MCB Email Content Delivery
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sommariva, E.
Right arrow Articles by Dalgaard, J. Z.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sommariva, E.
Right arrow Articles by Dalgaard, J. Z.
Molecular and Cellular Biology, April 2005, p. 2770-2784, Vol. 25, No. 7
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.7.2770-2784.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Schizosaccharomyces pombe Swi1, Swi3, and Hsk1 Are Components of a Novel S-Phase Response Pathway to Alkylation Damage{dagger}

Elena Sommariva,1 Till K. Pellny,1 Nilay Karahan,1 Sanjay Kumar,2 Joel A. Huberman,2 and Jacob Z. Dalgaard1*

Marie Curie Research Institute, Oxted, Surrey, United Kingdom,1 Department of Cancer Genetics, Roswell Park Cancer Institute, Buffalo, New York2

Received 26 August 2004/ Returned for modification 23 September 2004/ Accepted 20 December 2004


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The Swi1 and Swi3 proteins are required for mat1 imprinting and mating-type switching in Schizosaccharomyces pombe, where they mediate a pause of leading-strand replication in response to a lagging-strand signal. In addition, Swi1 has been demonstrated to be involved in the checkpoint response to stalled replication forks, as was described for the Saccharomyces cerevisiae homologue Tof1. This study addresses the roles of Swi1 and Swi3 during a replication process perturbed by the presence of template bases alkylated by methyl methanesulfonate (MMS). Both the swi1 and swi3 mutations have additive effects on MMS sensitivity and on the MMS-induced damage checkpoint response when combined with chk1 and cds1, but they are nonadditive with hsk1. Cells with swi1, swi3, or hsk1 mutations are also defective in slowing progression through S phase in response to MMS damage. Moreover, swi1 and swi3 strains show increased levels of genomic instability even in the absence of exogenously induced DNA damage. Chromosome fragmentation, increased levels of single-stranded DNA, increased recombination, and instability of replication forks stalled in the presence of hydroxyurea are observed, consistent with the possibility that the replication process is affected in these mutants. In conclusion, Swi1, Swi3, and Hsk1 act in a novel S-phase checkpoint pathway that contributes to replication fork maintenance and to survival of alkylation damage.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In Schizosaccharomyces pombe there are two known checkpoint pathways responding to DNA damage, both depending on the upstream ATR/ATM/MEC1-like kinase Rad3 (reviewed in references 43 and 39). The Rad3 kinase activates one or both of the transducing kinases, Cds1 and Chk1. Each of these phosphorylates a variety of targets to activate appropriate downstream checkpoint responses. The cds1 pathway can be activated during S phase by stalled forks and by different types of DNA damage (perhaps as a consequence of fork stalling at damaged sites). The chk1 pathway can be activated by DNA damage during the G2 phase and, in the absence of cds1, in the S phase as well (5, 30). Different forms of DNA damage display differential activation of the two pathways. Hydroxyurea (HU) causes activation of the cds1 pathway (26). Camptothecin and UV damage mainly activate the chk1 pathway (7), while methyl methanesulfonate (MMS) exposure leads to activation of both (26, 55).

The Dfp1-Hsk1 kinase complex has less well defined roles in the S-phase checkpoint and in response to DNA damage (15, 47, 48). Hsk1 is a serine-threonine kinase homologous to Cdc7 in Saccharomyces cerevisiae, and it is necessary for activating origins of replication during S phase by phosphorylating target proteins such as the MCM complex (23). Hsk1 also has checkpoint functions; it is necessary for complete activation of the Cds1 response to HU (47, 49), and it acts as a substrate for Cds1 in vivo and in vitro (47). The Hsk1 regulatory subunit, Dfp1 (also called Him1), binds Hsk1 and stimulates its kinase activity. Dfp1 is essential for the G1-to-S transition, and its expression and phosphorylation state are cell cycle regulated (6). In a cds1-dependent reaction, Dfp1 is hyperphosphorylated by Hsk1 in response to HU-induced early-S-phase arrest (49, 6). A particular mutant, dfp11-376, which carries a C-terminal deletion (removing amino acids 377 to 545), is sensitive to MMS but surprisingly not to UV, ionizing radiation, or HU. This mutant also shows fragmentation of the DNA and genomic instability, but the mutation does not affect the protein's role in origin activation. The C terminus of Dfp1 is therefore involved in the alkylation damage response but is outside the known damage response pathways (15).

Swi1 and Swi3 possess several replication-related functions at the mating-type locus mat1: they are involved in replication pausing at the MPS1 (for "mat1 pause site 1") barrier and in replication termination at RTS1 (for "replication termination site 1") (10). In addition, reports have implicated Swi1 and Swi3 in a cds1-dependent checkpoint (37, 38) and the Swi1 and Swi3 S. cerevisiae homologues, Tof1 and Csm3, in a RAD53-dependent checkpoint (14, 51).

Here we show that swi1 and swi3 strains are sensitive to MMS. A genetic analysis of the MMS sensitivity defines the products of these genes as a group that has nonoverlapping functions with the checkpoint proteins Chk1 and Cds1. A septation index analysis reveals that swi1 and swi3 display additive effects with cds1 and chk1 in the checkpoint that inhibits mitosis in response to MMS damage. In contrast, we observed that the hsk1-1312 temperature-sensitive mutation displays, at the permissive temperature, MMS sensitivity that is nonadditive with that of swi1 and swi3. Like the swi1 and swi3 strains, hsk1 mutants are deficient in the ability to retard progression through S phase in response to MMS damage. In swi1 and swi3 mutants, we detected genomic instability, high levels of single-stranded DNA (ssDNA), chromosome fragmentation, increased recombination, and instability of stalled replication forks. Our data suggest that Swi1, Swi3, and Hsk1/Dfp1 cooperate in a novel S-phase alkylation damage response pathway. They also may act to coordinate leading- and lagging-strand replication at damaged bases or at replication forks stalled at other barriers.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
S. pombe strains. Strains were constructed using standard methods (33): AFY7, h+ leu1::(dfp11-376-6his3HA-leu1+) dfp1-D1 ura4-D18 ade6-M216 (15); E111, h90 ade6-M210 his2 swi1-111 (11); E146, h90 ade6-M210 his2 swi3-146 (11); ES81, h90 ade6-M210 leu1-32 ura4-D18 swi1-111 swi3-146; ES87, h90 ade6-M216 ura4-D18 swi3-146 {Delta}chk1::ura4+; ES88, h90 ade6-M210 ura4-D18 swi3-146 {Delta}cds1::ura4+; ES105, h90 ade6-M216 ura4-D18 swi1-111 {Delta}cds1::ura4+; ES143, h90 ade6-M210 swi1-111 {Delta}chk1::ura4+; ES210, h90 ade6-M210 leu1-32 swi3-146 chk1-HA tagged; ES219, h90 ade6-M216 leu1-32 swi1-111 chk1-HA tagged; ES268, h– ura4-D18 {Delta}cds1::ura4+ swi3-146 chk1-HA tagged; ES 274, h90 ade6-M210 leu1-32 ura4-D18 swi3-146 {Delta}cds1:: ura4+ + pTE2 (swi3+); ES277, h90 ade6-M210 ura4-D18 swi3-146 cdc10-M17; ES281, h90 ade6-M210 swi1-111 cdc10-M17; ES296, h90 ade6-M216 leu1-32 ura4-D18 hsk1ts-1312; ES299, h90 ade6-M210 leu1-32 ura4-D18 hsk1ts-1312 swi1-111; ES300, h90 ade6-M210 ura4-D18 hsk1ts-1312 swi3-146; ES331, h+ leu1::(dfp11-376-6his3HA leu1+) dfp1-D1 ura4-D18 ade6-M216 swi1-111; ES333, h+ leu1::(dfp11-376-6his3HA leu1+) dfp1-D1 ura4-D18 ade6-M216 swi3-146; FO163, h– his3-D1 ura4-D18 leu1-32 ade6 M375int::pUC8/his3+/ade6-L469; JZ236, {Delta}RTS1-h90-SspI::RTS1 ade6-M210 leu1-32 ura4-D18 swi1-rtf3; JZ366, h90 ade6-M216 leu1-32 cdc10-M17; JZ473, h90::LEU2 ade6-M210 leu1-32 ura4-D18 {Delta}chk1::ura4+; JZ474, h90::LEU2 ade6-M210 leu1-32 ura4-D18 {Delta}rad3::ura4+; JZ475, h90::LEU2 ade6-M210 leu1-32 ura4-D18 {Delta}cds1::ura4+; JZ483, h90 ade6-M216 ura4-D18 swi3-146 his2; JZ538, h– his3-D1 ura4-D18 leu1-32 ade6-M375int::pUC8/his3+/ade6-L469 {Delta}swi1::ura4+; JZ542, h– his3-D1 ura4-D18 leu1-32 ade6-M375int::pUC8/his3+/ade6-L469 {Delta}swi1::ura4+; NI358, h+ leu1-32 ura4-D18 hsk1-89 >> ura4+ (49); NW222, h– chk1-HA tagged ade6-M216 leu1-32 (56); SK02, h– ura4-D18 leu1-32 ade6-704 cds1::kan-MX6; SP812, Msmt0 ade6-M216 leu1-32 (A. Klar, unpublished data); SP976, h90 ade6-M210 leu1-32 ura4-D18 swi6-mod+ (Klar, unpublished); 501, h– ura4-D18 leu1-32 ade6-704 (28).

E. coli strains. The following strains were used: pES28—TOPO F' containing TY01, 25S rRNA gene TA-cloned forward in the vector pCR 2.1-TOPO; pES29—TOPO F' containing TY02, 25S rRNA gene TA-cloned backwards in the vector pCR 2.1-TOPO; and pES30—TOPO F' containing empty vector pCR 2.1-TOPO (Invitrogen).

Media. Strains were grown in yeast extract (YE) medium, YE medium supplemented with 225 mg of adenine per liter (YEA medium), YES medium (21), or EMM–Leu, –Ade, or –His medium (33), as indicated. When not otherwise stated, YEA medium was used.

Two-dimensional gel electrophoresis. Cells were harvested in log phase, and chromosomal DNA was purified as described previously (18) and digested with HindIII-KpnI or BamHI. Two-dimensional gel analysis (10 µg of DNA) was performed as described previously (4).

MMS sensitivity. YEA medium plates were made with the indicated concentrations of MMS. Serial dilutions of logarithmically growing cultures were spotted onto the plates, which were then incubated for 3 days at 33°C. Only one dilution is shown for each concentration of MMS.

HU sensitivity. For the short-term assay, 12 mM HU was added to logarithmically growing cultures, and samples were taken every 1 h, plated on YEA medium plates, and incubated for 3 days at 33°C. Surviving colonies were counted and compared with the number of untreated colonies. For the long-term assay, serial dilutions of logarithmically growing cultures were spotted onto plates containing 0, 2, or 5 mM HU. The plates were incubated for 3 days at 33°C and then photographed.

Western analysis. Cells from 10-ml cultures at an optical density at 600 nm of 0.5 to 1.0 were resuspended in 200 µl of 20% trichloroacetic acid and broken with glass beads. After addition of 400 µl of 5% TCA, the liquid phase was collected and centrifuged to pellet the proteins. The pellet was then resuspended in sodium dodecyl sulfate loading buffer (pH 8.0), boiled for 5 min, and loaded on an 8% polyacrylamide gel. The proteins were transferred to a nitrocellulose membrane, and the filters were probed with commercial anti-HA antibodies (12CA5; Roche) at a concentration of 2 µg/ml in phosphate-buffered saline-0.5% Tween 20-5% dried milk. Secondary antibodies (GAM; Bio-Rad) were used at a dilution of 1:2,500. The filters were treated with SuperSignal Peroxidase Solution (Pierce) for detection.

Flow cytometry analysis. For flow cytometry analysis (21), cells were fixed in 70% ethanol, pelleted, washed in 50 mM sodium citrate (pH 7.0), and incubated for 2 h in 50 mM sodium citrate containing 0.1 mg of RNase A per ml. After sonication, the cells were stained with 1 µM Sytox Green (Molecular Probes). The analysis was performed on either the Beckman-Coulter Epics XL or Becton-Dickinson FACScan flow cytometer system. FlowJo software (Treestar, Inc.) was used to generate the histograms.

Pulsed-field gel electrophoresis. DNA was purified in agarose plugs as described previously (9). The electrophoresis was performed in a Bio-Rad gel apparatus as specified by the manufacturer for S. pombe chromosomes.

Determination of the recombination rate. The strains were streaked for single colonies on YEA medium to permit the growth of unrecombined Ade– cells. After 2 days of incubation at 33°C, unsectored pure red colonies were then streaked for single colonies on YEA medium plates and incubated for 2 days at 33°C; a whole colony (13 colonies for each strain) was then diluted in water, and an aliquot was spread on YEA medium and EMM–Ade plates. After 2 to 3 days of incubation, the colonies were counted, and the colonies on the EMM-Ade plates were replica plated onto EMM–His plates to distinguish between conversions (Ade+ His+) and deletions (Ade+ His–). Recombination rates were determined by the method of Luria and Delbrück (27), using the equation, r = aN ln (NaC), where r is the total number of recombined cells in all 13 colonies, a is the recombination rate, N represents the average total number of cells in each colony, and C is 13, the number of colonies we used to determine the recombination rate for each strain. After the final incubations, the numbers of colonies on the YEA medium plates were determined and multiplied by the dilution factor to obtain N. The numbers of Ade+ His+ colonies (from the EMM-His plates) or Ade+ His– colonies (colonies on the EMM-Ade plates minus colonies on the EMM-His plates) were multiplied by the dilution factor to determine the appropriate value of r. The corresponding value of a was estimated by iteration.

Quantification of single-stranded regions present in chromosomal DNA by using hydrophobic membranes. Restriction enzyme-digested DNA was either directly applied by slot blot to a hydrophobic membrane or separated on a 0.7% agarose gel before being transferred to the membrane. In both methods, one set of samples was transferred under denaturing conditions, to determine the amount of total DNA, while the other set was transferred under native conditions, to determine the amount of fragments with single-stranded regions.

For the denaturing slot blots, 1 µg of digested DNA was denatured in a total volume of 100 µl of 0.25 M NaOH-0.5 M NaCl for 10 min at room temperature. The volume of the solution was then adjusted to 200 µl using wash buffer (0.015 M NaCl, 0.0015 M sodium citrate, 0.125 M NaOH), and the sample was applied by a vacuum apparatus (Milliblot; Millipore) to the GeneScreen membrane (NEN). For native slot blots, 10 µg of digested DNA in 200 µl of 0.015 M NaCl-0.0015 M sodium citrate (wash buffer) was applied directly to the membrane. In both experiments, the filter was washed with 2 volumes of wash buffer after sample application.

In the experiment where samples were first separated on agarose gels, the gels transferred under denaturing conditions were sequentially incubated twice in 0.25 N HCl for 20 min, 0.5 M NaOH-1 M NaCl for 20 min, and 0.5 M Tris-HCl (pH 8.0)-3 M NaCl for 30 min. The gels transferred under native conditions were incubated twice for 30 min in 0.5 M Tris-HCl (pH 8.0)-3 M NaCl. For both types of pretreated gels, the DNA was transferred to the membrane by Southern transfer using a 1.5 M NaCl-0.15 M sodium citrate solution.

Septation and binucleated-cell index. Cultures were treated with either 0.01 or 0.03% MMS as stated. Cell samples (200 µl) were collected, and the cells were pelleted by centrifugation, washed in 200 µl of PEM (100 mM PIPES [pH 6.9], 1 mM EGTA, 1 mM MgSO4), and resuspended in 50 µl of PEM. Septa and nuclei were visualized by fluorescence microscopy by adding 1 µl of 1-mg/ml Calcofluor (Sigma) with 10 µl of 4',6-diamidino-2 phenylindole (DAPI)/Antifade solution (1 µl of DAPI [1 mg/ml] and 99 µl of Antifade [Vysis] [10 mg/ml] in phosphate-buffered saline) to a 10-µl aliquot of cell suspension. Three hundred or more cells were analyzed for each sample.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
swi1 and swi3 mutants are sensitive to DNA alkylation damage. Initially we investigated the UV, HU and MMS sensitivities of the swi1 and swi3 strains. For this purpose we used the swi1-111 and swi3-146 alleles, which are functionally null mutants for replication pausing and imprinting at mat1 (10, 22). These swi mutant strains resemble the dfp11-376 mutant (15) in showing very low sensitivity to UV damage (38; our unpublished data) and HU (see below) but being sensitive to MMS (Fig. 1A; also see Fig. 6A and B). Swi1 and Swi3 probably function in the same biochemical pathway, because the swi1 swi3 double mutant displays MMS sensitivity similar to that of the single mutants (Fig. 1A). The {Delta}swi1 mutant displays the same sensitivity as the swi1-111 mutant used in the experiment in Fig. 1A (data not shown). To address whether swi1 and swi3 act through a previously described S-phase checkpoint pathway, a double-mutant analysis was performed with the checkpoint mutants chk1 and cds1 (Fig. 1A). The swi1 and swi3 mutants are less sensitive to MMS than the chk1 mutant but are more sensitive than the cds1 strain. The swi1 and swi3 mutations have additive effects when combined with the chk1 and cds1 mutations, indicating that Swi1 and Swi3 have functions in the MMS damage response distinct from those of Cds1 and Chk1. Importantly, in this experiment, colonies of the swi1 chk1 and swi3 chk1 double mutants were observed only in the absence of MMS. Thus, Swi1 and Swi3 are likely to play major roles in the cellular response to alkylation damage. Western analysis of swi1 and swi3 mutant strains carrying HA-tagged Chk1 showed that Swi1 and Swi3 are not necessary for activation of Chk1 by phosphorylation (Fig. 1B). In fact, Chk1 phosphorylation was detectable in unperturbed mutant cells (especially in the swi1 strain), and the level of phosphorylation was increased by MMS treatment of the swi3 strain (Fig. 1B). In combination, these results suggest that Swi1 and Swi3 participate in an MMS-damage response pathway separate from Cds1 and Chk1 but with possible overlapping functions.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 1. swi1 and swi3 constitute a single MMS sensitivity pathway that has nonoverlapping functions with chk1 and cds1. (A) Analysis of the MMS sensitivity of swi1 and swi3 and the cds1 and chk1 checkpoint mutants. Approximately 100 cells from mutant and wild-type (wt) strains were plated on medium containing MMS at the concentrations given above the columns. rad3 is also shown as a control. Strains: wt (SP976), cds1 (JZ475), chk1 (JZ473), rad3 (JZ474), swi1 (E111), swi1 cds1 (ES105), swi1 chk1 (ES143), swi3 (E146), swi3 chk1 (ES87), swi3 cds1 (ES88), and swi1 swi3 (ES81). (B) Western analysis of HA-tagged Chk1p. The characteristic shift of Chk1p due to activation by phosphorylation is observed in swi1 (ES219) and swi3 (ES210) strains in the absence, and (in the case of swi3) is increased in the presence, of MMS. A wild-type strain containing HA-tagged Chk1p (NW222) is shown as a control.

 


View larger version (33K):
[in this window]
[in a new window]
  		FIG.6. hsk1 is nonadditive with respect to swi1 and swi3 in the response to MMS. (A) MMS sensitivities of the hsk1-1312, swi1 hsk1-1312, and swi3 hsk1-1312 strains. Approximately 100 cells from mutant and wild-type (wt) strains were plated on medium containing MMS at the concentrations given above the columns. The plates were incubated at 25°C. Strains: wt (SP976), swi1 (E111), swi3 (E146), hsk1-1312 (ES296), swi1 hsk1-1312 (ES299), swi3 hsk1-1312 (ES300). (B) A time course showing cell survival in cultures treated for the indicated times with 0.03% MMS. Viability was determined by taking samples from treated cultures and plating them on rich medium. The average of two independent experiments is shown. The vertical lines indicate the two data points obtained. (C) The hsk1 mutant displays a proficient cell cycle arrest in response to MMS damage. A similar cell cycle arrest is observed in the swi1 hsk1 and swi3 hsk1 double mutants. The percentages of septated cells in 0.03% MMS-treated asynchronous cultures were determined. Strain names are given in panel A. (D and E) The swi1, swi3, and hsk1-89 strains lack the intra-S-phase checkpoint. Log-phase cultures in YES medium were treated with 0.0075% MMS (low) or 0.015% MMS (high) or left untreated. Samples were collected for flow cytometry at the indicated times. The experiment in panel D was carried out at 30°C, while the one in panel E was carried out at 25°C. For brevity, the data for the cds1 strain are not shown in panel E. The strains were as follows: wt, 501; cds1, SK02; swi1, E111; swi3, E146; hsk1-89, NI358. (F) MMS sensitivities of the swi1, swi3, and dfp11-376 single and double mutants. The assay was carried out as described in panel A, except that 25-fold more cells were spotted for the strains containing the dfp11-376 mutation (see the text for details).

 
Swi1 and Swi3 are involved in the checkpoint that inhibits mitosis in response to MMS damage. To further characterize the swi1 and swi3 response to MMS, we analyzed whether passage through mitosis is inhibited when an asynchronous culture is exposed to MMS. Since formation of a septum is dependent on prior mitosis, we measured the percentage of cells with a septum as an indication of passage through mitosis (15). In wild-type cultures, the proportion of septated cells was dramatically reduced within 2 h of exposure to MMS (Fig. 2A). While the cds1 strain displayed an almost wild-type decrease in septation index in response to MMS damage, the chk1 strain failed to do so. Therefore, Chk1 is a major contributor to the induction of mitotic arrest and chk1 mutant cells continue dividing in the presence of DNA damage. The swi1 (Fig. 2A, top panel) and swi3 (middle panel) single mutants and the swi1 swi3 double mutant (bottom panel) all displayed wild-type mitotic arrest in response to MMS. However, as with the general sensitivity shown in Fig. 1A, additive effects were observed when the two swi mutations were combined with the checkpoint mutations chk1 or cds1. First, even in the absence of MMS damage, 3.7 and 4.7%, of the swi1 chk1 and swi3 chk1 double mutant cells displayed the "cut" phenotype, respectively (Fig. 2B and C). The "cut" (for "cell untimely torn" [17]) phenotype is characteristic of a failure to undergo mitotic arrest. It consists of a septated cell with nuclear division defects, and it can be detected by staining the cells with DAPI and Calcofluor. Second, when these double mutants were exposed to MMS, the number of cells with the "cut" phenotype increased to 10% of the total number (Fig. 2B and C). Furthermore, the septation index of swi1 chk1 and swi3 chk1 double mutants was somewhat higher than that of the chk1 mutant alone (Fig. 2A). Finally, we observed that when the swi3 mutation, but not the swi1 mutation, was combined with the cds1 mutation, the double mutant failed to undergo a robust mitotic arrest (Fig. 2A, middle panel). The observed checkpoint defect could be complemented by a swi3+-containing plasmid, verifying that it is the swi3 mutation that is responsible for arrest failure (Fig. 2D). DAP1 and Calcofluor staining of MMS-treated cells from the swi3 cds1 strain revealed the "cut" phenotype in 2.9% of the cells (Fig. 2B and C).



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 2. swi1 and swi3 single mutants display a proficient cell cycle arrest in response to MMS damage, but additive effects are observed when swi1 and swi3 are combined with cds1 or chk1 mutations. (A) Septation index of single and double mutants. Changes in the percentage of septated cells as a function of time after the addition of 0.03% MMS were determined (15). The data points shown were obtained as the average of the values from two independent measurements. The vertical lines extend upward to the higher of the original data points. The top, middle, and bottom panels display the analyses of the swi1, swi3, and swi1 swi3 mutants, respectively. In each panel the indices of wild-type (wt), cds1, chk1, and rad3 strains are given for comparison. Strain names for each genotype are given in the legend to Fig. 1. (B) The "cut" phenotype can be detected in swi1 chk1, swi3 chk1, and swi3 cds1 double mutants. Images of the cells stained with DAPI and Calcofluor were taken at the 6-h time point for both the treated and untreated cultures. Black arrows indicate cut, anucleate, or multinucleated cells. (C) Quantification of the percentage of cells displaying the cut phenotype for the indicated strains, for both treated and untreated cultures. (D) The swi3 cds1 mitotic checkpoint defect can be complemented by a plasmid containing the swi3+ gene. The swi3 cds1 strain was transformed with a swi3+-containing plasmid (pTE2, strain ES274). The septation index was determined for asynchronously growing cultures exposed to 0.03% MMS. The experimentwas performed as described above, except that cells were grown in medium allowing plasmid selection. (E) Chk1 can be activated in a swi3 cds1 strain. Western analysis of HA-tagged Chk1 is shown. Chk1 is activated by phosphorylation in a swi3 cds1 strain (ES268) even in the absence of damage, but 0.03% MMS treatment for 2 h increases the intensity of the phosphorylated band.

 
To test these observations in more detail, we synchronized cells in G2 by using a lactose gradient (1), exposed them to MMS, and monitored checkpoint-dependent inhibition of mitosis by comparing the relative levels of septation in treated and untreated cultures as a function of time (Fig. 3). The strains were exposed to two different concentrations of MMS, 0.01 and 0.03%. In this experimental setup, strains with a functional mitotic checkpoint are expected, depending on the amount of MMS, either to delay or to prevent mitosis and septation. At 0.01% MMS, wild-type cells were delayed but managed to progress through one division, and then the cell cycle arrested. At 0.03% MMS, even the first mitosis was prevented (Fig. 3). In contrast, the rad3 and chk1 mutants, which lack the mitotic checkpoint, showed progression through mitosis and accumulation of septated cells. The rad3 and chk1 cells completed at least one cell cycle at 0.01% MMS, and they appeared to arrest with a septum in the second cycle. At 0.03% MMS, these strains accumulated septa after the first mitosis (Fig. 3A). Importantly, the swi1, swi3, and swi1 swi3 mutants displayed wild-type-like mitotic arrest in response to MMS, confirming that these Swi proteins are not essential for the mitotic checkpoint. Likewise, as expected, the swi1 chk1 and swi3 chk1 double mutants displayed a phenotype similar to that of the chk1 strain (Fig. 3A).



View larger version (40K):
[in this window]
[in a new window]
  		FIG.3. Confirmation of checkpoint defects in synchronous cultures. (A) Small G2 cells were purified using a lactose gradient (1) and then released into medium containing MMS or no MMS. Two different concentrations of MMS were used: 0.01 and 0.03%. The percentage of septated cells is plotted as a function of time for the single and double mutants. The grey dashed lines indicate the treated cultures; the black lines indicate the untreated cultures. Strain names are given in the legend to Fig. 1. (B) Drawing of the growth stages of a small G2 cell. The phases of the cell cycle and the relative content of DNA are given on the left. The grey arrow indicates the starting point of the experiment, when MMS was added, and the black arrow indicates the stage at which the cells were counted. Cells that do not pass through mitosis due to an active mitotic checkpoint do not form a septum.

 
Consistent with the results from unsynchronized cells in the experiment in Fig. 2A, the swi3 cds1 double mutant also failed to completely suppress septation when exposed to MMS in the G2 phase (Fig. 3A). At 0.01% MMS, septation was not at all suppressed—the cells continued cycling at the same rate as untreated cells. At 0.03% MMS, septation was partially inhibited. The septation index results for swi3, cds1, and swi3 cds1 cells in Fig. 3A were confirmed by independent measurements of the index of binucleate cells in cultures synchronized in G2 as above. The wild-type and single-mutant cells were able to complete one division before cell cycle arrest in 0.01% MMS. However, the double-mutant cells continued cycling for at least two divisions (Fig. 4). The failure of the swi3 cds1 cells to undergo cell cycle arrest was not due to a failure to activate Chk1, since the level of Chk1 activation increased in response to MMS damage (Fig. 2E). Thus, neither Swi3 nor Cds1 is required for Chk1 activation, and these two proteins have nonoverlapping checkpoint functions. However, these data do not exclude the possibility that Swi3 could also play a very minor role in the Chk1-dependent mitotic response to MMS damage. This is the first characterization of a Swi3 function not shared by Swi1.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 4. The swi3 cds1 double mutant lacks a functional G2 checkpoint. Cells were synchronized in G2 as described in the legend to Fig. 3 and treated with 0.01% MMS or left untreated. The percentage of binucleate cells, determined by DAPI staining, is plotted as a function of time for each strain. Like the septation index, the percentage of binucleate cells is a measure of the extent of passage through mitosis (Fig. 3B).

 
Previous studies indicated that Swi1 and Swi3 are important for replication fork pausing and stability (10, 37, 38, 53). The analysis presented above suggests that these Swi proteins also play roles in survival of alkylation damage to DNA and minor roles in activation of the mitotic checkpoint. The genetic experiments indicate that, in these roles, the Swi proteins have nonoverlapping functions with Cds1 and Chk1. These results suggest that Swi1 and Swi3 may act in a novel S-phase damage response pathway parallel to the Cds1 and Chk1 pathways.

swi1 and swi3 mutants are defective in S-phase entry and in the intra-S-phase checkpoint. The observations described above prompted us to test whether Swi1 and Swi3 proteins are required for the intra-S-phase checkpoint, which slows DNA synthesis in response to DNA damage (14, 21, 28, 42). To test this possibility, we constructed strains that contain the temperature-sensitive cdc10 mutation in addition to the swi mutations. These strains can be synchronized in G1 by using a temporary shift from the permissive temperature, 25°C, to the nonpermissive 37°C. By synchronizing cells in G1, exposing them to MMS, and subsequently releasing them into the cell cycle, the direct effects of these mutations on S-phase entry and progression could be monitored by flow cytometry. Synchronized cdc10 strains carrying the swi1 and swi3 mutations were compared to an otherwise wild-type cdc10 strain in Fig. 5. In the absence of MMS, both mutant strains showed a delay in S-phase entry but then completed S phase at a higher rate than the control cdc10 strain did. In the presence of MMS, these characteristics were exacerbated (Fig. 5). An even greater delay in S-phase entry was observed in the swi1 and swi3 strains. The synchrony of the swi mutant cultures deteriorated at later time points, and cells containing greatly variable amounts of DNA were observed. A significant portion of the swi1 and swi3 cells replicated a greater fraction of their genomes than did wild-type cells. The time interval over which significant replication occurred was approximately the same as for untreated cells, indicating that—once started—replication was not significantly delayed in a major portion of the mutant cells. Analysis of chromosomal DNA by pulsed-field gel electrophoresis showed that DNA replication continued, at least partially, at the late time points, as the chromosomal DNA was retained in the well and did not enter the gel (data not shown). Therefore, these mutant cells did not complete S phase, even though they progressed further than the control cells. This progression into S phase indicated that the swi1 and swi3 strains failed to appropriately retard DNA replication in response to alkylation damage. The importance of Swi1 and Swi3 for slowing progression through S phase in response to MMS treatment was subsequently confirmed by an independent method (see Fig. 6D).



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 5. swi1 and swi3 mutants show defects in S-phase progression and a defective intra-S-phase checkpoint in response to MMS damage. cdc10 (JZ366, wt), cdc10 swi1 (ES281, swi1), and cdc10 swi3 (ES277, swi3) cells were synchronized in G1 by shifting the temperature from 25 to 37°C for 4.5 h, incubated for 1.0 h at 37°C in the presence or absence of 0.03% MMS, and then shifted back to 25°C in fresh medium without MMS. Samples were taken at the indicated time points after the shift to 25°C and analyzed by flow cytometry.

 
The Hsk1-Dfp1 complex acts in the same MMS damage-response pathways as Swi1 and Swi3. While the checkpoint functions of Chk1 and Cds1 are well defined, the essential S. pombe kinase complex Hsk1-Dfp1 has a less well understood replication checkpoint role. Interestingly, strains carrying certain mutations in dfp1 or hsk1 display phenotypes similar to those described above for swi1 and swi3. The dfp11-376 mutant is MMS sensitive, but not UV or HU sensitive (15). The temperature-sensitive hsk1-89 strain fails to restrain mitosis when replication is incomplete at the semipermissive temperature (49). The hsk1-1312 strain is sensitive to MMS even at the permissive temperature and shows Chk1p activation even in the absence of DNA damage (47).

We decided to carry out further tests of the damage responses of hsk1 mutants. Using a plating assay, in which cells are grown in the presence of increasing amounts of MMS (Fig. 6A), we found that the MMS sensitivities of the hsk1-1312, swi1 hsk1-1312, and swi3 hsk1-1312 mutants are similar to each other and slightly greater than those of the swi1 and swi3 strains. When we measured the rate at which cells were killed by increasing times of exposure to 0.03% MMS, we found that the single and double mutants were all killed at about the same rate (Fig. 6B). A similar result was obtained with the hsk1-89 allele (data not shown). In addition, like the swi1 and swi3 mutants, the hsk1-1312 allele and its double mutants with swi1 and swi3 were competent for the MMS-induced mitotic checkpoint (Fig. 6C). Thus, Hsk1 appears to be in the same functional group as Swi1 and Swi3 with regard to MMS sensitivity.

We also tested whether the hsk1 mutants, like the swi1 and swi3 mutants, are deficient in the intra-S-phase checkpoint in response to MMS damage (Fig. 6D and E). For this purpose, we employed a new assay for the intra-S-phase checkpoint in unsynchronized fission yeast cells (21). Because cytokinesis usually takes place after the S phase, G1- and S-phase cells are binucleate (Fig. 3B), and untreated exponentially growing cultures show a single major peak with 2C DNA content when analyzed by flow cytometry. The 2C position is indicated by the thin vertical lines in Fig. 6D and E. MMS treatment of checkpoint-proficient cells slows the passage through S phase, thus prolonging most of the S phase until after cytokinesis (see Fig. 3B for rationale; note that cytokinesis occurs at a fixed time interval after mitosis; this time interval is unaffected by DNA replication). The post-cytokinesis S-phase cells form a shoulder or second peak of cells to the left of the main peak, with DNA content between 1C and 2C (wt in Figure 6D and E [21]). In contrast, cds1 cells, which are deficient in the intra-S-phase checkpoint, do not slow S phase, and therefore only a 2C peak can be observed (Fig. 6D). Importantly, swi1 and swi3 cells did not accumulate in S phase (Fig. 6D), confirming the evidence in Fig. 5 that these strains are defective in the intra-S-phase checkpoint. The possibility that the failure of the cds1, swi1, and swi3 cells to accumulate in S phase after MMS treatment (Fig. 6D) might be a consequence of checkpoint-induced cell cycle arrest in G2 was excluded in several ways. First, we demonstrated the ability of cds1, swi1, and swi3 cells to progress through one complete division at similar low concentrations of MMS (Fig. 3A and 4). Second, with the minor exception of cds1 cells after addition of 0.015% MMS, these mutant cells increased in number after addition of MMS to an extent comparable to wild-type cells (see Table S1 in the supplemental material). Third, after addition of MMS, these mutant cells displayed decreases in septation index and binucleate-cell index that were similar to the decreases displayed by wild-type cells (see Table S1 in the supplemental material). Thus, neither the wild-type nor mutant cells arrested in G2 during the first division after addition of 0.0075 or 0.015% MMS; however, the wild-type cells accumulated in S phase whereas the mutant cells did not.

Similarly, at 25°C (a temperature which is permissive for cell proliferation), the hsk1-89 culture showed complete absence of the second peak (Fig. 6E). Using the same assay, we found that hsk1-1312 cells are at most only slightly defective in forming a second peak at 25°C (see Fig. S1 in the supplemental material). We conclude that—like Swi1 and Swi3 function—proper Hsk1 function is essential for slowing the S phase in response to DNA damage by MMS.

As a further test of this conclusion, we synchronized wild-type, hsk1-1312, and hsk1-89 cells in early S phase with 12 mM HU (6h at 25°C), treated the cells with 0.03% MMS for 1 h, and then released them into fresh medium lacking HU and MMS (data not shown). Under these conditions, both hsk1 mutants progressed through S phase without slowing. Thus, our data strongly suggest that Swi1, Swi3, and Hsk1 are all essential for the intra-S-phase checkpoint.

The C-terminal truncation mutant, dfp11-376, is very sensitive to MMS (15). However, this mutant displays slow growth and low cell viability even in the absence of MMS, suggesting that essential functions are affected. We tested the MMS sensitivity of the swi1 dfp11-376 and swi3 dfp11-376 double mutants by using the plate assay described above (Fig. 6F). We found that, to get an equal number of colonies on the plates lacking MMS, 25-fold more of the dfp1 single-mutant cells and the swi1 dfp1 and swi3 dfp1 double-mutant cells had to be spotted compared to the swi1, swi3, and wild-type cells. In addition, the slow growth of the dfp1, swi1 dfp1, and swi3 dfp1 strains required the plates of these strains to be incubated for 5 days (rather than the usual 3 days) at 33°C. Taking these differences into account, the dfp1, swi1 dfp1, and swi3 dfp1 strains displayed sensitivities to MMS that were roughly similar to those of the swi1 and swi3 strains (Fig. 6F). However, we observed that the swi1 dfp1 and swi3 dfp1 double-mutant strains were slightly more sensitive than the dfp1 single mutant, suggesting that the roles of Dfp1, Swi1, and Swi3 in protecting cells against MMS damage may be partly nonoverlapping.

Swi1 and Swi3 are necessary to maintain genomic stability. Checkpoints are mechanisms that cells have developed to maintain genome integrity. Certain checkpoint mutants have been shown to suffer from chromosome fragmentation and loss and to have a high frequency of genomic rearrangements (34). Interestingly, we observed by Western analysis that swi1 and swi3 mutants show Chk1 phosphorylation, characteristic of the presence of elevated ssDNA levels or DNA damage, in untreated cells (Fig. 1B). To investigate this in more detail, we prepared intact chromosomal DNAs in agarose plugs and analyzed them on a pulsed-field gel. Equal numbers of cells were used for each strain. Chromosomes that are undergoing DNA replication or that have aberrant structures fail to enter the gel (20, 25). While three intact S. pombe chromosomes could easily be visualized in the case of wild-type cells, the intensities of the swi1 and swi3 chromosome bands were highly reduced (Fig. 7A). Also, the size of chromosome III DNA was reduced in swi1 and swi3 mutants (Fig. 7A). This chromosome varies in size between strains, since it contains the rRNA gene array (Fig. 7A, lanes wt 1 and wt 2). However, greater reduction of chromosome III size is observed in mutants that show increased rates of mitotic recombination (25). The number of rRNA gene copies in the wild type has been estimated to be approximately 120 (29). The reductions in sizes of the swi1 and swi3 chromosome III DNAs correspond to a loss of approximately 35 copies of the rRNA gene repeat, and extrachromosomal rDNA can be observed in these strains (data not shown). However, we cannot exclude the possibility that internal deletions may account for some of the chromosome III size reduction in these strains. Interestingly, abolishing only the activity of Swi1 that is necessary for replication termination at RTS1 does not cause this genomic instability, since the swi1-rtf3 mutant strain displayed wild-type chromosome III size and signal intensity (Fig. 7A). Snaith et al. found that the hsk1-1312 mutant strain shows similar chromosome III abnormalities (47), consistent with a relationship between Swi1, Swi3, and Hsk1.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 7. swi1 and swi3 mutant strains display genomic instability. (A) swi1 and swi3 mutants show chromosome fragmentation. The pulsed-field gel shows chromosomal DNAs from the strains wt 1 (SP976), wt 2 (SP812), swi1 (E111), swi1-rtf3 (JZ236), and swi3 (E146). The positions of the three intact S. pombe chromosomal DNAs are given to the right. (B) Schematic drawing of the intrachromosomal recombination substrate and of the possible outcomes of recombination events (adapted from reference 40). The red circles indicate the approximate positions of the mutations. (C) swi1 and swi3 cells have increased mitotic recombination. The histogram displays the recombination rates for the substrate in panel B. Rates of conversion- anddeletion-type recombination events are shown as shaded or open boxes, respectively. The given values are the means of two independent experiments, and the range of the values obtained is given as a vertical line. (D) swi1 and swi3 mutants display increased sectoring on YE medium (33) due to recombination at the direct repeats of ade6 in the substrate. The average number of sectors observed per colony are 0.5, 5.0, and 4.6 for wt, swi1, and swi3, respectively (~50 colonies analyzed). Cells with an ade6– mutation turn red on YE medium due to the accumulation of a slightly toxic red pigment. Cells with the wild-type (wt) ade6+ gene do not accumulate the pigment, are white, and grow slightly faster.

 
The reduced size and intensity of the chromosome III band in the swi mutants suggested increased levels of mitotic recombination. To analyze this in more detail, we determined the frequency of recombination events in the swi mutants. We used a recombination substrate that consisted of a direct repeat of two ade6– heteroalleles separated by a wild-type his3+ gene (40) (Fig. 7B). The recombination substrate was introduced into the swi1 and swi3 backgrounds by genetic crosses. Two types of recombination events can be detected. First, gene conversions, involving the formation of Holliday junctions, restore one fully active ade6+ gene while leaving the his3+ gene unaltered. Thus, gene conversions produce cells with an Ade+ His+ phenotype. Second, deletions, caused by single-strand annealing, replication slippage, or intrachromatid crossing over, recreate a wild-type ade6+ at the cost of deletion of the his3+ gene. Thus, deletion events produce Ade+ His– cells. Much higher levels of both conversions and deletions were observed in the swi1 and swi3 mutant strains (Fig. 7C). The increased rate of mitotic recombination in these cells could also be observed on YE (rich) plates. On YE medium, the two ade6 mutations confer a red color to the colony due to accumulation of a slightly toxic intermediate in the adenine biosynthesis pathway. Sectors containing ade6+ cells appear larger and white due to reduced abundance of the red intermediate. The swi1 and swi3 mutants displayed an increased frequency of white sectors compared to the wild-type strain (Fig. 7D), indicative of higher recombination rates. Taken together, these data show that swi1 and swi3 mutants have increased levels of recombination and genomic instability in the absence of exogenous DNA-damaging agents.

Effects of hydroxyurea on swi mutants. We wanted to test whether the relationships between swi mutations and mutations in the cds1 and chk1 genes that we observed for MMS sensitivity (Fig. 1A and 6B) also applied to HU sensitivity. For this purpose, we used two assays. The first assay (Fig. 8A; short-term assay) tested the abilities of wild-type and mutant strains to survive on rich HU-free medium following exposure to HU for periods of 0 to 6 h. Since exposure to HU is relatively brief, it is likely that this assay primarily measures the ability of cells to prevent irreversible replication fork alterations in the presence of HU. Indeed, survival of brief exposure to HU has previously been called "S-phase recovery" (26) and has been found to be highly dependent on Cds1 (35). We confirmed that cds1 mutant cells are very sensitive to brief exposures to HU (Fig. 8A). In contrast, the swi1, swi3, and swi1 swi3 (and also the chk1 and wild-type) strains were insensitive to HU in this short-term assay (Fig. 8A). Like the swi1-111 strain (Fig. 8A), the swi1{Delta} strain was insensitive to HU under these conditions (data not shown). Additive effects were observed when the swi1 or swi3 mutation was combined with the chk1 mutation (Fig. 8A). In addition, when the swi1 mutation was combined with the cds1 mutation, a partial complementation of the cds1 phenotype was observed (Fig. 8A).



View larger version (52K):
[in this window]
[in a new window]
  		FIG. 8. Characterization of the response of the swi1, swi3, and hsk1 mutants to HU treatment. (A) Sensitivities of single and double mutants to short-term exposure to HU. The top, middle, and bottom panels display the survival of the swi1 mutant strains, the swi3 mutant strains, and the swi1 swi3 double mutant, respectively. Wild-type (wt), rad3, cds1, and chk1 survival curves are shown as controls. Samples were taken at the indicated time points after addition of 12 mM HU and plated on YEA medium plates. Shown are the percentages (average values of two independent experiments) of colonies observed from treated cultures relative to untreated cultures. The vertical lines extend to the upper data points. (B) Plate assay displaying the sensitivities of the strains to long-term exposure to HU. The experiment was performed as described in the legend to Fig. 1, except that the plates contained the indicated concentrations of HU. (C) Characterization of HU-stalled replication intermediates from wt (SP976), swi1 (E111), and hsk1-1312 (ES296) strains. Cultures were incubated or not in the presence of 12 mM HU for 2h at 33°C (6 h at 25°C for hsk1-1312). DNA was purified and digested with HindIII-KpnI for subsequent analysis on two-dimensional gels. Probes that detect the HindIII-KpnI origin-containing fragment of S. pombe rDNA (44) were used for hybridization. A drawing of the region analyzed is provided. Black arrows mark X-shaped intermediates. Open arrows indicate weak signals from intermediates possibly formed by breaks within single-stranded regions. ETS, external transcribed spacer (44). (D) Line drawing of the two-dimensional gel signals and their names. (E) One of many possible mechanisms by which single-stranded regions, formed as a consequence of strand uncoupling at replication forks, could lead to replication intermediates with altered mobilities. In this diagram, a large single-stranded region is assumed to have been generated as a result of continuing leading-strand synthesis during blocked lagging-strand synthesis. A nuclease- or shear-induced break at any position within the single-stranded region (some of the possible break sites are indicated by arrows) would lead to a branched restriction fragment (between sites "R") with a single-stranded tail of variable length. Families of similar structures could be responsible for the shadow signals indicated by open arrows in panel C.

 
The second assay tested the ability of cells to produce visible colonies on plates containing increasing concentrations of HU (37, 38) (long-term assay). It is likely that cell survival and growth under these conditions requires more than replication fork stabilization, because in this assay (in contrast to the short-term assay [Fig. 8A]) cds1 cells were significantly less sensitive than rad3 cells (Fig. 8B) or rad26 cells (37, 38) (note that rad26 encodes the fission yeast homolog of ATRIP and rad26 mutants usually have the same phenotypes as rad3 mutants), and the swi mutants were nearly as sensitive as the cds1 mutant (Figure 8B) (37, 38). All of the tested mutant strains were sensitive to 5 mM HU. However, the swi1, swi3, and chk1 strains were relatively resistant to 2 mM HU. The cds1 strain was somewhat sensitive to 2 mM HU, while the rad3 strain was hypersensitive to 2 mM HU. The double mutants swi1 cds1, swi1 chk1, swi3 chk1, and swi1 swi3 were no more sensitive to 2 mM HU than were the corresponding single mutants, but the swi3 cds1 mutant was slightly more sensitive than the single mutant (Fig. 8B). The simplest interpretation of these combined results is that Cds1 is required for both short- and long-term survival in HU, while Swi1, Swi3, and Chk1 are less important for short-term survival and are required only for long-term surival. The swi1 and swi3 mutations display additive effects with chk1 in the short-term assay and swi1 complements cds1 in the stort-term assay, while swi3 is mildly additive with cds1 in the long-term assay. Overall, the results suggest that in response to HU, as in response to MMS, Swi1 and Swi3 act in pathways that are at least partially nonoverlapping with pathways involving Cds1 and Chk1.

Noguchi et al. (37) described instability in a swi1 mutant strain of HU-stalled replication forks at the rRNA gene locus. We detected similar instability in both swi1 and hsk1 mutant strains (Fig. 8C). We found that, when replication was slowed in HU, unusual replication intermediates, which formed a diffuse smear, accumulated in the swi1 and hsk1 mutant strains (open arrows in Fig. 8C). The signal from X-shaped structures, presumably recombination intermediates, also increased in the HU-treated mutant cells (solid arrows in Fig. 8C). We suggest that the diffuse smear may be constituted by intermediates containing single-stranded regions of variable length. One of the ways in which such single-stranded regions could arise would be if coupling between leading- and lagging-strand synthesis were lost at the paused replication forks in the mutant strains, as predicted from the roles of Swi1 and Swi3 in replication fork pausing at mat1 (53). One possible explanation (of many) for the smear signal is that endonuclease scission of the single-stranded regions at variable positions would generate a family of altered replication intermediates (Fig. 8D and E). The faint diffuse linear signals that emanate from the 1N spots in the mutant strains (open arrows) might be a consequence of preferred pause regions, which would generate enhanced signals within the overall smear pattern (Fig. 8D and E).

Cells with swi1 or swi3 mutations have increased ssDNA levels during S phase. To test the possibility that the swi1 and swi3 mutations would lead to increased ssDNA at replication forks, we measured the levels of ssDNA during S phase (Fig. 9). We tested for the presence of single-stranded regions within rRNA gene restriction fragments. For this purpose, we used a method that is based on the fact that ssDNA can bind much more efficiently to a hydrophobic membrane than double-stranded DNA can (see Materials and Methods) (57). To increase the signal-to-noise ratio, the DNA was restricted with a frequently cutting enzyme that generates blunt ends. Subsequently, the DNA was transferred, either directly or after separation on an agarose gel, to a membrane under both native and denaturing conditions. The ratio of the signals obtained indicated the relative number of fragments that contained single-stranded regions. Significantly more ssDNA-containing fragments were detected in the samples from swi1 and swi3 cells than in those from wild-type cells, when DNA was analyzed from cultures progressing through S phase (Fig. 9). Elevated levels of ssDNA were present in the samples from wild-type cells at only one time point after release from the G1 block, presumably the time at which the target sequence was replicated. In contrast, every time point from swi1 and swi3 cells displayed increased ssDNA levels. Thus, our observations are consistent with the hypothesis that replication in the swi1 and swi3 strains is accompanied by increased ssDNA levels, presumably at stalled forks, and that treating cells with HU emphasizes this by increasing the proportion of stalled forks.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 9. Increased levels of ssDNA in swi1 and swi3 strains during S phase. (A) Single-mutant cdc10 (ES366) and double-mutant swi1 cdc10 (ES281) cells were blocked in G1 by a temperature shift to 37°C for 4.5 h. After the shift back to 25°C, the cultures progressed synchronously through the cell cycle. Samples were taken at the indicated time points, and DNA was extracted as for two-dimensional gels. DNA samples were digested with EcoRV. Half of each sample was denatured in NaOH. The native and denatured samples were slot blotted onto a membrane under native or denaturing conditions, respectively. The filter was probed with a 32P-labeled HindIII-KpnI RNA fragment (Fig. 8B) and quantified using a phosphorimager. The ratios of the signals obtained from native DNA relative to denatured DNA, normalized to the ratio for wild-type cells in G1 phase, are given. The experiment was done in duplicate, and the average and range are reported in the graph. (B) In an independent experiment, single-mutant cdc10 (ES366) and double-mutant swi1 cdc10 (ES281) and swi3 cdc10 (ES277) cells were synchronized in G1 and released into S phase as above. DNA was extracted as for two-dimensional gels, digested with AluI, and separated on a 0.7% agarose gel in duplicate. Half of the samples were transferred by Southern blotting to a membrane under native conditions, and half were transferred under denaturing conditions. The filters were probed with a 32P-labeled HaeII-BamHI probe to the ribosomal repeats. The signals were quantified, and the ratio was calculated as above. For comparison, the level of ssDNA in native DNA from wild-type (wt) cells in G1 is highlighted by a lighter gray box.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Repair of alkylation damage involves multiple pathways, which act both during the S and G2 phases. Alkylation of the bases can occur when the DNA is single stranded or when it is double stranded. When MMS reacts with double-stranded DNA, methylation preferentially occurs at the nitrogen at position 7 of guanine and the nitrogen at position 3 of adenine. More infrequently, bases can be modified at other positions, including at the oxygen at position 6 and the nitrogen at position 3 of guanine, at the nitrogen at position 7 of adenine, and at the oxygen at position 7 of thymine (3). Such damaged bases are repaired by both base and nucleotide excision repair pathways (reviewed in reference 32). In contrast, when MMS reacts with ssDNA it generates mainly 1-methyladenine and 3-methylcytosine. These modifications can be repaired directly via an oxidative demethylation (52). Importantly, due to the changes in molecular structure, some alkylation-damaged bases, including 3-methyladenine and 3-methylguanine, act as barriers for DNA replication (13, 45). Thus, alkylation-specific repair and checkpoint processes are likely to be present to allow the cells to recover from this type of event in order to ensure maintenance of replication fork integrity, checkpoint activation, and recruitment of repair enzymes. Here we identify a novel pathway, involving Swi1, Swi3, and Hsk1-Dpf1, that has S-phase-specific functions in response to MMS. This pathway maintains replication fork integrity and slows S-phase progression as part of the intra-S-phase checkpoint response to the DNA damage.

The swi1 and swi3 genes were originally identified in a screen for mutants that affect S. pombe mating-type switching (12). Recently, we have shown that at the mat1 replication pause site, MPS1, Swi1 and Swi3 mediate a leading-strand replication pause, from a signal read during lagging-strand synthesis (53). Thus, at MPS1, Swi1 and Swi3 coordinate the two replication processes in such a way that the integrity of the replication process is maintained. There are several observations indicating that, in addition to their roles at mat1, Swi1 and Swi3 might act during general replication. First, Swi1 is similar to S. cerevisiae Tof1 (46), which is implicated in the Mec1p-, Rad53p-dependent intra-S-phase checkpoint (14). Tof1 colocalizes with the replication machinery and coimmunoprecipitates, preferentially during S phase, with the replication protein Cdc45 (19); it also acts, with Mrc1p, to anchor replication proteins to sites of HU-stalled replication forks (19). Second, both swi1 and swi3 mutant strains are sensitive to camptothecin (10), a drug known to cause cytoxicity in S phase by inhibiting topoisomerase 1 (Top1). The top1 swi1 and top1 swi3 double mutants display a slow-growth phenotype suggestive of genetic interaction (10). Finally, Swi1 and Swi3 function in S-phase checkpoint activation upstream of Cds1 when cells are exposed to HU for prolonged periods, and Swi1 and Swi3 appear to move along with the replication fork (37, 38).

We set out to test if Swi1 and Swi3 play general roles at replication barriers in S phase such as alkylation-damaged bases introduced using MMS. Our results demonstrate involvement of these proteins in survival of (Fig. 1A and 6A, B, and F) and the mitotic checkpoint response to (Fig. 2 to 4 and 6C) MMS damage, with functions that are nonoverlapping with those of both Cds1 and Chk1. Additional results (Fig. 6 and 8C) suggest that Swi1 and Swi3 act in the same pathway as the Hsk1-Dfp1 kinase complex. Both hsk1 and dfp1 belong to the same functional group as swi1 and swi3 with regard to MMS sensitivity (Fig. 6A and F). In addition, the swi1, swi3, and hsk1 mutants resemble each other with regard to competence for the mitotic checkpoint (Fig. 6C) and lack of the intra-S-phase checkpoint (Fig. 6D and E). In combination, our results suggest that Swi1p, Swi3p, and Hsk1p cooperate with each other in a novel S-phase damage response pathway that is at least partially independent of the pathways mediated by Cds1 and Chk1 and probably parallel to them. The data also clearly show that this novel pathway is of minor importance compared to the Chk1 pathway for the induction of the mitotic cell cycle arrest that occurs in response to MMS damage.

The importance of this new S-phase DNA damage response pathway and the genomic instability of the swi1 and swi3 strains suggest a role for these proteins in controlling replication progression in the presence of replication fork impediments. Our data suggest that Swi1, Swi3, and Hsk1 act in concert to detect stalled replication forks, to induce slowing of S phase, and perhaps to maintain coordination of leading- and lagging-strand replication as observed at mat1 (53). Indeed, our analysis of the role of the swi1 and swi3 mutants in pausing replication at mat1 (53), as well as their genetic interaction with DNA polymerase alpha (11), suggest that this novel pathway might specifically act during lagging-strand replication. If these genes were involved in coordination of leading- and lagging-strand replication, mutant strains would be predicted to uncouple the two replication processes, resulting in uncoordinated polymerase progression. This would lead to drastic consequences, similar to those observed here: (i) increased amounts of ssDNA during S phase (Fig. 9); (ii) activation of Chk1p in swi1, swi3 (Fig. 1C), and hsk1 (47) strains in the absence of exogenously induced DNA damage; (iii) increased frequency of mitotic recombination (Fig. 7C, D and 8C); and (iv) instability of stalled forks in HU (Fig. 8C).

Such an uncoupling of leading- and lagging-strand replication has been observed in bacteria, when replication forks encounter lesions (41). Higuchi et al. reported that an abasic lesion on the lagging strand blocked the production of Okazaki fragments but did not affect the progress of leading-strand replication (16). It is possible that eukaryotes developed a system to avoid such uncoupling in response to replication block (in our case due to alkylated bases), and our data point to roles for Swi1 and Swi3 in such a system.

The data presented here suggest a model for the role of Hsk1 in the S-phase damage response. Since swi1 and swi3 are in the same functional group as hsk1, and since both dfp1 and hsk1 mutant cells display genomic instability (15, 47), we propose on the basis of the function for Swi1 and Swi3 at the mat1 locus, that a Swi1- and Swi3-dependent signal for the presence of damage on the lagging strand could be sensed or transduced by the Hsk1 kinase, to pause the leading-strand replication machinery. If Hsk1 and Cds1 played roles in parallel pathways during S phase, one would expect that cross talk could occur between the two kinases. This would explain the observations that Hsk1 acts as a substrate for Cds1 in vitro and in vivo whereas Hsk1 acts upstream of Cds1 at HU-stalled forks (47, 49). Such cross talk between Hsk1 and Cds1 kinases could also explain the decrease in Cds1 activation observed by Noguchi et al. in the swi1 and swi3 background when cells are exposed to HU for prolonged periods (37, 38). Indeed, we observed that both the Swi1-, Swi3-, Hsk1-dependent pathway and the Cds1 pathway are required for slowing of the S phase in response to the DNA damage. The proposed roles for S. pombe Hsk1p and Cds1p proteins would fit well with the model by Tercero et al. (50), in which replication forks are both activators and primary effectors of the S. cerevisiae S-phase checkpoint. A role for checkpoint proteins in coordinating replication is also consistent with the effects of mrc1 and tof1 mutations in budding yeast (19). These mutations uncouple the localization of Cdc45, Pol1, Pol3, Dpb3, Rfa1, Cdc47, and Cdc54 from the position of DNA synthesis, when replication forks are stalled in HU.

Homologues of Swi1, Swi3, Dfp1, and Hsk1 exist in the genomes of organisms from yeasts to metazoans. Mutations in the S. cerevisiae homologues of swi1 and swi3, TOF1 and CSM3, respectively, and also mutations in the Caenorhabditis elegans swi1 homologue, CeTim1, all affect sister chromatid cohesion (8, 31), which is thought to be established during replication (24, 36). Similarly, mutations in S. pombe hsk1 also affect sister chromatid cohesion (2). In S. cerevisiae, Tof1 and Csm3 play a role in the S-phase DNA damage response (14, 51). Thus, the novel fission yeast S-phase checkpoint and alkylation damage response pathway described here is likely to be conserved.

In support of the proposed model, recent data (Karim Labib, personal communication) suggest that Tof1 and Csm3 are in a complex with the MCM proteins (the presumed replicative helicase), the GINS complex, and polymerase {alpha}. Importantly, some of the MCM proteins are downstream targets of the Hsk1p kinase (23), underlining the connections between these proteins.


   ACKNOWLEDGMENTS
 
Special thanks are due to Avelino Bueno, Tony Carr, Susan Forsburg, Amar Klar, Hisao Masai, Fekret Osman, Tadayuki Takeda, and Nancy Walworth for providing strains and to Natalie Mansfield and Douglas Drummond for technical assistance. We thank our colleagues at the Marie Curie Research Institute for helpful discussions and suggestions.

Research in the Dalgaard laboratory was sponsored by the Association of International Cancer Research and the Marie Curie Cancer Care, and research in the Huberman laboratory was sponsored by grants (GM49294, CA84302, and CA95908) from the National Institutes of Health. The Flow Cytometry facility of Roswell Park Cancer Institute is supported by grant P30CA16056 from the National Institutes of Health.


   FOOTNOTES
 
* Corresponding author. Mailing address: Marie Curie Research Institute, The Chart, Oxted, Surrey RH8 0TL, United Kingdom. Phone: 44 (0) 1883-722-306. Fax: 44 (0) 1883-714-375. E-mail: j.dalgaard{at}mcri.ac.uk. Back

{dagger} Supplemental material for this article may be found at http://mcb.asm.org/. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Al-Khodairy, F., E. Fotou, K. S. Sheldrick, D. J. Griffiths, A. R. Lehmann, and A. M. Carr. 1994. Identification and characterization of new elements involved in checkpoint and feedback controls in fission yeast. Mol. Biol. Cell 5:147-160.[Abstract]

2. Bailis, J. M., P. Bernard, R. Antonelli, R. C. Allshire, and S. L. Forsburg. 2003. Hsk1-Dfp1 is required for heterochromatin-mediated cohesion at centromeres. Nat. Cell Biol. 5:1111-1116.[CrossRef][Medline]

3. Beranek, D. T. 1990. Distribution of methyl and ethyl adducts following alkylation with mono functional alkylating agents. Mutat. Res. 231:11-30.[Medline]

4. Brewer, B. J., and W. L. Fangman. 1987. The localization of replication origins on ARS plasmids in S. cerevisiae. Cell 51:463-471.[CrossRef][Medline]

5. Brondello, J. M., M. N. Boddy, B. Furnari, and P. Russell. 1999. Basis for the checkpoint signal specificity that regulates Chk1 and Cds1 protein kinases. Mol. Cell. Biol. 19:4262-4269.[Abstract/Free Full Text]

6. Brown, G. W., and T. J. Kelly. 1999. Cell cycle regulation of Dfp1, an activator of the Hsk1 protein kinase. Proc. Natl. Acad. Sci. USA 96:8443-8448.[Abstract/Free Full Text]

7. Capasso, H., C. Palermo, S. Wan, H. Rao, U. P. John, M. J. O'Connell, and N. C. Walworth. 2002. Phosphorylation activates Chk1 and is required for checkpoint-mediated cell cycle arrest. J. Cell Sci. 115:4555-4564.[Abstract/Free Full Text]

8. Chan, R. C., A. Chan, M. Jeon, T. F. Wu, D. Pasqualone, A. E. Rougvie, and B. J. Meyer. 2003. Chromosome cohesion is regulated by a clock gene paralogue TIM-1. Nature 424:1002-1009.[CrossRef][Medline]

9. Dalgaard, J. Z., and A. J. Klar. 1999. Orientation of DNA replication establishes mating-type switching pattern in S. pombe. Nature 400:181-184.[CrossRef][Medline]

10. Dalgaard, J. Z., and A. J. Klar. 2000. Swi1 and Swi3 perform imprinting, pausing, and termination of DNA replication in S. pombe. Cell 102:745-751.[CrossRef][Medline]

11. Egel, R., D. H. Beach, and A. J. Klar. 1984. Genes required for initiation and resolution steps of mating-type switching in fission yeast. Proc. Natl. Acad. Sci. USA 81:3481-3485.[Abstract/Free Full Text]

12. Egel, R., and B. Eie. 1987. Cell lineage asymmetry for Schizosaccharomyces pombe: unilateral transmission of a high-frequency state of mating-type switching of a homothallic fission yeast. Curr. Genet. 12:429-433.[CrossRef]

13. Evensen, G., and E. Seeberg. 1982. Adaptation to alkylation resistance involves the induction of a DNA glycosylase. Nature 296:773-775.[CrossRef][Medline]

14. Foss, E. J. 2001. Tof1p regulates DNA damage responses during S phase in Saccharomyces cerevisiae. Genetics 157:567-577.[Abstract/Free Full Text]

15. Fung, A. D., J. Ou, S. Bueler, and G. W. Brown. 2002. A conserved domain of Schizosaccharomyces pombe dfp1(+) is uniquely required for chromosome stability following alkylation damage during S phase. Mol. Cell. Biol. 22:4477-4490.[Abstract/Free Full Text]

16. Higuchi, K., T. Katayama, S. Iwai, M. Hidaka, T. Horiuchi, and H. Maki. 2003. Fate of DNA replication fork encountering a single DNA lesion during oriC plasmid DNA replication in vitro. Genes Cells 8:437-449.[Abstract]

17. Hirano, T., S. Funahashi, T. Uemura, and M. Yanagida. 1986. Isolation and characterization of Schizosaccharomyces pombe cut mutants that block nuclear division but not cytokinesis. EMBO J. 5:2973-2979.[Medline]

18. Huberman, J. A., L. D. Spotila, K. A. Nawotka, S. M. el-Assouli, and L. R. Davis. 1987. The in vivo replication origin of the yeast 2 microns plasmid. Cell 51:473-481.[CrossRef][Medline]

19. Katou, Y., Y. Kanoh, M. Bando, H. Noguchi, H. Tanaka, T. Ashikari, K. Sugimoto, and K. Shirahige. 2003. S-phase checkpoint proteins Tof1 and Mrc1 form a stable replication-pausing complex. Nature 424:1078-1083.[CrossRef][Medline]

20. Kelly, T. J., G. S. Martin, S. L. Forsburg, R. J. Stephen, A. Russo, and P. Nurse. 1993. The fission yeast cdc18+ gene product couples S phase to START and mitosis. Cell 74:371-382.[CrossRef][Medline]

21. Kumar, S., and J. A. Huberman. 2004. On the slowing of S phase in response to DNA damage in fission yeast. J. Biol. Chem. 279:43574-43580.[Abstract/Free Full Text]

22. Lee, B. S., S. I. Grewal, and A. J. Klar. 2004. Biochemical interactions between proteins and mat1 cis-acting sequences required for imprinting in fission yeast. Mol. Cell. Biol. 24:9813-9822.[Abstract/Free Full Text]

23. Lee, J. K., Y. S. Seo, and J. Hurwitz. 2003. The Cdc23 (Mcm10) protein is required for the phosphorylation of minichromosome maintenance complex by the Dfp1-Hsk1 kinase. Proc. Natl. Acad. Sci. USA 100:2334-2339.[Abstract/Free Full Text]

24. Lee, J. Y., and T. L. Orr-Weaver. 2001. The molecular basis of sister-chromatid cohesion. Annu. Rev. Cell Dev. Biol. 17:753-777.[CrossRef][Medline]

25. Liang, D. T., J. A. Hodson, and S. L. Forsburg. 1999. Reduced dosage of a single fission yeast MCM protein causes genetic instability and S phase delay. J. Cell Sci. 112:559-567.[Abstract]

26. Lindsay, H. D., D. J. Griffiths, R. J. Edwards, P. U. Christensen, J. M. Murray, F. Osman, N. Walworth, and A. M. Carr. 1998. S-phase-specific activation of Cds1 kinase defines a subpathway of the checkpoint response in Schizosaccharomyces pombe. Genes Dev. 12:382-395.[Abstract/Free Full Text]

27. Luria, S. E., and M. Delbrück. 1943. Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28:491-511.[Free Full Text]

28. Marchetti, M. A., S. Kumar, E. Hartsuiker, M. Maftahi, A. M. Carr, G. A. Freyer, W. C. Burhans, and J. A. Huberman. 2002. A single unbranched S- phase DNA damage and replication fork blockage checkpoint pathway. Proc. Natl. Acad. Sci. USA 99:7472-7477.[Abstract/Free Full Text]

29. Maleszka, R., and G. D. Clark-Walker. 1993. Yeasts have a four-fold variation in ribosomal DNA copy number. Yeast 9:53-58.[CrossRef][Medline]

30. Martinho R. G., H. D. Lindsay, G. Flaggs, A. J. DeMaggio, M. F. Hoekstra, A. M. Carr, and N. J. Bentley. 1998. Analysis of Rad3 and Chk1 protein kinases defines different c