|Home | About | Journals | Submit | Contact Us | Français|
Rad53 of Saccharomyces cerevisiae is a checkpoint kinase whose structure and function are conserved among eukaryotes. When a cell detects damaged DNA, Rad53 activity is dramatically increased, which ultimately leads to changes in DNA replication, repair, and cell division. Despite its central role in checkpoint signaling, little is known about Rad53 substrates or substrate specificity. A number of proteins are implicated as Rad53 substrates; however, the evidence remains indirect. Previously, we have provided evidence that Swi6, a subunit of the Swi4/Swi6 late-G1-specific transcriptional activator, is a substrate of Rad53 in the G1/S DNA damage checkpoint. In the present study we identify Rad53 phosphorylation sites in Swi6 in vitro and demonstrate that at least one of them is targeted by Rad53 in vivo. Mutations in these phosphorylation sites in Swi6 shorten but do not eliminate the Rad53-dependent delay of the G1-to-S transition after DNA damage. We derive a consensus for Rad53 site preference at positions −2 and +2 (−2/+2) and identify its potential substrates in the yeast proteome. Finally, we present evidence that one of these candidates, the cohesin complex subunit Scc1 undergoes DNA damage-dependent phosphorylation, which is in part dependent on Rad53.
Checkpoints or surveillance mechanisms of eukaryotic cells ensure that critical stages of the cell cycle are initiated and completed in the proper order. These mechanisms become critical for survival when DNA is damaged and must be repaired before chromosome segregation lest genetic information is lost in cell division. A checkpoint is comprised of damage sensors and transducers, which convey a regulatory signal to the cell cycle and repair machinery (34, 64). In yeast cells, as in higher eukaryotes, there are three recognized DNA damage checkpoints (17). When DNA is damaged, cells can delay the G1-to-S transition, decelerate S phase progression, and pause before mitosis. Operation of the checkpoint involves activation and deactivation of the checkpoint kinases Mec1, Rad53, Chk1, and Dun1 (3, 42, 44, 45, 56). The key regulators, Rad53 and Mec1 kinases, are involved in all three of the checkpoints via partially overlapping circuits (17). While the phosphorylation preference of Mec1 is known (31), Rad53 specificity remains undetermined. Also, the targets of the Rad53 kinase in budding yeast are not well established. A number of proteins are reported to undergo phosphorylation in a Rad53- and DNA damage-dependent manner in vivo; however, it remains to be established whether these proteins are direct substrates of Rad53 and, if so, what specific residues are phosphorylated (64).
The sensitivity of the three DNA damage checkpoints to the levels of DNA damage appears to be quite different. For example, the same dose of gamma rays leads to a higher level of activation of Rad53 in G2 cells than in G1 cells (21). One double-strand break can hold cells at the G2/M boundary for many hours (46), whereas G1 cells do not activate the checkpoint in response to a single double-strand break (42).
G1 progression includes two waves of G1 cyclin (Cln) transcription, where CLN3 transcription peaks at the end of M and in early G1 and CLN1 and CLN2 mRNA levels peak in late G1 (37). Cln3/Cdk is an unstable activator of the Swi4/Swi6 transcription factor complex via a yet-unidentified mechanism (9, 55). When active, this complex brings about a rapid burst of transcription of dozens of genes (27), including CLN1 and CLN2, which enables budding and the G1-to-S transition (38).
Methyl methanesulfonate (MMS) treatment and UV and gamma irradiation during G1 can delay both budding and the onset of DNA replication. The treated cells remain sensitive to the mating pheromone, alpha factor (53, 54), and have low levels of CLN mRNA (51) and low Cln/Cdk activity (16). CLN1 and CLN2 transcription is downregulated upon addition of MMS (51). The rate of recovery of CLN1 and CLN2 mRNAs, which determines the timing of the G1-to-S transition, depends on at least one of the central checkpoint kinases, Rad53, and on the Swi4/Swi6 transcription factor complex, which activates CLN1 and CLN2 transcription. Moreover, Swi6 undergoes a DNA damage-inducible, Rad53-dependent phosphorylation in vivo and can be phosphorylated by Rad53 immunoprecipitates in vitro (51).
In the present study, we demonstrate that a short pulse of MMS administered to elutriated G1 cells delays the G1-to-S transition. MMS-treated G1 cells remain in G1 for a longer time and thus grow substantially larger than untreated cells before they transit into S phase. This delay of S phase is substantially reduced in rad53-11 checkpoint mutant cells. G1 cells lacking Swi6 also have a shorter delay of the G1-to-S transition after MMS treatment. Swi6 is phosphorylated by Rad53 in vitro, and we have identified five sites of this phosphorylation. At least one of these sites in Swi6 is phosphorylated by Rad53 in response to DNA damage in vivo. Alignment of the identified sites reveals conservation of flanking sequences, which we show to be critical for phosphorylation of Swi6 by Rad53 in vitro. This enables us to derive a consensus site for Rad53 phosphorylation and to implicate a number of other proteins as Rad53 targets. One of these new candidate proteins, the cohesin complex subunit Scc1, was tested and shown to undergo damage-induced phosphorylation in a Rad53-dependent manner. In cells with the Rad53 phosphorylation sites eliminated from Swi6, the checkpoint response to MMS in G1 is reduced but not abolished. This may be explained by the large number of other potential Rad53 substrates in G1 cells.
The yeast strains BY2006 (MATa ura3 leu2 trp1 his3), BY2007 (MATa ura3 leu2 trp1 his3 rad53-11::URA3), and BY2390 (MATa ura3 leu2 trp1 his3 rad53-11) have been described elsewhere (50, 51). BY2917 and BY2981 are swi6::LEU2 and swi6::URA3 derivatives of BY2006. BY3256 and BY3257 are RAD53-HA-TRP1 and rad53-11-HA-TRP1 derivatives of BY2006 and BY2390, respectively, with Rad53 tagged with a hemagglutinin (HA) epitope tag (13). BY3258 is a swi6::LEU2 derivative of BY2390. The genotype of BY2918 is MATα ura3 leu2 trp1 his3 rad53-11::LEU2. SBY376 (MATa ura3 leu2 trp1::LacO his::pCUP1-GFP12-LacI12 can1 ade2 bar1 MCD1-HA3) (4) is a kind gift of S. Biggins. BY3242 and BY3243 are RAD+ and rad53-11 spores from the cross of strains SBY376 and BY2918. The genotype of BY479 is MATa dbf4-1 ura3 trp1 ade5.
The plasmids pBD1378 and pBD1265 are pRS316 and pZUC12 vectors, respectively, with the SWI6 gene promoter and open reading frame (ORF) (52). All mutations were introduced into the plasmid-borne SWI6 by using site-directed mutagenesis (32). The mutant constructs are listed in Table Table1.1. Wild-type and alanine mutant constructs were expressed in swi6Δ strain BY2917 or BY2981. To generate glutathione S-transferase (GST)-Swi6 fusions, a 2.0-kb EcoRI fragment containing the SWI6 ORF encoding amino acids 1 to 573 was cloned into pGEX2T-1 (Pharmacia, Piscataway, N.J.), yielding pBD1998. A 1.0-kb EcoRI fragment containing SWI6 ORF codons for amino acids 573 to 803 was cloned into pGEX3X-1 (Pharmacia), yielding pBD2685. All mutant SWI6 genes were cloned into these plasmids, and the resulting constructs are listed in Table Table1.1. Plasmid YQ118 encoding HA-tagged Dbf4 is a kind gift of R. Sclafani.
All rich (YEP) and minimal (YC) media, growth conditions, and alpha factor synchronizations were as described before (6). Cultures used for elutriation were grown in YC-Ura glucose media and then inoculated into YEPD media and grown to an optical density at 600 nm of 2.5 to 3. Elutriation and cell volume measurements were performed essentially as described earlier (50) in YEPD. For MMS treatments, elutriated cultures were split at time point 0, and MMS was added to one-half of these cultures. MMS was removed by filtration, and cultures were resuspended in YEPD and allowed to grow. Samples were taken every 10 or 15 min, and cell size and the percentage of budded cells were determined. The [32P]orthophosphate labeling conditions were as described before (52).
Extraction of the labeled Swi6 protein from yeast was done as described previously (52). The full-length untagged recombinant Swi6 was described previously (51). GST fusions of Swi6 were purified out of 50-ml cultures of the BL21(DE3) Escherichia coli strain. Expression of the GST-Swi6 fusions was induced with 100 mM IPTG (isopropyl-β-d-thiogalactopyranoside) for 1 h. Bacteria were lysed by sonication in phosphate-buffered saline (PBS) supplemented with 0.1% Triton X-100, 10 mM phenylmethylsulfonyl fluoride, 1 μg of leupeptin/ml, and 1 μg of pepstatin A/ml. Precipitation with glutathione beads and elution of the precipitated proteins with PBS-glutathione solution were done according to the Pharmacia Biotech GST gene fusion system protocol. Fusions were eluted in a volume of 30 μl and adjusted to 1× HEPES kinase buffer (see below) and 10% glycerol by using 10× stock solutions. Then, 2 to 4 μl of these eluates was used in a kinase assay with Rad53.
To isolate HA-tagged Scc1, protein extracts from unlabeled or [32P]orthophosphate-labeled cells were made in the lysis buffer [50 mM Tris HCl (pH 8.0), 0.3 M (NH4)2SO4, 5% glycerol, 0.1% NP-40, 0.05% deoxycholate, 1 mM dithiothreitol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 μg of pepstatin A/ml, 1 μg of leupeptin/ml, 20 mM β-glycerol phosphate, 20 mM NaF, 1 mM sodium orthovanadate]. These extracts were either loaded onto 7 or 8% (59:1 cross-link ratio) sodium dodecyl sulfate (SDS)-polyacrylamide gels or immunoprecipitated with 12CA5 antibodies. Immunoprecipitates were washed three times with lysis buffer, five times with the PBS containing 1% Triton X-100 and 10% glycerol (PBS-Triton-glycerol), and three times with PBS-Triton-glycerol containing 0.8 M NaCl. Immunoprecipitates were resolved on SDS-polyacrylamide gels and autoradiographed. Labeled Scc1 bands were excised and processed as described for Swi6 (52). Western blotting with 12CA5 antibodies was performed as described earlier for Swi6 (49) with horseradish peroxidase-conjugated goat anti-mouse secondary antibodies (Gibco-BRL, Rockville, Md.).
Immunoprecipitation of untagged or HA-tagged Rad53 out of yeast cell extracts was done as described previously (51) with rabbit polyclonal antibody to Rad53 (a gift of S. J. Elledge) or 12CA5 mouse monoclonal antibodies to the HA epitope. Baculovirus-expressed, HA-tagged Rad53 purified from insect cells was provided by S. J. Elledge. Kinase assays were performed with 50 to 100 ng of recombinant full-length (51) or GST-tagged and truncated Swi6 (see above) by using ca. 1 to 5 ng of baculovirus-expressed HA-Rad53 or Rad53 immunoprecipitated out of ca. 1 to 5 × 107 cells treated with either 4 μg of 4-nitroquinoline-N-oxide (4NQO)/ml or 0.1% MMS for 40 min. Reactions were carried out in HEPES kinase buffer (20 mM HEPES-NaOH [pH 7.5], 10 mM MgCl2, 10 mM MnCl2) for 20 to 30 min at 29°C. Products were resolved by SDS-polyacrylamide gel electrophoresis (PAGE) and autoradiographed. Phosphorylated Swi6 was then excised out of dried SDS-PAGE gels and processed as described below.
Mapping and phosphoamino acid analysis were done as described previously (5) and as described for Swi6 (52). Chymotryptic digestion of Swi6 was conducted with 1 mg of chymotrypsin (Sigma, St. Louis, Mo.)/ml. Phosphopeptide maps of Scc1 digested with trypsin (Sigma) were run in the first dimension at pH 1.9 and in the second dimension in isobutyric buffer (5).
The searches were performed by using the Pattern Matching tool of the Saccharomyces cerevisiae database at Stanford (SGD). The output list of S. cerevisiae ORFs is available online (www.fhcrc.org/labs/breeden/Rad53). The most stringent consensus suggested by the analysis of Swi6 includes basic residues in the −3 position and L, M, V, or I at positions −2 and +2 and excludes proline from the +1 position. This site, [RK][LMVI]x[ST][noP][LMVI], is encountered at least once in 1,958 yeast ORFs, so we list only the 519 that contain two or more such sites.
We used elutriation to study cell cycle progression in small G1 cells after DNA damage. Typically, these elutriated cells grow at a specific rate and execute a G1-to-S transition when they reach the critical volume predetermined by medium conditions and cell's genotype (43). A pulse of MMS reduced the rate of volume enlargement of G1 cells as measured by plotting their mean volume increase over time (Fig. (Fig.1A).1A). This growth retardation was dose dependent (data not shown), and it may result from a reduction of biosynthetic capacity, since transcription of many of the ribosomal protein genes is downregulated by this agent (28).
As a consequence of the reduced rate of growth caused by MMS treatment, treated cells required more time to attain the same volume as untreated controls (Fig. 1A and D). However, MMS-treated wild-type cells also remained in G1 beyond the volume that was sufficient for budding in the untreated controls and grew up to 5 fl larger before they budded (Fig. (Fig.1B1B and and5C)5C) and initiated DNA replication (data not shown). The rad53-11 mutation shortened the damage-induced delay of budding but did not eliminate it. rad53-11 cells treated with MMS initiated budding sooner than MMS-treated wild-type cells and did not grow as large as the wild type before budding (Fig. 1B and D), but they exhibited the same degree of growth retardation caused by MMS (Fig. (Fig.1A).1A). In both types of cells, wild type and rad53-11, the response to MMS was the same whether MMS was added to small, early-G1 cells with the 18-fl mean volume (Fig. (Fig.1A)1A) or to larger, mid-G1 cells with a 22-fl mean volume (data not shown).
Since Swi6 is a potential target of Rad53 phosphorylation and is necessary for the prolonged repression of CLN1 and CLN2 transcription after MMS (51), we sought to determine whether the MMS-induced delay of the G1-to-S transition in elutriated cells required Swi6. The swi6Δ cells are large at birth compared to the wild type (30 to 35 fl versus 18 to 22 fl, Fig. 1A and C). However, they undergo the G1-to-S transition coordinately, as is evident from the relatively steep slope of the budding curves in Fig. Fig.1C.1C. Importantly, pulse treatment with MMS caused a reduced delay of the G1-to-S transition in swi6Δ cells. Similar to rad53-11 cells, the MMS-treated swi6Δ cells started to bud when they were only 2 to 3 fl larger than untreated controls (Fig. (Fig.1C;1C; see also Fig. Fig.5C5C).
Finally, the MMS-treated swi6Δ rad53-11 double mutant cells budded at the same or an even smaller volume than the untreated controls (Fig. (Fig.1C).1C). The growth retardation caused by MMS in the double mutant was retained (Fig. (Fig.1A);1A); however, there was no delay of budding. In fact, 50% budding was attained at a smaller volume and 10 min earlier in the treated swi6Δ rad53-11 culture than in the untreated control. This suggests that rad53-11 swi6Δ cells have lost the ability to delay in G1 in response to DNA damage. This synthetic phenotype of the swi6Δ rad53-11 strain may arise if the rad53-11 allele retains residual checkpoint activity, which is further compromised by the absence of Swi6 (e.g., RAD53 transcription may be Swi6 dependent ). It is also possible that there is a Rad53-independent branch of the checkpoint that operates through Swi6. Either way, these data indicate that both Rad53 and Swi6 contribute to the proper execution of the G1-to-S transition after DNA damage.
Swi6 undergoes DNA damage-inducible, Rad53-dependent phosphorylation in vivo (51). In addition, Swi6 can be phosphorylated in vitro with the immunoprecipitates of activated Rad53 isolated from cells which received DNA damage or with immunoprecipitates of Rad53 overproduced in undamaged yeast cells (51). To detemine whether Swi6 is a direct substrate of Rad53, we performed kinase reactions with purified recombinant Rad53 and Swi6 and compared phosphopeptide maps of Swi6 phosphorylated by recombinant Rad53 to the maps obtained by using immunoprecipitated wild-type and checkpoint-deficient Rad53.As seen in Fig. Fig.2C,2C, wild-type, damage-activated Rad53 from yeast cells phosphorylates Swi6 on a number of sites (1, 1a, 2, 3, 7, V, and VII). Neither mutant Rad53 from damaged cells (Fig. (Fig.2A),2A), nor the wild-type Rad53 isolated from cells with no DNA damage (Fig. (Fig.2B)2B) was capable of efficiently phosphorylating Swi6 on these sites. However, recombinant Rad53 provided in excess in vitro gave rise to a phosphorylation pattern on Swi6 that was qualitatively identical to the one generated by the damage-activated Rad53 from yeast cells (Fig. (Fig.2D),2D), indicating that the damage-induced phosphorylations could be directly attributed to Rad53 activity. The phosphorylated peptides were subjected to phosphoamino acid analysis, and only phosphoserine and phosphothreonine residues were detected (data not shown).
Swi6 is a 91-kDa protein with 104 serine and threonine residues. In order to map the sites of phosphorylation of Swi6 by Rad53, we used deletion derivatives of Swi6 purified as GST-tagged proteins. The tag did not affect the phosphorylation pattern (data not shown). Figure Figure33 shows a comparison of the phosphopeptide maps of the full-length untagged Swi6 and its two deletion derivatives, aa1-573 and aa573-803, that were phosphorylated by recombinant Rad53. The map positions of these Rad53-generated tryptic phosphopeptides were compared to the predicted positions of the tryptic phosphopeptides of Swi6 (5). Serine and/or threonine residues in the peptides that closely matched the observed migration patterns were mutated, and mutant substrates were purified and tested in a kinase reaction with the recombinant Rad53.
Figure Figure33 summarizes the results of this analysis. We identified five sites targeted by Rad53 in Swi6, which account for the major phophopeptides 1, 1a, 2, 3, 7, V, and VII. Peptides 1 and 1a were produced by the same sequence, [KKLKIDT(169)SVIDAE…] (Fig. (Fig.3B).3B). Both serine and threonine residues can be phosphorylated in this sequence since the phosphoamino acid composition of this peptide is a mixture of both phosphoserine and phosphothreonine (data not shown). However, the mobility of these peptides suggests that they carry only one phosphate, either on T169 or on S170.
Another major phosphorylation site, [KAKKIRS(547)QLLKN…], accounts for at least two peptides, 3 and 7, where peptide 7 has a sequence of S(547)QLLK, and peptide 3 is most likely an incompletely digested peptide with the sequence S(547)QLLKNPPET(556)TSLINDVQNLLNS (Fig. (Fig.3C).3C). Within this sequence, S547 remains a predominant site of phosphorylation and/or a primary recognition site by Rad53, since its substitution to aspartic acid eliminates both peptides 3 and 7 (Fig. (Fig.3C).3C). However, in a fraction of molecules, the threonines 556 and 557 may be phosphorylated instead of S547, since peptide 3 contains both phosphothreonine and phosphoserine (Fig. (Fig.4A).4A). Accordingly, substitution of threonines 556 and 557 to alanines does not abolish the phosphorylation of peptides 3 and 7, but it does eliminate threonine as a phosphoacceptor in peptide 3 (Fig. (Fig.4A4A).
The site […QEMPT(267)SLNN…] maps to peptide 2, where mainly phosphothreonine is detected (Fig. (Fig.3D3D and data not shown). Peptide V has the sequence IQS(715)MLPPTVLLK, with S715 as the phosphorylatable residue (Fig. (Fig.3G),3G), and peptide VII [HLTNVLDT(743)ISTK] has T743 as the phosphorylatable residue (Fig. (Fig.3H3H).
Alignment of these Rad53 phosphorylation sites reveals a strong preference for hydrophobic amino acids with long aliphatic side chains, I, L, V, or M at positions +2 and −2, where 0 and +1 are, respectively, the first and the second phosphorylatable residues (Fig. (Fig.4B).4B). Also, a preference for a basic or hydrophilic residue at the −3 position is evident. We refer to this type of site as a −2/+2 consensus site. Inspection of the Swi6 protein sequence showed that in addition to the five identified −2/+2 sites there were five more sites of this type. However, based on the pattern of the Swi6 phosphopeptide maps, these other sites are unlikely to be phosphorylated by Rad53 to any significant level. Of these, only S160 is definitely located on the surface of the Swi6 protein and is accessible, since it is phosphorylated in vivo (52). However, it carries a proline residue in +1 position. At least in the context of short peptide libraries, an S/TP sequence was found to be a poor substrate for a human Rad53 homolog, Chk2 (41). It is therefore not surprising that S160 is not targeted by Rad53.
The weakly phosphorylated ET(556)TSLI site is a “half” site, with a hydrophobic aliphatic residue flanking the phosphorylatable residue only on one side (+2, Fig. Fig.4B).4B). We tested several half sites for phosphorylation by Rad53, and detected only weak reactivity at S149 and S152 of the sequence QDMS(149)LDS (data not shown), which contains two adjacent half sites (Fig. (Fig.4B4B).
As mentioned above, our analysis revealed conservation of the −2 and +2 positions in the Rad53 phosphorylation sites mapped on Swi6. Moreover, in two cases in which only one of these positions was occupied by the conserved residue I, L, M, or V, we observed only weak phosphorylation by Rad53. To address whether these hydrophobic aliphatic residues are important for the recognition of the site by Rad53, we replaced the −2 isoleucine or the +2 leucine around the S547 site with alanine, which is also nonpolar but less hydrophobic. The resulting mutants were phosphorylated in vitro by Rad53 immunoprecipitated out of 4NQO-treated yeast and digested with chymotrypsin. Chymotrypsin cleaves at least 10 amino acids away from I545 and L549 to release large peptides, and as such their mobilities will be less affected by the alanine substitutions. These digests show that the two major peptides that are generated due to S547 phosphorylation (compare the wild-type and S547A maps in Fig. Fig.4C)4C) disappear when either I545 or L549 are mutated to alanine (Fig. (Fig.4C).4C). This result and the strong conservation we observed in the major Rad53 sites of Swi6 lead us to conclude that hydrophobic aliphatic residues at both the −2 and +2 positions are critical determinants of the Rad53 phosphorylation specificity.
Only a subset of in vitro Rad53 phosphorylation sites can be readily observed on an in vivo phosphopeptide map of Swi6. The in vitro peptides 3 and 7 are the closest in their phosphopeptide map positions to the DNA damage-inducible peptides previously detected in vivo (51). These in vivo peptides are also Rad53 dependent, as we have shown previously (see above). We therefore asked whether S547 of Swi6, which maps to the peptides 3 and 7, is the site of a detectable phosphorylation by Rad53 in vivo. As shown before (51), the wild-type Swi6 isolated from MMS-treated RAD+ cells exhibits two damage-inducible phosphopeptides that are well separated from the bulk of the peptides on the two-dimensional map (Fig. (Fig.5A).5A). When S547 is substituted for A, these phosphopeptides are no longer observed upon MMS treatment (Fig. (Fig.5B).5B). We conclude that S547 is a site of in vivo phosphorylation by Rad53 under conditions of DNA damage. The maps in Fig. Fig.55 also show other changes in the in vivo map upon MMS treatment. This suggests that additional, yet-unidentified sites on Swi6 may be targets of damage-inducible phosphorylation. However, we cannot attribute them to Rad53 at this time (Fig. (Fig.2)2) (51).
The MMS-induced G1/S checkpoint activity of wild-type cells was compared to that of cells carrying alanine substitutions in serine and threonine residues at sites of Rad53 phosphorylation in Swi6 (Fig. (Fig.5C).5C). G1 cells were collected by elutriation, and the G1-to-S transition was monitored by counting budding over time with or without MMS treatment, as in Fig. Fig.1.1. All mutants tested behaved as wild type in driving the G1-to-S transition when no DNA damage was applied. That is, they transited to the S phase at the same critical volume as the wild type and thus were not measurably defective under normal growth conditions (data not shown). When the single mutants, S547A or I545A, were compared to the wild type under DNA damage conditions, there was a slight acceleration of budding (Fig. (Fig.5C5C and data not shown). Further alanine substitutions of serines and threonines in up to five mapped Rad53-phosphorylatable sites had a similarly modest effect, and none of these mutants reduced the damage-induced delay as much as rad53-11 or swi6Δ mutants (Fig. (Fig.5C).5C). To follow this up, we used alpha factor as means of synchronizing the cells in G1. Seven and six independent measurements were done on strains with the wild-type Swi6 and a mutant Swi6 with alanines substituted for serines and threonines in all five Rad53 sites (m5, see example in Fig. Fig.5D).5D). These measurements showed that, on average, the wild type achieved 50% budding in 84.8 ± 2.53 min after a 30-min pulse of 0.1% MMS during alpha factor arrest, whereas the mutant had 50% buds after 73.4 ± 3.2 min. Therefore, in the absence of Rad53 phosphorylation sites in Swi6, cells were somewhat less capable of delaying the G1-to-S transition after DNA damage. In rad53-11 cells, eliminating multiple Rad53 phosphorylation sites in Swi6 had no additional effect on the G1-to-S transition delay compared to rad53-11 cells alone (data not shown). Overall, these data are consistent with the notion that phosphorylation of Swi6 by Rad53 contributes to the mechanism through which Rad53 delays S phase in response to DNA damage but is not its only component.
The residues T169 and S170 are adjacent to the Swi6 nuclear localization sequence, which is inactivated by phosphorylation of S160 in late G1 (52), or by substituting S160 with aspartic acid. We thus asked whether analogous substitutions of T169 and S170 affected nuclear localization of Swi6 in G1. We found that Swi6 remained nuclear throughout the MMS-induced delay of the G1-to-S transition, whether T169 or S170 were mutated or not (data not shown). In addition, the S160D mutant of Swi6, which is not restricted to the nucleus in G1 (52), was still found in the cytoplasm in MMS and had no effect on the MMS-induced delay of the G1-to-S transition (data not shown). We conclude that changes in the subcellular localization of Swi6 do not have a role in the MMS-induced pausing of the cell cycle in G1.
The most stringent consensus suggested by the analysis of Swi6 includes basic residues (lysine or arginine) in the −3 position, excludes proline from +1 position, and requires L, M, V, or I at positions −2 and +2. Using the version of this consensus −2/+2 site, we sought to identify new candidate Rad53 targets. A total of 519 proteins were identified that carried two or more such sites (see Materials and Methods). The search returned all of the mismatch repair Msh proteins (Msh1 to Msh6) and some Mlh proteins (Mlh1 to Mlh3) (35). −2/+2 sites were found in the three proteins critical for telomere tethering to the nuclear periphery, Nup145, Mlp1, and Mlp2 (18). Both Swi6 partners, the Mbp1 and Swi4 proteins, contain −2/+2 sites, and four of these in Swi4 are located within the Swi6-binding domain of Swi4. Finally, we noted that all SMC (for structural maintenance of chromosomes) proteins and the SMC-like Rad50 contained Rad53 sites. The six SMC proteins form the cohesin (SMC1 and -3), condensin (SMC2 and -4), and the SMC5-SMC6 complex that is involved in DNA repair (22). Cohesin and condensin complexes have additional subunits (Scc1/Mcd1, Scc3 and Ycs4, Ycs5, and Brn1, respectively) (22). Remarkably, all of these subunits also carry one or more stringent −2/+2 sites.
Scc1 is a cohesin subunit whose proteolysis at the G2/M boundary allows sister chromatid separation (40). G2/M-specific phosphorylation of Scc1 by the Polo-like kinase Cdc5 promotes this proteolysis (2). There are three perfect −2/+2 sites in Scc1: S112, S273, and S367. Interestingly, two of these −2/+2 sites in Scc1 are adjacent to or overlap the mapped Cdc5 phosphorylation sites (2) (Fig. (Fig.6A).6A). In SDS-PAGE, Scc1 from untreated cells migrates predominantly as a single band, with trace amounts of a slower-migrating form (Fig. (Fig.6B).6B). This latter form can be attributed to Cdc5 phosphorylation. A 20- to 30-min pulse of MMS or 4NQO triggers the appearance of a more prominent slowly migrating form of Scc1 (Fig. 6B and C). In the rad53-11 cells treated with the same doses of 4NQO, the appearance of this slowest-migrating, damage-inducible form of Scc1 was virtually eliminated and only a band of intermediate mobility was detected (Fig. (Fig.6C).6C). The cell cycle distribution of both wild-type and rad53-11 cells was comparable before and after this short 4NQO treatment (data not shown). Thus, the observed differences in the modification pattern of Scc1 cannot be attributed to the damage-induced accumulation of cells at the G2/M border in the wild-type cells and not in the rad53 cells. Rather, it is likely that at least a subset of DNA damage-induced modifications on Scc1 in vivo are occurring in a Rad53-dependent manner.
To further test this, we generated in vivo phosphopeptide maps of HA-tagged Scc1 immunoprecipitated from wild-type and rad53-11 cells treated with 0.5 μg of 4NQO/ml. The arrow in Fig. Fig.77 denotes one peptide within Scc1 that is more prominently phosphorylated after 4NQO treatment in RAD+ but not in rad53-11 cells (Fig. 7A and B). This finding is consistent with the Rad53 and damage-specific mobility shift observed in Scc1 in SDS-PAGE and suggests that Scc1 undergoes DNA damage-induced phosphorylation, a part of which is attributable to Rad53.
We have identified five sites of Rad53-mediated phosphorylation on Swi6 by using both recombinant Rad53 and the kinase immunoprecipitated out of yeast cells undergoing DNA damage. Importantly, Rad53 from both of these sources produces nearly identical profiles of phosphorylation on Swi6. This suggests that Rad53 alone and not a copurifying yeast kinase is responsible for Swi6 phosphorylation in our assays. Moreover, it indicates that DNA damage does not alter the site specificity of Rad53 but only increases kinase activity. This result is consistent with our previous observation that Rad53 overproduced in undamaged yeast cells can also phosphorylate Swi6 in vitro.
At least one of the identified sites, S547, is phosphorylated in vivo in an MMS-inducible, Rad53-dependent manner, suggesting that Rad53 recognizes Swi6 as one of its substrates in a cell undergoing DNA damage. Phosphorylation of this site is substoichiometric, suggesting that only a subpopulation of Swi6 molecules can be phosphorylated. For example, only DNA-bound or Swi4-bound fractions of Swi6 could be targeted by Rad53. We did not detect in vivo phosphorylation of the other sites mapped in vitro on Swi6. These sites may not be accessible, or they may be phosphorylated in a even more transient or substochiometric manner. Alternatively, there may be inducible phosphopeptides that are obscured by overlapping constitutive phosphopeptides on the peptide map.
Alignment of the five Rad53 phosphorylation sites of Swi6 reveals a remarkable degree of conservation, particularly at −2 and +2 positions (Fig. (Fig.4B).4B). Both of these positions are occupied only by hydrophobic aliphatic amino acids, and conservative substitution of these residues to alanine prevents phosphorylation by Rad53. Position −3 shows a preference for basic or hydrophilic residues, and proline may be unfavorable at +1. There may also be a bias toward basic residues at the −6 position. Finally, it appears that either position 0 or +1 can be phosphorylated within a given site, if occupied by serine or threonine. Interestingly, comparison of the Swi6 sites to the in vivo phosphorylation site recognized by Chk2, the mammalian homolog of Rad53 (36), on human BRCA1 (S988) (33) shows exactly the same sequence preference at −2 and +2, as the one identified in the present study (Fig. (Fig.4A).4A). Another in vivo Chk2-recognized site, S20 of p53 (8, 23, 48), a reportedly less-than-optimal site, has a leucine only in the +2 position. This is consistent with our finding that “half” sites can be phosphorylated by Rad53 in vitro, albeit weakly. However, the other reported in vivo Chk2 phosphorylation site, S123 of CDC25A (15), has no obvious similarity to the −2/+2 consensus (Fig. (Fig.4A4A).
The specificities of the Chk1 and Chk2 kinases have been studied by using peptide libraries (41). The peptide substrate preferences derived for human Chk2 are similar to our consensus in that they include hydrophobic residues at positions following the phosphorylatable residue, and a basic residue, arginine, at position −3. However, there is no preference seen for the −2 position in the peptide study. These differences may be attributed to the use of peptide versus protein substrates or to the necessarily limited array of sequences present in any peptide library. It is also possible that human Chk2 and yeast Rad53 have somewhat different substrate preferences or that the identified Swi6 sites are not the highest affinity sites for Rad53 phosphorylation, just as S988 in BRCA1 is not the best Chk2 substrate (41). The in vivo targets of checkpoint kinases may carry multiple suboptimal sites as a way of transforming their phosphorylation into an on-off switch actuated by a substantial increase of the kinase activity, as is the case for Sic1 phosphorylation (39).
The −2/+2 consensus allows us to derive insights into Rad53 activity and regulation. All of the proteins implicated as targets or binding partners of Rad53 (Crt1, Dbf4, Dun1, Cdc5, Rad55, Bfa1, Asf1, Rad9, and Mrc1 [1, 13, 14, 25, 26, 57, 59, 64]) carry one to nine matches to the −2/+2 consensus. In Rad53 itself there are eight potential −2/+2 sites, most of which are within 30 amino acids of S/TQ sites, which are potential sites for phosphorylation by the upstream checkpoint kinases Mec1 and Tel1. The association of potential Rad53 and Mec1/Tel1 sites suggests that both transcatalytic (via Mec1 and Tel1) and autocatalytic pathways may activate Rad53 in response to damage (20).
In addition, three −2/+2 sites are located between the β-sheets of the FHA2 domain of Rad53 (12). FHA1 and FHA2 are protein domains that bind to phosphorylated threonine residues embedded within the FHA1- or FHA2-binding motifs (TxxD or TxxL/I, Fig. Fig.88 ). Autophosphorylation of Rad53 on these sites could change the properties of the FHA2 domain. For example, the phosphorylated FHA2 domain of Rad53 may no longer associate with Rad9. This could provide a mechanistic explanation for the observed release of Rad53 from the complex with Rad9 after autophosphorylation (20).
Interestingly, four of five Swi6 −2/+2 sites overlap or adjoin sequences that match the FHA2-binding motif (Fig. (Fig.8).8). This is analogous to the in vivo Chk2 phosphorylation site S20 in p53, which overlaps the FHA1-binding motif that binds the FHA1 domain of Rad53 (10). In Rad9, there are three −2/+2 sites within the cluster of the FHA-binding motifs that were shown to be necessary for Rad53 binding and activation in vivo (47). Among these, two −2/+2 sites completely overlay both the putative Mec1/Tel1 phosphorylation sites (S/TQ) and the FHA-binding motifs. The third −2/+2 site in Rad9 adjoins the FHA-binding motif (Fig. (Fig.8).8). It is tempting to speculate that Rad53 phosphorylation adjacent to FHA-binding motifs affects their ability to bind to FHA domains.
Our search for consensus sites for Rad53 phosphorylation in other proteins revealed a number of potentially interesting new candidate Rad53 targets among cell cycle and DNA metabolism proteins, in particular, the cohesin complex. This complex is known to be targeted by the DNA damage checkpoint in G2 both in yeast (60) and in mammals (30, 61). We have shown that one of the cohesin complex subunits, Scc1, undergoes DNA damage-induced phosphorylation in vivo. We also established that at least one of the damage-inducible phosphorylation events on Scc1 is largely Rad53 dependent. Although an in vitro kinase assay is needed to prove that Scc1 is indeed a direct Rad53 target, our data are certainly consistent with this possibility.
Given the fact that the DNA damage checkpoint pathways in budding yeast (19, 26, 44, 60) consist of multiple parallel branches, it is not surprising that elimination of the Rad53 phosphorylation sites in Swi6 has only a minor effect on the length of the checkpoint-mediated delay in the G1-to-S transition after damage. It is possible that Rad53 acts upon other components of the late-G1 transcription machinery, e.g., Swi4 and Mbp1, since both of these proteins carry multiple Rad53 consensus sites. Other kinases, such as Dun1, Chk1, and Mec1, are also activated and could phosphorylate Swi6 and additional G1 targets. This is consistent with our findings that SWI6-null mutants have a G1 checkpoint defect and that the rad53-11 swi6Δ double mutant has a more extreme G1 checkpoint-deficient phenotype than either mutation alone.
It is unclear at present what biochemical activities of Swi6 are affected by Rad53 phosphorylation and DNA damage in general. MMS-induced damage does not abolish Swi6 nuclear localization in G1. Nor does it prevent Swi6 from binding to the CLN1 promoter in vivo in RAD+ or rad53-11 cells (J. M. Sidorova, unpublished data). Loss of Sin3 repressor, or of Stb1, which may connect Swi6 to the Sin3 repressor complex (24, 29), does not give rise to a G1/S checkpoint defect (Sidorova, unpublished). Perhaps other, yet-unidentified interacting factors may confer the effects of Swi6 phosphorylation by Rad53 on the G1-to-S transition.
Checkpoints are likely to include multiple layers of control imposed on key events, with dozens of substrates targeted by more than one kinase and phosphorylated either in an additive or competitive fashion. A network thus organized would be highly responsive to changing conditions and failsafe. Understanding of the Rad53 phosphorylation site preferences provides a new tool for dissecting the complexity of checkpoint controls in budding yeast.
We gratefully acknowledge members of the Breeden lab for support and numerous discussions. We particularly thank Bernard Mai for generating pBD1998; Steve Elledge for helpful discussions and for providing baculovirus-expressed Rad53 and anti-Rad53 serum; and Andrew Emili, Robert Sclafani, and Sue Biggins for strains and constructs.
This work was supported by the NIH grant GM41073 to L.L.B. and by a Leukemia and Lymphoma Society senior fellowship to J.M.S.