sap1 promotes the association of the replication fork protection … · 2007-07-26 · department...

Copyright � 2007 by the Genetics Society of AmericaDOI: 10.1534/genetics.106.065334

Sap1 Promotes the Association of the Replication Fork ProtectionComplex With Chromatin and Is Involved in the Replication

Checkpoint in Schizosaccharomyces pombe

Chiaki Noguchi and Eishi Noguchi1

Department of Biochemistry and Molecular Biology, Drexel University College of Medicine, Philadelphia, Pennsylvania 19102

Manuscript received August 29, 2006Accepted for publication November 29, 2006

ABSTRACT

Sap1 is involved in replication fork pausing at rDNA repeats and functions during mating-type switchingin Schizosaccharomyces pombe. These two roles are dependent on the ability of Sap1 to bind specific DNAsequences at the rDNA and mating-type loci, respectively. In S. pombe, Swi1 and Swi3 form the replicationfork protection complex (FPC) and play important roles in the activation of the replication checkpoint andthe stabilization of stalled replication forks. Here we describe the roles of Sap1 in the replication check-point. We show that Sap1 is involved in the activation of the replication checkpoint kinase Cds1 and thatsap1 mutant cells accumulate spontaneous DNA damage during the S- and G2-phases, which is indicative offork damage. We also show that sap1 mutants have a defect in the resumption of DNA replication after forkarrest. Sap1 is localized at the replication origin ori2004 and this localization is required for the associationof the FPC with chromatin. We propose that Sap1 is required to recruit the FPC to chromatin, therebycontributing to the activation of the replication checkpoint and the stabilization of replication forks.

ACCURATE DNA replication in eukaryotic genomesis a challenging task for the cellular machinery

involved. Cells are constantly under stress from ex-posure to intrinsic and extrinsic agents that can inter-fere with the progression of the replication machinery(replisome). To manage and counter these events, cellsare equipped with a quality control system, the DNAreplication checkpoint, which is activated by stalledreplication forks (Nyberg et al. 2002). Stalled forksthreaten genomic integrity because they can potentiallyenable the rearrangement, breakage, or collapse throughdisassembly of the replisome complex (McGlynn andLloyd 2002). In human cells, defects in the replicationcheckpoint can result in genomic instability, leading toboth developmental and neurological defects and, sig-nificantly, a predisposition to cancer (Paulovich et al.1997; Zhou and Elledge 2000; Abraham 2001; Nyberg

et al. 2002; Kastan and Bartek 2004). In the fission yeastSchizosaccharomyces pombe, the Cds1 kinase enforces thereplication checkpoint by coordinating both cell cyclearrest and DNA repair pathways (Boddy and Russell

2001). We have also shown in a previous study that Cds1prevents fork collapse in response to deoxyribonucleo-tide triphosphate (dNTP) starvation (Noguchi et al.2003), indicating that Cds1 is required to maintain rep-lication forks in a replication-competent state. Likewise,

in the budding yeast Saccharomyces cerevisiae, the inactiva-tion of Rad53, a Cds1 homolog, is associated with repli-cation fork collapse and loss of genomic integrity inresponse to dNTP starvation (Lopes et al. 2001; Tercero

and Diffley 2001; Kolodner et al. 2002; Sogo et al.2002).

Recently, we have shown that Swi1 and Swi3 form a‘‘replication fork protection complex’’ (FPC) that isrequired for efficient Cds1 activation and the sub-sequent stabilization of replication forks in fission yeast(Noguchi et al. 2004). The Swi1–Swi3 complex isevolutionally conserved and is homologous to the Tof1–Csm3 complex in budding yeast and the Timeless–Tipincomplex in humans (Gotter 2003; Lee et al. 2004;Mayer et al. 2004; Noguchi et al. 2004). In both bud-ding and fission yeast, the FPC travels with the repli-cation fork and probably acts during the very earlystages of replication checkpoint signaling (Katou et al.2003; Noguchi et al. 2004). The FPC has also beensuggested to function in a novel S-phase checkpointpathway that contributes to replication fork mainte-nance and survival following alkylation damage infission yeast (Sommariva et al. 2005). In human cells,Timeless has been demonstrated to interact with Chk1(a functional mammalian homolog of Cds1/Rad53)and ATR to control Chk1 activity (Unsal-Kacmaz et al.2005). In addition, the downregulation of Timeless inhuman cells compromises replication and disrupts theintra-S-phase checkpoint (Unsal-Kacmaz et al. 2005),suggesting that there is functional conservation of theFPC across diverse species.

1Corresponding author: Department of Biochemistry and MolecularBiology, Drexel University College of Medicine, 245 N. 15th St., MS497, Philadelphia, PA 19102. E-mail: [email protected]

Genetics 175: 553–566 (February 2007)

Swi1 and Swi3 are also required for programmed forkpausing near both the mating-type (mat1) locus andrDNA loci in fission yeast (Dalgaard and Klar 2000;Krings and Bastia 2004). This Swi1–Swi3-dependentfork pausing event at mat1 is required to initiate thegene conversion events that mediate mating-type switch-ing (Dalgaard and Klar 2000). This mating-typeswitching process also requires switch-activating sites,SAS1 and SAS2, near the mat1 locus (Arcangioli andKlar 1991). SAS1 is known to interact with the Sap1protein, another trans-acting factor that is also thoughtto be required for mating-type switching (Arcangioli

et al. 1994a). Interestingly, recent studies have shownthat Sap1 also binds the Ter1/RFB1 sequence, whichhas strong homology to SAS1 and is required for forkpausing at the rDNA loci (Krings and Bastia 2005;Mejia-Ramirez et al. 2005). The mechanisms by whichthe FPC and Sap1 contribute to programmed fork pau-sing are largely unknown, however. A previous structuralstudy has shown that Sap1 acts as a dimer and bendsDNA upon binding to its specific recognition site (Bada

et al. 2000). Moreover, Sap1 appears to be essential forcell growth, and the overexpression of its C-terminaldomain is associated with chromosomal fragmentation,the loss of minichromosomes, and the cut phenotype(Arcangioli et al. 1994a; de Lahondes et al. 2003),suggesting also that Sap1 plays important roles duringDNA replication and in the maintenance of genomicintegrity.

In this study, we report the identification of the sap11

gene as a dosage suppressor of the swi3-1 mutation. Wefurther show that Sap1 plays an important role in theactivation of the replication checkpoint and in thestabilization of replication forks. Sap1 appears to be as-sociated with the replication origin ori2004 and to berequired for the recruitment of the FPC to chromatin,which contributes to the preservation of genomicintegrity.

MATERIALS AND METHODS

General techniques: The methods used for genetic andbiochemical analyses of fission yeast have been describedpreviously (Moreno et al. 1991; Alfa et al. 1993). Immuno-blotting, Cds1 kinase assay, and ultraviolet (UV) sensitivityassay were performed as described in our earlier studies(Noguchi et al. 2004). To visualize nuclear DNA, cells werefixed in 70% ethanol for 20 min, washed in PBS, and stainedwith 49,6-diamidino-2-phenylindole (DAPI). Pulsed-field gelelectrophoresis was performed as described (Nakamura et al.2002; Noguchi et al. 2002).

Chromatin immunoprecipitation assay: The cell extractsused for chromatin immunoprecipitation (ChIP) assays wereprepared from 5 3 108 S. pombe cells as described previously(Ogawa et al. 1999; Noguchi et al. 2004). Briefly, S. pombe cells(5 3 108) were fixed in 1% formaldehyde for 20 min at roomtemperature and then quenched in 125 mm glycine for 5 min.Cells were washed in TBS and disrupted in lysis buffer (50 mm

Hepes–KOH pH 7.5, 140 mm NaCl, 1 mm EDTA, and 1%

Triton X-100) supplemented with protease inhibitors [0.2 mm

p-(Amidinophenyl) methanesulfonyl fluoride (p-APMSF) andRoche protease inhibitor cocktail]. The broken cells weresonicated six times for 20 sec each with a Misonix Sonicator3000 until chromatin DNA was sheared into 500- to 700-bpfragments. Cell lysate was clarified by two rounds of maximum-speed centrifugation in an Eppendorf 5415C microcentrifugeat 4�. Immunoprecipitations were performed in these cell ex-tracts using either anti-FLAG M2 agarose (Sigma–Aldrich, St.Louis) or anti-GFP agarose (Santa Cruz Biotechnology, SantaCruz, CA). PCR amplification conditions and the specific prim-ers used in these studies have also been described previously(Ogawa et al. 1999).

Isolation of the sap11 gene: swi3-1 cells were transformedwith an S. pombe genomic library cloned into a pUR18 vector(Barbet et al. 1992) with a ura41 selection marker. Trans-formed cells were selected on medium lacking uracil and sup-plemented with 10 mm camptothecin (CPT). Library-derivedgenomic DNAs cloned into pUR18 were then isolated fromCPT-resistant colonies. We isolated clone CR13, among others,that appeared to contain only one open reading frame en-coding the Sap1 protein, which we designated as pSap1.

Isolation of sap1 temperature-sensitive mutants: GenomicDNA was isolated from S. pombe cells containing the sap1-3FLAG-Kanr gene. The sap1-3FLAG fragment was then ampli-fied from this genomic DNA preparation by EXtaq polymerase(TaKaRa, Ohtsu, Japan) using the primers P294 and P296 andsubsequently cloned into the AhdI site of pBluescript II TKS(1), to generate the pTKS-Sap1-3FLAG-Kanr construct. Thesap1-3FLAG-Kanr gene was mutagenized by error-prone PCRusing pTKS-Sap1-3FLAG as the template and the primers P294and P296. Error-prone PCR was performed using five- andthreefold higher than recommended concentrations of EXtaqand dNTPs, respectively. The wild-type sap11 gene was thenreplaced with the mutagenized sap1-3FLAG-Kanr gene at thesap1 locus by a standard transformation method. Kanamycin-resistant colonies were isolated and their growth was examinedat both 25� and 35� to select for temperature-sensitive mutants.

Gene cloning, plasmids, primers, and S. pombe strain con-struction: The S. pombe strains used in this study were con-structed using standard techniques (Alfa et al. 1993), andtheir genotypes are listed in Table 1. sap1-3FLAG (sap1-3FLAG-Kanr) and sap1-GFP (sap1-GFP-Kanr) were generated by a two-step PCR method (Krawchuk and Wahls 1999) usingprimers P295, P296, P300, and P301 to construct a 3xFLAGand a GFP tag at the C terminus of sap1, respectively. pFA6a-3FLAG-KanMX6 (Noguchi et al. 2004) and pFA6a-GFP-KanMX6 (Bahler et al. 1998) were used as the templates forthe PCR-based gene tagging. The primers used in these pro-cedures are available on request. Mutations and epitope-tagged genes have previously been described for cdc25-22(Fantes 1979), swi1D (swi1TKanr) (Noguchi et al. 2003),swi3D (swi3TKanr) (Noguchi et al. 2004), swi3-1 (Gutz andSchmidt 1985), cds1D (cds1Tura41) (Boddy et al. 1998), chk1D(chk1Tura41) (al-Khodairy et al. 1994), rad3D (rad3Tura41),rad22-YFP (rad22-YFP-Kanr) (Noguchi et al. 2003), swi1-GFP(swi1-GFP-Kanr) (Noguchi et al. 2003), swi3-GFP (swi3-GFP-Kanr) (Noguchi et al. 2004), swi1-3FLAG (swi1-3FLAG-Kanr)(Noguchi et al. 2004), uve1D (uve1TLEU2) (Yonemasu et al.1997), rad13D (rad13Tura41) (Yonemasuet al. 1997), and orp1-5FLAG (orp1-5FLAG-Kanr) (Ogawa et al. 1999).

RESULTS

Identification of sap11 as a dosage suppressor of theswi3-1 mutation: To understand how the FPC is regu-lated during its role in the stabilization of replication

554 C. Noguchi and E. Noguchi

forks, we performed a genetic screen to isolate multi-copy suppressors of the swi3-1 mutation. The swi3-1mutation introduces an extra adenine at nucleotideposition 126 in the open reading frame of the gene,causing a frameshift and a truncation of the Swi3protein (Noguchi et al. 2004). However, overproduc-tion of the swi3-1 allele can still partially suppress themating-type switching defect of swi3D cells (data notshown), suggesting that this mutant gene might expressreduced levels of functional Swi3 protein by transla-tional readthrough or that the truncated protein prod-uct has weak biological activity. Nevertheless, the swi3-1mutation renders fission yeast cells sensitive to a varietyof genotoxic agents including CPT and hydroxyurea(HU) (Figure 1, A and B). CPT traps topoisomerase Iby forming a single-strand nick in DNA. This causesbreaks in the replication forks when they collide withthe CPT-stabilized topoisomerase I–DNA complex (Tsao

et al. 1993; Pommier 1996, 2004). HU depletes the dNTPpool by inhibiting ribonucleotide reductase and therebyleads to a replisome progression arrest.

We transformed swi3-1 mutant cells with an S. pombegenomic library to screen for clones that could grow inthe presence of CPT and identified the sap11 gene(Figure 1A). Similar levels of suppression were also ob-served in the presence of HU (Figure 1B). To determinethe mechanism underlying the suppression of swi3-1mutation by sap11, we investigated whether a dosageincrease in the sap11 gene could also suppress the CPTsensitivities of the swi1D or swi3D mutants. As shown inFigure 1C, sap11 partially rescued the CPT sensitivity ofswi1D and suppressed swi3D only very weakly, indicatingthat the presence of Swi3 is important for this partialsuppression. These results together suggest that Sap1may physically interact with Swi3 to facilitate a damageresponse to the effects of CPT. However, we performed

TABLE 1

The S. pombe strains used in this study

Strains Genotypea Reference/source

Y1 h� PR109, Noguchi et al. (2004)Y2 h1 PR110, P. RussellY211 h� swi1TKanr EN3182, Noguchi et al. (2003)Y212 h� rad3Tura41 This studyY149 h� cds1Tura41 KT2751, Noguchi et al. (2003)Y266 h1 cds1Tura41 chk1Tura41 BF2115, Noguchi et al. (2003)Y492 h� chk1Tura41 EN3181, Noguchi et al. (2003)Y570 h� swi1-GFP-Kanr This studyY576 h� swi3-1 ade6-210 his7-366 This studyY625 h� swi1-3FLAG-Kanr cdc25-22 EN3377, Noguchi et al. (2004)Y635 h� swi3-GFP-Kanr EN3384, Noguchi et al. (2004)Y640 h� orp1-5FLAG-Kanr cdc25-22 This studyY668 h� swi3TKanr EN3366, Noguchi et al. (2004)Y669 h1 rad22-YFP-Kanr This studyY670 h� rad22-YFP-Kanr EN3322, Noguchi et al. (2003)Y976 h� sap1-GFP-Kanr This studyY724 h� sap1-3FLAG-Kanr This studyY1113 h1 sap1-1-3FLAG-Kanr This studyY1119 h1 sap1-27-3FLAG-Kanr This studyY1125 h1 sap1-48-3FLAG-Kanr This studyY1182 h� uve1TLEU2 rad13Tura41 EN3286, Noguchi et al. (2003)Y1422 h� sap1-3FLAG-Kanr cdc25-22 This studyY1455 h1 sap1-1-3FLAG-Kanr rad3Tura41 This studyY1457 h1 sap1-1-3FLAG-Kanr chk1Tura41 This studyY1459 h� sap1-1-3FLAG-Kanr cds1Tura41 This studyY1468 h1 sap1-48-3FLAG-Kanr rad3Tura41 This studyY1469 h1 sap1-48-3FLAG-Kanr chk1Tura41 This studyY1470 h� sap1-48-3FLAG-Kanr cds1Tura41 This studyY1505 h1 sap1-48-3FLAG-Kanr rad22-YFP-Kanr This studyY1506 h� sap1-48-3FLAG-Kanr rad22-YFP-Kanr This studyY1643 h� swi1-GFP-Kanr sap1-48-3FLAG-Kanr This studyY1645 h1 swi3-GFP-Kanr sap1-48-3FLAG-Kanr This studyY1824 h� sap1-48-3FLAG-Kanr uve1TLEU2 rad13Tura41 This studyY1928 smt0 rad22-YFP-Kanr This studyY1930 smt0 sap1-48-3FLAG-Kanr rad22-YFP-Kanr This study

a All strains are also leu1-32 ura4-D18.

Sap1 in the Replication Checkpoint 555

a number of immunoprecipitation studies but did notdetect any physical interaction between Swi3 and Sap1(data not shown). Moreover, we did not identify anySap1 peptides in our previous mass spectrometric anal-ysis of Swi1-associated proteins (Noguchi et al. 2004).These results suggest that the Swi3–Sap1 interactionmay be unstable and/or transient in the cell or that theassociation between these proteins is indirect.

Isolation of temperature-sensitive mutants of sap1:To facilitate genetic analysis of sap11, we isolated a num-ber of temperature-sensitive (ts) mutants of the sap1gene using the error-prone PCR method. Among thesemutants, sap1-1 and sap1-48 show severe growth defectsat 32�, whereas sap1-27 fails to grow at 35� (Figure 2A).We determined mutation sites of the sap1 ts alleles andfound that sap1-1 contained a substitution at the 181stcodon (TTA to TCA), leading to the amino acid changefrom leucine to serine (L181S). sap1-27 contained asubstitution at the 41st codon (ATG to ACG), resultingin the amino acid change from methionine to threo-nine (M41T). sap1-48 contained a substitution at the188th codon (ATG to AAG), leading to the amino acidchange from methionine to lysine (M188K). An earlierstudy has shown that Sap1 has functional domains (do-mains I, II, and III) essential for SAS1-specific DNAinteraction (supplemental Figure S1B at http://www.genetics.org/supplemental/). This study has also iden-tified two additional domains (IV and V) required forefficient Sap1–DNA interaction and has found that do-main IV is involved in Sap1 oligomerization (Arcangioli

et al. 1994b) (supplemental Figure S1B). Interestingly,

our mutational analysis revealed that the sap1-27 (M41T)mutation resides in domain I. This is identical to the mu-tation shown to have a defect in DNA binding reportedin an earlier study (Arcangioli et al. 1994b). sap1-1(L181S) and sap1-48 (M188K) mutations are novel andare both in domain IV required for oligomerizationand efficient DNA binding (supplemental Figure S1B).

Sap1 plays an important role in survival duringS-phase following either replication fork arrest or DNAdamage: We first investigated whether Sap1 is involvedin promoting cell survival following the generation ofDNA lesions that block replication forks. UV irradiationcauses the formation of cyclobutane dimers and otherlesions that ultimately lead to an arrest in replisome pro-gression. Serial dilutions of cells plated on agar growthmedium were exposed to increasing doses of UV irradia-tion. These experiments were done at 25�, a permissivetemperature for the sap1 ts mutations, and we observedthat sap1-1 and sap1-27 cells were not sensitive to UVirradiation (data not shown), whereas sap1-48 cellsdisplay only weak sensitivity to UV irradiation (Figure2C, top). This suggests that Sap1 does not have a directrole in the repair of UV lesions. To confirm this pre-diction, we next examined whether Sap1 contributes tocell survival in a mutant strain defective for both nucle-otide excision repair (rad13D) and UV damage excisionrepair (uve1D). These two pathways are responsible for alldetectable UV damage repair in fission yeast (Yonemasu

et al. 1997). Moreover, strains that are defective for bothpathways are acutely sensitive to UV irradiation, andtheir survival in these conditions depends upon eitherhomologous recombination or translesion synthesispathways that can bypass the UV damage during DNAreplication (Friedberg et al. 1995; Woodgate 1999;Boddy et al. 2000). sap1-48 mutants were not sensitiveto UV irradiation at low doses but this mutation repro-ducibly enhanced the UV sensitivity of a rad13D uve1D

double mutant (Figure 2B). These results suggest thatSap1 is critical when UV-damaged DNA is not repairedand is therefore important for UV damage tolerance inS. pombe.

We also performed additional UV survival assays todetect genetic interactions between sap1-48 and muta-tions in checkpoint kinases. The checkpoint kinasesCds1 and Chk1 have overlapping roles during theS-phase stress responses (Rhind and Russell 2000;Boddy and Russell 2001). We found from our currentexperiments that sap1-48 chk1D cells were much moresensitive to UV irradiation than either of the singlemutant cells. In contrast, there were no evident geneticinteractions between sap1-48 and cds1D. The sap1-48cds1D cells were equally sensitive to UV irradiation asthe sap1-48 single mutant (Figure 2C, top). Chk1 is aneffector kinase that functions during the G2-M DNAdamage checkpoint and is activated by DNA damageduring G2. Hence, our current findings suggest that thesap1-48 mutation leads to an increased level of DNA

Figure 1.—The isolation of sap11 as a dosage suppressor ofswi3-1. (A) Fivefold serial dilutions of swi3-1 cells harboringthe indicated plasmids were incubated on yeast extract withsupplements (YES) agar medium supplemented with 0 or 5mm CPT for 2–3 days at 32�. (B) swi3-1 cells harboring theindicated plasmid were incubated on YES medium containing0 or 5 mm HU. (C) swi1D or swi3D cells were transformed withthe indicated plasmid and plated onto YES medium contain-ing 0 or 5 mm CPT for 2–3 days at 32�. Representative imagesof repeat experiments are shown.


Figure 2.—Sap1 is required for cellsurvival after replication fork arrest.(A) Temperature-sensitive phenotypesof sap1-1, sap1-27, and sap1-48 mutants.Fivefold serial dilutions of cells wereplated on YES agar medium and incu-bated at the indicated temperatures for2–3 days. (B) Sap1 contributes to cell sur-vival following UV irradiation in a rad13Duve1D strain that cannot excise UV dam-age. Cells of the indicated genotypeswere spread onto YES agar mediumand exposed to the indicated doses ofUV irradiation. Agar plates were incu-bated for 5 days at 25� and colony num-bers were counted to determine the UVsurvival rate. Experiments were per-formed three times, and error bars indi-cate the standard deviations. (C)Synergistic interactions between thesap1 and chk1 mutations indicate thatSap1 plays important roles in the cellulartolerance to fork arrest. For UV survivalassays, fivefold serial dilutions of cellsof the indicated genotypes were spottedonto YES agar medium and exposed tothe indicated doses of UV irradiation.The plates were then incubated for 2–3days at 25�. For HU, MMS, and CPT sen-sitivity assays, fivefold serial dilutions ofcells were incubated on YES agar me-dium supplemented with the indicatedamounts of these drugs for 3–5 days at25�. Representative images of repeat ex-periments are shown.


damage following UV exposure, creating an increasedrequirement for Chk1 activity during the G2-M check-point. These data also lend support to the idea that Sap1is important for cellular tolerance to replication forkarrest caused by UV lesions. We obtained similar resultsusing another ts mutant, sap1-1 (Figure 2C, top).

We next assessed whether Sap1 contributes to the re-sponse to stalled replication forks caused by HU ormethylmethanesulfonate (MMS). HU is a ribonucleo-tide reductase inhibitor leading to an arrest in rep-lisome progression. MMS causes the alkylation oftemplate DNA bases and thereby disrupts the replica-tion process. sap1-48 shows significant sensitivity to HUand weak sensitivity to MMS (Figure 2C, middle andbottom). These sensitivities were further strengthenedby introduction of the chk1D background into the sap1-48 strain. In contrast, the sap1-48 mutation did not showsynergistic interaction with cds1D in HU and MMS sen-sitivity assays (Figure 2C, middle and bottom). Takentogether, our present data suggest that Sap1 is impor-tant for survival during a replication fork arrest that isprovoked by exposure to HU and MMS. Interestingly,we also found synergistic interactions involving rad3D

and either sap1-1 or sap1-48 in HU and MMS sensitivityassays. Rad3 is responsible for cell cycle arrest that iscontrolled by checkpoint kinase Chk1 and Cds1 (Boddy

and Russell 2001; Nyberg et al. 2002). Therefore, thesynergistic interaction between rad3D and sap1 suggeststhat Sap1 has an important function that is independentof cell cycle arrest, and this function may contribute tocell survival following a replication fork arrest inducedby HU or MMS.

We also examined whether Sap1 is involved in the S-phase DNA damage response. For this purpose, weinduced fork breakage during S-phase by exposure toCPT. sap1-48 displays significant sensitivity to CPT, sug-gesting that Sap1 does play an important role in thetolerance to DNA damage during S-phase. In addition,sap1-48 again showed a genetic interaction with chk1D

but not with cds1D (Figure 2C, bottom).In all of the above sensitivity assays for sap1 mutants,

we performed similar experiments in h� and h1 back-grounds and observed no significant difference betweenthe two mating types (data not shown), indicating thatthe mating type has little or no effect on Sap1’s functionin tolerance to DNA damage and fork block.

Sap1 is involved in the replication checkpoint: In allof our current sensitivity assays that we have alreadydescribed, we observed synergistic genetic interactionsbetween the sap1 and chk1D mutations. In contrast, sap1cds1D mutants did not display a stronger sensitivity toS-phase stress agents, compared with either single mu-tant. Therefore, it is possible that Sap1 is involved in aCds1-dependent checkpoint pathway. To address thispossibility, we assessed the role of Sap1 in HU-induceddivision arrest. Because Cds1 and Chk1 have redundantroles in this arrest pathway (Boddy and Russell 2001;

Rhind and Russell 2000), the role of Sap1 was eval-uated in both a cds1D and a chk1D background. In-terestingly, in the absence of HU, sap1-48 chk1D cellsshow a slight increase in the frequency of aberrantmitoses (6.3%) when compared with sap1-48 (2.1%) orsap1-48 cds1D (2.8%) (Figure 3, A and C). Moreover,when cells are treated with HU, the wild-type and sap1-48, cds1D and chk1D mutant strains undergo cell divisionarrest, whereas rad3D or cds1D chk1D mutant cells, whichare unable to activate either Chk1 or Cds1, display amitotic catastrophe or ‘‘cut phenotype’’ that is indicativeof checkpoint failure (Figure 3, B and C). In addition, asignificantly increased population of double-mutantsap1-48 chk1D cells fails to undergo cell division arrestfollowing exposure to HU (19.2% aberrant mitoses),whereas this arrest pathway was functional in sap1-48cds1D mutant cells (1.6% aberrant mitoses) (Figure 3, Band C). The data from these experiments thus suggestthat Sap1 is involved in the HU-induced cell cycle arrestpathway that is mediated by Cds1, thus contributing toactivation of the replication checkpoint. To further ad-dress the role of Sap1 in the activation of the replicationcheckpoint, Cds1 was immunoprecipitated and assayedusing myelin basic protein as a substrate. HU was foundto cause potent activation of Cds1 in wild-type cells,whereas Cds1 activation was strongly diminished inswi1D cells. A significant reduction in Cds1 activationwas also observed in sap1-48 cells (Figure 3D), support-ing our conclusion that Sap1 is involved in activation ofthe replication checkpoint in fission yeast.

To provide a molecular explanation of how the plasmid-borne Sap1 suppresses the drug sensitivities of the swi3-1mutation, we examined whether the Sap1 plasmid isable to reactivate Cds1 kinase activity in swi3-1 and swi1D

strains. However, we did not observe significant reactiva-tion of Cds1 (data not shown). This result suggests that thesuppression of swi3-1 by the Sap1 plasmid might not bedue to reactivation of the Cds1 kinase. It is also possiblethat Cds1 kinase reactivation is not readily detectable inour kinase assay system. Nevertheless, Sap1 overexpres-sion may be sufficient to protect replication forks whenthey are arrested in the presence of HU.

Sap1-48 mutant cells accumulate spontaneous Rad22DNA repair foci: Our data thus far show that Sap1 playsan important role in the replication checkpoint path-way. Interestingly, mutations in Sap1 were also found tosignificantly elevate the sensitivity levels to HU and MMSin a rad3D mutant that is defective in the activation ofboth the Cds1 and the Chk1 pathways (Figure 2C), sug-gesting that Sap1 has a checkpoint-independent rolein the cellular tolerance to replication fork arrest fol-lowing HU or MMS treatment. Moreover, we observeda slight increase in the frequency of aberrant mitosesin the unperturbed sap1-48 chk1D cells (Figure 3, A andC), suggesting that sap1-48 mutant cells may undergospontaneous DNA damage, which should be repairedby the Chk1-dependent G2-M DNA damage checkpoint


Figure 3.—Sap1 is required for HU-induced cell cycle arrest enforced byCds1. The indicated strains were incu-bated in YES liquid medium supple-mented with 0 (A) or 12 mm HU (B)for 6 hr at 25� and then stained withDAPI to visualize nuclear DNA. A differ-ential interference contrast (DIC) imageof the same field has been superim-posed onto the DAPI image. Represen-tative images of repeat experiments areshown. Arrowheads indicate cells show-ing aberrant mitosis. (C) To determinethe ratio of cells that undergo aberrantmitosis, at least 200 cells were countedfor each strain shown in A and B. Theaverages of three independent experi-ments are shown and error bars corre-spond to the standard deviations. (D)Cds1 kinase activation is defective insap1-48 cells. Cells of the indicated geno-types were incubated in YES liquid me-dium supplemented with 12 mm HUfor 0, 1, 2, and 4 hr at 25�. Kinase activityof immunoprecipitated Cds1 was mea-sured using myelin basic protein (MBP)as a substrate. The radiolabeled MBPwas detected after gel electrophoresis(top). The radioactivity levels (countsper minute, CPM) of MBP were then de-termined in a liquid scintillation counter(bottom). Representative results fromrepeat experiments are shown.


pathway. To address this possibility, we monitored theformation of Rad22 DNA repair foci in sap1-48 cells. Thedesign of this experiment was based on previous studiesusing FPC-deficient cells, which display a large increasein the numbers of spontaneous Rad22 foci (Noguchi

et al. 2003, 2004). The sap1-48 strain was engineered toexpress Rad22-YFP under its endogenous promoter.Fission yeast Rad22 is homologous to Rad52 in buddingyeast, which is a protein that binds to single-strandedDNA (ssDNA) during homologous recombination(Paques and Haber 1999). Moreover, Rad22 is knownto form nuclear foci at double-strand breaks and othersites that have exposed ssDNA segments (Kim et al. 2000;

Du et al. 2003). We predicted therefore that if the sap1-48 cells were to incur spontaneous DNA damage as theresult of malfunctioning replication forks, Rad22 wouldbe recruited to the abnormal DNA structures contain-ing ssDNA regions. As expected, a large increase inspontaneous Rad22-YFP foci was detectable in sap1-48cells at 25�, a permissive temperature for sap1-48 (Figure4A). In wild-type cells, 10.7% of the nuclei contained asingle Rad22-YFP focus and only 0.4% contained two ormore foci (Figure 4, A and B). In contrast, we observeda dramatic increase in the number of Rad22-YFP fociin sap1-48 cells whereby 37.7% of the sap1-48 mutantnuclei displayed at least one focus, and 6.3% harbored

Figure 4.—Sap1 is involved inthe stabilization of replicationforks. (A) Cells expressing Rad22-YFP were grown in YES mediumat 25� until midlog phase. A largeincrease inRad22-YFPfociaccumu-lation was observed in the sap1-48cells. All images shown are repre-sentative of repeat experiments.(B) Quantification of Rad22-YFPfoci according to cell cycle stage.Rad22-YFP foci form during S-phase and persist throughout G2-phase. The percentages of nucleithat have at least one focus or har-bor two or more foci are shown.At least 200 cells were counted foreach strain. Error bars correspondto thestandard deviationsobtainedfrom three independent experi-ments. (C) Sap1 is required forthe resumption of replication fol-lowing fork arrest. Chromosomesamples from either wild-type orsap1-48 cells were examined bypulsed-field gel electrophoresis.Cells were grown until midlogphase and then incubated in thepresence of 12 mm HU for 5 hrat 25�. Then cells were thenwashed andreleased into fresh me-dium. Chromosome samples wereprepared at the indicated times.Representative results from repeatexperiments are shown.


two or more foci (Figure 4, A and B). To explorewhether Rad22 foci arose from replication abnormali-ties, the cell cycle position of cells containing Rad22-YFPfoci was estimated by analyzing cell length, nucleinumber and position, and the presence of a divisionplate (Figure 4B). This cell cycle analysis demonstratedthat the Rad22-YFP foci formed during S-phase andwere maintained during G2 phase in the sap1-48 cells(Figure 4B). These results suggest that sap1-48 cellsmay experience replication fork abnormalities duringS-phase. Our data may also indicate that the repair of forkabnormalities is delayed in the sap1-48 cells, althoughfurther investigations will be needed to confirm this.

Sap1 is known to be involved in mating-type switch-ing. To address whether this function at the mating-typelocus affects the formation of Rad22 foci, we examinedRad22-YFP localization in the smt0 background. In thisbackground, S. pombe cells do not experience a double-strand break (DSB) at the mating-type locus. We com-pared smt0 sap11, h� sap11, and h1 sap11 cells and foundno significant difference in Rad22 foci formationamong these strains (supplemental Figure S2A at http://www.genetics.org/supplemental/). As expected, smt0sap1-48 cells displayed a significant increase in Rad22-YFP formation compared to smt0 cells. However, this in-crease was further enhanced in h� sap1-48 and h1 sap1-48cells. Considering the fact that cells experience a DSB atthe mating-type locus in h� and h1 cells, our data suggestthat Sap1 may be required for repair of DSB at the mating-type locus.

We have previously shown that inactivation of Swi1 orSwi3 causes a large increase in Rad22-YFP foci formation(Noguchi et al. 2003, 2004). Therefore, to rationalizethe suppression of swi3-1 by the plasmid-borne Sap1(Figure 1), we examined whether the Sap1 plasmid wasable to suppress the accumulation of Rad22-YFP DNArepair foci formed in swi3-1 and swi1D cells. Althoughthe effect is not dramatic, we observed a significant re-duction of Rad22-YFP foci formation in swi3-1 and swi1D

cells by the plasmid-borne Sap1 (supplemental FigureS2B at http://www.genetics.org/supplemental/). Thisresult suggests that Sap1 contributes to fork protectionby facilitating the function of Swi3.

Sap1 is required for the resumption of the replica-tion fork after arrest: To further explore the role ofSap1 in the maintenance of replication forks, we ex-amined the recovery of DNA replication after fork arrestinduced by HU exposure. Wild-type and sap1-48 cellswere cultured at a permissive temperature (25�) andchromosome samples were prepared prior to (log) andat 5 hr post-HU treatment and also at different timesduring recovery after the removal of HU. These chro-mosomes were then resolved by pulsed-field gel electro-phoresis (PFGE). Chromosomes from logarithmicallygrowing cells in both wild type and sap1-48 were ob-served to have migrated into the gel, indicating that thesap1-48 cells have no significant defect in replicating

DNA at 25� (Figure 4C, log). HU treatment caused anarrest of DNA replication, which in turn reduced thequantity of the chromosomes that could enter in the gelfor both strains (Figure 4C, HU, 5 hr). Wild-type chro-mosomes migrated into the gel at 2 hr after the HUremoval due to the completion of DNA synthesis. How-ever, chromosomes from the sap1-48 cells did not mi-grate at this time point and displayed a reduced capacityto enter the gel at either 4 or 6 hr during recovery(Figure 4C). This indicates that Sap1 functions duringthe recovery of DNA replication after fork arrest. Hence,given the fact that sap1-48 cells also accumulate sponta-neous DNA damage during DNA replication, we con-cluded that Sap1 is involved in the stabilization ofreplication forks.

It is noteworthy that we repeatedly observed that thesap1-48 strains harbor a shorter chromosome III (Figure4C, log), which in S. pombe contains rDNA repeats. Thisresult suggests that there is an increased mitotic re-combination rate at the rDNA repeats in the sap1-48cells. Similar findings have been reported in a numberof replication mutants including swi1D and swi3D

(Sommariva et al. 2005), further suggesting that Sap1plays an important role in maintaining proper DNAreplication and genomic integrity.

Sap1 is associated with the replication origin ori2004:FPC has been shown previously to be a component ofreplication forks (Katou et al. 2003; Noguchi et al.2004; Calzada et al. 2005). To determine whether Sap1also functions at the replication forks, we examined itslocalization by ChIP analysis. The cdc25-22 strain wasengineered to express a FLAG-tagged Sap1 protein(Sap1-3FLAG) via its endogenous gene promoter. Allsap1-3FLAG strains showed normal growth rates and nodetectable sensitivity to HU, MMS, CPT, or UV irradia-tion, indicating that Sap1-3FLAG is functional. Thecdc25-22 allele was utilized to synchronize the cells at theG2-M boundary. We subsequently examined the locali-zation of Sap1-3FLAG at the well-characterized replica-tion origin ori2004 on S. pombe chromosome II (Ogawa

et al. 1999). The movement of Sap1 along the chromo-some was monitored at ori2004 and at two positions 14and 30 kb away (Figure 5, A and H). Sap1 showed only aweak association with ori2004 in G2-arrested cells. Uponrelease from the cdc25-22 arrest at the G2-M boundary,Sap1-3FLAG was observed to strongly associate with thisreplication origin at 60–80 min, which subsequentlydeclined by 180 min. There was no significant Sap1 as-sociation with the chromatin at the distal 14- and 30-kbpositions throughout the experiment (Figure 5A).

We also examined septation to monitor cell cycleprogression, which in fission yeast occurs in S-phase.The level of Sap1 association with ori2004 was found tocorrelate with an increase in the septation index, whichalso coincided with the onset of S-phase, and this in-teraction was found to decline as the septation in-dex decreased (Figure 5A), indicating that Sap1 tightly


Figure 5.—Sap1 associates with the replication origins ori2004 and is required for the recruitment of the FPC to chromatin. (A)ChIP assays of Sap1-3FLAG were performed at ori2004 and at sites located 14 or 30 kb away from this origin. cdc25-22 cells weresynchronized at the G2-M boundary by incubation at 36� for 4 hr and then released into fresh YES media at 25�. ChIP assays wereperformed at the indicated times. An increase in the septation index indicates the onset of S-phase. (B) ChIP assays of Sap1-3FLAGwere performed as described above, except that the cells were released into YES medium containing 10 mm HU to slow downS-phase. (C and D) ChIP assay of Orp1-5FLAG (C) and Rad11-3FLAG (D), performed as described in A. Representative resultsfrom repeat experiments are shown. (E) ChIP assays of Sap1 temperature-sensitive mutants were performed in asynchronouscultures at 25�, a permissive temperature for sap1 mutant cells. Immunoblots of Sap1 mutant proteins show that all of the mutantswere expressed at equivalent levels. Immunoblotting of tubulin was also performed as a loading control. (F) Both Swi1 and Swi3fail to associate with chromatin in sap1-48 cells. ChIP assays of Swi1-GFP and Swi3-GFP were performed at the ori2004 region.Immunoblotting analysis shows that Swi1 and Swi3 are equally expressed in both the wild-type and the sap1-48 background.(G) ChIP assays of Swi1-GFP and Swi3-GFP were performed as described above, except that cells were grown in the present of12 mm HU. Swi1 and Swi3 are shown to be equally expressed by immunoblotting analysis. (H) Diagram of the ori2004 regionon S. pombe chromosome II used in ChIP is shown.


associates with ori2004 during S-phase. As controls, weperformed ChIP analyses of Orp1-FLAG and Rad11-3FLAG. Orp1 is a component of the origin recognitioncomplex and associates with replication origins through-out the cell cycle (Ogawa et al. 1999), whereas Rad11 isthe large subunit of the replication protein A (RPA) com-plex that migrates with the replication forks (Noguchi

et al. 2004). Consistent with previous reports (Ogawa

et al. 1999), Orp1-5FLAG was also found to localize atori2004, and this association persisted throughout thecell cycle (Figure 5C). In contrast, Rad11 associated withori2004 at the onset of S-phase and then relocated fromori2004 to the 14- and 30-kb positions during the courseof the experiment (Figure 5D). The association of Rad11with ori2004 peaks between 60 and 80 min. Rad11 moststrongly associates with the 14-kb position at 80 min andwith the 30-kb position between 80 and 120 min, con-firming that Rad11 travels with replication forks. This isin striking contrast with Orp1 and Sap1, both of whichlocalize only at ori2004. These results thus strengthenour conclusion that Sap1 tightly associates with thereplication origin ori2004 during S-phase.

To confirm Sap1 association with ori2004, we reducedthe rate of fork movement by releasing cdc25-22 sap1-3FLAG cells from G2-M block in the presence of hydroxy-urea. As shown in Figure 5B, Sap1 association with ori2004again becomes prominent as the septation index in-creased. This association persisted throughout the ex-tended S-phase due to the hydroxyurea treatment. Again,we found no significant increase in the level of Sap1association with the 14- and 30-kb positions. Thus, weconcluded that Sap1 tightly associates with ori2004throughout S-phase.

Since Sap1 has only a weak association with ori2004outside of S-phase, we further examined localization ofSap1-GFP within the cells. Strains carrying the sap1-GFPgene from its native promoter showed normal growthrate and no significant sensitivity to HU and CPT, sug-gesting that Sap1-GFP is functional. Sap1-GFP colocal-ized with DAPI-stained DNA in cells at all stages of thecell cycle and in HU-treated cells (data not shown). Insitu chromatin-binding assays (Noguchi et al. 2003)were used to determine whether Sap1-GFP interactswith chromatin. After Triton X-100 extraction of solublenuclear proteins, Sap1-GFP was still colocalized withDNA in all cells (data not shown). This might representthe localization of Sap1 to nonorigin DNA sequencessuch as the rDNA and mating-type loci as previouslyreported (Arcangioli et al. 1994a; Krings and Bastia

2005; Mejia-Ramirez et al. 2005). However, we speculatethat Sap1 has a weak affinity to origins outside ofS-phase, but that this association becomes very tightduring S-phase as ChIP analysis clearly showed thatSap1’s association with ori2004 increased during S-phase.

Sap1 is required for the association between FPCand chromatin: To further understand how the Sap1protein contributes to the stabilization of replication

forks, we examined whether Sap1 mutants can alsoassociate with ori2004. FLAG-tagged Sap1 temperature-sensitive mutant proteins were expressed from theendogenous sap1 promoter and ChIP analyses wereperformed with cells expressing these mutants at apermissive temperature. As shown in Figure 5E, wild-type Sap1 displayed a tight association with ori2004 in anasynchronous culture. However, all three Sap1 mutantproteins displayed a strongly reduced level of associa-tion with ori2004, suggesting that the chromatin associ-ation of Sap1 is important for the stabilization of thereplication fork. This result is consistent with our muta-tional analysis showing that sap1 mutants contain muta-tions in domains required for Sap1–DNA interaction(supplemental Figure S1B at http://www.genetics.org/supplemental/).

We then determined whether the association of Sap1with chromatin is required for FPC localization tochromatin. A sap1-48 strain was engineered to expresseither Swi1-GFP or Swi3-GFP from the endogenousgene promoter. Moreover, the cells expressing thesefusion products were found to be resistant to S-phasestressing agents, indicating that Swi1-GFP and Swi3-GFPare fully functional. When ChIP was performed withSwi1-GFP or Swi3-GFP in a sap11 wild-type background,both the GFP-tagged proteins associated strongly withthe 14- and 30-kb positions and only weakly with ori2004in asynchronous cultures (Figure 5F). These results indi-cate that, under these conditions, Swi1 and Swi3 havealready relocated away from ori2004 as the replicationfork progresses. In contrast, when ChIP was performedwith Swi1-GFP or Swi3-GFP in a sap1-48 background, weobserved a dramatic decrease in the levels of Swi1–Swi3associated with the ori2004 flanking region (Figure 5F).We also obtained similar results in the presence of HU(Figure 5G). In this condition most of the replicationorigins have already fired, explaining why Swi1 and Swi3were not observed to be concentrated at ori2004.Similarly, Swi1 and Swi3 failed to associate with chroma-tin in the sap1-48 background (Figure 5G). In contrast,the association of Sap1 with ori2004 was unaffected bythe absence of Swi1 or Swi3 (data not shown). Takentogether with the evidence that the Sap1-48 protein failsto associate with ori2004 (Figure 5E), our results suggestthat the association between Sap1 and chromatin isrequired for FPC recruitment to the replication forks.

DISCUSSION

In our study, we have described the results of a seriesof experiments that establish the involvement of Sap1in replication (S-phase) stress response pathways infission yeast. These analyses include the assessment ofthe role of Sap1 in the stabilization of replication forksand in the Cds1-dependent replication checkpointpathway. We demonstrate from these experiments thatSap1 associates with ori2004 and is required for FPC


recruitment to the ori2004-flanking region. On the basisof our present data, we thus propose that Sap1 assists inthe maintenance of genomic integrity probably byrecruiting the replication fork protection complex tothe origins of replication.

The role of Sap1 in the activation of the replicationcheckpoint and in fork stabilization: Recent studieshave identified a group of proteins that are required foractivation of the replication checkpoint and the stabi-lization of replication forks in fission yeast. Mrc1, amediator of the replication checkpoint, is known to beessential for Cds1 activation in a Rad3-dependent man-ner (Tanaka and Russell 2001; Zhao et al. 2003). Swi1,a Tof1/Timeless-related protein, which is required forreplication fork pausing at both mating-type and rDNAloci, is also important for proper activation of Cds1(Noguchi et al. 2003). In addition, Swi1 forms a FPCwith Swi3 and is involved in replication fork stabilization(Noguchi et al. 2004). Hsk1-Dfp1, the Cdc7-Dbf4-related kinase that interacts with FPC, is also importantfor the activation of Cds1 (Takedaet al. 2001; Matsumoto

et al. 2005). In our present study, we describe an ad-ditional factor, Sap1, which is required for the activationof the replication checkpoint and the stabilization ofreplication forks. Sap1-48 mutant cells show decreasedlevels of Cds1 activation in response to HU and furthershow increased sensitivities to a variety of agents thatlead to a replication fork arrest (Figures 2 and 3). Fur-thermore, in a similar manner to the FPC (Noguchi

et al. 2004), Sap1 is required for the resumption of DNAreplication after HU-induced fork arrest (Figure 4C).Interestingly, our data also suggest that Sap1 may as-sociate with the replication origins (Figure 5) and thatthis is a prerequisite for the localization of FPC to thereplication forks (Figure 5, F and G), suggesting thatSap1 recruits FPC to these forks. This function requiresDNA binding and oligomerizing activities of Sap1 sincemutations in DNA-binding and oligomerization do-mains greatly reduced the level of the Sap1-ori2004association (supplemental Figure S1B at http://www.genetics.org/supplemental/ and Figure 5E). In our cur-rent model, upon the initiation of replication, Sap1 re-mains localized at the replication origins while theFPC migrates with the replication forks. Hence, Sap1may not have any direct activity in fork stabilization.Although we have not successfully shown that there is aphysical interaction between Sap1 and FPC, our presentgenetic data suggest that there may well be a weak ortransient association between these factors. Taken to-gether, the results of our current study suggest that Sap1recruits the FPC to the replication forks and plays animportant role in mediating the timely activation of thereplication checkpoint. Our findings also indicate thatSap1 functions in replication fork stabilization.

The role of Sap1 in replication fork pausing: To date,Sap1 has been reported to interact with at least two cis-elements, SAS1 of the mating-type locus and Ter1/RFB1

of the rDNA array (Arcangioli et al. 1994a; Krings andBastia 2005; Mejia-Ramirez et al. 2005). Sap1 has alsobeen shown to be involved in the induction of a stalledreplication fork at rDNA repeats (Krings and Bastia

2005; Mejia-Ramirez et al. 2005). In addition, thestalled fork at the Ter1/RFB1 site also requires Swi1and Swi3 (Krings and Bastia 2004), both of which arecomponents of the replication forks (Noguchi et al.2004). Hence, we speculate that the FPC associates withSap1 at the Ter1/RFB1 site to halt fork progression.However, it has been shown that Sap1 is not required forthe stalled replication fork at SAS1 of the mating-typelocus (Dalgaard and Klar 2000), suggesting that morecomplicated mechanisms control the halting of replica-tion forks, involving Swi1–Swi3 and Sap1.

Fork pausing may inevitably involve the stabilization ofreplication forks, which would be required to avoid col-lapse or rearrangements at programmed pausing sites,which are spread throughout the genome. In our presentstudy, we provide evidence that Sap1 may also associatewith replication origins. In addition, since Sap1 recruitsthe FPC to chromatin, we propose that the Sap1–FPCinteraction is critical for the recruitment of the FPC tofacilitate fork stabilization and fork pausing.

Sap1’s roles in cell viability: Earlier studies have sug-gested that Sap1 acts downstream of Swi1 and Swi3during the process of mating-type switching in fissionyeast (Dalgaard and Klar 2000). However, Sap1 isessential for cell viability, indicating its important rolesin other general cellular functions (Arcangioli et al.1994a; de Lahondes et al. 2003). Moreover, we haveconsistently found putative Sap1 homologs in a varietyof other organisms that do not employ mating-typeswitching. For example, Sap1 shows significant similar-ity to Girdin (also known as APE, GIV, or Hook-relatedprotein) in humans and mice (Anai et al. 2005;Enomoto et al. 2005; Le-Niculescu et al. 2005; Simpson

et al. 2005) and also to the Xenopus LOC431973 protein(supplemental Figure S1A at http://www.genetics.org/supplemental/), indicating its conservation throughoutevolution. However, we could not identify a Sap1-likeprotein in budding yeast. Sap1 comprises a DNA-bind-ing domain, a dimerization domain, and a C-terminaldomain (Arcangioli et al. 1994b) (supplemental Fig-ure S1A) and it is significant that the regions of greatesthomology reside in the DNA-binding and oligomeriza-tion domains, indicating that the functions of Sap1 inthe replication checkpoint and during fork stabiliza-tion may be conserved. Interestingly, in human, Girdinhas been implicated in the regulation of DNA synthe-sis (Anai et al. 2005), suggesting its possible role in thereplication checkpoint. In S. pombe, de Lahondes et. al.have reported that the overexpression of the Sap1C-terminal domain during DNA replication stronglyinduces minichromosome loss, which is associated withthe cut phenotype (de Lahondes et al. 2003). This sug-gests that Sap1 may play an important conserved role


in DNA replication per se. In this study, we have providedevidence that Sap1 tightly associates with the replicationorigin ori2004 during S-phase. Since Sap1 appears to belocated on chromatin outside the S-phase, it is possiblethat Sap1 is an ancillary component of the prereplica-tive complex (pre-RC). Taken together, our data suggestthat Sap1 might be essential for the initiation of DNAreplication. It would therefore be interesting to exam-ine in a future study whether Sap1 is a component ofpre-RC that is required for the initiation of DNAreplication.

We thank T. Wang for generously providing Cds1 polyclonal antiseraand H. Masukata and P. Russell for donating the S. pombe strains. Wealso thank B. Arcangioli, T. Nakamura, J. Nickels, and P. Russell fortheir helpful discussions. This work was supported by a LeukemiaResearch Foundation grant (to E.N.) and a Drexel University Collegeof Medicine start-up fund (to E.N.).

LITERATURE CITED

Abraham, R. T., 2001 Cell cycle checkpoint signaling through theATM and ATR kinases. Genes Dev. 15: 2177–2196.

Alfa, C., P. Fantes, J. Hyams, M. McLeod and E. Warbrick,1993 Experiments With Fission Yeast. Cold Spring Harbor Labora-tory Press, Cold Spring Harbor, NY.

al-Khodairy, F., E. Fotou, K. S. Sheldrick, D. J. Griffiths, A. R.Lehmann et al., 1994 Identification and characterization ofnew elements involved in checkpoint and feedback controls infission yeast. Mol. Biol. Cell 5: 147–160.

Anai, M., N. Shojima, H. Katagiri, T. Ogihara, H. Sakoda et al.,2005 A novel protein kinase B (PKB)/AKT-binding protein en-hances PKB kinase activity and regulates DNA synthesis. J. Biol.Chem. 280: 18525–18535.

Arcangioli, B., and A. J. Klar, 1991 A novel switch-activating site(SAS1) and its cognate binding factor (SAP1) required for effi-cient mat1 switching in Schizosaccharomyces pombe. EMBO J. 10:3025–3032.

Arcangioli, B., T. D. Copeland and A. J. Klar, 1994a Sap1, a pro-tein that binds to sequences required for mating-type switching,is essential for viability in Schizosaccharomyces pombe. Mol. Cell.Biol. 14: 2058–2065.

Arcangioli, B., M. Ghazvini and V. Ribes, 1994b Identification ofthe DNA-binding domains of the switch-activating-protein Sap1from S. pombe by random point mutations screening in E. coli.Nucleic Acids Res. 22: 2930–2937.

Bada, M., D. Walther, B. Arcangioli, S. Doniach and M. Delarue,2000 Solution structural studies and low-resolution model ofthe Schizosaccharomyces pombe Sap1 protein. J. Mol. Biol. 300:563–574.

Bahler, J., J. Q. Wu, M. S. Longtine, N. G. Shah, A. McKenzie, 3rd

et al., 1998 Heterologous modules for efficient and versatilePCR-based gene targeting in Schizosaccharomyces pombe. Yeast 14:943–951.

Barbet, N., W. J. Muriel and A. M. Carr, 1992 Versatile shuttle vec-tors and genomic libraries for use with Schizosaccharomyces pombe.Gene 114: 59–66.

Boddy, M. N., and P. Russell, 2001 DNA replication checkpoint.Curr. Biol. 11: R953–R956.

Boddy, M. N., B. Furnari, O. Mondesert and P. Russell, 1998 Rep-lication checkpoint enforced by kinases Cds1 and Chk1. Science280: 909–912.

Boddy, M. N., A. Lopez-Girona, P. Shanahan, H. Interthal, W. D.Heyer et al., 2000 Damage tolerance protein Mus81 associateswith the FHA1 domain of checkpoint kinase Cds1. Mol. Cell.Biol. 20: 8758–8766.

Calzada, A., B. Hodgson, M. Kanemaki, A. Bueno and K. Labib,2005 Molecular anatomy and regulation of a stable replisomeat a paused eukaryotic DNA replication fork. Genes Dev. 19:1905–1919.

Dalgaard, J. Z., and A. J. Klar, 2000 swi1 and swi3 perform im-printing, pausing, and termination of DNA replication in S.pombe. Cell 102: 745–751.

de Lahondes, R., V. Ribes and B. Arcangioli, 2003 Fission yeastSap1 protein is essential for chromosome stability. Eukaryot. Cell2: 910–921.

Du, L.-L., T. Nakamura, B. A. Moser and P. Russell, 2003 Re-tention but not recruitment of Crb2 at double-strand breaks re-quires Rad1 and Rad3 complexes. Mol. Cell. Biol. 23: 6150–6158.

Enomoto, A., H. Murakami, N. Asai, N. Morone, T. Watanabe et al.,2005 Akt/PKB regulates actin organization and cell motility viaGirdin/APE. Dev. Cell 9: 389–402.

Fantes, P., 1979 Epistatic gene interactions in the control of divisionin fission yeast. Nature 279: 428–430.

Friedberg, E. C., G. C. Walker and W. Siede, 1995 DNA Repair andMutagenesis. ASM Press, Washington, DC.

Gotter, A. L., 2003 Tipin, a novel timeless-interacting protein, isdevelopmentally co-expressed with timeless and disrupts itsself-association. J. Mol. Biol. 331: 167–176.

Gutz, H., and H. Schmidt, 1985 Switching genes in Schizosaccharo-myces pombe. Curr. Genet. 9: 325–331.

Kastan, M. B., and J. Bartek, 2004 Cell-cycle checkpoints and can-cer. Nature 432: 316–323.

Katou, Y., Y. Kanoh, M. Bando, H. Noguchi, H. Tanaka et al.,2003 S-phase checkpoint proteins Tof1 and Mrc1 form a stablereplication-pausing complex. Nature 424: 1078–1083.

Kim, W. J., S. Lee, M. S. Park, Y. K. Jang, J. B. Kim et al., 2000 Rad22protein, a rad52 homologue in Schizosaccharomyces pombe, binds toDNA double-strand breaks. J. Biol. Chem. 275: 35607–35611.

Kolodner, R. D., C. D. Putnam and K. Myung, 2002 Maintenanceof genome stability in Saccharomyces cerevisiae. Science 297: 552–557.

Krawchuk, M. D., and W. P. Wahls, 1999 High-efficiency gene tar-geting in Schizosaccharomyces pombe using a modular, PCR-basedapproach with long tracts of flanking homology. Yeast 15:1419–1427.

Krings, G., and D. Bastia, 2004 swi1- and swi3-dependent andindependent replication fork arrest at the ribosomal DNA ofSchizosaccharomyces pombe. Proc. Natl. Acad. Sci. USA 101: 14085–14090.

Krings, G., and D. Bastia, 2005 Sap1p binds to Ter1 at the ribo-somal DNA of Schizosaccharomyces pombe and causes polar replica-tion fork arrest. J. Biol. Chem. 280: 39135–39142.

Lee, B. S., S. I. Grewal and A. J. Klar, 2004 Biochemical in-teractions between proteins and mat1 cis-acting sequences re-quired for imprinting in fission yeast. Mol. Cell. Biol. 24:9813–9822.

Le-Niculescu, H., I. Niesman, T. Fischer, L. DeVries and M. G.Farquhar, 2005 Identification and characterization of GIV, anovel Galpha i/s-interacting protein found on COPI, endoplas-mic reticulum-Golgi transport vesicles. J. Biol. Chem. 280: 22012–22020.

Lopes, M., C. Cotta-Ramusino, A. Pellicioli, G. Liberi, P. Plevani

et al., 2001 The DNA replication checkpoint response stabilizesstalled replication forks. Nature 412: 557–561.

Matsumoto, S., K. Ogino, E. Noguchi, P. Russell and H. Masai,2005 Hsk1-Dfp1/Him1, the Cdc7-Dbf4 kinase in Schizosaccharo-myces pombe, associates with Swi1, a component of the replicationfork protection complex. J. Biol. Chem. 280: 42536–42542.

Mayer, M. L., I. Pot, M. Chang, H. Xu, V. Aneliunas et al.,2004 Identification of protein complexes required for efficientsister chromatid cohesion. Mol. Biol. Cell 15: 1736–1745.

McGlynn, P., and R. G. Lloyd, 2002 Recombinational repair andrestart of damaged replication forks. Nat. Rev. Mol. Cell. Biol.3: 859–870.

Mejia-Ramirez, E., A. Sanchez-Gorostiaga, D. B. Krimer, J. B.Schvartzman and P. Hernandez, 2005 The mating typeswitch-activating protein Sap1 Is required for replication forkarrest at the rRNA genes of fission yeast. Mol. Cell. Biol. 25:8755–8761.

Moreno, S., A. Klar and P. Nurse, 1991 Molecular genetic analysisof fission yeast Schizosaccharomyces pombe. Methods Enzymol. 194:795–823.

Nakamura, T. M., B. A. Moser and P. Russell, 2002 Telomere bind-ing of checkpoint sensor and DNA repair proteins contributes to


maintenance of functional fission yeast telomeres. Genetics 161:1437–1452.

Noguchi, E., P. Shanahan, C. Noguchi and P. Russell, 2002 CDKphosphorylation of Drc1 regulates DNA replication in fissionyeast. Curr. Biol. 12: 599–605.

Noguchi, E., C. Noguchi, L.-L. Du and P. Russell, 2003 Swi1 pre-vents replication fork collapse and controls checkpoint kinaseCds1. Mol. Cell. Biol. 23: 7861–7874.

Noguchi, E., C. Noguchi, W. H. McDonald, J. R. Yates, 3rd andP. Russell, 2004 Swi1 and Swi3 are components of a replicationfork protection complex in fission yeast. Mol. Cell. Biol. 24:8342–8355.

Nyberg, K. A., R. J. Michelson, C. W. Putnam and T. A. Weinert,2002 Toward maintaining the genome. DNA damage and rep-lication checkpoints. Annu. Rev. Genet. 36: 617–656.

Ogawa, Y., T. Takahashi and H. Masukata, 1999 Association of fis-sion yeast Orp1 and Mcm6 proteins with chromosomal replica-tion origins. Mol. Cell. Biol. 19: 7228–7236.

Paques, F., and J. Haber, 1999 Multiple pathways of recombinationinduced by double-strand breaks in Saccharomyces cerevisiae.Microbiol. Mol. Biol. Rev. 63: 349–404.

Paulovich, A. G., D. P. Toczyski and L. H. Hartwell, 1997 Whencheckpoints fail. Cell 88: 315–321.

Pommier, Y., 1996 Eukaryotic DNA topoisomerase I: genomegatekeeper and its intruders, camptothecins. Semin. Oncol.23: 3–10.

Pommier, Y., 2004 Camptothecins and topoisomerase I: a foot in thedoor. Targeting the genome beyond topoisomerase I with camp-tothecins and novel anticancer drugs: importance of DNA rep-lication, repair and cell cycle checkpoints. Curr. Med. Chem.Anti-Canc. Agents 4: 429–434.

Rhind, N., and P. Russell, 2000 Chk1 and Cds1: linchpins of theDNA damage and replication checkpoint pathways. J. Cell Sci.113(22): 3889–3896.

Simpson, F., S. Martin, T. M. Evans, M. Kerr, D. E. James et al.,2005 A novel hook-related protein family and the characteriza-tion of hook-related protein 1. Traffic 6: 442–458.

Sogo, J. M., M. Lopes and M. Foiani, 2002 Fork reversal and ssDNAaccumulation at stalled replication forks owing to checkpointdefects. Science 297: 599–602.

Sommariva, E., T. K. Pellny, N. Karahan, S. Kumar, J. A. Huberman

et al., 2005 Schizosaccharomyces pombe Swi1, Swi3, and Hsk1 arecomponents of a novel S-phase response pathway to alkylationdamage. Mol. Cell. Biol. 25: 2770–2784.

Takeda, T., K. Ogino, K. Tatebayashi, H. Ikeda, K. Arai et al.,2001 Regulation of initiation of S phase, replication checkpointsignaling, and maintenance of mitotic chromosome structuresduring S phase by Hsk1 kinase in the fission yeast. Mol. Biol. Cell12: 1257–1274.

Tanaka, K., and P. Russell, 2001 Mrc1 channels the DNA replicationarrest signal to checkpoint kinase Cds1. Nat. Cell Biol. 3: 966–972.

Tercero, J. A., and J. F. Diffley, 2001 Regulation of DNA replica-tion fork progression through damaged DNA by the Mec1/Rad53 checkpoint. Nature 412: 553–557.

Tsao, Y. P., A. Russo, G. Nyamuswa, R. Silber and L. F. Liu,1993 Interaction between replication forks and topoisomeraseI-DNA cleavable complexes: studies in a cell-free SV40 DNA rep-lication system. Cancer Res. 53: 5908–5914.

Unsal-Kacmaz, K., T. E. Mullen, W. K. Kaufmann and A. Sancar,2005 Coupling of human circadian and cell cycles by the time-less protein. Mol. Cell. Biol. 25: 3109–3116.

Woodgate, R., 1999 A plethora of lesion-replicating DNA poly-merases. Genes Dev. 13: 2191–2195.

Yonemasu, R., S. J. McCready, J. M. Murray, F. Osman, M. Takao

et al., 1997 Characterization of the alternative excision repairpathway of UV-damaged DNA in Schizosaccharomyces pombe.Nucleic Acids Res. 25: 1553–1558.

Zhao, H., K. Tanaka, E. Noguchi, C. Noguchi and P. Russell,2003 Replication checkpoint protein Mrc1 is regulated byRad3 and Tel1 in fission yeast. Mol. Cell. Biol. 23: 8395–8403.

Zhou, B. B., and S. J. Elledge, 2000 The DNA damage response:putting checkpoints in perspective. Nature 408: 433–439.

Communicating editor: M. D. Rose


sap1 promotes the association of the replication fork protection … · 2007-07-26 · department...

Documents