMX6 leu1-32 ura4-D18 h–), Sp359 (dcc1::natMX6 h–), Sp360 (dcc1::natMX6 h+S), Sp365 (ctf18::natMX6 h–), Sp378 (ctf8::natMX6 h–), Sp380 (ctf8::natMX6 h+S), Sp382 (ctf18::natMX6 h+S) and Sp402 (elg1-TAP-natMX6 h–). A wild-type diploid strain (leu1-32/leu1-32 ura4-D18/ura4-D18 ade6-M210/ade6-M216 h–/h+) was created by standard methods by mating haploids Sp292 and Sp293 and used to construct various rfc1 mutants (see below). The following rfc1 strains were generated by subsequent matings: HSp1 (rfc1-44 h+), HSp9 (rfc1-54 h+), HSp17 (rfc1-44 h–), HSp23 (rfc1-54 h–), HSp25 (rfc1-44 elg1::natMX6 h+), HSp26 (rfc1-44 elg1::natMX6 h–), HSp27 (rfc1-44 elg1::kanMX6 ctf18::natMX6 h+) and KSp1 (elg1::kanMX6 ctf18::natMX6 h–). Yeast growth media (YE, EMM and ME) were used essentially as described (34). 5-Fluoroorotic acid (5-FOA) selection was performed using YE medium supplemented with 1 mg/ml 5-FOA. G418 (Geneticin; from Invitrogen/GibcoBRL) and nourseothricin (ClonNAT; from Werner BioAgents) were used at 100 mg/l in YE agar. Transformations were performed by electroporation (35) or lithium acetate methods (36) as appropriate. Tetrad dissection was accomplished using a micromanipulator (Singer Instruments). Flow cytometry on fixed propidium-iodide stained cells was performed by using a FACS Calibur instrument and CellQuest software (BD Biosciences) as described previously (37), cell number counts by using a Z2 particle counter (Beckman Coulter), and pulse-field gel electrophoresis (PFGE) as described by Kelly et al. (38) using a BioRad CHEF DRII system.
Construction and analysis of rfc1+/rfc1-C strain
For plasmid integration into the genome, a series of constructs based on plasmid pJKE1 were used. pJKE1 corresponds to pUR19 (39) from which ars1+ has been deleted by ClaI digestion and re-ligation. Initially, sequences encoding amino acids 590–935 of Rfc1 (corresponding to the CTD plus 130 amino acids N-terminal to the CTD) were amplified by PCR using pUR19-Rfc1 as template and cloned into plasmid pJKE1 to generate plasmid pJKE11. The PCR overlap extension method (39) was then used, with pJKE11 as template and complementary 50mer oligonucleotides, to introduce stop codons in all three frames into the rfc1+ sequence, so that the encoded protein would terminate after Pro719. The product of the second round of PCR was cloned into pJKE32 (see below) to generate pJKE59. Linearized plasmid DNA was then transformed into an S.pombe diploid strain (leu1-32/leu1-32 ura4-D18/ura4-D18 ade6-M210/ade6-M216 h–/h+) and stable integrative transformants isolated and characterized by PCR using genomic DNA as template. Sequencing of PCR products generated in this manner allowed confirmation of the genotypes. The sequences of all the oligonucleotides used can be obtained from the authors on request. Diploids were then induced to sporulate on ME medium and asci dissected using a micromanipulator (Singer Instruments). Following re-growth, the genotypes of viable colonies were confirmed by replica plating.
Pentapeptide insertion mutagenesis
To mutagenize rfc1+ using the pentapeptide insertion method, sequences encoding amino acids 590–935 of Rfc1 (corresponding to the CTD plus 130 amino acids N-terminal to the CTD) were amplified by PCR and cloned into plasmid pJKE31, a derivative of pJKE1 with the polylinker KpnI site removed (see below), to generate pJKE32. Next, pJKE32 was mutagenized by the PCR overlap extension method (40) to remove the KpnI site present in the rfc1+ sequence (the sequence G GTA CC was changed to G GTT CC, with no effect on encoded amino acid sequence), to make pJKE42. Pentapeptide insertion mutagenesis was performed using the Tn4430 system (41,42). pJKE42 was transformed into E.coli FH1046 and individual transformant colonies were mated with E.coli DS941 after which the mating mixtures were plated on the medium to select for clones containing pJKE42 carrying the Tn4430 transposon. Isolates containing a Tn4430 insertion within the rfc1 region were then identified by colony PCR and plasmid DNA prepared. This was restricted by KpnI and re-ligated to remove the bulk of the transposon, before being transformed into E.coli DH5. Seventeen individual transposon-free clones were isolated and the position of the 15 bp insertion determined by KpnI restriction mapping and DNA sequencing. Six duplicated mutants were discarded at this point, leaving a collection of twelve different pentapeptide insertion alleles (rfc1-43–rfc1-54, carried by plasmids pJKE43–pJKE54). Site-specific mutations (rfc1-33–rfc1-41, carried by plasmids pJKE33–pJKE41) were generated by PCR overlap extension mutagenesis (40) using complementary 36mer oligonucleotides and plasmid pJKE11 as a template. The sequences of the oligonucleotides used can be obtained from the authors upon request. Following the second round of PCR the products were cloned into pJKE32. Subsequently, linearized plasmid DNA was transformed into diploid S.pombe and integrative transformants isolated and analysed as described above. For further analysis, 5-FOA was used to select for rfc1-44 and rfc1-54 strains in which the plasmid had recombined out of the genome leaving a mutated copy of the rfc1 gene on the chromosome. These strains were then backcrossed to wild type to remove auxotrophic markers, resulting in strains HSp1 (rfc1-44 h+), HSp9 (rfc1-54 h+), HSp17 (rfc1-44 h–) and HSp23 (rfc1-54 h–).
Suppressor screening
To screen for suppressors of rfc1-44, a 1.3 kb fragment was amplified by PCR from the rfc5+ chromosomal region and subcloned into plasmid pJKE31 (above). This corresponded to 200 bp 5' to the rfc5+ open reading frame (ORF) and 1.2 of sequence encoding the first 289 (of 358) amino acids of Rfc5. The resulting plasmid pJKE55 was subjected to mutagenesis with Tn4430 as described above except that, following mating with E.coli DS941, a library of transposon-containing plasmids was prepared, digested with KpnI, re-ligated and transformed in E.coli DH5. Plasmid DNA was prepared from the >3000 pooled colonies, linearized and used to transform rfc1-44.
Gene deletions
The elg1+, ctf18+, dcc1+ and ctf8+ genes were deleted in wild-type 972 h– (SP347) and h+S (Sp348) strains using the PCR-mediated gene targeting method for fission yeast as described by B?hler et al. (36) using either pFA6a-kanMX6 or pFA6a-natMX6 as PCR templates. The latter (a gift from P. Hentges and T. Carr, University of Sussex) confers resistance to nourseothricin (also known as clonNAT, supplied by Werner BioAgents), which was added to YE plates at 100 μg/ml. Oligonucleotides (100mer with 80 nt of gene-specific sequence) were supplied by DNA Technology A/S. The deletions were confirmed by PCR using oligonucleotide primers flanking the cassette. The sequences of all the oligonucleotides used can be obtained from the authors upon request.
Purification of the Elg1–RFC
To facilitate purification of the Elg1–RFC, the chromosomal elg1+ gene was tagged at its 3' end with sequences encoding the TAP module marked with the natMX6 cassette. For this, the kanMX6 module was first excised from pFA6a-CTAP-kanMX6 (43) and replaced with the natMX6 cassette from pFA6a-natMX6 (above). The resulting plasmid, pFA6a-CTAP-natMX6, was then used as the template for PCR using 100mer oligonucleotides (sequences from the authors upon request), the PCR product transformed into S.pombe strain Sp347 and nourseothricin-resistant colonies identified. Successful tagging was confirmed by PCR and direct sequencing of amplified chromosomal DNA fragments. For purification, a 12 litres culture of the elg1-TAP-natMX6 h– strain (Sp402) was grown to an OD600 nm of 0.8 (1 x 107 cells/ml) at 32°C, before the cells were harvested and processed as described previously (43) except that the cells were broken by grinding in liquid nitrogen using an RM100 mortar grinder (Retsch) for 15 min. Following TCA precipitation, the samples were resolved by SDS–PAGE and silver stained. Trypsin digestion of excised bands and mass spectrometric analysis was carried out commercially at the Proteomics Facility at the University of Dundee, UK. (http://www.dundee.ac.uk/biocentre/FingerPrints/).
RESULTS
The conserved CTD of fission yeast Rfc1 is required for RFC function
In all eukaryotic species, the large subunit of RFC differs from the small subunits in having extended N- and C-terminal regions (Figure 1A). In order to probe the function of the CTD in greater detail, one copy of the chromosomal rfc1+ gene in a diploid S.pombe strain was altered to encode a truncated Rfc1 protein lacking the entire 215 amino acid CTD (Materials and Methods). This rfc1+/rfc1-C diploid was sporulated and the properties of haploid rfc1-C cells analysed following meiotic tetrad dissection. Spores expressing the truncated Rfc1-C protein were capable of germination and 1–5 cell divisions, but incapable of colony formation (data not shown). Similar to S.pombe rfc1 cells (44), the arrested rfc1-C cells were often elongated, indicative of a cell cycle defect. Transformation of the rfc1+/rfc1-C diploid prior to sporulation with a plasmid carrying wild-type rfc1+ rescued the rfc1-C haploid, allowing efficient colony formation (data not shown). The CTD is therefore required for RFC function in vivo, consistent with the earlier observation that the deletion of sequences from the C-terminus of the human Rfc1 orthologue hRFC140 abolished RFC complex formation in vitro (28).
Figure 1 (A) Schematic structure of the RFC subunits. Each small subunit (Rfc2–Rfc5) is a member of the AAA+ family of proteins and comprises three structural domains I–III (indicated above the Rfc2 protein). The large subunit Rfc1 is also an AAA+ protein and possesses the same domains but has additional N- and C-terminal regions. The N-terminal region contains a PIP (PCNA-interacting protein) motif and a single BRCT motif of unknown function. The function of the CTD is not known. Each RFC subunit also contains a number of conserved sequence blocks, RFC boxes II–VIII (grey boxes, shown numbered on the Rfc1 protein). (B) Location of mutants within the Rfc1 CTD. An expanded view of part of domain III and the CTD is shown, spanning amino acids 590–935 (the CTD spans 720–935). The mutants generated in this study are indicated as follows: open circles (pentapeptide insertion mutants, no significant effect on protein function), grey circles (pentapeptide insertion mutants, temperature-sensitive), closed circles (pentapeptide insertion mutants, non-functional), open squares (amino acid substitutions, no significant effect on protein function). See Supplementary Information for more details.
Mutational analysis of the CTD
To further probe its function, the CTD was subjected to mutational analysis. Both random and directed mutagenesis strategies were used to generate a total of 21 mutant rfc1 alleles. Random mutagenesis was accomplished using the Tn4430-based pentapeptide insertion method (41,42). Twelve alleles were constructed by this method (Figure 1B, circles), seven of which contained insertions within the CTD. The remaining five insertions mapped N-terminal to the CTD, within the RFC collar domain (5). Oligonucleotide-directed in vitro mutagenesis was used to generate a further nine rfc1 alleles with mutations within the CTD (Figure 1B, squares). All nine mutations changed individual conserved charged amino acids to alanine (see Supplementary Information for details).
As above, the mutant alleles were targeted to one copy of the rfc1+ gene in a diploid strain (Materials and Methods). Following meiosis and sporulation, the properties of the rfc1 mutants were analysed by tetrad dissection and re-growth. The results are summarized in the Supplementary Information. Briefly, 13 of the 21 alleles were viable in haploid form and essentially indistinguishable from wild type (Figure 1B, open circles or squares) whereas only one, rfc1-43, was inviable (Figure 1B, closed circle). The latter encodes an Rfc1 protein with a pentapeptide insertion within the CTD. Two of the mutant alleles, rfc1-44 and rfc1-54, displayed a temperature-sensitive phenotype (Figure 1B, grey circles and Figure 2A) that could be rescued by wild-type rfc1+ (data not shown).
Figure 2 Characterization of temperature-sensitive rfc1 alleles. (A) Serial dilutions of wild-type (rfc1+), rfc1-44 and rfc1-54 cells incubated at 18, 25 and 35°C, respectively, for 3–4 days. (B) Flow cytometry of propidium iodide stained wild-type and rfc1-44 cells following shift from 25 to 35°C with or without the addition of HU. Samples for analysis were taken every hour from 0 to 8 h. (C) PFGE analysis of chromosome structure in wild type or rfc1-44 following shift from 25 to 35°C. Samples for the analysis were taken every 2 h from 0 to 8 h. Note that chromosome III in rfc1-44 cells appears somewhat shorter than in the wild-type control, indicative of the loss of rDNA repeats. (D) Sensitivity of wild-type, rfc1-44 and rfc1-54 cells to camptothecin (CPT, 10 μM), hydroxyurea (HU, 5 mM), methylmethane sulphonate (MMS, 0.005%) and UV light (300 J/m2). Serial dilutions were spotted and incubated at 25°C for 4 days.
One of these alleles, rfc1-44, was characterized further. When shifted to the restrictive temperature of 35°C, rfc1-44 cells arrest with a 2C DNA content (determined by flow cytometry of propidium iodide stained cells, Figure 2B) indicating that bulk DNA replication is completed prior to arrest. Note that the apparent increase in DNA content seen following shift to the restrictive temperature (seen as a gradual rightwards drift in the position of the 2C peak over the timecourse of the experiment shown in Figure 2B) is the result of ongoing mitochondrial DNA replication in elongating (i.e. cell cycle arrested) S.pombe cells, as described previously (37). Similar peak movements are shown in Figure 6 for rfc1-44 single-, double- and triple-mutant combinations (see below), but not in cells of normal length.
Figure 6 Regulated overproduction of RFC/RFC-like complex large subunits in wild-type rfc1-44, ctf18 and elg1 cells. (A) Expression in wild-type and rfc1-44 cells growing on medium containing thiamine (promoter repressed). (B) Expression in wild-type and rfc1-44 cells growing on medium lacking thiamine (promoter derepressed). (C) Expression in ctf18 and elg1 cells growing on medium containing thiamine. (D) Expression in ctf18 and elg1 cells growing on medium lacking thiamine. Cells were transformed with the indicated plasmids (pREP3X, etc.) and serial dilutions plated on EMM plates either with or without 5 μg/ml thiamine, to either repress or induce large subunit expression from the nmt1 promoter. Plates were incubated at 25 or 35°C as indicated for 3–5 days.
To analyse DNA structure in the arrested cells, PFGE of chromosomal DNA from wild-type and mutant cells was performed. Many previously isolated yeast DNA replication mutants display a 2C DNA content by flow cytometry but their chromosomes cannot enter a PFGE gel (45). Similar results are obtained when chromosomes are prepared from the cells arrested in S-phase with HU, for example, and are taken as an indication that the chromosomes remain at least partly unreplicated (45). Significantly, by 4 h after temperature shift, the chromosomes from rfc1-44 cells failed to enter the PFGE gel, indicating that the replication is not complete under these circumstances (Figure 2C). At the permissive temperature of 25°C, rfc1-44 cells display increased sensitivity compared with wild type to the DNA replication inhibitor HU and to the DNA damaging agents, MMS and UV, consistent with the previous data implicating RFC and PCNA in various DNA repair mechanisms (Figure 2D). In addition, rfc1-44 is synthetically lethal at 25°C with a number of temperature-sensitive DNA replication mutants (data not shown), including pol3-ts3, cdc1-P13, cdc27-P11, cdc27-D1 and dna2-C2 (46–49). The pol3+, cdc1+ and cdc27+ genes encode the catalytic, B and C subunits, respectively, of DNA polymerase in S.pombe, while dna2+ encodes an endonuclease-helicase implicated in Okazaki fragment processing.
The rfc1-44 mutant can be rescued by mutations in rfc5
The structure of the budding yeast RFC complex has been solved recently (5). The structure shows the ATPase domains of all five subunits forming a right-handed spiral whose pitch matches that of duplex DNA and the CTD forming a physical link between the two ends of the spiral, bridging Rfc1 and Rfc5. This suggested to us that the RFC complex in rfc1-44 cells might be defective in this linking function and that it might be possible to identify suppressors of rfc1-44 by randomly mutagenizing rfc5+. The cloned rfc5+ gene was therefore subjected to pentapeptide insertion mutagenesis and a library of mutagenized plasmids constructed (Materials and Methods). The library DNA was then linearized by restriction enzyme digestion and transformed into rfc1-44 cells. Approximately 2400 integrative transformants obtained at 25°C were tested for their ability to grow at 35°C and six clones capable of growth selected for further analysis, including sequencing of the chromosomal rfc5 gene (Figure 3). Remarkably, all six contained pentapeptide insertions within a very short stretch of the Rfc5 protein, despite the random distribution of insertions in individual library clones as determined by restriction mapping (data not shown). Indeed, five of the six insertions (named rfc5-S1 through rfc5-S5, see Figure 3B) were identical, containing an insertion of the pentapeptide sequence PSRGT between amino acids 75 and 76. The sixth allele (rfc5-S6) contained an insertion of the sequence SRGTP between amino acids 76 and 77. Unfortunately, amino acids 68–85 of budding yeast Rfc5 are absent from the published three-dimensional structure, making it impossible to locate these in relation to the CTD (5). However, these genetic observations suggest that there may be a close functional relationship between those regions of the Rfc1 and Rfc5 proteins defined by rfc1-44 and rfc5-S and support a model in which the biochemical defect in the RFC (Rfc1–44) complex lies in bridging the ends of the RFC spiral.
Figure 3 Suppression of rfc1-44 by rfc5-S1. (A) Upper parts: serially diluted cultures of wild-type, rfc1-44 and rfc1-44 rfc5-S1 double mutant cells plated at 25°C (left panel, 4 days) or 35°C (right panel, 3 days). Lower parts: the same serial dilutions plated on either 5 mM HU (left panel) or 0.005% MMS (right panel) and incubated at 25°C for 4 days. Although rfc5-S1 can suppress rfc1-44 at 35°C, it does not fully restore rfc1-44 activity as the double mutant cells are still sensitive to HU and MMS. See the text for details. (B) Sequence alignment of S.pombe, S.cerevisiae and human Rfc5 proteins showing location of rfc1-44 suppressing insertion alleles. The structure of this region is unknown, except for a short ?-strand (1SXJ ?22) encompassing amino acids 86–88. Conserved residues are indicated by dots beneath the aligned sequences.
Fission yeast Elg1–RFC
During the course of this work, several groups reported the isolation and genetic analysis of the budding yeast ELG1 gene encoding the large subunit of a novel RFC-like complex (20–27). To investigate whether a similar complex existed in S.pombe, the fission yeast elg1+ gene was identified by BLAST searching (E-value 6.7 x 10–16). elg1+ (systematic name SPBC947.11c, see Table 2 and Figure 4A) lies on S.pombe chromosome II and encodes a 2760 bp ORF interrupted by two introns of 44 and 47 nt, resulting in a predicted protein of 920 amino acids and Mr of 103.9 kDa.
Table 2 RFC and RFC-like complex components in fission yeast
Figure 4 Fission yeast Elg1. (A) Chromosome context and schematic structure of the elg1+ gene on chromosome II. (B) TAP of Elg1–RFC. Protein extracts were prepared from elg1-TAP cells and subject to TAP. Following TCA precipitation, the purified proteins were subjected to SDS–PAGE and silver staining (lane 1). The bands indicated were excised and identified by mass spectrometry (see Supplementary Information). Wild-type cells (lane 2) were processed in parallel. The species marked with an asterisk corresponds to N-terminally degraded Elg1. (C) The elg1 cells display increased sensitivity to MMS. Serial dilutions were plated on YE medium containing 0, 0.002, 0.004 or 0.006% MMS and incubated for 4 days at 32°C. (D) The rfc1-44 elg1 cells are viable at 28°C. Eight tetratype tetrads (labelled 1–8) are shown. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate.
To determine the composition of the Elg1–RFC complex, the chromosomal elg1+ gene was tagged at its 3' end with sequences encoding the tandem affinity purification (TAP) tag and the Elg1–RFC purified by the TAP technique (43). Silver staining and subsequent mass spectrometric analysis of the purified material identified Elg1 in both full-length and N-terminally degraded forms, as well as all four small RFC subunits (see Figure 4B and Supplementary Material for the coverage of individual proteins). No additional proteins were identified in these purified preparations, strongly suggesting that Elg1–RFC comprises the Elg1, Rfc2, Rfc3, Rfc4 and Rfc5 proteins only. If the Elg1–RFC complex contains additional proteins, perhaps similar to Dcc1 and Ctf8 in Ctf18–RFC, these must be present in sub-stoichiometric amounts relative to full-length Elg1, be refractive to silver staining, or lost during the TAP purification. Further analysis will be required to address these issues.
To study its function, elg1+ was deleted from the chromosome and replaced with the natMX6 cassette by the PCR-mediated gene targeting method (36). Cells carrying the elg1 allele (elg1::natMX6) were viable and displayed mild sensitivity to MMS (Figure 4C) but were insensitive to UV, HU and the topoisomerase inhibitor camptothecin (data not shown) as is the case for budding yeast elg1 cells (21–23).
Next, rfc1-44 was crossed to elg1. The double mutant rfc1-44 elg1 was viable at 28°C (Figure 4D) and essentially indistinguishable from rfc1-44. However, closer inspection suggested that at 36.5°C the rfc1-44 elg1 cells grew marginally better than the rfc1-44 mutant alone, suggesting that elg1+ might exert a negative effect on rfc1-44 cells (data not shown). This is explored further below.
Fission yeast Ctf18–RFC
To complement our studies on the S.pombe Elg1–RFC, the role of the fission yeast Ctf18–RFC in rfc1-44 cells was examined. In S.cerevisiae, the Ctf18–RFC comprises seven subunits: Ctf18, Rfc2, Rfc3, Rfc4, Rfc5, Dcc1 and Ctf8 (2). The same appears to be true of the human Ctf18–RFC also (17). To investigate whether S.pombe possessed Ctf18, Dcc1 and Ctf8 proteins and to test their function in relation to RFC, the fission yeast genome sequence was probed for genes encoding putative fission yeast Ctf18, Dcc1 and Ctf8 orthologues. Searches using the budding yeast proteins as the query sequences readily identified putative Ctf18 and Dcc1 orthologues (BLAST E-scores of 3.7 x 10–70 and 5.1 x 10–05, respectively; Table 2). ctf18+ (systematic name SPBC902.02c) lies on chromosome II and is predicted to encode a 960 amino acid protein with an Mr of 108.6 kDa (Figure 5A). This is somewhat larger than the S.cerevisiae protein (741 amino acids, Mr 84.3 kDa). The two proteins are 28% identical over their C-terminal 720 amino acids. The dcc1+ gene on chromosome I (SPAC31A2.15c) is predicted to encode a 349 amino acid protein with predicted Mr 40.9 kDa, that is, 22% identical to S.cerevisiae Dcc1 (Figure 5A).
Figure 5 Fission yeast Ctf18, Dcc1 and Ctf8. (A) Chromosome context and schematic structure of the ctf18+, dcc1+ and ctf8+ genes. (B) Multiple sequence alignment of Ctf8 homologues from various species. Key: Hs (human, AI301014 ), At (Arabadopsis thaliana, BAA97465 1), Sp (S.pombe, this study), Sc (S.cerevisiae, accession number NP_012061 ), Eg (Eremothecium gossypii, NP_985024 ), Gl (Giardia lamblia, EAA38993 , Nc (Neurospora crassa, XP_328185 ), Gz (Gibberella zeae, EAA67941 and Mg (Magnaporthe grisea, EAA51651 (C) Sensitivity of wild-type ctf18, dcc1 and ctf8 strains to the microtubule inhibitor benomyl. Serial dilutions were plated on YE plates with/without 10 μg/ml benomyl and incubated at 32°C for 4 days. (D) All three double mutants (ctf18 rfc1-44, dcc1 rfc1-44 and ctf8 rfc1-44) are synthetically lethal. Four to six tetratype tetrads are shown for each cross. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate.
A gene encoding a Ctf8 homologue was also identified. This gene, designated ctf8+, contains a single 50 bp intron (confirmed by cDNA sequencing; data not shown) and encodes a 109 amino acid protein. The ctf8+ is located on chromosome I between the SPAC19D5.06c and SPAC19D5.05c genes. Owing to its small size, the presence of the intron, and the low sequence similarity between Ctf8 family members (S.pombe Ctf8 is only 20% identical to its S.cerevisiae counterpart, see Figure 5B), this region of the chromosome was not previously predicted to encode a protein. The gene has now been assigned the systematic name SPAC19D5.11c (Table 2).
To investigate the role of these three proteins in S.pombe, all three genes were deleted from the chromosome and replaced by the natMX6 cassette. Cells carrying one of the ctf18::natMX6 (ctf18), dcc1::natMX6 (dcc1) or ctf8::natMX6 (ctf8) alleles were viable and indistinguishable from wild type under normal growth conditions. Deletion of CTF18 in S.cerevisiae renders cells sensitive to the microtubule inhibitor benomyl (16). This was also true of S.pombe ctf18, dcc1 and ctf8 cells (Figure 5C), suggesting that the Ctf18–RFC complexes may perform similar, if not identical, functions in the two yeasts. Consistent with this, while this work was in progress, Watanabe and co-workers (51) reported the isolation of ctf18 and dcc1 mutant alleles in a screen for meiotic chromosome non-disjunction mutants. Inactivation of either gene resulted for the most part in sister chromatid non-disjunction during the second meiotic division. Taken together, these data suggest that the identified genes above encode components of the S.pombe Ctf18–RFC.
Next, it was intended to examine the properties of double mutants of rfc1-44 and ctf18. However, when rfc1-44 was crossed to ctf18, no rfc1-44 ctf18 double mutant progeny were obtained at 25°C, the permissive temperature for rfc1-44 (Figure 5D). Thus in contrast to elg1+, ctf18+ is required for viability of rfc1-44, indicating that Ctf18–RFC plays an essential role when RFC function is impaired. Whether ctf8+ and dcc1+ were required for rfc1-44 viability was also tested, by crossing dcc1 and ctf8 strains to rfc1-44, but in each case the double mutant, rfc1-44 dcc1 or rfc1-44 ctf8, was inviable (Figure 5B), as was the case for rfc1-44 ctf18. We conclude that the full Ctf18–RFC activity is required for the rescue of rfc1-44.
Overproduction of RFC and RFC-like complex large subunits in rfc1-44
In order to test the effects of overproducing the large subunits, each gene was cloned into a fission yeast expression plasmid, 3' to the thiamine-repressible nmt1 promoter (52). The resulting pREP plasmids were then transformed into wild-type, rfc1-44, elg1 and ctf18 strains. With one exception, the transformed cells were viable and suitable for further analysis (Figure 6). The exception was rfc1-44 transformed with pREP3X-Elg1. Only microcolonies could be obtained following transformation and plating onto minimal medium with or without 5 μg/ml thiamine (i.e. with the nmt1 promoter repressed or fully derepressed) at the permissive temperature of 25°C. Microscopic examination revealed that the transformed cells were highly elongated and only marginally viable, as even after prolonged incubation the formation of larger colonies was not possible. Strikingly, transformation of wild-type, elg1 or ctf18, strains with the same pREP3X-Elg1 plasmid did not affect their growth, either in the presence or in the absence of thiamine. Thus, cells in which RFC activity is impaired by mutation within the CTD are highly sensitive to increased Elg1 protein levels. The extent of Elg1 overproduction under these circumstances is not known, as our recent attempts to generate anti-Elg1 antibodies have not been successful (data not shown). Nevertheless, it is tempting to speculate that in these cells excess Elg1 out-competes Rfc1–44 for binding to the small RFC subunits, resulting in an increase in the levels of Elg1–RFC. Additional dominant-negative phenotypes, albeit non-lethal ones, were seen when rfc1+ was overexpressed in wild-type, rfc1-44, elg1 and ctf18, cells grown in the absence of thiamine (promoter fully derepressed), and when ctf18+ was overexpressed in rfc1-44 cells under the same conditions at 25°C (Figure 6).
Inactivation of the Elg1–RFC restores viability to rfc1-44 ctf18 cells
The results above suggested that the Elg1–RFC complex might play a negative role in rfc1-44 cells, since elevated levels of elg1+ expression prevented rfc1-44 colony formation. To test this further, construction of an rfc1-44 elg1 ctf18 triple mutant was attempted by crossing rfc1-44 elg1 to ctf18 elg1, for the reason that viability of rfc1-44 ctf18 might be restored by inactivation of the negatively acting elg1+. As predicted, the rfc1-44 elg1 ctf18 triple mutant was viable (Figure 7A). Fifteen tetrads, each of which gave rise to four viable colonies, were obtained following tetrad dissection of meiotic asci from this cross. Further analysis of the 24 meiotic products from six of these asci (using PCR and replica plating to nourseothricin or G418 to identify ctf18 and elg1, respectively, and replica plating to 35°C to identify rfc1-44; Materials and Methods) identified six triple mutants, as predicted, among four tetratypic asci, one parental ditype and one non-parental ditype. Viable rfc1-44 elg1 ctf18 triple mutants, together with inviable rfc1-44 ctf18 double mutants, were also obtained by crossing rfc1-44 with elg1 ctf18 (data not shown).
Figure 7 (A) The triple mutant rfc1-44 elg1 ctf18 is viable. Four tetratype tetrads derived from a cross of rfc1-44 elg1 x ctf18 elg1 are shown. In each case, all four meiotic products were capable of colony formation. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate. (B) PFGE analysis of chromosome structure following shift of rfc1-44 and rfc1-44 elg1 cells to 35°C for 4 h, with samples being taken for PFGE analysis every hour. (C) Flow cytometric analysis of propidium-iodide stained cells following shift of the indicated strains to 35°C for 6 h. The asterisk in the triple mutant indicates the position of a sub-population of cells with 1C DNA content.
Further evidence in support of the notion that the Elg1–RFC plays a negative role in rfc1-44 cells came from PFGE analysis of chromosome structure following shift of rfc1-44 and rfc1-44 elg1 cells to the restrictive temperature (Figure 7B). As shown previously, the ability of the chromosomes from rfc1-44 cells to enter a PFGE gel is significantly reduced by 2 h after shift to the restrictive temperature. In rfc1-44 elg1 cells, however, the chromosomes can still enter the gel at later time points (Figure 7B, compare 3 and 4 h timepoints), suggesting that their structure is stabilized by the loss of Elg1–RFC activity. However, note that rfc1-44 elg1 cells were still incapable of colony formation at the restrictive temperature, i.e. the deletion of elg1+ does not fully rescue rfc1-44 (data not shown).
In the rfc1-44 elg1 ctf18 triple mutant cells, all three PCNA-interacting RFC and RFC-like complexes are functionally impaired or absent. The DNA content of these cells following shift to the rfc1-44 restrictive temperature was examined and compared to the two viable double mutants rfc1-44 elg1 and elg1 ctf18, the single mutants rfc1-44, elg1 and ctf18, and wild type (Figure 7C). No significant differences in DNA content profile were apparent between wild type, elg1 and ctf18, and the double mutant elg1 ctf18. As shown in Figure 2C, shifting rfc1-44 to 36.5°C results in rightward movement and broadening of the DNA content peak as the cells arrest in elongated form (37). However, no cells with 1C DNA content are apparent. Similar behaviour is seen with rfc1-44 elg1 cells. However, when the rfc1-44 elg1 ctf18 triple mutant cells are shifted to 36.5°C, there is a transient accumulation of cells with <2C DNA content, suggesting that simultaneous impairment or loss of RFC and Ctf18–RFC results in a severe failure in the replication process, consistent with the inviability of the rfc1-44 ctf18 double mutant.
DISCUSSION
RFC and the RFC-like complexes, Ctf18–RFC and Elg1–RFC, play vital roles in maintaining genome integrity in eukaryotic cells (2,6). RFC loading of PCNA is essential for chromosomal replication and optimal DNA repair, Ctf18–RFC is required for the establishment of sister chromatid cohesion in S-phase, and Elg1–RFC has been implicated in DNA replication, replication fork re-start, recombination and repair (2,6). This describes the results of experiments aimed at investigating the role of the CTD of the large subunit of RFC in fission yeast, Rfc1. Deletion of the entire CTD was shown to abolish RFC function in vivo and a collection of CTD mutant rfc1 alleles generated by random or directed mutagenesis techniques were examined. Twelve alleles were constructed by random pentapeptide insertion. Perhaps surprisingly, only one of the twelve insertions disrupted RFC function, while nine of the twelve had no significant effect. The non-functional mutant protein Rfc143 carries an insertion of the sequence QGVPL between amino acids 753 and 754 (see Supplementary Data for details). Comparison with the S.cerevisiae Rfc1 structure (5) suggests that this region of the Rfc1 protein (amino acids 737–757) is likely to form a lengthy -helix that would be disrupted by the pentapeptide insertion. The remaining two alleles (rfc1-44 and rfc1-54) caused a temperature-sensitive phenotype (Figure 2A). In this case, the encoded mutant proteins carried insertions of the sequences IRGTP and PVGVP between amino acids 811 and 812, and 780 and 781, respectively. Neither insertion can be mapped onto the S.cerevisiae Rfc1 structure as both fall within a region whose structure could not be determined. None of the nine site-directed mutants created in this study, all of which replaced well-conserved charged amino acids within the CTD with alanine (see Supplementary Information), disrupted RFC function.
Further analysis of rfc1-44 showed that these cells were capable of bulk DNA replication after being shifted to the restrictive temperature (Figure 2B) but that the DNA was incapable of entering an agarose gel in PFGE experiments (Figure 2C). These properties are shared by a number of other DNA replication mutants in S.pombe, including the essential DNA helicases encoded by pfh1 and dna2 (49,53) and the Dna2 interacting factor cdc24 (54,55). The rfc1-44 is also more sensitive than wild-type to the DNA damaging agents MMS, HU and UV (Figure 2D) and displays synthetic lethality with conditional lethal pol3, cdc1 and cdc27 alleles, indicative of a close involvement in lagging strand DNA synthesis (46). In an unpublished work, the ability of a range of DNA replication proteins, including the four small RFC subunits, when overproduced, to suppress rfc1-44 at restrictive or semi-restrictive temperatures was tested. However, none of the proteins tested was capable of rescuing rfc1-44. In a subsequent screening for multi-copy suppressors of rfc1-44 using genomic DNA libraries, the same suppressing gene sks2+ (SPBC1709.05) was identified from two of the three libraries screened (J. Kim and S.A. MacNeill, unpublished data). The sks2+ (formerly known as hsc1+) encodes an HSP70 family member (56,57) that may act as a chaperone promoting the folding or assembly of the RFC complex in rfc1-44 cells. Whether Sks2 plays a role in RFC function in wild-type cells awaits further analysis.
That bulk DNA replication was possible in rfc1-44 cells at the restrictive temperature (Figure 2B) suggested that either the RFC complex was not completely inactive under these conditions or that other clamp loaders could partially compensate for the loss of RFC function. To investigate this, analysis of the RFC-like complexes Elg1–RFC and Ctf18–RFC was initiated. The Elg1–RFC complex was purified (Figure 4B) and genes encoding the unique Elg1–RFC and Ctf18–RFC subunits Elg1, Ctf18, Dcc1 and Ctf8 were identified and deleted from the genome (Figures 4 and 5). All three double mutants rfc1-44 ctf18, rfc1-44 dcc1 and rfc1-44 ctf8 are non-viable at the permissive temperature for rfc1-44 (Figure 5D). Biochemical evidence indicates that although Ctf18–RFC lacking Dcc1 and Ctf8 can load PCNA onto DNA in vitro, its activity is 10-fold less than that of the seven subunit complex including Dcc1 and Ctf8 (17). Our genetic results suggest that maximal Ctf18–RFC activity is required when RFC function is impaired, as the both dcc1+ and ctf8+ are required for rfc1-44 viability.
In sharp contrast, the double mutant rfc1-44 elg1 was viable at the permissive temperature (Figure 4D) and PFGE results suggests that inactivation of the Elg1–RFC has a positive effect on rfc1-44 at the restrictive temperature, as the chromosomes were able to enter the gel at later timepoints compared to the single mutant (Figure 7B). This negative effect of Elg1–RFC was most clearly seen when ctf18 elg1 was crossed with rfc1-44 elg1 to generate a viable rfc1-44 elg1 ctf18 triple mutant. This result shows that inactivation of the Elg1–RFC restores viability to the rfc1-44 ctf18 double mutant, reinforcing the view that the Elg1–RFC plays an opposing role to RFC and Ctf18–RFC in S.pombe cells.
Previously, various functions have been proposed for Elg1–RFC (6). For example, Elg1–RFC might function as a specialized PCNA clamp loader for DNA substrates associated with, for example, replication fork stalling or collapse. It is possible that in rfc1-44 ctf18 cells, recognition of such structures and subsequent PCNA loading by Elg1–RFC has a harmful effect, whereas loss of Elg1–RFC function in the triple mutant rfc1-44 ctf18 elg1 allows replication fork recovery to be achieved by an alternative mechanism such as homologous recombination. Intriguingly, in S.cerevisiae elg1 is synthetically lethal with genes involved in homologous recombination . Alternatively, it has been suggested that the Elg1–RFC could function as a PCNA clamp unloader (6). Recycling of PCNA is likely to be vital for efficient replication but exactly how this occurs is unclear. If Elg1–RFC acts as a dedicated clamp unloader, its inactivation might counteract, at least in part, the consequences of impairing RFC activity by the rfc1-44 mutation. Finally, Elg1–RFC might have a role in loading or unloading a modified form of PCNA onto DNA. Ubiquitination and sumoylation of PCNA has been demonstrated to play an important role in replication and repair of damaged DNA but whether PCNA modification impinges on clamp loading and unloading is not known. Further detailed genetic and biochemical studies will be required to resolve these issues.
SUPPLEMENTARY MATERIAL
Supplementary Material is available at NAR Online.
ACKNOWLEDGEMENTS
We would like to thank our friends and colleagues in Edinburgh and elsewhere for their help during the course of this work, in particular Dr Hiroyuki Tanaka for many helpful discussions, Pierre Hentges and Professor Tony Carr (University of Sussex) for the pFA6A-natMX6 plasmid and Dr Finbarr Hayes (UMIST) for supplying the Tn4430 system components. S.M. and F.G. were supported by a Wellcome Trust Senior Research Fellowship in Basic Biomedical Science, J.K. by a Wellcome Trust Prize Studentship, K.M. by the Wellcome Trust 4 year PhD Programme Genetics and Disease, I.C. by a Darwin Trust of Edinburgh vacation scholarship and K.R. by the European Union (contract QLK3-CT-2002-02071). Funding to pay the Open Access publication charges for this article was provided by the University of Copenhagen.
REFERENCES
Maga, G. and Hubscher, U. (2003) Proliferating cell nuclear antigen (PCNA): a dancer with many partners J. Cell Sci., 116, 3051–3060 .
Majka, J. and Burgers, P.M. (2004) The PCNA-RFC families of DNA clamps and clamp loaders Prog. Nucleic Acid Res. Mol. Biol., 78, 227–260 .
Tsurimoto, T. and Stillman, B. (1989) Purification of a cellular replication factor, RF-C that is required for coordinated synthesis of leading and lagging strands during simian virus-40 DNA replication in vitro Mol. Cell. Biol., 9, 609–619 .
Tsurimoto, T. and Stillman, B. (1990) Functions of replication factor C and proliferating cell nuclear antigen—functional similarity of DNA polymerase accessory proteins from human cells and bacteriophage T4 Proc. Natl Acad. Sci. USA, 87, 1023–1027 .
Bowman, G.D., O'Donnell, M., Kuriyan, J. (2004) Structural analysis of a eukaryotic sliding DNA clamp–clamp loader complex Nature, 429, 724–730 .
Kim, J. and MacNeill, S.A. (2003) Genome stability: a new member of the RFC family Curr. Biol., 13, R873–R875 .
Parrilla-Castellar, E.R., Arlander, S.J., Karnitz, L. (2004) Dial 9-1-1 for DNA damage: the Rad9-Hus1-Rad1 (9-1-1) clamp complex DNA Repair (Amst.), 3, 1009–1014 .
Shimomura, T., Ando, S., Matsumoto, K., Sugimoto, K. (1998) Functional and physical interaction between Rad24 and Rfc5 in the yeast checkpoint pathways Mol. Cell. Biol., 18, 5485–5491 .
Naiki, T., Shimomura, T., Kondo, T., Matsumoto, K., Sugimoto, K. (2000) Rfc5, in cooperation with rad24, controls DNA damage checkpoints throughout the cell cycle in Saccharomyces cerevisiae Mol. Cell. Biol., 20, 5888–5896 .
Green, C.M., Erdjument-Bromage, H., Tempst, P., Lowndes, N.F. (2000) A novel Rad24 checkpoint protein complex closely related to replication factor C Curr. Biol., 10, 39–42 .
Bermudez, V.P., Lindsey-Boltz, L.A., Cesare, A.J., Maniwa, Y., Griffith, J.D., Hurwitz, J., Sancar, A. (2003) Loading of the human 9-1-1 checkpoint complex onto DNA by the checkpoint clamp loader hRad17-replication factor C complex in vitro Proc. Natl Acad. Sci. USA, 100, 1633–1638 .
Majka, J. and Burgers, P.M. (2003) Yeast Rad17/Mec3/Ddc1: a sliding clamp for the DNA damage checkpoint Proc. Natl Acad. Sci. USA, 100, 2249–2254 .
Ellison, V. and Stillman, B. (2003) Biochemical characterization of DNA damage checkpoint complexes: clamp loader and clamp complexes with specificity for 5' recessed DNA PLoS Biol., 1, E33 .
Mayer, M.L., Gygi, S.P., Aebersold, R., Hieter, P. (2001) Identification of RFC(Ctf18p, Ctf8p, Dcc1p): an alternative RFC complex required for sister chromatid cohesion in S.cerevisiae Mol. Cell, 7, 959–970 .
Naiki, T., Kondo, T., Nakada, D., Matsumoto, K., Sugimoto, K. (2001) Chl12 (Ctf18) forms a novel replication factor C-related complex and functions redundantly with Rad24 in the DNA replication checkpoint pathway Mol. Cell. Biol., 21, 5838–5845 .
Hanna, J.S., Kroll, E.S., Lundblad, V., Spencer, F.A. (2001) Saccharomyces cerevisiae CTF18 and CTF4 are required for sister chromatid cohesion Mol. Cell. Biol., 21, 3144–3158 .
Bermudez, V.P., Maniwa, Y., Tappin, I., Ozato, K., Yokomori, K., Hurwitz, J. (2003) The alternative Ctf18-Dcc1-Ctf8-replication factor C complex required for sister chromatid cohesion loads proliferating cell nuclear antigen onto DNA Proc. Natl Acad. Sci. USA, 100, 10237–10242 .
Shiomi, Y., Shinozaki, A., Sugimoto, K., Usukura, J., Obuse, C., Tsurimoto, T. (2004) The reconstituted human Chl12–RFC complex functions as a second PCNA loader Genes Cells, 9, 279–290 .
Ben Aroya, S. and Kupiec, M. (2005) The Elg1 replication factor C-like complex: a novel guardian of genome stability DNA Repair (Amst.), 4, 409–417 .
Scholes, D.T., Banerjee, M., Bowen, B., Curcio, M.J. (2001) Multiple regulators of Ty1 transposition in Saccharomyces cerevisiae have conserved roles in genome maintenance Genetics, 159, 1449–1465 .
Bellaoui, M., Chang, M., Ou, J., Xu, H., Boone, C., Brown, G.W. (2003) Elg1 forms an alternative RFC complex important for DNA replication and genome integrity EMBO J., 22, 4304–4313 .
Ben Aroya, S., Koren, A., Liefshitz, B., Steinlauf, R., Kupiec, M. (2003) ELG1, a yeast gene required for genome stability, forms a complex related to replication factor C Proc. Natl Acad. Sci. USA, 100, 9906–9911 .
Kanellis, P., Agyei, R., Durocher, D. (2003) Elg1 forms an alternative PCNA-interacting RFC complex required to maintain genome stability Curr. Biol., 13, 1583–1595 .
Huang, M.E., Rio, A.G., Nicolas, A., Kolodner, R.D. (2003) A genomewide screen in Saccharomyces cerevisiae for genes that suppress the accumulation of mutations Proc. Natl Acad. Sci. USA, 100, 11529–11534 .
Smolikov, S., Mazor, Y., Krauskopf, A. (2004) ELG1, a regulator of genome stability, has a role in telomere length regulation and in silencing Proc. Natl Acad. Sci. USA, 101, 1656–1661 .
Smith, S., Hwang, J.Y., Banerjee, S., Majeed, A., Gupta, A., Myung, K. (2004) Mutator genes for suppression of gross chromosomal rearrangements identified by a genome-wide screening in Saccharomyces cerevisiae Proc. Natl Acad. Sci. USA, 101, 9039–9044 .
Banerjee, S. and Myung, K. (2004) Increased genome instability and telomere length in the elg1-deficient Saccharomyces cerevisiae mutant are regulated by S-phase checkpoints Eukaryot. Cell, 3, 1557–1566 .
Uhlmann, F., Cai, J., Gibbs, E., O'Donnell, M., Hurwitz, J. (1997) Deletion analysis of the large subunit p140 in human replication factor C reveals regions required for complex formation and replication activities J. Biol. Chem., 272, 10058–10064 .
Podust, V.N., Tiwari, N., Ott, R., Fanning, E. (1998) Functional interactions among the subunits of replication factor C potentiate and modulate its ATPase activity J. Biol. Chem., 273, 12935–12942 .
Gomes, X.V., Gary, S.L., Burgers, P.M. (2000) Overproduction in Escherichia coli and characterization of yeast replication factor C lacking the ligase homology domain J. Biol. Chem., 275, 14541–14549 .
Warbrick, E. (1998) PCNA binding through a conserved motif Bioessays, 20, 195–199 .
Bork, P., Hofmann, K., Bucher, P., Neuwald, A.F., Altschul, S.F., Koonin, E.V. (1997) A superfamily of conserved domains in DNA damage-responsive cell cycle checkpoint proteins FASEB J., 11, 68–76 .
Allen, B.L., Uhlmann, F., Gaur, L.K., Mulder, B.A., Posey, K.L., Jones, L.B., Hardin, S.H. (1998) DNA recognition properties of the N-terminal DNA binding domain within the large subunit of replication factor C Nucleic Acids Res., 26, 3877–3882 .
Moreno, S., Klar, A., Nurse, P. (1991) Molecular genetic analysis of fission yeast Schizosaccharomyces pombe. Methods Enzymol, 194, 795–823 .
Prentice, H.L. (1992) High-efficiency transformation of Schizosaccharomyces pombe by electroporation Nucleic Acids Res., 20, 621 .
Bahler, J., Wu, J.Q., Longtine, M.S., Shah, N.G., McKenzie, A., III, Steever, A.B., Wach, A., Philippsen, P., Pringle, J.R. (1998) Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomyces pombe Yeast, 14, 943–951 .
Sazer, S. and Sherwood, S.W. (1990) Mitochondrial growth and DNA synthesis occur in the absence of nuclear DNA replication in fission yeast J. Cell Sci., 97, 509–516 .
Kelly, T.J., Martin, G.S., Forsburg, S.L., Stephen, R.J., Russo, A., Nurse, P. (1993) The fission yeast cdc18+ gene-product couples S-phase to start and mitosis Cell, 74, 371–382 .
Barbet, N., Muriel, W.J., Carr, A.M. (1992) Versatile shuttle vectors and genomic libraries for use with Schizosaccharomyces pombe Gene, 114, 59–66 .
Sambrook, J. and Russell, D.W. Molecular Cloning: A Laboratory Manual, 3rd edn, (2001) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press .
Hallet, B., Sherratt, D.J., Hayes, F. (1997) Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein Nucleic Acids Res., 25, 1866–1867 .
Hayes, F. and Hallet, B. (2000) Pentapeptide scanning mutagenesis: encouraging old proteins to execute unusual tricks Trends Microbiol., 8, 571–577 .
Tasto, J.J., Carnahan, R.H., McDonald, W.H., Gould, K.L. (2001) Vectors and gene targeting modules for tandem affinity purification in Schizosaccharomyces pombe Yeast, 18, 657–662 .
Kai, M., Tanaka, H., Wang, T.S. (2001) Fission yeast Rad17 associates with chromatin in response to aberrant genomic structures Mol. Cell. Biol., 21, 3289–3301 .
Hennessy, K.M., Lee, A., Chen, E., Botstein, D. (1991) A group of interacting yeast DNA replication genes Genes Dev., 5, 958–969 .
Francesconi, S., Park, H., Wang, T.S.F. (1993) Fission yeast with DNA polymerase temperature-sensitive alleles exhibits cell division cycle phenotype Nucleic Acids Res., 21, 3821–3828 .
MacNeill, S.A., Moreno, S., Reynolds, N., Nurse, P., Fantes, P.A. (1996) The fission yeast Cdc1 protein, a homolog of the small subunit of DNA polymerase , binds to Pol3 and Cdc27 EMBO J., 15, 4613–4628 .
Tanaka, H., Ryu, G.H., Seo, Y.S., MacNeill, S.A. (2004) Genetics of lagging strand DNA synthesis and maturation in fission yeast: suppression analysis links the Dna2–Cdc24 complex to DNA polymerase Nucleic Acids Res., 32, 6367–6377 .
Kang, H.Y., Choi, E., Bae, S.H., Lee, K.H., Gim, B.S., Kim, H.D., Park, C., MacNeill, S.A., Seo, Y.S. (2000) Genetic analyses of Schizosaccharomyces pombe dna2+ reveal that dna2 plays an essential role in Okazaki fragment metabolism Genetics, 155, 1055–1067 .
Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E., Wilm, M., Seraphin, B. (2001) The tandem affinity purification (TAP) method: a general procedure of protein complex purification Methods, 24, 218–229 .
Petronczki, M., Chwalla, B., Siomos, M.F., Yokobayashi, S., Helmhart, W., Deutschbauer, A.M., Davis, R.W., Watanabe, Y., Nasmyth, K. (2004) Sister-chromatid cohesion mediated by the alternative RF-CCtf18/Dcc1/Ctf8, the helicase Chl1 and the polymerase--associated protein Ctf4 is essential for chromatid disjunction during meiosis II J. Cell Sci., 117, 3547–3559 .
Maundrell, K. (1990) nmt1 of fission yeast. A highly transcribed gene completely repressed by thiamine J. Biol. Chem., 265, 10857–10864 .
Tanaka, H., Ryu, G.H., Seo, Y.S., Tanaka, K., Okayama, H., MacNeill, S.A., Yuasa, Y. (2002) The fission yeast pfh1+ gene encodes an essential 5' to 3' DNA helicase required for the completion of S-phase Nucleic Acids Res., 30, 4728–4739 .
Gould, K.L., Burns, C.G., Feoktistova, A., Hu, C.P., Pasion, S.G., Forsburg, S.L. (1998) Fission yeast cdc24+ encodes a novel replication factor required for chromosome integrity Genetics, 149, 1221–1233 .
Tanaka, H., Tanaka, K., Murakami, H., Okayama, H. (1999) Fission yeast Cdc24 is a replication factor C- and proliferating cell nuclear antigen-interacting factor essential for S-phase completion Mol. Cell Biol., 19, 1038–1048 .
Usui, T., Yoshida, M., Kasahara, K., Honda, A., Beppu, T., Horinouchi, S. (1997) A novel HSP70 gene of Schizosaccharomyces pombe that confers K-252a resistance Gene, 189, 43–47 .
Oishi, K., Sugiura, R., Shuntoh, H., Kuno, T. (1996) Cloning and characterization of hsc1+, a heat shock cognate gene of the fission yeast Schizosaccharomyces pombe Gene, 181, 45–49 .
Prakash, S., Johnson, R.E., Prakash, L. (2004) Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function Annu. Rev. Biochem., 74, 317–353 .(Jiyoung Kim1, Kathryn Robertson1, Katie )
*To whom correspondence should be addressed. Tel: +45 35 32 20 05; Fax: +45 35 32 20 40; Email: s.a.macneill@mermaid.molbio.ku.dk
ABSTRACT
Proliferating cell nuclear antigen loading onto DNA by replication factor C (RFC) is a key step in eukaryotic DNA replication and repair processes. In this study, the C-terminal domain (CTD) of the large subunit of fission yeast RFC is shown to be essential for its function in vivo. Cells carrying a temperature-sensitive mutation in the CTD, rfc1-44, arrest with incompletely replicated chromosomes, are sensitive to DNA damaging agents, are synthetically lethal with other DNA replication mutants, and can be suppressed by mutations in rfc5. To assess the contribution of the RFC-like complexes Elg1–RFC and Ctf18–RFC to the viability of rfc1-44, genes encoding the large subunits of these complexes have been deleted and overexpressed. Inactivation of Ctf18–RFC by the deletion of ctf18+, dcc1+ or ctf8+ is lethal in an rfc1-44 background showing that full Ctf18–RFC function is required in the absence of fully functional RFC. In contrast, rfc1-44 elg1 cells are viable and overproduction of Elg1 in rfc1-44 is lethal, suggesting that Elg1–RFC plays a negative role when RFC function is inhibited. Consistent with this, the deletion of elg1+ is shown to restore viability to rfc1-44 ctf18 cells.
INTRODUCTION
The ring-shaped homotrimer proliferating cell nuclear antigen (PCNA) plays key roles in many aspects of DNA metabolism in eukaryotic cells. PCNA encircles DNA to form a sliding clamp that can move freely along the duplex and acts to tether a host of interacting proteins implicated in DNA replication, DNA repair and DNA modification, such as DNA ligase I, the nucleases Fen1 and XP-G, uracil-N-glycosylase and cytosine-5-methyltransferase. However, the central function of PCNA is to act as a processivity factor for the replicative DNA polymerases. In the absence of polymerase tethering by PCNA, chromosomal DNA replication cannot be completed efficiently, if at all .
The loading of PCNA onto DNA requires that the PCNA ring be opened and then closed around the duplex. This reaction is catalysed in an ATP-dependent manner by a clamp loader complex. Clamp loaders have been identified in all three kingdoms of life. In eukaryotic cells, the first clamp loader to be identified was replication factor C (RFC), discovered in the studies of SV40 viral DNA replication in vitro (3,4). RFC is a pentameric complex comprising a large subunit (Rfc1) and four smaller subunits (Rfc2–Rfc5). All the five subunits are members of the AAA+ protein superfamily of ATPase and ATPase-related proteins, and are essential for chromosomal DNA replication and cell survival in yeast. Recently, the structure of the yeast RFC complex was solved, revealing a complex with spiral geometry that closely matches to that of the DNA duplex (5).
In addition to RFC, three RFC-like complexes have been identified in recent years, two of which have been demonstrated to possess clamp loading activity, albeit with different sliding clamps . In this study, the fission yeast protein nomenclature is used when describing these complexes, although it should be noted that much of the information that has been gathered on their functions has come from the analysis of budding yeast (see Table 1 for nomenclature summary).
Table 1 Yeast RFC/RFC-like complex and sliding clamp nomenclature
The first RFC-like complex to be identified (Rad17–RFC in fission yeast) performs its function in DNA structure checkpoints . The Rad17–RFC comprises the same four small subunits as RFC, Rfc2–Rfc5, but the large subunit Rfc1 is replaced by the related Rad17 protein (8–10). The effect of substituting Rad17 for Rfc1 is to alter the biochemical properties of the complex, notably its DNA substrate specificity (11–13). However, the Rad17–RFC also functions with a different sliding clamp. Instead of PCNA, Rad17–RFC loads the 9-1-1 complex onto DNA. This alternative sliding clamp comprises the Rad9, Rad1 and Hus1 DNA checkpoint proteins (Table 1) (11–13). It is thought that the Rad17–RFC and 9-1-1 complex act together as sensors of DNA structure perturbations.
The second RFC-like complex to be identified was Ctf18–RFC (14,15). In this complex, the large subunit of RFC is substituted with Ctf18. Two additional factors, Dcc1 and Ctf8, also associate with Ctf18, resulting in a seven subunit complex. The Ctf18–RFC is required for the efficient establishment of chromosome cohesion during S-phase, the process by which the newly replicated sister chromatids are held together until anaphase. Although all the three genes are non-essential, Saccharomyces cerevisiae cells carrying ctf18, dcc1 or ctf8 mutations display genome instability and chromosome loss phenotypes (14,16). Biochemical evidence indicates that Ctf18–RFC functions as an ATP-dependent PCNA loader, loading PCNA onto primed templates 10-fold less efficiently than RFC (17,18).
The third RFC-like complex sees Rfc1 substituted by Elg1 . Mutants in the budding yeast ELG1 gene were identified in a number of independent genetic screens for factors affecting genome stability in diverse ways (20–27): elg1 mutants display increased levels of mitotic recombination, chromosome loss, transposition and gross chromosomal rearrangement. The elg1 mutants also display telomere lengthening and are synthetically lethal or sick when combined with mutations in various DNA processing pathways, such as homologous recombination (synthetic interactions with rad50, rad51, rad52, rad54, rad55, rad57, mre11 and xrs2), replication fork re-start (mus81 and mms4) and DNA replication (pol32, dna2 and rad27). The elg1 cells are also mildly sensitive to the alkylating agent methylmethanesulphonate (MMS) but not to hydroxyurea (HU) or UV treatment. Despite this analysis, the precise function of the Elg1–RFC is unknown, although it has been shown that the complex co-immunoprecipitates with PCNA (23). The complete subunit composition of the Elg1–RFC is also unknown.
Although the small and large subunits of the clamp loader complexes are related to one another, the large subunits have extended N- and C-terminal regions. In the case of Rfc1, the function of the N-terminal extended region has been analysed previously (28–30). This region includes both a conserved PCNA-binding motif with a potential role in targeting RFC to replication factories (31) and a BRCT domain (32), and possesses DNA-binding activity (28,33) but is not required either for PCNA loading in vitro or for RFC function in vivo. The function of the C-terminal domain (CTD) has been less well characterized, although in vitro reconstitution experiments point to an important role for this domain in RFC complex assembly (28). Recent structural studies suggest a role for this domain in stabilizing the RFC complex by bridging the ends of the spiral assembly of RFC subunits but the functional consequences of this are not known (5).
In this study, the CTD of Schizosaccharomyces pombe Rfc1 is shown to be essential for RFC function in vivo, and the isolation and the analysis of novel mutant rfc1 alleles generated by random and targeted mutagenesis of the CTD, including a temperature-sensitive mutant, rfc1-44, is described. Analysis of rfc1-44 shows the cells to be incapable of completing DNA replication at their restrictive temperature, sensitive to DNA damaging agents, and synthetically lethal with a number of DNA replication mutants. To investigate this further, genes encoding S.pombe orthologues of Elg1, Ctf18, Dcc1 and Ctf8 are identified, elg1, ctf18, dcc1 and ctf8 strains constructed and analysed, and rfc1-44 ctf18, rfc1-44 dcc1 and rfc1-44 ctf8 double mutants shown to be inviable, implying that Ctf18–RFC function is essential for cell viability in the absence of fully functional RFC. The purification of the Elg1–RFC from fission yeast is also described, demonstrating for the first time in any organism that this complex has a five subunit structure. In support of a model in which Elg1–RFC plays a negative role when RFC function is impaired, elevated levels of Elg1–RFC are shown to be lethal in rfc1-44 and the deletion of elg1+ shown to restore viability to rfc1-44 ctf18 cells. Finally, the identification of suppressors of rfc1-44 mapping within the rfc5 gene is reported, suggesting that the functional defect in RFC in rfc1-44 cells may lie in the Rfc1–Rfc5 interaction revealed by the crystal structure.
MATERIALS AND METHODS
Yeast strains, media and methods
Haploid S.pombe strains used for strain construction were as follows: Sp236 (leu1-32 ura4-D18 h+), Sp292 (leu1-32 ura4-D18 ade6-M210 h–), Sp293 (leu1-32 ura4-D18 ade6-M216 h+), Sp322 (leu1-32 ura4-D18 h–), Sp347 (972 h–) and Sp348 (h+S). PCR-mediated gene targeting was used to construct the following strains: Sp323 (elg1::natMX6 leu1-32 ura4-D18 h–), Sp325 (elg1::kanMX6 leu1-32 ura4-D18 h–), Sp359 (dcc1::natMX6 h–), Sp360 (dcc1::natMX6 h+S), Sp365 (ctf18::natMX6 h–), Sp378 (ctf8::natMX6 h–), Sp380 (ctf8::natMX6 h+S), Sp382 (ctf18::natMX6 h+S) and Sp402 (elg1-TAP-natMX6 h–). A wild-type diploid strain (leu1-32/leu1-32 ura4-D18/ura4-D18 ade6-M210/ade6-M216 h–/h+) was created by standard methods by mating haploids Sp292 and Sp293 and used to construct various rfc1 mutants (see below). The following rfc1 strains were generated by subsequent matings: HSp1 (rfc1-44 h+), HSp9 (rfc1-54 h+), HSp17 (rfc1-44 h–), HSp23 (rfc1-54 h–), HSp25 (rfc1-44 elg1::natMX6 h+), HSp26 (rfc1-44 elg1::natMX6 h–), HSp27 (rfc1-44 elg1::kanMX6 ctf18::natMX6 h+) and KSp1 (elg1::kanMX6 ctf18::natMX6 h–). Yeast growth media (YE, EMM and ME) were used essentially as described (34). 5-Fluoroorotic acid (5-FOA) selection was performed using YE medium supplemented with 1 mg/ml 5-FOA. G418 (Geneticin; from Invitrogen/GibcoBRL) and nourseothricin (ClonNAT; from Werner BioAgents) were used at 100 mg/l in YE agar. Transformations were performed by electroporation (35) or lithium acetate methods (36) as appropriate. Tetrad dissection was accomplished using a micromanipulator (Singer Instruments). Flow cytometry on fixed propidium-iodide stained cells was performed by using a FACS Calibur instrument and CellQuest software (BD Biosciences) as described previously (37), cell number counts by using a Z2 particle counter (Beckman Coulter), and pulse-field gel electrophoresis (PFGE) as described by Kelly et al. (38) using a BioRad CHEF DRII system.
Construction and analysis of rfc1+/rfc1-C strain
For plasmid integration into the genome, a series of constructs based on plasmid pJKE1 were used. pJKE1 corresponds to pUR19 (39) from which ars1+ has been deleted by ClaI digestion and re-ligation. Initially, sequences encoding amino acids 590–935 of Rfc1 (corresponding to the CTD plus 130 amino acids N-terminal to the CTD) were amplified by PCR using pUR19-Rfc1 as template and cloned into plasmid pJKE1 to generate plasmid pJKE11. The PCR overlap extension method (39) was then used, with pJKE11 as template and complementary 50mer oligonucleotides, to introduce stop codons in all three frames into the rfc1+ sequence, so that the encoded protein would terminate after Pro719. The product of the second round of PCR was cloned into pJKE32 (see below) to generate pJKE59. Linearized plasmid DNA was then transformed into an S.pombe diploid strain (leu1-32/leu1-32 ura4-D18/ura4-D18 ade6-M210/ade6-M216 h–/h+) and stable integrative transformants isolated and characterized by PCR using genomic DNA as template. Sequencing of PCR products generated in this manner allowed confirmation of the genotypes. The sequences of all the oligonucleotides used can be obtained from the authors on request. Diploids were then induced to sporulate on ME medium and asci dissected using a micromanipulator (Singer Instruments). Following re-growth, the genotypes of viable colonies were confirmed by replica plating.
Pentapeptide insertion mutagenesis
To mutagenize rfc1+ using the pentapeptide insertion method, sequences encoding amino acids 590–935 of Rfc1 (corresponding to the CTD plus 130 amino acids N-terminal to the CTD) were amplified by PCR and cloned into plasmid pJKE31, a derivative of pJKE1 with the polylinker KpnI site removed (see below), to generate pJKE32. Next, pJKE32 was mutagenized by the PCR overlap extension method (40) to remove the KpnI site present in the rfc1+ sequence (the sequence G GTA CC was changed to G GTT CC, with no effect on encoded amino acid sequence), to make pJKE42. Pentapeptide insertion mutagenesis was performed using the Tn4430 system (41,42). pJKE42 was transformed into E.coli FH1046 and individual transformant colonies were mated with E.coli DS941 after which the mating mixtures were plated on the medium to select for clones containing pJKE42 carrying the Tn4430 transposon. Isolates containing a Tn4430 insertion within the rfc1 region were then identified by colony PCR and plasmid DNA prepared. This was restricted by KpnI and re-ligated to remove the bulk of the transposon, before being transformed into E.coli DH5. Seventeen individual transposon-free clones were isolated and the position of the 15 bp insertion determined by KpnI restriction mapping and DNA sequencing. Six duplicated mutants were discarded at this point, leaving a collection of twelve different pentapeptide insertion alleles (rfc1-43–rfc1-54, carried by plasmids pJKE43–pJKE54). Site-specific mutations (rfc1-33–rfc1-41, carried by plasmids pJKE33–pJKE41) were generated by PCR overlap extension mutagenesis (40) using complementary 36mer oligonucleotides and plasmid pJKE11 as a template. The sequences of the oligonucleotides used can be obtained from the authors upon request. Following the second round of PCR the products were cloned into pJKE32. Subsequently, linearized plasmid DNA was transformed into diploid S.pombe and integrative transformants isolated and analysed as described above. For further analysis, 5-FOA was used to select for rfc1-44 and rfc1-54 strains in which the plasmid had recombined out of the genome leaving a mutated copy of the rfc1 gene on the chromosome. These strains were then backcrossed to wild type to remove auxotrophic markers, resulting in strains HSp1 (rfc1-44 h+), HSp9 (rfc1-54 h+), HSp17 (rfc1-44 h–) and HSp23 (rfc1-54 h–).
Suppressor screening
To screen for suppressors of rfc1-44, a 1.3 kb fragment was amplified by PCR from the rfc5+ chromosomal region and subcloned into plasmid pJKE31 (above). This corresponded to 200 bp 5' to the rfc5+ open reading frame (ORF) and 1.2 of sequence encoding the first 289 (of 358) amino acids of Rfc5. The resulting plasmid pJKE55 was subjected to mutagenesis with Tn4430 as described above except that, following mating with E.coli DS941, a library of transposon-containing plasmids was prepared, digested with KpnI, re-ligated and transformed in E.coli DH5. Plasmid DNA was prepared from the >3000 pooled colonies, linearized and used to transform rfc1-44.
Gene deletions
The elg1+, ctf18+, dcc1+ and ctf8+ genes were deleted in wild-type 972 h– (SP347) and h+S (Sp348) strains using the PCR-mediated gene targeting method for fission yeast as described by B?hler et al. (36) using either pFA6a-kanMX6 or pFA6a-natMX6 as PCR templates. The latter (a gift from P. Hentges and T. Carr, University of Sussex) confers resistance to nourseothricin (also known as clonNAT, supplied by Werner BioAgents), which was added to YE plates at 100 μg/ml. Oligonucleotides (100mer with 80 nt of gene-specific sequence) were supplied by DNA Technology A/S. The deletions were confirmed by PCR using oligonucleotide primers flanking the cassette. The sequences of all the oligonucleotides used can be obtained from the authors upon request.
Purification of the Elg1–RFC
To facilitate purification of the Elg1–RFC, the chromosomal elg1+ gene was tagged at its 3' end with sequences encoding the TAP module marked with the natMX6 cassette. For this, the kanMX6 module was first excised from pFA6a-CTAP-kanMX6 (43) and replaced with the natMX6 cassette from pFA6a-natMX6 (above). The resulting plasmid, pFA6a-CTAP-natMX6, was then used as the template for PCR using 100mer oligonucleotides (sequences from the authors upon request), the PCR product transformed into S.pombe strain Sp347 and nourseothricin-resistant colonies identified. Successful tagging was confirmed by PCR and direct sequencing of amplified chromosomal DNA fragments. For purification, a 12 litres culture of the elg1-TAP-natMX6 h– strain (Sp402) was grown to an OD600 nm of 0.8 (1 x 107 cells/ml) at 32°C, before the cells were harvested and processed as described previously (43) except that the cells were broken by grinding in liquid nitrogen using an RM100 mortar grinder (Retsch) for 15 min. Following TCA precipitation, the samples were resolved by SDS–PAGE and silver stained. Trypsin digestion of excised bands and mass spectrometric analysis was carried out commercially at the Proteomics Facility at the University of Dundee, UK. (http://www.dundee.ac.uk/biocentre/FingerPrints/).
RESULTS
The conserved CTD of fission yeast Rfc1 is required for RFC function
In all eukaryotic species, the large subunit of RFC differs from the small subunits in having extended N- and C-terminal regions (Figure 1A). In order to probe the function of the CTD in greater detail, one copy of the chromosomal rfc1+ gene in a diploid S.pombe strain was altered to encode a truncated Rfc1 protein lacking the entire 215 amino acid CTD (Materials and Methods). This rfc1+/rfc1-C diploid was sporulated and the properties of haploid rfc1-C cells analysed following meiotic tetrad dissection. Spores expressing the truncated Rfc1-C protein were capable of germination and 1–5 cell divisions, but incapable of colony formation (data not shown). Similar to S.pombe rfc1 cells (44), the arrested rfc1-C cells were often elongated, indicative of a cell cycle defect. Transformation of the rfc1+/rfc1-C diploid prior to sporulation with a plasmid carrying wild-type rfc1+ rescued the rfc1-C haploid, allowing efficient colony formation (data not shown). The CTD is therefore required for RFC function in vivo, consistent with the earlier observation that the deletion of sequences from the C-terminus of the human Rfc1 orthologue hRFC140 abolished RFC complex formation in vitro (28).
Figure 1 (A) Schematic structure of the RFC subunits. Each small subunit (Rfc2–Rfc5) is a member of the AAA+ family of proteins and comprises three structural domains I–III (indicated above the Rfc2 protein). The large subunit Rfc1 is also an AAA+ protein and possesses the same domains but has additional N- and C-terminal regions. The N-terminal region contains a PIP (PCNA-interacting protein) motif and a single BRCT motif of unknown function. The function of the CTD is not known. Each RFC subunit also contains a number of conserved sequence blocks, RFC boxes II–VIII (grey boxes, shown numbered on the Rfc1 protein). (B) Location of mutants within the Rfc1 CTD. An expanded view of part of domain III and the CTD is shown, spanning amino acids 590–935 (the CTD spans 720–935). The mutants generated in this study are indicated as follows: open circles (pentapeptide insertion mutants, no significant effect on protein function), grey circles (pentapeptide insertion mutants, temperature-sensitive), closed circles (pentapeptide insertion mutants, non-functional), open squares (amino acid substitutions, no significant effect on protein function). See Supplementary Information for more details.
Mutational analysis of the CTD
To further probe its function, the CTD was subjected to mutational analysis. Both random and directed mutagenesis strategies were used to generate a total of 21 mutant rfc1 alleles. Random mutagenesis was accomplished using the Tn4430-based pentapeptide insertion method (41,42). Twelve alleles were constructed by this method (Figure 1B, circles), seven of which contained insertions within the CTD. The remaining five insertions mapped N-terminal to the CTD, within the RFC collar domain (5). Oligonucleotide-directed in vitro mutagenesis was used to generate a further nine rfc1 alleles with mutations within the CTD (Figure 1B, squares). All nine mutations changed individual conserved charged amino acids to alanine (see Supplementary Information for details).
As above, the mutant alleles were targeted to one copy of the rfc1+ gene in a diploid strain (Materials and Methods). Following meiosis and sporulation, the properties of the rfc1 mutants were analysed by tetrad dissection and re-growth. The results are summarized in the Supplementary Information. Briefly, 13 of the 21 alleles were viable in haploid form and essentially indistinguishable from wild type (Figure 1B, open circles or squares) whereas only one, rfc1-43, was inviable (Figure 1B, closed circle). The latter encodes an Rfc1 protein with a pentapeptide insertion within the CTD. Two of the mutant alleles, rfc1-44 and rfc1-54, displayed a temperature-sensitive phenotype (Figure 1B, grey circles and Figure 2A) that could be rescued by wild-type rfc1+ (data not shown).
Figure 2 Characterization of temperature-sensitive rfc1 alleles. (A) Serial dilutions of wild-type (rfc1+), rfc1-44 and rfc1-54 cells incubated at 18, 25 and 35°C, respectively, for 3–4 days. (B) Flow cytometry of propidium iodide stained wild-type and rfc1-44 cells following shift from 25 to 35°C with or without the addition of HU. Samples for analysis were taken every hour from 0 to 8 h. (C) PFGE analysis of chromosome structure in wild type or rfc1-44 following shift from 25 to 35°C. Samples for the analysis were taken every 2 h from 0 to 8 h. Note that chromosome III in rfc1-44 cells appears somewhat shorter than in the wild-type control, indicative of the loss of rDNA repeats. (D) Sensitivity of wild-type, rfc1-44 and rfc1-54 cells to camptothecin (CPT, 10 μM), hydroxyurea (HU, 5 mM), methylmethane sulphonate (MMS, 0.005%) and UV light (300 J/m2). Serial dilutions were spotted and incubated at 25°C for 4 days.
One of these alleles, rfc1-44, was characterized further. When shifted to the restrictive temperature of 35°C, rfc1-44 cells arrest with a 2C DNA content (determined by flow cytometry of propidium iodide stained cells, Figure 2B) indicating that bulk DNA replication is completed prior to arrest. Note that the apparent increase in DNA content seen following shift to the restrictive temperature (seen as a gradual rightwards drift in the position of the 2C peak over the timecourse of the experiment shown in Figure 2B) is the result of ongoing mitochondrial DNA replication in elongating (i.e. cell cycle arrested) S.pombe cells, as described previously (37). Similar peak movements are shown in Figure 6 for rfc1-44 single-, double- and triple-mutant combinations (see below), but not in cells of normal length.
Figure 6 Regulated overproduction of RFC/RFC-like complex large subunits in wild-type rfc1-44, ctf18 and elg1 cells. (A) Expression in wild-type and rfc1-44 cells growing on medium containing thiamine (promoter repressed). (B) Expression in wild-type and rfc1-44 cells growing on medium lacking thiamine (promoter derepressed). (C) Expression in ctf18 and elg1 cells growing on medium containing thiamine. (D) Expression in ctf18 and elg1 cells growing on medium lacking thiamine. Cells were transformed with the indicated plasmids (pREP3X, etc.) and serial dilutions plated on EMM plates either with or without 5 μg/ml thiamine, to either repress or induce large subunit expression from the nmt1 promoter. Plates were incubated at 25 or 35°C as indicated for 3–5 days.
To analyse DNA structure in the arrested cells, PFGE of chromosomal DNA from wild-type and mutant cells was performed. Many previously isolated yeast DNA replication mutants display a 2C DNA content by flow cytometry but their chromosomes cannot enter a PFGE gel (45). Similar results are obtained when chromosomes are prepared from the cells arrested in S-phase with HU, for example, and are taken as an indication that the chromosomes remain at least partly unreplicated (45). Significantly, by 4 h after temperature shift, the chromosomes from rfc1-44 cells failed to enter the PFGE gel, indicating that the replication is not complete under these circumstances (Figure 2C). At the permissive temperature of 25°C, rfc1-44 cells display increased sensitivity compared with wild type to the DNA replication inhibitor HU and to the DNA damaging agents, MMS and UV, consistent with the previous data implicating RFC and PCNA in various DNA repair mechanisms (Figure 2D). In addition, rfc1-44 is synthetically lethal at 25°C with a number of temperature-sensitive DNA replication mutants (data not shown), including pol3-ts3, cdc1-P13, cdc27-P11, cdc27-D1 and dna2-C2 (46–49). The pol3+, cdc1+ and cdc27+ genes encode the catalytic, B and C subunits, respectively, of DNA polymerase in S.pombe, while dna2+ encodes an endonuclease-helicase implicated in Okazaki fragment processing.
The rfc1-44 mutant can be rescued by mutations in rfc5
The structure of the budding yeast RFC complex has been solved recently (5). The structure shows the ATPase domains of all five subunits forming a right-handed spiral whose pitch matches that of duplex DNA and the CTD forming a physical link between the two ends of the spiral, bridging Rfc1 and Rfc5. This suggested to us that the RFC complex in rfc1-44 cells might be defective in this linking function and that it might be possible to identify suppressors of rfc1-44 by randomly mutagenizing rfc5+. The cloned rfc5+ gene was therefore subjected to pentapeptide insertion mutagenesis and a library of mutagenized plasmids constructed (Materials and Methods). The library DNA was then linearized by restriction enzyme digestion and transformed into rfc1-44 cells. Approximately 2400 integrative transformants obtained at 25°C were tested for their ability to grow at 35°C and six clones capable of growth selected for further analysis, including sequencing of the chromosomal rfc5 gene (Figure 3). Remarkably, all six contained pentapeptide insertions within a very short stretch of the Rfc5 protein, despite the random distribution of insertions in individual library clones as determined by restriction mapping (data not shown). Indeed, five of the six insertions (named rfc5-S1 through rfc5-S5, see Figure 3B) were identical, containing an insertion of the pentapeptide sequence PSRGT between amino acids 75 and 76. The sixth allele (rfc5-S6) contained an insertion of the sequence SRGTP between amino acids 76 and 77. Unfortunately, amino acids 68–85 of budding yeast Rfc5 are absent from the published three-dimensional structure, making it impossible to locate these in relation to the CTD (5). However, these genetic observations suggest that there may be a close functional relationship between those regions of the Rfc1 and Rfc5 proteins defined by rfc1-44 and rfc5-S and support a model in which the biochemical defect in the RFC (Rfc1–44) complex lies in bridging the ends of the RFC spiral.
Figure 3 Suppression of rfc1-44 by rfc5-S1. (A) Upper parts: serially diluted cultures of wild-type, rfc1-44 and rfc1-44 rfc5-S1 double mutant cells plated at 25°C (left panel, 4 days) or 35°C (right panel, 3 days). Lower parts: the same serial dilutions plated on either 5 mM HU (left panel) or 0.005% MMS (right panel) and incubated at 25°C for 4 days. Although rfc5-S1 can suppress rfc1-44 at 35°C, it does not fully restore rfc1-44 activity as the double mutant cells are still sensitive to HU and MMS. See the text for details. (B) Sequence alignment of S.pombe, S.cerevisiae and human Rfc5 proteins showing location of rfc1-44 suppressing insertion alleles. The structure of this region is unknown, except for a short ?-strand (1SXJ ?22) encompassing amino acids 86–88. Conserved residues are indicated by dots beneath the aligned sequences.
Fission yeast Elg1–RFC
During the course of this work, several groups reported the isolation and genetic analysis of the budding yeast ELG1 gene encoding the large subunit of a novel RFC-like complex (20–27). To investigate whether a similar complex existed in S.pombe, the fission yeast elg1+ gene was identified by BLAST searching (E-value 6.7 x 10–16). elg1+ (systematic name SPBC947.11c, see Table 2 and Figure 4A) lies on S.pombe chromosome II and encodes a 2760 bp ORF interrupted by two introns of 44 and 47 nt, resulting in a predicted protein of 920 amino acids and Mr of 103.9 kDa.
Table 2 RFC and RFC-like complex components in fission yeast
Figure 4 Fission yeast Elg1. (A) Chromosome context and schematic structure of the elg1+ gene on chromosome II. (B) TAP of Elg1–RFC. Protein extracts were prepared from elg1-TAP cells and subject to TAP. Following TCA precipitation, the purified proteins were subjected to SDS–PAGE and silver staining (lane 1). The bands indicated were excised and identified by mass spectrometry (see Supplementary Information). Wild-type cells (lane 2) were processed in parallel. The species marked with an asterisk corresponds to N-terminally degraded Elg1. (C) The elg1 cells display increased sensitivity to MMS. Serial dilutions were plated on YE medium containing 0, 0.002, 0.004 or 0.006% MMS and incubated for 4 days at 32°C. (D) The rfc1-44 elg1 cells are viable at 28°C. Eight tetratype tetrads (labelled 1–8) are shown. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate.
To determine the composition of the Elg1–RFC complex, the chromosomal elg1+ gene was tagged at its 3' end with sequences encoding the tandem affinity purification (TAP) tag and the Elg1–RFC purified by the TAP technique (43). Silver staining and subsequent mass spectrometric analysis of the purified material identified Elg1 in both full-length and N-terminally degraded forms, as well as all four small RFC subunits (see Figure 4B and Supplementary Material for the coverage of individual proteins). No additional proteins were identified in these purified preparations, strongly suggesting that Elg1–RFC comprises the Elg1, Rfc2, Rfc3, Rfc4 and Rfc5 proteins only. If the Elg1–RFC complex contains additional proteins, perhaps similar to Dcc1 and Ctf8 in Ctf18–RFC, these must be present in sub-stoichiometric amounts relative to full-length Elg1, be refractive to silver staining, or lost during the TAP purification. Further analysis will be required to address these issues.
To study its function, elg1+ was deleted from the chromosome and replaced with the natMX6 cassette by the PCR-mediated gene targeting method (36). Cells carrying the elg1 allele (elg1::natMX6) were viable and displayed mild sensitivity to MMS (Figure 4C) but were insensitive to UV, HU and the topoisomerase inhibitor camptothecin (data not shown) as is the case for budding yeast elg1 cells (21–23).
Next, rfc1-44 was crossed to elg1. The double mutant rfc1-44 elg1 was viable at 28°C (Figure 4D) and essentially indistinguishable from rfc1-44. However, closer inspection suggested that at 36.5°C the rfc1-44 elg1 cells grew marginally better than the rfc1-44 mutant alone, suggesting that elg1+ might exert a negative effect on rfc1-44 cells (data not shown). This is explored further below.
Fission yeast Ctf18–RFC
To complement our studies on the S.pombe Elg1–RFC, the role of the fission yeast Ctf18–RFC in rfc1-44 cells was examined. In S.cerevisiae, the Ctf18–RFC comprises seven subunits: Ctf18, Rfc2, Rfc3, Rfc4, Rfc5, Dcc1 and Ctf8 (2). The same appears to be true of the human Ctf18–RFC also (17). To investigate whether S.pombe possessed Ctf18, Dcc1 and Ctf8 proteins and to test their function in relation to RFC, the fission yeast genome sequence was probed for genes encoding putative fission yeast Ctf18, Dcc1 and Ctf8 orthologues. Searches using the budding yeast proteins as the query sequences readily identified putative Ctf18 and Dcc1 orthologues (BLAST E-scores of 3.7 x 10–70 and 5.1 x 10–05, respectively; Table 2). ctf18+ (systematic name SPBC902.02c) lies on chromosome II and is predicted to encode a 960 amino acid protein with an Mr of 108.6 kDa (Figure 5A). This is somewhat larger than the S.cerevisiae protein (741 amino acids, Mr 84.3 kDa). The two proteins are 28% identical over their C-terminal 720 amino acids. The dcc1+ gene on chromosome I (SPAC31A2.15c) is predicted to encode a 349 amino acid protein with predicted Mr 40.9 kDa, that is, 22% identical to S.cerevisiae Dcc1 (Figure 5A).
Figure 5 Fission yeast Ctf18, Dcc1 and Ctf8. (A) Chromosome context and schematic structure of the ctf18+, dcc1+ and ctf8+ genes. (B) Multiple sequence alignment of Ctf8 homologues from various species. Key: Hs (human, AI301014 ), At (Arabadopsis thaliana, BAA97465 1), Sp (S.pombe, this study), Sc (S.cerevisiae, accession number NP_012061 ), Eg (Eremothecium gossypii, NP_985024 ), Gl (Giardia lamblia, EAA38993 , Nc (Neurospora crassa, XP_328185 ), Gz (Gibberella zeae, EAA67941 and Mg (Magnaporthe grisea, EAA51651 (C) Sensitivity of wild-type ctf18, dcc1 and ctf8 strains to the microtubule inhibitor benomyl. Serial dilutions were plated on YE plates with/without 10 μg/ml benomyl and incubated at 32°C for 4 days. (D) All three double mutants (ctf18 rfc1-44, dcc1 rfc1-44 and ctf8 rfc1-44) are synthetically lethal. Four to six tetratype tetrads are shown for each cross. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate.
A gene encoding a Ctf8 homologue was also identified. This gene, designated ctf8+, contains a single 50 bp intron (confirmed by cDNA sequencing; data not shown) and encodes a 109 amino acid protein. The ctf8+ is located on chromosome I between the SPAC19D5.06c and SPAC19D5.05c genes. Owing to its small size, the presence of the intron, and the low sequence similarity between Ctf8 family members (S.pombe Ctf8 is only 20% identical to its S.cerevisiae counterpart, see Figure 5B), this region of the chromosome was not previously predicted to encode a protein. The gene has now been assigned the systematic name SPAC19D5.11c (Table 2).
To investigate the role of these three proteins in S.pombe, all three genes were deleted from the chromosome and replaced by the natMX6 cassette. Cells carrying one of the ctf18::natMX6 (ctf18), dcc1::natMX6 (dcc1) or ctf8::natMX6 (ctf8) alleles were viable and indistinguishable from wild type under normal growth conditions. Deletion of CTF18 in S.cerevisiae renders cells sensitive to the microtubule inhibitor benomyl (16). This was also true of S.pombe ctf18, dcc1 and ctf8 cells (Figure 5C), suggesting that the Ctf18–RFC complexes may perform similar, if not identical, functions in the two yeasts. Consistent with this, while this work was in progress, Watanabe and co-workers (51) reported the isolation of ctf18 and dcc1 mutant alleles in a screen for meiotic chromosome non-disjunction mutants. Inactivation of either gene resulted for the most part in sister chromatid non-disjunction during the second meiotic division. Taken together, these data suggest that the identified genes above encode components of the S.pombe Ctf18–RFC.
Next, it was intended to examine the properties of double mutants of rfc1-44 and ctf18. However, when rfc1-44 was crossed to ctf18, no rfc1-44 ctf18 double mutant progeny were obtained at 25°C, the permissive temperature for rfc1-44 (Figure 5D). Thus in contrast to elg1+, ctf18+ is required for viability of rfc1-44, indicating that Ctf18–RFC plays an essential role when RFC function is impaired. Whether ctf8+ and dcc1+ were required for rfc1-44 viability was also tested, by crossing dcc1 and ctf8 strains to rfc1-44, but in each case the double mutant, rfc1-44 dcc1 or rfc1-44 ctf8, was inviable (Figure 5B), as was the case for rfc1-44 ctf18. We conclude that the full Ctf18–RFC activity is required for the rescue of rfc1-44.
Overproduction of RFC and RFC-like complex large subunits in rfc1-44
In order to test the effects of overproducing the large subunits, each gene was cloned into a fission yeast expression plasmid, 3' to the thiamine-repressible nmt1 promoter (52). The resulting pREP plasmids were then transformed into wild-type, rfc1-44, elg1 and ctf18 strains. With one exception, the transformed cells were viable and suitable for further analysis (Figure 6). The exception was rfc1-44 transformed with pREP3X-Elg1. Only microcolonies could be obtained following transformation and plating onto minimal medium with or without 5 μg/ml thiamine (i.e. with the nmt1 promoter repressed or fully derepressed) at the permissive temperature of 25°C. Microscopic examination revealed that the transformed cells were highly elongated and only marginally viable, as even after prolonged incubation the formation of larger colonies was not possible. Strikingly, transformation of wild-type, elg1 or ctf18, strains with the same pREP3X-Elg1 plasmid did not affect their growth, either in the presence or in the absence of thiamine. Thus, cells in which RFC activity is impaired by mutation within the CTD are highly sensitive to increased Elg1 protein levels. The extent of Elg1 overproduction under these circumstances is not known, as our recent attempts to generate anti-Elg1 antibodies have not been successful (data not shown). Nevertheless, it is tempting to speculate that in these cells excess Elg1 out-competes Rfc1–44 for binding to the small RFC subunits, resulting in an increase in the levels of Elg1–RFC. Additional dominant-negative phenotypes, albeit non-lethal ones, were seen when rfc1+ was overexpressed in wild-type, rfc1-44, elg1 and ctf18, cells grown in the absence of thiamine (promoter fully derepressed), and when ctf18+ was overexpressed in rfc1-44 cells under the same conditions at 25°C (Figure 6).
Inactivation of the Elg1–RFC restores viability to rfc1-44 ctf18 cells
The results above suggested that the Elg1–RFC complex might play a negative role in rfc1-44 cells, since elevated levels of elg1+ expression prevented rfc1-44 colony formation. To test this further, construction of an rfc1-44 elg1 ctf18 triple mutant was attempted by crossing rfc1-44 elg1 to ctf18 elg1, for the reason that viability of rfc1-44 ctf18 might be restored by inactivation of the negatively acting elg1+. As predicted, the rfc1-44 elg1 ctf18 triple mutant was viable (Figure 7A). Fifteen tetrads, each of which gave rise to four viable colonies, were obtained following tetrad dissection of meiotic asci from this cross. Further analysis of the 24 meiotic products from six of these asci (using PCR and replica plating to nourseothricin or G418 to identify ctf18 and elg1, respectively, and replica plating to 35°C to identify rfc1-44; Materials and Methods) identified six triple mutants, as predicted, among four tetratypic asci, one parental ditype and one non-parental ditype. Viable rfc1-44 elg1 ctf18 triple mutants, together with inviable rfc1-44 ctf18 double mutants, were also obtained by crossing rfc1-44 with elg1 ctf18 (data not shown).
Figure 7 (A) The triple mutant rfc1-44 elg1 ctf18 is viable. Four tetratype tetrads derived from a cross of rfc1-44 elg1 x ctf18 elg1 are shown. In each case, all four meiotic products were capable of colony formation. The genotypes of colonies were confirmed by replica plating and/or PCR with genomic DNA templates as appropriate. (B) PFGE analysis of chromosome structure following shift of rfc1-44 and rfc1-44 elg1 cells to 35°C for 4 h, with samples being taken for PFGE analysis every hour. (C) Flow cytometric analysis of propidium-iodide stained cells following shift of the indicated strains to 35°C for 6 h. The asterisk in the triple mutant indicates the position of a sub-population of cells with 1C DNA content.
Further evidence in support of the notion that the Elg1–RFC plays a negative role in rfc1-44 cells came from PFGE analysis of chromosome structure following shift of rfc1-44 and rfc1-44 elg1 cells to the restrictive temperature (Figure 7B). As shown previously, the ability of the chromosomes from rfc1-44 cells to enter a PFGE gel is significantly reduced by 2 h after shift to the restrictive temperature. In rfc1-44 elg1 cells, however, the chromosomes can still enter the gel at later time points (Figure 7B, compare 3 and 4 h timepoints), suggesting that their structure is stabilized by the loss of Elg1–RFC activity. However, note that rfc1-44 elg1 cells were still incapable of colony formation at the restrictive temperature, i.e. the deletion of elg1+ does not fully rescue rfc1-44 (data not shown).
In the rfc1-44 elg1 ctf18 triple mutant cells, all three PCNA-interacting RFC and RFC-like complexes are functionally impaired or absent. The DNA content of these cells following shift to the rfc1-44 restrictive temperature was examined and compared to the two viable double mutants rfc1-44 elg1 and elg1 ctf18, the single mutants rfc1-44, elg1 and ctf18, and wild type (Figure 7C). No significant differences in DNA content profile were apparent between wild type, elg1 and ctf18, and the double mutant elg1 ctf18. As shown in Figure 2C, shifting rfc1-44 to 36.5°C results in rightward movement and broadening of the DNA content peak as the cells arrest in elongated form (37). However, no cells with 1C DNA content are apparent. Similar behaviour is seen with rfc1-44 elg1 cells. However, when the rfc1-44 elg1 ctf18 triple mutant cells are shifted to 36.5°C, there is a transient accumulation of cells with <2C DNA content, suggesting that simultaneous impairment or loss of RFC and Ctf18–RFC results in a severe failure in the replication process, consistent with the inviability of the rfc1-44 ctf18 double mutant.
DISCUSSION
RFC and the RFC-like complexes, Ctf18–RFC and Elg1–RFC, play vital roles in maintaining genome integrity in eukaryotic cells (2,6). RFC loading of PCNA is essential for chromosomal replication and optimal DNA repair, Ctf18–RFC is required for the establishment of sister chromatid cohesion in S-phase, and Elg1–RFC has been implicated in DNA replication, replication fork re-start, recombination and repair (2,6). This describes the results of experiments aimed at investigating the role of the CTD of the large subunit of RFC in fission yeast, Rfc1. Deletion of the entire CTD was shown to abolish RFC function in vivo and a collection of CTD mutant rfc1 alleles generated by random or directed mutagenesis techniques were examined. Twelve alleles were constructed by random pentapeptide insertion. Perhaps surprisingly, only one of the twelve insertions disrupted RFC function, while nine of the twelve had no significant effect. The non-functional mutant protein Rfc143 carries an insertion of the sequence QGVPL between amino acids 753 and 754 (see Supplementary Data for details). Comparison with the S.cerevisiae Rfc1 structure (5) suggests that this region of the Rfc1 protein (amino acids 737–757) is likely to form a lengthy -helix that would be disrupted by the pentapeptide insertion. The remaining two alleles (rfc1-44 and rfc1-54) caused a temperature-sensitive phenotype (Figure 2A). In this case, the encoded mutant proteins carried insertions of the sequences IRGTP and PVGVP between amino acids 811 and 812, and 780 and 781, respectively. Neither insertion can be mapped onto the S.cerevisiae Rfc1 structure as both fall within a region whose structure could not be determined. None of the nine site-directed mutants created in this study, all of which replaced well-conserved charged amino acids within the CTD with alanine (see Supplementary Information), disrupted RFC function.
Further analysis of rfc1-44 showed that these cells were capable of bulk DNA replication after being shifted to the restrictive temperature (Figure 2B) but that the DNA was incapable of entering an agarose gel in PFGE experiments (Figure 2C). These properties are shared by a number of other DNA replication mutants in S.pombe, including the essential DNA helicases encoded by pfh1 and dna2 (49,53) and the Dna2 interacting factor cdc24 (54,55). The rfc1-44 is also more sensitive than wild-type to the DNA damaging agents MMS, HU and UV (Figure 2D) and displays synthetic lethality with conditional lethal pol3, cdc1 and cdc27 alleles, indicative of a close involvement in lagging strand DNA synthesis (46). In an unpublished work, the ability of a range of DNA replication proteins, including the four small RFC subunits, when overproduced, to suppress rfc1-44 at restrictive or semi-restrictive temperatures was tested. However, none of the proteins tested was capable of rescuing rfc1-44. In a subsequent screening for multi-copy suppressors of rfc1-44 using genomic DNA libraries, the same suppressing gene sks2+ (SPBC1709.05) was identified from two of the three libraries screened (J. Kim and S.A. MacNeill, unpublished data). The sks2+ (formerly known as hsc1+) encodes an HSP70 family member (56,57) that may act as a chaperone promoting the folding or assembly of the RFC complex in rfc1-44 cells. Whether Sks2 plays a role in RFC function in wild-type cells awaits further analysis.
That bulk DNA replication was possible in rfc1-44 cells at the restrictive temperature (Figure 2B) suggested that either the RFC complex was not completely inactive under these conditions or that other clamp loaders could partially compensate for the loss of RFC function. To investigate this, analysis of the RFC-like complexes Elg1–RFC and Ctf18–RFC was initiated. The Elg1–RFC complex was purified (Figure 4B) and genes encoding the unique Elg1–RFC and Ctf18–RFC subunits Elg1, Ctf18, Dcc1 and Ctf8 were identified and deleted from the genome (Figures 4 and 5). All three double mutants rfc1-44 ctf18, rfc1-44 dcc1 and rfc1-44 ctf8 are non-viable at the permissive temperature for rfc1-44 (Figure 5D). Biochemical evidence indicates that although Ctf18–RFC lacking Dcc1 and Ctf8 can load PCNA onto DNA in vitro, its activity is 10-fold less than that of the seven subunit complex including Dcc1 and Ctf8 (17). Our genetic results suggest that maximal Ctf18–RFC activity is required when RFC function is impaired, as the both dcc1+ and ctf8+ are required for rfc1-44 viability.
In sharp contrast, the double mutant rfc1-44 elg1 was viable at the permissive temperature (Figure 4D) and PFGE results suggests that inactivation of the Elg1–RFC has a positive effect on rfc1-44 at the restrictive temperature, as the chromosomes were able to enter the gel at later timepoints compared to the single mutant (Figure 7B). This negative effect of Elg1–RFC was most clearly seen when ctf18 elg1 was crossed with rfc1-44 elg1 to generate a viable rfc1-44 elg1 ctf18 triple mutant. This result shows that inactivation of the Elg1–RFC restores viability to the rfc1-44 ctf18 double mutant, reinforcing the view that the Elg1–RFC plays an opposing role to RFC and Ctf18–RFC in S.pombe cells.
Previously, various functions have been proposed for Elg1–RFC (6). For example, Elg1–RFC might function as a specialized PCNA clamp loader for DNA substrates associated with, for example, replication fork stalling or collapse. It is possible that in rfc1-44 ctf18 cells, recognition of such structures and subsequent PCNA loading by Elg1–RFC has a harmful effect, whereas loss of Elg1–RFC function in the triple mutant rfc1-44 ctf18 elg1 allows replication fork recovery to be achieved by an alternative mechanism such as homologous recombination. Intriguingly, in S.cerevisiae elg1 is synthetically lethal with genes involved in homologous recombination . Alternatively, it has been suggested that the Elg1–RFC could function as a PCNA clamp unloader (6). Recycling of PCNA is likely to be vital for efficient replication but exactly how this occurs is unclear. If Elg1–RFC acts as a dedicated clamp unloader, its inactivation might counteract, at least in part, the consequences of impairing RFC activity by the rfc1-44 mutation. Finally, Elg1–RFC might have a role in loading or unloading a modified form of PCNA onto DNA. Ubiquitination and sumoylation of PCNA has been demonstrated to play an important role in replication and repair of damaged DNA but whether PCNA modification impinges on clamp loading and unloading is not known. Further detailed genetic and biochemical studies will be required to resolve these issues.
SUPPLEMENTARY MATERIAL
Supplementary Material is available at NAR Online.
ACKNOWLEDGEMENTS
We would like to thank our friends and colleagues in Edinburgh and elsewhere for their help during the course of this work, in particular Dr Hiroyuki Tanaka for many helpful discussions, Pierre Hentges and Professor Tony Carr (University of Sussex) for the pFA6A-natMX6 plasmid and Dr Finbarr Hayes (UMIST) for supplying the Tn4430 system components. S.M. and F.G. were supported by a Wellcome Trust Senior Research Fellowship in Basic Biomedical Science, J.K. by a Wellcome Trust Prize Studentship, K.M. by the Wellcome Trust 4 year PhD Programme Genetics and Disease, I.C. by a Darwin Trust of Edinburgh vacation scholarship and K.R. by the European Union (contract QLK3-CT-2002-02071). Funding to pay the Open Access publication charges for this article was provided by the University of Copenhagen.
REFERENCES
Maga, G. and Hubscher, U. (2003) Proliferating cell nuclear antigen (PCNA): a dancer with many partners J. Cell Sci., 116, 3051–3060 .
Majka, J. and Burgers, P.M. (2004) The PCNA-RFC families of DNA clamps and clamp loaders Prog. Nucleic Acid Res. Mol. Biol., 78, 227–260 .
Tsurimoto, T. and Stillman, B. (1989) Purification of a cellular replication factor, RF-C that is required for coordinated synthesis of leading and lagging strands during simian virus-40 DNA replication in vitro Mol. Cell. Biol., 9, 609–619 .
Tsurimoto, T. and Stillman, B. (1990) Functions of replication factor C and proliferating cell nuclear antigen—functional similarity of DNA polymerase accessory proteins from human cells and bacteriophage T4 Proc. Natl Acad. Sci. USA, 87, 1023–1027 .
Bowman, G.D., O'Donnell, M., Kuriyan, J. (2004) Structural analysis of a eukaryotic sliding DNA clamp–clamp loader complex Nature, 429, 724–730 .
Kim, J. and MacNeill, S.A. (2003) Genome stability: a new member of the RFC family Curr. Biol., 13, R873–R875 .
Parrilla-Castellar, E.R., Arlander, S.J., Karnitz, L. (2004) Dial 9-1-1 for DNA damage: the Rad9-Hus1-Rad1 (9-1-1) clamp complex DNA Repair (Amst.), 3, 1009–1014 .
Shimomura, T., Ando, S., Matsumoto, K., Sugimoto, K. (1998) Functional and physical interaction between Rad24 and Rfc5 in the yeast checkpoint pathways Mol. Cell. Biol., 18, 5485–5491 .
Naiki, T., Shimomura, T., Kondo, T., Matsumoto, K., Sugimoto, K. (2000) Rfc5, in cooperation with rad24, controls DNA damage checkpoints throughout the cell cycle in Saccharomyces cerevisiae Mol. Cell. Biol., 20, 5888–5896 .
Green, C.M., Erdjument-Bromage, H., Tempst, P., Lowndes, N.F. (2000) A novel Rad24 checkpoint protein complex closely related to replication factor C Curr. Biol., 10, 39–42 .
Bermudez, V.P., Lindsey-Boltz, L.A., Cesare, A.J., Maniwa, Y., Griffith, J.D., Hurwitz, J., Sancar, A. (2003) Loading of the human 9-1-1 checkpoint complex onto DNA by the checkpoint clamp loader hRad17-replication factor C complex in vitro Proc. Natl Acad. Sci. USA, 100, 1633–1638 .
Majka, J. and Burgers, P.M. (2003) Yeast Rad17/Mec3/Ddc1: a sliding clamp for the DNA damage checkpoint Proc. Natl Acad. Sci. USA, 100, 2249–2254 .
Ellison, V. and Stillman, B. (2003) Biochemical characterization of DNA damage checkpoint complexes: clamp loader and clamp complexes with specificity for 5' recessed DNA PLoS Biol., 1, E33 .
Mayer, M.L., Gygi, S.P., Aebersold, R., Hieter, P. (2001) Identification of RFC(Ctf18p, Ctf8p, Dcc1p): an alternative RFC complex required for sister chromatid cohesion in S.cerevisiae Mol. Cell, 7, 959–970 .
Naiki, T., Kondo, T., Nakada, D., Matsumoto, K., Sugimoto, K. (2001) Chl12 (Ctf18) forms a novel replication factor C-related complex and functions redundantly with Rad24 in the DNA replication checkpoint pathway Mol. Cell. Biol., 21, 5838–5845 .
Hanna, J.S., Kroll, E.S., Lundblad, V., Spencer, F.A. (2001) Saccharomyces cerevisiae CTF18 and CTF4 are required for sister chromatid cohesion Mol. Cell. Biol., 21, 3144–3158 .
Bermudez, V.P., Maniwa, Y., Tappin, I., Ozato, K., Yokomori, K., Hurwitz, J. (2003) The alternative Ctf18-Dcc1-Ctf8-replication factor C complex required for sister chromatid cohesion loads proliferating cell nuclear antigen onto DNA Proc. Natl Acad. Sci. USA, 100, 10237–10242 .
Shiomi, Y., Shinozaki, A., Sugimoto, K., Usukura, J., Obuse, C., Tsurimoto, T. (2004) The reconstituted human Chl12–RFC complex functions as a second PCNA loader Genes Cells, 9, 279–290 .
Ben Aroya, S. and Kupiec, M. (2005) The Elg1 replication factor C-like complex: a novel guardian of genome stability DNA Repair (Amst.), 4, 409–417 .
Scholes, D.T., Banerjee, M., Bowen, B., Curcio, M.J. (2001) Multiple regulators of Ty1 transposition in Saccharomyces cerevisiae have conserved roles in genome maintenance Genetics, 159, 1449–1465 .
Bellaoui, M., Chang, M., Ou, J., Xu, H., Boone, C., Brown, G.W. (2003) Elg1 forms an alternative RFC complex important for DNA replication and genome integrity EMBO J., 22, 4304–4313 .
Ben Aroya, S., Koren, A., Liefshitz, B., Steinlauf, R., Kupiec, M. (2003) ELG1, a yeast gene required for genome stability, forms a complex related to replication factor C Proc. Natl Acad. Sci. USA, 100, 9906–9911 .
Kanellis, P., Agyei, R., Durocher, D. (2003) Elg1 forms an alternative PCNA-interacting RFC complex required to maintain genome stability Curr. Biol., 13, 1583–1595 .
Huang, M.E., Rio, A.G., Nicolas, A., Kolodner, R.D. (2003) A genomewide screen in Saccharomyces cerevisiae for genes that suppress the accumulation of mutations Proc. Natl Acad. Sci. USA, 100, 11529–11534 .
Smolikov, S., Mazor, Y., Krauskopf, A. (2004) ELG1, a regulator of genome stability, has a role in telomere length regulation and in silencing Proc. Natl Acad. Sci. USA, 101, 1656–1661 .
Smith, S., Hwang, J.Y., Banerjee, S., Majeed, A., Gupta, A., Myung, K. (2004) Mutator genes for suppression of gross chromosomal rearrangements identified by a genome-wide screening in Saccharomyces cerevisiae Proc. Natl Acad. Sci. USA, 101, 9039–9044 .
Banerjee, S. and Myung, K. (2004) Increased genome instability and telomere length in the elg1-deficient Saccharomyces cerevisiae mutant are regulated by S-phase checkpoints Eukaryot. Cell, 3, 1557–1566 .
Uhlmann, F., Cai, J., Gibbs, E., O'Donnell, M., Hurwitz, J. (1997) Deletion analysis of the large subunit p140 in human replication factor C reveals regions required for complex formation and replication activities J. Biol. Chem., 272, 10058–10064 .
Podust, V.N., Tiwari, N., Ott, R., Fanning, E. (1998) Functional interactions among the subunits of replication factor C potentiate and modulate its ATPase activity J. Biol. Chem., 273, 12935–12942 .
Gomes, X.V., Gary, S.L., Burgers, P.M. (2000) Overproduction in Escherichia coli and characterization of yeast replication factor C lacking the ligase homology domain J. Biol. Chem., 275, 14541–14549 .
Warbrick, E. (1998) PCNA binding through a conserved motif Bioessays, 20, 195–199 .
Bork, P., Hofmann, K., Bucher, P., Neuwald, A.F., Altschul, S.F., Koonin, E.V. (1997) A superfamily of conserved domains in DNA damage-responsive cell cycle checkpoint proteins FASEB J., 11, 68–76 .
Allen, B.L., Uhlmann, F., Gaur, L.K., Mulder, B.A., Posey, K.L., Jones, L.B., Hardin, S.H. (1998) DNA recognition properties of the N-terminal DNA binding domain within the large subunit of replication factor C Nucleic Acids Res., 26, 3877–3882 .
Moreno, S., Klar, A., Nurse, P. (1991) Molecular genetic analysis of fission yeast Schizosaccharomyces pombe. Methods Enzymol, 194, 795–823 .
Prentice, H.L. (1992) High-efficiency transformation of Schizosaccharomyces pombe by electroporation Nucleic Acids Res., 20, 621 .
Bahler, J., Wu, J.Q., Longtine, M.S., Shah, N.G., McKenzie, A., III, Steever, A.B., Wach, A., Philippsen, P., Pringle, J.R. (1998) Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomyces pombe Yeast, 14, 943–951 .
Sazer, S. and Sherwood, S.W. (1990) Mitochondrial growth and DNA synthesis occur in the absence of nuclear DNA replication in fission yeast J. Cell Sci., 97, 509–516 .
Kelly, T.J., Martin, G.S., Forsburg, S.L., Stephen, R.J., Russo, A., Nurse, P. (1993) The fission yeast cdc18+ gene-product couples S-phase to start and mitosis Cell, 74, 371–382 .
Barbet, N., Muriel, W.J., Carr, A.M. (1992) Versatile shuttle vectors and genomic libraries for use with Schizosaccharomyces pombe Gene, 114, 59–66 .
Sambrook, J. and Russell, D.W. Molecular Cloning: A Laboratory Manual, 3rd edn, (2001) Cold Spring Harbor, NY Cold Spring Harbor Laboratory Press .
Hallet, B., Sherratt, D.J., Hayes, F. (1997) Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein Nucleic Acids Res., 25, 1866–1867 .
Hayes, F. and Hallet, B. (2000) Pentapeptide scanning mutagenesis: encouraging old proteins to execute unusual tricks Trends Microbiol., 8, 571–577 .
Tasto, J.J., Carnahan, R.H., McDonald, W.H., Gould, K.L. (2001) Vectors and gene targeting modules for tandem affinity purification in Schizosaccharomyces pombe Yeast, 18, 657–662 .
Kai, M., Tanaka, H., Wang, T.S. (2001) Fission yeast Rad17 associates with chromatin in response to aberrant genomic structures Mol. Cell. Biol., 21, 3289–3301 .
Hennessy, K.M., Lee, A., Chen, E., Botstein, D. (1991) A group of interacting yeast DNA replication genes Genes Dev., 5, 958–969 .
Francesconi, S., Park, H., Wang, T.S.F. (1993) Fission yeast with DNA polymerase temperature-sensitive alleles exhibits cell division cycle phenotype Nucleic Acids Res., 21, 3821–3828 .
MacNeill, S.A., Moreno, S., Reynolds, N., Nurse, P., Fantes, P.A. (1996) The fission yeast Cdc1 protein, a homolog of the small subunit of DNA polymerase , binds to Pol3 and Cdc27 EMBO J., 15, 4613–4628 .
Tanaka, H., Ryu, G.H., Seo, Y.S., MacNeill, S.A. (2004) Genetics of lagging strand DNA synthesis and maturation in fission yeast: suppression analysis links the Dna2–Cdc24 complex to DNA polymerase Nucleic Acids Res., 32, 6367–6377 .
Kang, H.Y., Choi, E., Bae, S.H., Lee, K.H., Gim, B.S., Kim, H.D., Park, C., MacNeill, S.A., Seo, Y.S. (2000) Genetic analyses of Schizosaccharomyces pombe dna2+ reveal that dna2 plays an essential role in Okazaki fragment metabolism Genetics, 155, 1055–1067 .
Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E., Wilm, M., Seraphin, B. (2001) The tandem affinity purification (TAP) method: a general procedure of protein complex purification Methods, 24, 218–229 .
Petronczki, M., Chwalla, B., Siomos, M.F., Yokobayashi, S., Helmhart, W., Deutschbauer, A.M., Davis, R.W., Watanabe, Y., Nasmyth, K. (2004) Sister-chromatid cohesion mediated by the alternative RF-CCtf18/Dcc1/Ctf8, the helicase Chl1 and the polymerase--associated protein Ctf4 is essential for chromatid disjunction during meiosis II J. Cell Sci., 117, 3547–3559 .
Maundrell, K. (1990) nmt1 of fission yeast. A highly transcribed gene completely repressed by thiamine J. Biol. Chem., 265, 10857–10864 .
Tanaka, H., Ryu, G.H., Seo, Y.S., Tanaka, K., Okayama, H., MacNeill, S.A., Yuasa, Y. (2002) The fission yeast pfh1+ gene encodes an essential 5' to 3' DNA helicase required for the completion of S-phase Nucleic Acids Res., 30, 4728–4739 .
Gould, K.L., Burns, C.G., Feoktistova, A., Hu, C.P., Pasion, S.G., Forsburg, S.L. (1998) Fission yeast cdc24+ encodes a novel replication factor required for chromosome integrity Genetics, 149, 1221–1233 .
Tanaka, H., Tanaka, K., Murakami, H., Okayama, H. (1999) Fission yeast Cdc24 is a replication factor C- and proliferating cell nuclear antigen-interacting factor essential for S-phase completion Mol. Cell Biol., 19, 1038–1048 .
Usui, T., Yoshida, M., Kasahara, K., Honda, A., Beppu, T., Horinouchi, S. (1997) A novel HSP70 gene of Schizosaccharomyces pombe that confers K-252a resistance Gene, 189, 43–47 .
Oishi, K., Sugiura, R., Shuntoh, H., Kuno, T. (1996) Cloning and characterization of hsc1+, a heat shock cognate gene of the fission yeast Schizosaccharomyces pombe Gene, 181, 45–49 .
Prakash, S., Johnson, R.E., Prakash, L. (2004) Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function Annu. Rev. Biochem., 74, 317–353 .(Jiyoung Kim1, Kathryn Robertson1, Katie )