Analysis of the Role of Phosphorylation in Fission Yeast Cdc13p/CyclinB Function*

The Cdk1p-cyclin B complex drives entry into mitosis in all eukaryotes. Cdc13p is the single essential cyclin in Schizosaccharomyces pombe and a member of the cyclin B family. Cdc13p abundance rises during G2-phase and falls as cells progress through mitosis and G1. Cdc13p degradation, mediated by the anaphase-promoting complex, is an important mechanism of Cdk1p inhibition and mitotic exit. Cdk1p-cyclin B1 complexes shuttle between the nucleus and cytoplasm, and preventing nuclear accumulation of Cdk1p-cyclin B1 in mammalian cells appears to be one mechanism of preventing entry into mitosis during a DNA damage-induced checkpoint delay. In vertebrates, phosphorylation plays a key role in regulating the intracellular distribution of cyclins. Previous mass spectrometric analysis identified sites of Cdc13p phosphorylation. Here, we have confirmed that these sites are the sole in vivo Cdc13p phosphorylation sites and have studied the role that phosphorylation plays in Cdc13p localization and function. Our data indicate that Cdc13p accumulates in the nucleolus in response to G2 checkpoint delays, rather than in the cytoplasm, and that phosphorylation plays no role in Cdc13p localization or function.

The Cdk1p-cyclinB complex drives entry into mitosis in all eukaryotes (1). In the fission yeast, Schizosaccharomyces pombe, progression through the cell cycle requires the function of its Cdk1p (encoded by the cdc2 ϩ gene) both in G 1 before the initiation of S-phase and at the G 2 -M boundary (2). The activity of S. pombe Cdk1p oscillates throughout the cell cycle, peaking as cells enter M-phase (reviewed in Ref. 3). This periodicity of Cdk1p is dependent on its cell cycle-specific association with various cyclins, its phosphorylation state, and the activity of an inhibitor of Cdk1p-cyclin function, Rum1p (reviewed in Ref. 1).
Unlike many other organisms, S. pombe cells produce relatively few cyclin partners for Cdk1p. Cdc13p is the single essential cyclin in S. pombe (4 -6) and a member of the cyclin B family (reviewed in (7)). Cdc13p abundance rises during G 2phase and falls slowly as cells progress through mitosis and G 1 (8). It functions both to drive the events of mitosis and prevent re-initiation of S-phase (9).
The degradation of Cdc13p as cells exit mitosis is an important mechanism of Cdk1p inhibition and occurs through ubiquitin-triggered proteolysis mediated by the anaphase-promoting complex ubiquitin ligase. The motif specifying Cdc13p degradation has been mapped to its N terminus, similar to other cyclin Bs (10). Truncation of the first 70 amino acids or deletion of the 9 amino acid destruction box stabilizes the protein in an in vitro assay for ubiquitin-mediated protein degradation. Overproduction of stable Cdc13p fragments in S. pombe cells results in high Cdk1p activity, an accumulation of cells in anaphase and a block to septation (10). These stable fragments of Cdc13p retain a conserved stretch of 150 amino acids that is a hallmark of cyclin proteins and is the region through which cyclins bind Cdks and promote their activation (1). Transiently expressed anaphase-promoting complex activators are probably involved in the recognition of the "destruction box" motif within the N terminus of Cdc13p. In S. pombe, five such anaphase-promoting complex activators, members of the CDC20 protein family, are predicted by the genome sequence and three have been studied as follows: Slp1p (present in mitosis) (11,12); Srw1p/Ste9p (present during G 1 ) (13,14); and Mfr1p/Fzr1p (present during meiosis and sporulation) (15,16). Indeed, Ste9p/Srw1p is required for Cdc13p degradation when cells arrest in G 1 in preparation for mating (13,14,17).
For some time, it has been appreciated that mitotic Cdk1pcyclin B1 complexes shuttle between the nucleus and cytoplasm, accumulating in the nucleus during mitosis (18). Preventing nuclear accumulation of Cdk1p-cyclin B1 appears to be one mechanism of preventing entry into mitosis during a DNA damage-induced checkpoint delay (19 -21). Cdk1p-Cdc13p accumulates in the nucleus during S-and G 2 -phases and also can be detected at the nucleolus, at the spindle pole body, and along the mitotic spindle (22)(23)(24)(25). Interestingly, the ability of Cdk1p to enter the nucleus is Cdc13p-dependent (23). In other organisms, phosphorylation plays a key role in regulating the intracellular distribution of cyclins. Cyclin B1, as an example, has a cytoplasmic retention sequence (26) that contains a nuclear export sequence (20,27,28). Phosphorylation of a single serine within the nuclear export sequence is sufficient to inhibit nuclear export (29). However, this is insufficient to explain the kinetics of mitotic entry. Additional phosphorylations within the cytoplasmic retention sequence are important to create a nuclear import signal that is essential for the timely onset of mitosis (29 -31). At least some of the phosphorylation sites within the cytoplasmic retention sequence are probably autophosphorylation sites of Cdk1-cyclin B (32)(33)(34).
In this study, we have examined the role of Cdc13p phosphorylation in its function and localization. We have confirmed the assignment of phosphorylation sites identified by mass spectrometry (35) and present evidence that Cdc13p phosphoryla-tion is not important for its function in normal cell cycle progression or in the cell response to checkpoint signals that monitor DNA damage or the completion of DNA replication.

EXPERIMENTAL PROCEDURES
Yeast Methods-S. pombe strains used in this study are listed in Table I. Strains were constructed by random spore analysis or by tetrad dissection when necessary. S. pombe strains were grown in minimal medium with the appropriate supplements in the presence or absence of 5 g/ml thiamine or in standard yeast extract medium (36). All of the temperature shift experiments were carried out by growing cells to mid-log phase at 25°C and shifting to 36°C for 4 h. Hydroxyurea was added to medium at a final concentration of 12 mM. Camptothecin was added at a final concentration of 40 M. S. pombe transformations were carried out by electroporation (37) or by a lithium acetate procedure (38). The cdc13 ϩ gene and cdc13 mutant alleles were tagged at their chromosomal locus at the 3Ј end of the open reading frame with sequences encoding green fluorescent protein (GFP) 1 by a PCR-mediated strategy as described previously (38). Proper integration of the epitope cassette was confirmed by PCR. The cdc13-GFP:kan R strains retained normal length at division.
Antibodies, Cell Labeling, and Immunoprecipitation-Rabbit polyclonal antibodies to Cdc13p were generated against the entire Cdc13 protein produced in Escherichia coli. The protein was purified from E. coli as inclusion bodies and resolved from contaminants by SDSpolyacrylamide gel electrophoresis. Serum from two bleeds (fifth and sixth) of one of the rabbits was subjected to ammonium sulfate precipitation and resuspension and is used throughout this study (GJG56). This aliquot of anti-Cdc13p serum has been used extensively in previous studies (39 -41). In vivo labeling of cells was accomplished as described (42). S. pombe cell lysates were prepared as detailed previously (43). Cdc13p was recovered from denatured cell lysates using the GJG56 serum described above.
Phosphoamino Acid Analysis and Phosphopeptide Mapping-[ 32 P]Cdc13p bands were cut from Immobilon-P membranes, and slices were hydrolyzed in 6 N HCl for 60 min at 110°C (44) or subjected to trypsinization as described (45). Partial acid hydrolysis products were separated by electrophoresis in two dimensions on thin-layer cellulose plates at pH 1.9 and 3.5 (45) using the Hunter thin-layer electrophoresis system (C.B.S. Scientific, Del Mar, CA). Phosphopeptides were separated on thin-layer cellulose plates by electrophoresis at pH 1.9 for 30 min at 1 kV and ascending chromatography (45). Phosphoamino acids and phosphopeptides were visualized by autoradiography or with Amersham Biosciences PhosphorImager screens.
Site-specific Mutagenesis-A 3.3-kb SalI-BamHI genomic fragment containing the cdc13 ϩ open reading frame was cloned into the pBluescript vector (Stratagene). Oligonucleotide-directed mutagenesis was performed using the Chameleon or QuikChange kits (Stratagene) according to the manufacturer's instructions. The sequence of the oligonucleotide used to alter Ser 177 , Ser 180 , and Ser 183 to alanines is 5Ј-CCC-AAACTTCATCGCGATGCTGTTGAGGCTCCCGAAGCTCAAGATTG-G-3Ј. The sequences of other oligonucleotides are available upon request. The oligonucleotides used for alanine substitutions also introduced a NruI restriction site without further altering amino acid specification. Those for aspartic acid introduced a BamHI site. The presence of the desired mutations was confirmed by DNA sequencing. The mutants were subcloned into the yeast expression vector pIRT2 (46). To create a gene replacement strain, this vector was transformed into a diploid heterozygous for the cdc13 deletion allele with the relevant genotype cdc13:ura4 ϩ /cdc13 ϩ leu1-32/leu1-32. Leu ϩ Ura ϩ transformants were selected and induced to sporulate. Haploid Ura4 ϩ Leu ϩ progeny then were isolated. They were grown in yeast extract for approximately five generations and plated on minimal plates containing leucine, uracil, adenine, and 57 M 5-fluoro-orotic acid. A haploid colony that was Leu-Ura-Ade-was then selected as a possible gene replacement strain.
The cdc13 ϩ open reading frame was amplified from this strain and sequenced to check that the mutations were present.
Microscopy Methods-S. pombe cells in liquid culture were collected by centrifugation, fixed with glutaraldehyde, and stained with a DNAspecific fluorescent dye (4Ј,6-diamidino-2-phenylindole) as described previously (47). GFP-tagged Cdc13p was visualized in live cells. Microscopy was performed using a spinning disk confocal microscope (Ultraview LCI, PerkinElmer Life Sciences) and Ultraview LCI software (version 5.2, PerkinElmer Life Sciences) for image acquisition. Images were processed using Velocity software (version 1.4.2, Improvision). Z-series optical sections were taken at 0.5-m spacing for three-dimensional reconstructions.

RESULTS
Cdc13p Is a Phosphoprotein-To test whether Cdc13p is a phosphoprotein, wild-type S. pombe cells were labeled in vivo with [ 32 P]orthophosphate and a denatured protein lysate was prepared. A single phosphoprotein of the expected size was immunoprecipitated specifically by anti-Cdc13p but not preimmune serum (Fig. 1A). Phosphoamino acid analysis of [ 32 P]labeled Cdc13p indicated that it was phosphorylated exclusively on serine residues (Fig. 1B). Only a small percentage of [ 32 P]Cdc13p was ever released from PDVF or nitrocellulose membranes or eluted from gel slices (data not shown). However, better recovery was obtained with chromytrptic digestion than with tryptic digestion and separation of Cdc13p chymotryptic phosphopeptides revealed a complex pattern (Fig. 1C), raising the possibility that Cdc13p might be phosphorylated on several sites (Fig. 1C).
Cdc13p is phosphorylated throughout the cell cycle-Cdc13p levels vary during the cell cycle (8,48). To determine whether the extent of Cdc13p phosphorylation varied independently of its abundance, which would be an indicator of a possible regulatory role, different cell cycle mutants, wild-type cells, wildtype cells treated with drugs that block cell cycle progression, and wild type cells overproducing Cdc13p were labeled with [ 32 P]orthophosphate and Cdc13p was immunoprecipitated from denatured cell lysates of each. The following mutants and drugs were chosen: a cdc10 -129 mutant (G 1 -phase arrest), hydroxyurea (S-phase arrest), cdc25-22 (G 2 arrest) and cdc25-22 cells that had been arrested in G 2 and then released for 20 min into nocadazole (M-phase arrest). In large measure, the level of Cdc13p phosphorylation paralleled its abundance ( Fig. 2A). This was confirmed by quantitation of three such experiments (Fig. 2B). Specifically, the amount of [ 32 P] incor- poration into Cdc13p in each instance was determined by scintillation counting of gel slices containing the immunoprecipitates and compared with its abundance that was determined by immunoblotting and densitometry of Cdc13p immunoprecipitates. The amount of label and protein in a cdc25-22 mutant was arbitrarily defined as 100%. We conclude that although early in the cell cycle there appeared to be a somewhat higher stoichiometry of Cdc13p phosphorylation, Cdc13p phosphorylation increased with increasing levels of Cdc13p and was not restricted to any particular cell cycle stage. Confirming Cdc13p phosphorylation sites-As mentioned above, it was not possible to recover enough [ 32 P]labeled Cdc13p from gel slices or PVDF membranes to enable the sites of phosphorylation to be determined by standard approaches (45). However, sites of Cdc13p phosphorylation were identified using a mass spectrometric approach in which Cdc2p complexes purified by a tandem affinity approach were analyzed (35). Careful analyses of the mass spectrometric data suggested that at least one and possibly all three serines at positions 177, 180, and 183 were in vivo phosphorylation sites, but it was not possible to identify the exact residue(s) (35). These assignments were particularly encouraging given that Ser 177 , Ser 180 , and Ser 183 are predicted to be contained in large hydrophobic tryptic or chymotryptic peptides that would resist removal from PVDF membranes. To confirm that one or more of these serines represented bona fide in vivo Cdc13p phosphorylation sites and to assess their importance to Cdc13p function, they were altered by site-directed mutagenesis individually and in combination to the non-phosphorylatable amino acid alanine to create cdc13-S177A, cdc13-S180A, cdc13-S183A, and cdc13-3A (in which all three sites were mutated). All were able to complement both the cdc13-117 mutant at its restrictive temperature (Fig. 3A) and the cdc13:ura4 null allele (data not shown and see below), indicating that phosphorylation at these sites is not essential for Cdc13p function. To determine whether there were any subtle changes to cell cycle regulation resulting from the absence of Cdc13p phosphorylation, gene replacement strains were constructed in which these mutants were integrated into the genome in the place of wild type cdc13 ϩ . These mutant strains all divided at normal rates and at a normal size. No cell cycle defect could be discerned (data not shown).
To validate the mass spectrometric identification of Cdc13p phosphorylation sites, wild-type and cdc13-3A strains were labeled with [ 32 P]orthophosphate and Cdc13p was immunoprecipitated from each (Fig. 3B). As before, 32 P-labeled Cdc13p was detected in wild-type cells. In contrast, no labeled Cdc13p was detected from cdc13-3A. This is not due to the absence of Cdc13p protein in the labeled cells (Fig. 3C). This result confirms that Ser 177 , Ser 180 , and/or Ser 183 represent in vivo phosphorylation sites. Moreover, this result indicates that there are no other phosphorylation sites within Cdc13p, at least under normal growth conditions. This was a possibility given that the 91.7% sequence coverage of Cdc13p by mass spectrometry left some regions of the protein unanalyzed (35). Given the lack of effect, we have not determined which of the three serines becomes phosphorylated or whether the phosphorylation occurs at two or all three sites.
The Absence of Cdc13p Phosphorylation Does Not Affect Cdc13p Intracellular Localization-Phosphorylation of mammalian cyclin B1 plays an important role in its intracellular localization (reviewed in Refs. 18 and 49). As one response to genotoxic stress, cyclin B1 is also retained or shuttled into the cytoplasm (19 -21). To determine whether phosphorylation or the lack thereof affected Cdc13p localization, cdc13 ϩ and cdc13-3A were tagged at their endogenous locus with sequences encoding GFP. Live cell imaging of these strains indicated that Cdc13p-3A localized normally (Fig. 4, upper right  panel). Similar to wild-type Cdc13p (Fig. 4, upper left panel), Cdc13p-3A was present in the nucleus with nucleolar concentration at the spindle pole bodies and also decorated the mitotic spindle. To determine whether checkpoint activation and cell cycle delay affected its localization, wild-type and cdc13-3A cells were incubated for 4 h in hydroxyurea and the localization of Cdc13 proteins was examined. After this treatment, it was clear that cells had arrested in interphase and were somewhat elongated (Fig. 4, lower panels). Moreover, Cdc13p and Cdc13p-3A had both accumulated at the spindle pole bodies and in the nucleus with nucleolar localization being particularly evident (Fig. 4, lower panels). We conclude that the absence of Ser 177 , Ser 180 , and/or Ser 183 phosphorylation does not adversely affect the ability of Cdc13p to gain entry into the nucleus or localize to the nucleolus either normally or upon activation of the DNA replication checkpoint.
Because we did not detect an alteration in cell growth or checkpoint response when Cdc13p was not phosphorylated, we considered the possibility that dephosphorylation rather than phosphorylation of Cdc13p might play a regulatory role. To test this theory, we altered the cdc13 ϩ cDNA by site-directed mutagenesis to encode proteins in which the serine phosphorylation sites were replaced with aspartic acid residues (S177D, S177D, and S180D and S177D, S180D, and S183D) in the hope of mimicking constitutive phosphorylation. Each of the three mutants were integrated into the genome in single copy by replacing the cdc13 null allele. All three mutant strains were viable and exhibited wild type morphology in liquid medium FIG. 1. Cdc13p is a phosphoprotein. A, wild-type 972 cells were labeled with [ 32 P]orthophosphate and lysed in SDS lysis buffer. Anti-Cdc13p serum was added to one-half of the lysate, and preimmune serum was added to the other half. The immunoprecipitates were resolved by SDS-PAGE and transferred to a PVDF membrane. Labeled proteins were detected by autoradiography. The position of the band corresponding to Cdc13p is indicated with an arrowhead. The positions of molecular mass standards are given in kilodaltons. B, the piece of PVDF membrane containing Cdc13p was analyzed for its phosphoamino acid content. The positions of the phosphothreonine (T) and phosphotyrosine (Y) standard are diagrammed. S, phosphoserine. C, 32 P-labeled Cdc13p obtained from wild-type cells as in A was digested with chymotrypsin. The chymotryptic phosphopeptides were separated in two dimensions as described under "Experimental Procedures." Electrophoresis was performed in the horizontal dimension at pH 1.9 with the anode on the left. The origin is marked with an arrowhead. PAA, phosphoamino acid analysis; MAP, mitogen-activated protein. (Fig. 5, A-C). Given the apparent lack of functional changes, we examined the localization of only the triple mutant that we named Cdc13p-3D. cdc13-3D was tagged at its endogenous locus with sequences encoding GFP. Live cell imaging of this strain indicated that it localized normally (Fig. 5D). Furthermore, cdc13-3D cells arrested normally in hydroxyurea and the Cdc13-3Dp accumulated in the nucleus, particularly the nucleolus (Fig. 5E). To determine whether DNA damage precluded Cdc13p concentration in the nucleus and whether this was altered by phosphorylation, the cdc13-GFP, cdc13-3A-GFP, and cdc13-3D-GFP strains were treated with camptothecin for 3 h (50). In each case, the Cdc13 protein accumulated in the nucleus, particularly the nucleolus (Fig. 5, F-H). These results indicate that neither phosphorylation nor dephosphorylation of Cdc13p influences its intracellular localization. DISCUSSION Numerous studies have determined that cyclin B phosphorylation in vertebrate cells is an important influence on its intracellular localization. As observed in other organisms, we have shown here that Cdc13p, the only essential B-type cyclin of S. pombe, is a phosphoprotein. It is phosphorylated on one to three sites in close proximity to one another that lie N-terminal to the cyclin box and C-terminal of the destruction box. However, our evidence suggests that phosphorylation of this region is unnecessary for the function or regulation of Cdc13p.
Although 32 P incorporation into Cdc13p was detected by us earlier, 2 advances in mass spectrometry were required to provide information on the residues involved (35). These residues lie next to one another at amino acid positions 177, 180, and 183, and it is still not clear whether a single amino acid or all three sites serve as the phosphoacceptor residues. In retrospect, it is clear why conventional approaches used successfully to identify phosphorylation sites in other proteins (45)  To arrest cells in S-phase, wild-type cells were treated with 12 mM hydroxyurea (HU) for the duration of the labeling period. To arrest cells in mitosis, 40 g/ml nocadazole (NOC) was added to arrested cdc25-22 cells for 30 min and they were shifted to 25°C for 40 min prior to lysis. For Cdc13p overproduction (O.P.), wild-type cells were transformed with pREP1cdc13 ϩ and transformants were selected with thiamine present in the medium. Transformants were grown in the absence of thiamine in minimal medium for 14 h prior to switching them into labeling medium. They were then labeled for 4 h at 36°C in the absence of thiamine. Following clarification, lysates were subjected to immunoprecipitation with anti-Cdc13p serum. The immunoprecipitates were resolved by SDS-PAGE and transferred to PVDF membranes, and labeled proteins were detected by autoradiography. B, quantification of Cdc13 phosphorylation. The experiments described in A were repeated three times with the indicated strains in the left graph and three times with the strains indicated in the right graph. Gel bands containing [ 32 P]Cdc13p were excised and counted. Anti-Cdc13p immunoprecipitates prepared in parallel from unlabeled strains were subjected to immunoblotting with anti-Cdc13p serum. Cdc13p bands were scanned by densitometry. The average and standard deviations are plotted using the levels in cdc25-22 cells as an arbitrary 100% standard. chymotryptic peptides containing these sites are large and hydrophobic and therefore poorly recoverable from either gel slices or PDVF membrane. Interestingly, despite the lack of functional effect upon mutation of these sites to either alanine or aspartic acid residues, two of these three residues (Ser 177 and Ser 180 ) are conserved within Cig2p, a closely related but non-essential S. pombe cyclin B. However, they are not generally conserved among cyclin B molecules. One of the serines (Ser 180 ) is followed by a proline residue and thus is a potential S-P Cdk1 autophosphorylation site. Although Cdk1p⅐Cdc13p complexes do undergo some autophosphorylation in vitro, this is not affected in quality or quantity by mutation of Ser 180 to alanine. 3 Thus, autophosphorylation sites are not related to the sites studied here and may be in vitro artifacts because the sites analyzed here are the only ones detected in Cdc13p molecules isolated from cells.
It appears from our data that cytoplasmic sequestration of the Cdc13p is not involved in the fission yeast cell response to DNA replication-or DNA damage-induced G 2 arrest. This contrasts with the situation in vertebrate cells in which Cyclin B accumulates in the cytoplasm in response to DNA damage (19 -21). Rather, our data suggests that Cdc13p is sequestered in the nucleolus when cells are delayed in G 2 -phase by activation of G 2 checkpoints and that nucleolar accumulation does not require changes in Cdc13 phosphorylation state. The different strategy of sequestering Cdk1p activity within the nucleolus in S. pombe may obviate the need for controlling nuclear access through a checkpoint signaling pathway. Nucleolar localization of Cdc13p was first detected more than a decade ago (22,23), but the sequences within Cdc13p responsible for this localization pattern have not been elucidated.
Whereas Cdc13p phosphorylation sites were identified by mass spectrometry, a substantial portion of the protein (Ͼ10%) was not accounted for in the mass spectrometric data and, even in the regions covered, it is possible to miss some sites of modification (35). Therefore, it remained possible that other Cdc13p phosphorylation sites existed in these unidentified parts of the protein. This possibility is strongly disfavored by the fact that a S177A, S180A, and S183A triple mutant protein fails to incorporate any 32 P when cells are labeled with orthophosphate, and it is highly probable that these serines represent the only in vivo phosphorylation sites in the protein.
Consistent with the lack of regulatory influence, our data indicate that the amount of Cdc13p phosphorylation parallels the 3 A. Feoktistova and K. L. Gould, unpublished data.  (1), containing cdc13 ϩ (2), cdc13-177A (3), cdc13-180A (4), or cdc13-183A (5), and cdc13-3A (6), and transformants were selected at 27°C. Transformants were streaked to selective plates at 25 or 36°C. B and C, wild type (KGY28) and cdc13-3A (KGY274) strains were labeled with [ 32 P]orthophosphate for 4 h at 32°C and lysed in SDS lysis buffer. Following clarification, lysates were subjected to immunoprecipitation with anti-Cdc13p serum. The immunoprecipitates were resolved by SDS-PAGE and transferred to PVDF membranes, and labeled proteins were detected by autoradiography (B). The PVDF membrane was also subjected to immunoblotting with anti-Cdc13p serum (C). The positions of Cdc13 proteins are indicated with arrowheads.
FIG. 4. Localization of Cdc13p is independent of its phosphorylation state and checkpoint signaling. The cdc13-GFP (KGY629) (left panels) and cdc13-3A-GFP (KGY630) (right panels) strains were imaged after growth at 25°C in the absence (upper panels) or presence (lower panels) of hydroxyurea (HU) in the growth medium. a, nucleolus; b, spindle; c, nuclear envelope. Arrows indicate spindle pole body and nucleolus.
FIG. 5. Phosphomimetic substitutions do not affect Cdc13p localization or function. A-C, the cdc13-S177D (A), cdc13-S177D S180D (B), and cdc13-S177D S180D S183D (C) strains were grown to mid-log phase, fixed, and stained with 4Ј,6-diamidino-2-phenylindole to visualize overall morphology. D and E, the cdc13-3D-GFP strain (KGY5049) was imaged after growth at 25°C in the absence (ϪHU) (D) or presence (ϩHU) (E) of hydroxyurea in the growth medium. F-H, the cdc13-GFP (KGY629) (F), cdc13-3A-GFP (KGY630) (G), and cdc13-3D-GFP (KGY5049) (H) strains were grown to mid-log phase and treated with 40 M camptothecin (CPT) for 3 h. Live cell images were captured. accumulation of the protein whether it occurs in a prolonged G 2 arrest induced by cdc25 mutation or G 2 checkpoint activation. Although the lack of functional consequence of Cdc13p phosphorylation is unexpected, this study has illustrated the power of combining a mass spectrometric approach with site-directed mutagenesis and in vivo labeling to comprehensively analyze the potential functions of protein phosphorylation events.