MOLECULAR AND CELLULAR BIOLOGY, June 1996, p. 27442755 0270-7306/96/$04.

00 0 Copyright 1996, American Society for Microbiology

Vol. 16, No. 6

The SAPs, a New Family of Proteins, Associate and Function Positively with the SIT4 Phosphatase
MAY M. LUKE,1 FLAVIO DELLA SETA,1 CHARLES J. DI COMO,1,2 HANA SUGIMOTO,1 RYUJI KOBAYASHI,1 AND KIM T. ARNDT1* Cold Spring Harbor Laboratory, Cold Spring Harbor, New York 11724-2212,1 and Graduate Program in Genetics, State University of New York, Stony Brook, New York 117942
Received 11 January 1996/Returned for modication 4 March 1996/Accepted 10 March 1996

SIT4 is the catalytic subunit of a type 2A-related protein phosphatase in Saccharomyces cerevisiae that is required for G1 cyclin transcription and for bud formation. SIT4 associates with several high-molecular-mass proteins in a cell cycle-dependent fashion. We puried two SIT4-associated proteins, SAP155 and SAP190, and cloned the corresponding genes. By sequence homology, we isolated two additional SAP genes, SAP185 and SAP4. Though such an association is not yet proven for SAP4, each of SAP155, SAP185, and SAP190 physically associates with SIT4 in separate complexes. The SAPs function positively with SIT4, and by several criteria, the loss of all four SAPs is equivalent to the loss of SIT4. The data suggest that the SAPs are not functional in the absence of SIT4 and likewise that SIT4 is not functional in the absence of the SAPs. The SAPs are hyperphosphorylated in cells lacking SIT4, raising the possibility that the SAPs are substrates of SIT4. By sequence similarity, the SAPs fall into two groups, the SAP4/SAP155 group and the SAP185/SAP190 group. Overexpression of a SAP from one group does not suppress the defects due to the loss of the other group. These ndings and others indicate that the SAPs have distinct functions. During the early G1 phase of the cell cycle, and in the presence of nutrients, Saccharomyces cerevisiae cells synthesize cellular components and grow in size. Prior to reaching a point in late G1 termed Start, the growing G1 cells can exit the mitotic cell cycle, in response to either nutrient limitation or the presence of mating pheromones (27). By contrast, cells that have executed Start are committed to a round of cell division, independent of nutrients and the presence of mating pheromones. After passage through Start, a cell initiates the processes required for the formation of the daughter cell, namely, DNA synthesis, bud formation, and duplication of the spindle pole body. The execution of Start requires a sufcient level of G1 cyclin/ CDC28 protein kinase activity (6, 28, 29). The main determinants of the amount of G1 cyclin/CDC28 kinase activity are the levels of the CLN1 and CLN2 proteins, which are the major budding yeast G1 cyclins (41). The CLN1 and CLN2 RNA and proteins are unstable, and it is thought that their rates of transcription largely determine the rates and levels of CLN1 and CLN2 protein accumulation during late G1 (41, 45). The DNA-binding factor SWI4 and the associated factor SWI6 have been implicated in the transcription of CLN1 and CLN2 (24, 25). However, CLN1 and CLN2 are transcribed at significant levels in the absence of either SWI4 or SWI6 (7, 20, 37). A major unanswered question is how nutrient growth signals are transduced and integrated into an increase in CLN1 and CLN2 transcription during late G1. The execution of Start also requires the function of SIT4, which is the catalytic subunit of a type 2A-related protein phosphatase (1). Temperature-sensitive sit4 strains arrest in late G1 at Start, prior to the initiation of DNA synthesis, bud formation, and spindle pole body duplication (38). This Start arrest is due to the requirement for SIT4 for the transcription
* Corresponding author. Mailing address: Cold Spring Harbor Laboratory, P.O. Box 100, Cold Spring Harbor, NY 11724-2212. Phone: (516) 367-8836. Fax: (516) 367-8369. Electronic mail address: arndt@cshl.org. 2744

of CLN1 and CLN2, which is at least partly due to the involvement of SIT4 in SWI4 transcription (13). If CLN2 is provided from a SIT4-independent promoter, temperature-sensitive sit4-102 cells arrest with replicated DNA but are still mostly blocked for bud initiation (13). Therefore, in addition to being required for G1 cyclin expression, SIT4 is required in late G1 for bud formation. How SIT4 performs its G1 functions is not yet clear. The levels of SIT4 protein are constant during the cell cycle. However, SIT4 associates with two high-molecular-mass phosphoproteins, of 155 and 190 kDa, in a cell cycle-dependent manner (38). In this report, we term these proteins SAP155 and SAP190, for SIT4-associated proteins. In G1 daughter cells, SIT4 is not associated with SAP155 and SAP190 but is found mostly in a free monomeric form (38). In late G1, SIT4 associates in separate complexes with SAP155 and SAP190 and remains associated until about the middle of mitosis (38). Possibly, certain late G1 processes might be regulated via the association of the SAPs with SIT4. To understand how the SAPs affect the functions of SIT4, we puried SAP155 and SAP190 and cloned the corresponding genes. Surprisingly, not only are SAP155 and SAP190 related proteins, but the cells also contain additional SAPs related to SAP155 and SAP190. At least one of these proteins, SAP185, also associates with SIT4. Our ndings show that SIT4 associates with a new family of high-molecular-mass proteins which function positively with SIT4. Although the SAPs have sequence similarity, they provide at least two distinct cellular functions.
MATERIALS AND METHODS Strains and media. Table 1 lists the yeast strains used in this study. Rich (YP), synthetic complete (SC), and minimal (SD) media were constituted as described elsewhere (33) except that the YP media also contained 10 mM KH2PO4, 0.1 g of tryptophan per liter, and 0.1 g of adenine per liter and the SC media contained 0.2 g of leucine, 0.1 g of all other amino acids, 0.1 g of uracil, and 0.1 g of adenine per liter. Purication of SAP155 and SAP190. CY202 cells, containing hemagglutinin (HA) epitope-tagged SIT4, were grown in YPD (D indicates 2% glucose) me-

VOL. 16, 1996

THE SIT4-ASSOCIATED PROTEIN FAMILY TABLE 1. Yeast strains used in this study

2745

Strain

Genotypea

Reference

CY144 CY202 CY832 CY1078 CY3938 CY4029 CY4102 CY4380 CY4796 CY4802 CY4917 CY4985 CY4987 CY5001 CY5067 CY5077 CY5169 CY5170 CY5196 CY5201 CY5206 CY5210 CY5211 CY5215 CY5217 CY5218 CY5220 CY5224 CY5234 CY5236 CY5254 ML58-4B ML59-15A
a

MATa sit4::HIS3 ssd1-d2 sit4-102/LEU2-CEN W303 MATa sit4::HIS3 ssd1-d2 SIT4:HA-COOH/LEU2-CEN W303 MATa ssd1::LEU2 W303 MAT sit4::HIS3 ssd1-d2 SIT4/YCp50 W303 MATa sit4::HIS3 SSD1-v1 W303 MATa SSD1-v1 W303 MAT cln3::URA3 SSD1-v1 W303 MATa sap190::TRP1 SSD1-v1 W303 MATa sit4::HIS3 sap190::TRP1 SAP190:HA/YCp50 SSD1-v1 W303 MATa sap190::TRP1 SAP190:HA/YCp50 SSD1-v1 W303 MATa sap185::ADE2 SSD1-v1 W303 MATa sap155::HIS3 sap185::ADE2 sap190::TRP1 SSD1-v1 W303 MAT sap155::HIS3 sap185::ADE2 sap190::TRP1 SSD1-v1 W303 MATa sap185::ADE2 sap190::TRP1 SSD1-v1 W303 MATa sap185::ADE2 sap190::TRP1 SAP185:HA/YCp50 SSD1-v1 W303 MATa sit4::HIS3 sap185::ADE2 sap190::TRP1 SAP185:HA/YCp50 SSD1-v1 W303 MATa sit4::HIS3 SIT4/LEU2-CEN SSD1-v1 W303 MATa sit4::HIS3 SIT4:PY/LEU2-CEN SSD1-v1 W303 CY5169 YEp24 CY5170 YEp24 MATa sit4::HIS3 sap155::HIS3 sap185::ADE2 sap190::TRP1 SIT4/LEU2-CEN YEp24 SSD1-v1 W303 MATa sit4::HIS3 sap155::HIS3 sap185::ADE2 sap190::TRP1 SIT4/LEU2-CEN SAP4/YEp24 SSD1-v1 W303 MATa sit4::HIS3 sap155::HIS3 sap185::ADE2 sap190::TRP1 SIT4:PY/LEU2-CEN YEp24 SSD1-v1 W303 MATa sit4::HIS3 sap155::HIS3 sap185::ADE2 sap190::TRP1 SIT4:PY/LEU2-CEN SAP4/YEp24 SSD1-v1 W303 MATa sap4::LEU2 SSD1-v1 W303 MATa sap155::HIS3 SSD1-v1 W303 MATa sap155::HIS3 sap4::LEU2 SSD1-v1 W303 MATa sap185::ADE2 sap190::TRP1 SSD1-v1 W303 MATa sap155::HIS3 sap185::ADE2 sap190::TRP1 SSD1-v1 W303 MATa sap155::HIS3 sap185::ADE2 sap190::TRP1 sap4::LEU2 SSD1-v1 W303 MATa sap155::HIS3 ssd1-d2 W303 MATa sis2::LEU2 SSD1-v1 ura3 leu2 his3 trp1 ade2 MAT bem2::LEU2 SSD1-v1 ura3 leu2 his3 trp1 ade2

38 38 38 38 This This This This This This This This This This This This This This This This This

study study study study study study study study study study study study study study study study study

This study This study This study This This This This This This This This This study study study study study study study study study

The W303 strain background is ura3-1 leu2-3,112 his3-11,15 trp1-1 ade2-1 can1-100 and is normally ssd1-d2 (38). Plasmids are indicated by angled brackets.

dium. At an optical density at 600 nm of about 2.5, the cells were harvested, frozen, and stored at 70 C. The remaining steps were done at 4 C. Forty grams of frozen cells, 35 ml of glass beads, and about 25 ml of breaking buffer (100 mM Tris-HCl [pH 7.5], 200 mM NaCl, 1 mM EDTA, 5% glycerol, 1 mM dithiothreitol [DTT], freshly added protease inhibitors to nal concentrations of 0.2 mM phenylmethylsulfonyl uoride and 1.25 g each of leupeptin, antipain, chymostatin, and pepstatin [Sigma] per ml) were placed in a 100-ml container within the ice water jacket of a Biospec bead beater set to go on for 10 s and off for 80 s until 90% of the cells were broken (total time of about 60 to 90 min). The lysate was cleared by two successive 13,000 g spins of 20 and 10 min, giving about 35 ml of an extract with a protein concentration of about 30 mg/ml. The HA-tagged SIT4 protein was immunoprecipitated by incubation of the extract with 60 l of ascites uid containing antibody 12CA5 (14) for 2 h and then binding the immune complexes with 300 l of protein G-Sepharose beads (Pharmacia) with gentle rocking overnight. The protein G beads were washed 10 times in 20 volumes of a buffer containing 25% radioimmunoprecipitation assay buffer (50 mM Tris-HCl [pH 7.5], 200 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], 1 mM DTT, the protease inhibitors specied above) and 75% breaking buffer and rinsed once in a buffer containing 50 mM Tris (pH 7.5), 50 mM NaCl, 1 mM DTT, and the protease inhibitor mix. An equal volume of 2 Laemmli buffer was added to the beads, and the mixture was boiled. The proteins in the immune complexes were resolved on an SDS8% polyacrylamide gel and visualized by Coomassie blue staining. Peptide sequencing of SAP155 and SAP190. Peptide sequencing was done essentially as described previously (5). Ten peptide sequences each for SAP155 and SAP190 were obtained, and all matched the amino acid sequences predicted from the DNA sequences of SAP155 and SAP190. Cloning and sequencing of the SAP genes. Degenerate primers were designed from peptide sequences of SAP155 and SAP190, taking into consideration the codon usage of S. cerevisiae (34, 44) to minimize degeneracy. Genomic DNA of CY202 was used as the template DNA for PCR. For SAP155, primers from three peptides produced 0.5-, 0.7-, and 0.8-kb PCR products. For SAP190, sense and

antisense degenerate primers from three peptides produced PCR products of 1.4, 0.7, and 2.1 kb. Southern blotting of these PCR products conrmed that the 1.4- and 0.7-kb fragments are two parts of the 2.1-kb fragment. The SAP155 PCR product hybridized to clones 6341 and 4584 on chromosome VI, and the SAP190 PCR product hybridized to clone 4163 on chromosome XI of the Olson library of yeast inserts in lambda phage (30). Although the SAP155 gene was obtained from a YCp50 library by using the cloned SAP155 PCR product as a probe, the SAP155 gene that was sequenced and used for all of the experiments in this report was obtained from a library of yeast genomic inserts in YEp24 (3). This SAP155-containing YEp24 clone was isolated in a genetic screen for high-copynumber suppressors of the temperature-sensitive phenotype of a sit4-102 strain (7a). The full-length SAP190 gene was obtained from a library of yeast genomic inserts in YCp50 (32) by colony hybridization, using the cloned SAP190 PCR products as probes. The location of the SAP155 gene on the yeast DNA insert was determined by hybridization using the cloned PCR product and by subcloning for the smallest complementing DNA fragment, taking advantage of the ability of high-copynumber SAP155 to suppress the temperature-sensitive phenotype of a sit4-102 strain. We sequenced both strands of our SAP155 clone from positions 481 to 3224 (relative to the A of the ATG initiation codon). The sequence of SAP190, corresponding to the open reading frame (ORF) YKR028, was reported previously (11). We sequenced from KpnI to HpaI, containing the amino terminus, and from EcoRV to BstBI, containing the carboxy terminus, of our SAP190 clone. Relative to our SAP190 DNA sequence (which predicts a 1,033-amino-acid protein), the reported SAP190 sequence has a frameshift at codon 1009 and predicts a 1,098-amino-acid protein. The SAP185 gene was isolated from the YEp24 library (3) by colony hybridization, using as a probe the 1.35-kb BstEII-PstI fragment from plasmid pCS3-1 (a generous gift of B. Osmond). The sequence for SAP185 was published in a report describing the sequencing of a 9.7-kb fragment of chromosome X (referred to as ORF J0840 in reference 23). To isolate SAP4, 1 g of genomic DNA from a strain (CY4985) lacking

2746

LUKE ET AL.

MOL. CELL. BIOL.

with 2 U of calf alkaline phosphatase (Boehringer Mannheim) in a 50- l reaction volume at 30 C for 15 min. The phosphatase inhibitor cocktail used was exactly as described previously (42). Antibodies. A monoclonal antibody for SAP155 (10/2-63) was generated by using SAP155 (from large-scale SIT4:HA immunoprecipitates) as the antigen. Antibody 12CA5, against the HA epitope (14), and an antibody against the polyomavirus middle T epitope (15) were from ascites uid. Nucleotide sequence accession numbers. The sequence data reported have been assigned GenBank accession numbers U50560 (SAP155), U50561 (SAP190), and U50562 (SAP4).

SAP155, SAP185, and SAP190 was used as the template DNA for PCR. One sense (5 -ATHMGKAARAAYAAYTCHGAYTAYGA-3 ) and two antisense (5 -ADDNKCCRTTRAADTYTGYTG-3 and 5 -TYDRYMAWRTGDCCCA TRTADCC-3 ) degenerate primers (H represents ACT, D represents AGT, M represents AC, W represents AT, R represents AG, Y represents CT, and K represents GT) were designed on the basis of three of the most conserved regions of the SAP155, SAP185, and SAP190 and used for PCR. Two DNA fragments, of 0 .7 and 0.92 kb, were amplied. The DNA sequences of these two PCR products showed that they resulted from a single gene and that they predicted a protein related to the other SAPs. The 0.92-kb PCR product hybridized to clones 2985 and 3525 on chromosome VII of the Olson library of yeast inserts in lambda phage (30). The full-length SAP4 gene was isolated from both the YCp50 (32) and the YEp24 (3) libraries by colony hybridization using the cloned PCR products as probes. A 5.1-kb ClaI-to-SfcI DNA fragment (from about 0.8 kb upstream of the SAP4 ATG initiation codon to about 1.8 kb downstream of the SAP4 stop codon) was sequenced on both strands. Epitope tagging of the SAPs. For SAP185 and SAP190, a NotI site was created by oligonucleotide-directed mutagenesis (18) immediately upstream of the stop codon (41). For SAP155, a 12-bp NotI linker was inserted into the BsaBI site that occurs 10 bp upstream of the stop codon. A pair of complementary oligonucleotides that encode the HA epitope was inserted in-frame at the NotI site to create SAP190:HA, SAP185:HA, and SAP155:HA. The constructs were conrmed by sequencing and Western blot (immunoblot) analysis. The SAP190:HA, SAP185:HA, and SAP155:HA genes fully complemented their respective SAP functions. Deletion of the SAP genes. For sap190::TRP1, a NotI site was inserted by oligonucleotide-directed mutagenesis just after the translation initiation codon and just before the stop codon, allowing the deletion of DNA sequences for codons 2 through 1033 (of a total of 1,033 codons) and replacement by a 0.83-kb EcoRI-PstI fragment containing the TRP1 gene. For sap185::ADE2, the DNA sequences between the ClaI and NsiI sites containing codons 29 to 1024 (of a total of 1,058 codons) were deleted and replaced with a 2.1-kb BglII fragment containing the ADE2 gene. For sap155::HIS3, a NotI site was inserted just after the translational initiation codon by oligonucleotide-directed mutagenesis, and another was inserted at the BsaBI site 10 bp upstream of the stop codon by NotI linker insertion (see description of epitope tagging of SAP155, above), allowing the deletion of DNA sequences for codons 2 to 997 (of a total of 1,000 codons) and replacement by a 1.8-kb BamHI fragment containing the HIS3 gene. For sap4::LEU2, DNA sequences between the KpnI and SacI sites containing codons 32 to 768 (of a total of 818 codons) were deleted and replaced with a 2.2-kb XhoI-SalI fragment containing the LEU2 gene. All deletions were conrmed by Southern analysis. High-copy-number SAP plasmids. For SAP190, a 4.65-kb PvuII-SacI fragment from 775 to 3880 relative to the A of the ATG initiation codon was cloned into the BamHI site of YEp24. For SAP185, a 5.8-kb FspI-SalI fragment from 1470 to 4342 relative to the A of the ATG initiation codon was cloned into the BamHI site of YEp24. For SAP155, a 6.7-kb XhoI-PstI fragment from 2877 to 3841 relative to the A of the ATG initiation codon was cloned into the BamHI site of YEp24. The SAP155 fragment contains another ORF upstream of the SAP155 ORF. A frameshift mutation introduced at the SalI site (at codon 187 of SAP155) renders the high-copy-number SAP155 plasmid inactive for suppression of the temperature-sensitive phenotype of the sit4-102 strain. The YEp24 plasmid carrying SAP4 is the original library clone containing about 10 kb of insert, including all of the sequenced region of SAP4. Preparation of SIT4:PY. Duplex oligonucleotides were inserted into the NaeI site of the SIT4 gene, changing the carboxyl terminus of the predicted SIT4 protein from . . .RAGYFL to . . .RAGYFLMEYMPMEAGMEYMPMEA. Cells containing polyomavirus epitope-tagged SIT4 (SIT4:PY) as the only source of SIT4 grow slightly more slowly than isogenic SIT4 strains. Construction of a SSD1-v1 W303 strain. Strain CY832 (MATa ssd1::LEU2 W303) was transformed with a mixture of the 6-kb BamHI-SphI yeast DNA fragment containing the entire SSD1-v1 gene (from pCB881 [38]) and a TRP1 CEN plasmid at a molar ratio of 100:1. Trp transformants were screened for those that were also Leu , presumably as a result of a recombination event that replaced the LEU2 marker with the intact SSD1-v1 gene. To test for SSD1-v1, several Trp Leu transformants were crossed to CY1078 (ssd1-d2 W303 sit4::HIS3 plus a SIT4/YCp50 plasmid to maintain viability). After loss of the SIT4/YCp50 plasmid, the diploids were sporulated and dissected. Because sit4 is lethal in the absence of an SSD1-v allele, any diploid that gave viable His ( sit4) haploid progeny must have the SSD1-v1 gene contributed from the Trp Leu parent. Crosses were used to obtain MATa and MAT SSD1-v1 W303 strains. Interestingly, SSD1-v1 W303 strains do not mate and sporulate as well as ssd1-d2 W303 strains. Of the laboratory strains surveyed, about half contain an ssd1-d allele and about half contain an SSD1-v allele (38). Possibly, sequential rounds of mating and sporulation would select for ssd1-d variants. 35 S labeling, immunoprecipitation, and Western blot analysis. 35S labeling of proteins, immunoprecipitation, and Western blot analysis were performed essentially as described previously (38). Phosphatase treatment of immunoprecipitated SAPs. SAP155:HA, SAP185: HA, and SAP190:HA were immunoprecipitated from strains with and without a deletion of SIT4. The SAPs from the sit4 strains were phosphatase treated essentially as described previously (42) except that the reactions were carried out

RESULTS Isolation of the SAP155 and SAP190 genes. The SAP155 and SAP190 proteins were puried by large-scale immunoprecipitation of SIT4:HA followed by SDS-polyacrylamide gel electrophoresis (see Materials and Methods). From amino acid sequences of SAP155 and SAP190 peptides, degenerate oligonucleotides were designed and used for PCR amplication of genomic yeast DNA (see Materials and Methods). The PCR products were from the SAP155 and SAP190 genes because their DNA sequences predicted other sequenced SAP155 and SAP190 peptides that were not used to design the PCR primers. The SAP155 and SAP190 PCR products were used to obtain the SAP155 and SAP190 genes by colony hybridization to libraries of yeast genomic inserts in YCp50 (32) or YEp24 (3). We sequenced the SAP155 gene (see Materials and Methods). The SAP190 gene (ORF SCYKR028w) was sequenced as part of the chromosome XI sequencing project (11). However, our SAP190 clone contains a frameshift at codon 1009 relative to the reported SAP190 sequence (see Materials and Methods). On the basis of our SAP190 DNA sequence, we prepared the SAP190:HA gene, which encodes a SAP190 protein with the HA epitope at the carboxyl terminus. This construct expresses a fusion protein detectable by Western blotting with the anti-HA monoclonal antibody. Interestingly, the predicted SAP155 and SAP190 proteins are similar to each other, with 26% identity over their entire lengths (Fig. 1 and 2). Identication and isolation of the SAP185 gene. Searches of the databases with SAP190 identied an ORF (SCYJL098w) next to BCK1 on chromosome X (ORF J0840 in reference 23) that could encode a protein with 14% identity to SAP155 and 42% identity to SAP190 (Fig. 1 and 2). We isolated this gene, which we call SAP185, and prepared a strain containing a chromosomal deletion of the gene (see Materials and Methods). If SAP185 associates with SIT4, it might be detectable in SIT4 immunoprecipitates. We previously showed that when extracts prepared from cells containing either SIT4 or SIT4:HA were immunoprecipitated with anti-HA antibody 12CA5, SAP155 and SAP190 specically coimmunoprecipitated with SIT4:HA (38). However, the immunoprecipitates contained a 185-kDa protein (called p188 in reference 38) regardless of whether SIT4 or SIT4:HA extracts were used. Moreover, the presence of the 185-kDa band was dependent on the addition of antibody 12CA5. Therefore, we had thought that, unlike SAP155 and SAP190, the 185-kDa protein was not a specic SIT4-associated protein. Interestingly, when SAP185 was deleted, the 185-kDa band was no longer present in 12CA5 immunoprecipitates of either SIT4 or SIT4:HA extracts (data not shown). Moreover, when the SAP185 gene was present on a high-copy-number plasmid, the 185-kDa band increased in levels when the anti-HA antibody 12CA5 was used for immunoprecipitation of either SIT4or SIT4:HA-containing extracts (data not shown). These ndings indicate that SAP185 is recognized by the anti-HA antibody 12CA5. Therefore, to determine if SAP185 associates

VOL. 16, 1996

THE SIT4-ASSOCIATED PROTEIN FAMILY

2747

FIG. 1. The SAPs are a new family of related proteins. Alignment of the predicted SAP protein sequences is shown. First, SAP185 was aligned with SAP190 and SAP155 was aligned with SAP4. Then, the alignment of all four proteins was determined by the alignment of SAP155 with SAP190. Identical amino acid residues are in reverse font. The predicted SAP proteins are based on DNA sequences as described in Materials and Methods. , residue conserved among all four SAPs.

with SIT4, we prepared a strain containing SIT4:PY and used an anti-polyomavirus epitope antibody for the immunoprecipitations (15) (see Materials and Methods). A 185-kDa protein was specically present in the immunoprecipitates when the extracts were prepared from a SIT4:PY strain but not when the extracts were prepared from a SIT4 strain (Fig. 3A, lanes 3 and 4). Moreover, the 185-kDa band was absent from the SIT4:PY immunoprecipitates when the extracts were prepared from strains containing a deletion of SAP185 (lane 5). Therefore, SAP185 specically coimmunoprecipitates with SIT4:PY and is the third high-molecular-mass SAP identied. Identication and isolation of the SAP4 gene. To determine if additional SAP-related genes are present within the S. cerevisiae genome, we designed degenerate oligonucleotide PCR primers based on conserved regions of SAP155, SAP185, and SAP190 (see Materials and Methods). A discrete PCR product was obtained by using genomic DNA from a sap155 sap185 sap190 strain (CY4985) as the template. The translated DNA

sequence of this PCR product predicts a SAP-related protein. The PCR product was used as a probe for colony hybridization to obtain the full-length gene, which we call SAP4 (we do not currently know the apparent molecular mass of SAP4 on SDSpolyacrylamide gels). The sequence of the SAP4 gene predicts an 818-amino-acid protein related to the other three SAPs (Fig. 1 and 2). The SAP genes encode a novel family of proteins whose members specically associate with SIT4. Alignment of the four predicted SAPs shows that they belong to the same family of related proteins. The similarity is moderate in the aminoterminal regions, highest in the central regions, and least in the carboxyl-terminal regions of the proteins (Fig. 1). Moreover, on the basis of their sequence similarity, the SAPs fall into two groups: the SAP4/SAP155 group and the SAP185/SAP190 group (Fig. 1 and 2). Although the SAPs are related to each other, they are not signicantly similar to any other proteins in the current databases. All four SAPs are acidic (Fig. 2A). In

2748

LUKE ET AL.

MOL. CELL. BIOL.

FIG. 2. Molecular data for the SAPs. The molecular data (A), percent identity (B), and similarity tree (C) were calculated on the basis of the protein sequences shown in Fig. 1, using the GeneWorks or IG Suite program from IntelliGenetics, Inc. For the similarity tree (dendrogram), the horizontal distance is proportional to the relative sequence difference (cost of alignment) and was calculated by the GeneWorks program. The vertical distances are arbitrary.

addition, SAP155 has a high asparagine content (Fig. 2A). Interestingly, SAP155, SAP185, and SAP190 migrate on SDS8% polyacrylamide gels as proteins with molecular masses of 155, 185, and 190 kDa, which are larger than their predicted molecular masses of 114.8, 121.3, and 117 kDa, respectively. SAP155, SAP185, and SAP190 are the major proteins that specically coimmunoprecipitate with SIT4:PY (Fig. 3A). SAP155, SAP185, and SAP190 were absent from the immunoprecipitates when the extracts were prepared from a sap155 sap185 sap190 strain (Fig. 3A). We could not detect SAP4 as a discrete band in SIT4:PY immunoprecipitates from 35Slabeled extracts, even when the SAP4 gene was present on a high-copy-number plasmid (Fig. 3B) and the SAP155, SAP185, and SAP190 genes were deleted (Fig. 3A). Therefore, if SAP4 does associate with SIT4, then either SAP4 is a very minor SAP under these growth conditions (minimal medium with 2% glucose) and/or SAP4 comigrates with a nonspecic band so that it is not detectable as a separate band on the gels. We also examined if SAP155, SAP185, and SAP190 associate with the type 1 or the type 2A protein phosphatase catalytic subunit, to which SIT4 is related. Under conditions in which the individual SAPs were detected as intense bands in SIT4 immunoprecipitates (by Western analysis using antibodies directed against the individual SAPs), we were not able to detect SAP155, SAP185, or SAP190 in immunoprecipitates of either epitope-tagged GLC7 (the single type 1 phosphatase catalytic subunit in S. cerevisiae [12, 26, 39]) or epitope-tagged PPH21 or PPH22 (the two type 2A phosphatase catalytic subunits in S. cerevisiae [31, 36]) (data not shown). Therefore, the SAPs specically associate with the SIT4 phosphatase but not with the catalytic subunit of either type 1 or type 2A phosphatase. The SAPs associate with SIT4 in separate complexes. When yeast whole cell extracts are fractionated on gel ltration columns, about one-half (38) and sometimes as much as 90% (20a) of the SIT4 is present in high-molecular-mass SIT4-SAP complexes. The SIT4-SAP155 complex can be partially resolved from the SIT4-SAP190 complex during gel ltration (38). Moreover, we cannot detect any SAP155 in large-scale SAP190:HA immunoprecipitates (data not shown), further indicating that the SAP190-SIT4 complex is separate from the SAP155-SIT4 complex.

FIG. 3. The SAPs associate with SIT4 in separate complexes. (A) Immunoprecipitation of 35S-labeled proteins from strains containing untagged SIT4 (lanes 1 to 3) or SIT4:PY (lanes 4 to 6), using a monoclonal antibody against the polyomavirus epitope. Strains for lanes 3 and 4 have wild-type SAP genes, while those for lanes 1, 2, 5, and 6 lack SAP155, SAP185, and SAP190. The strains for lanes 1 and 6 contain SAP4/YEp24, while the strains for lanes 2 through 5 have YEp24. The strains used were CY5210 (lane 1), CY5206 (lane 2), CY5196 (lane 3), CY5201 (lane 4), CY5211 (lane 5), and CY5215 (lane 6). A 50-kDa protein, which is present at higher levels in lane 4, is present in both the SIT4 and SIT4:PY lanes and varies in amount from lane to lane between experiments. (B) Immunoprecipitation of 35S-labeled proteins from strains containing untagged SIT4 (lanes 1 through 5) or SIT4:PY (lanes 6 through 10), using a monoclonal antibody against the polyomavirus epitope. The strains contained either YEp24 (lanes 5 and 6), SAP190/YEp24 (lanes 4 and 7), SAP185/YEp24 (lanes 3 and 8), SAP155/YEp24 (lanes 2 and 9), or SAP4/YEp24 (lanes 1 and 10) and were transformants of CY5169 (SIT4; lanes 1 through 5) or CY5170 (SIT4:PY; lanes 6 through 10). Why increased levels of SAP185 do not seem to compete with SAP155 and SAP190 for binding to SIT4 is not currently known. For all strains, the growth medium was SD containing tryptophan, adenine, and 2% glucose.

SAP155 and SAP190 are present in excess relative to SIT4. For example, 5 to 10 times more SAP155 was present in the cell extracts compared with the amount of SAP155 that coimmunoprecipitated with SIT4 (data not shown). Therefore, it is possible that in vivo, SAP155 and SAP190 compete with each other for binding to SIT4. Overexpression of either SAP155 or SAP190 caused an increase in the amount of that particular SAP in the SIT4:PY immunoprecipitates (Fig. 3B). Importantly, the increased levels of either SAP155 or SAP190 decreased the levels of the other SAP in the SIT4:PY immunoprecipitates (Fig. 3B). Western analysis of unlabeled extracts showed that overexpression of SAP155 did not decrease the levels of SAP190 and that overexpression of SAP190 did not decrease the levels of SAP155 in the extracts (data not shown). Therefore, SAP155 and SAP190 compete for binding to SIT4. Overexpression of any of the SAPs suppresses the temperature-sensitive phenotype of a sit4-102 strain. The rst indication that the SAPs act positively with SIT4 was that any one of the four SAP genes (SAP4, SAP155, SAP185, or SAP190),

VOL. 16, 1996

THE SIT4-ASSOCIATED PROTEIN FAMILY

2749

FIG. 4. Effects due to the loss or gain of SAP function on growth rates. (A and B) Strains of the indicated genotypes were streaked either onto YPD plates (from a fresh streakout of the same strains on a YPD plate) or onto YPGE plates (from a fresh streakout of the same strains on a YPGE plate) and grown for the indicated number of days (D) at 30 C. For panel A, the times (1.5 days for YPD plates and 2.7 days for YPGE plates) were chosen so that the colony sizes of the wild-type (wt) strain were about the same on the YPGE plates as on the YPD plates. For panel B, the time (2.2 days) was chosen to compare growth of the sap185 sap190 strain on the YPD plate with that on the YPGE plate. The strains for panels A and B were all SSD1-v1 W303: CY4029 (wild type) CY5218 ( sap155), CY4917 ( sap185), CY4380 ( sap190), CY5224 ( sap185 sap190), CY5220 ( sap155 sap4), CY5217 ( sap4), CY3938 ( sit4), CY5236 ( sap155 sap185 sap190 sap4), and CY5234 ( sap155 sap185 sap190). (C) Strain CY144 (sit4-102 ssd1-d2) was transformed with either YEp24 or the indicated SAP genes on YEp24. The transformants were patched onto a SC-uracil plate, grown overnight at 24 C, and replica plated onto a YPD plate. Growth is shown after 1.5 days at 37 C.

when present on a high-copy-number plasmid, partially suppressed the temperature-sensitive phenotype of a sit4-102 mutant (Fig. 4C). Therefore, even though we have not yet been able to detect physical association of SAP4 with SIT4 in 35Slabeled extracts, high-copy-number SAP4 is equivalent to highcopy-number SAP155, SAP185, and SAP190 in terms of suppression of the sit4-102 mutation. Loss of the SAP genes causes either a slow growth rate or lethality. The second indication that the SAPs act positively with SIT4 was that, like cells without SIT4, cells without the SAP genes either had a slow growth rate or were inviable. The effects due to sit4 mutations are dependent on the SSD1 gene.

In ssd1-d strains, deletion of SIT4 is lethal and sit4-102 mutants are temperature sensitive and arrest in late G1. In SSD1-v strains, sit4 mutants are viable but grow very slowly and sit4-102 mutants are not temperature sensitive for growth (38). Because of the genetic interactions between SIT4 and SSD1, it is important to determine the phenotypes of the sap mutants in both SSD1-v and ssd1-d backgrounds. We prepared isogenic MATa and MAT SSD1-v1 derivatives of strain W303 (40), which normally has an ssd1-d2 allele (38) (see Materials and Methods) (interestingly, the SSD1-v1 W303 strains do not mate and sporulate as well as ssd1-d2 W303 strains). Strains containing deletions of the SAP genes and/or the SIT4 gene

2750

LUKE ET AL.

MOL. CELL. BIOL.

were prepared: one set in the SSD1-v1 W303 background and a second set in the ssd1-d2 W303 background. In the SSD1-v1 background, deletion of SAP185 alone or SAP190 alone caused no signicant growth rate defect on either YPD or YPGE (GE represents 2% glycerol plus 2% ethanol) medium (Fig. 4A). However, cells with deletions of both SAP185 and SAP190 had a moderate slow-growth-rate phenotype on YPD medium (Fig. 4A). This nding suggests that SAP185 and SAP190 have full or nearly full functional overlap and that their combined function is important when the cells are grown on YPD medium. By contrast, on YPGE medium, the sap185 sap190 cells had only a slight defect in growth rate (Fig. 4A and B). In fact, in contrast to wild-type cells, the sap185 sap190 cells grew almost as fast on the YPGE plates as on the YPD plates (Fig. 4B). Therefore, SAP185 and SAP190 are especially important when the cells are grown with glucose as the carbon source. Deletion of SAP155 (in the SSD1-v1 background) caused a slight to moderate growth rate defect on both YPD and YPGE media (Fig. 4A). Combining a deletion of SAP155 with a deletion of either SAP185 or SAP190 caused a slightly slower growth rate than deletion of SAP155 alone, but the effect was not as severe as deletion of both SAP185 and SAP190 (data not shown). Therefore, either the function of SAP155 overlaps slightly with that of SAP185 and SAP190 or the cells require full SAP185/SAP190 activity when SAP155 is absent. In addition, on either YPD or YPGE medium, deletion of SAP4 did not alter the growth rate of an otherwise wild-type strain (Fig. 4A) or of sap155 strains and sap185 sap190 strains (data not shown). However, the loss of SAP4 did cause a sap155 strain (but not a sap185 sap190 strain) to grow more slowly on SC-glycerol-ethanol (SC-GE) medium (data not shown). This nding suggests that, consistent with the sequence similarity, the function of SAP4 may be closer to that of SAP155 than to that of SAP185 or SAP190. The quadruple sap mutant ( sap4 sap155 sap185 sap190 SSD1-v1) grew as slowly as the isogenic sit4 mutant on either YPD or YPGE plates (Fig. 4A). This nding is consistent with the idea that the SIT4 pathway is nonfunctional in the absence of all four SAPs. When SAP4 was the only intact SAP gene, a slightly different result was obtained: the strain ( sap155 sap185 sap190 SSD1-v1) grew nearly as slowly as the sit4 mutant on YPD medium but signicantly faster than the sit4 mutant on YPGE medium (Fig. 4A). Therefore, as assayed in the absence of SAP155, SAP185, and SAP190, SAP4 is more important when the cells are grown on glycerol-ethanol medium than when they are grown on glucose medium. In the ssd1-d2 W303 background, deletion of SIT4 is lethal (38). Signicantly, deletion of the three major SAP genes (SAP155, SAP185, and SAP190) or deletion of all four SAP genes (SAP4, SAP155, SAP185, and SAP190) was also lethal in the ssd1-d2 background (data not shown). Cells containing any other combination of sap deletions in the ssd1-d2 background were viable and grew at a rate similar to that of cells with the same combination of sap deletions in the SSD1-v1 background (data not shown). In summary, in both the ssd1-d2 and SSD1-v1 backgrounds, loss of all SAP function is equivalent to loss of SIT4 function by growth rate and viability tests. Deletion of the SAP genes causes a G1 delay. Compared with wild-type SIT4 cells, sit4 SSD1-v1 cells have a defect in CLN1 and CLN2 G1 cyclin accumulation, spend a longer time in G1, and have a delay in the initiation of DNA replication and bud formation (13). For exponentially growing wild-type cultures, most of the cells have a 2n DNA content, about 80% of the cells are budded, and most of the 1n cells are small (Fig. 5), indicating that the G1 phase is short relative to the other stages

FIG. 5. Effects due to the loss of SAP function on percentage budded cells, cell size, and DNA content. Strains of the indicated genotypes, which were the same as those used for Fig. 4, were grown to exponential phase in SC-glucose medium and prepared for ow cytometry analysis as described previously (13). The percentage of budded cells (upper right corner in the panels on the left) was determined, by counting at least 400 cells for each of three independent cultures, from the sample of cells used for the ow cytometry. For the plots on the left, cell count (vertical axis) is plotted versus uorescence intensity, which is proportional to DNA content (horizontal axis). For the plots on the right, the vertical axis is DNA content, the horizontal axis is scattering, which is proportional to cell size, and the intensity and density of the datum points are proportional to the cell count.

of the cell cycle. By contrast, sit4 SSD1-v1 cells are delayed in G1: there are more 1n than 2n cells, there is a marked reduction in the percentage of budded cells, and there are many very large 1n cells (Fig. 5). The large 1n sit4 SSD1-v1 cells are most likely from two sources: (i) the large 1n mother cells that have just separated from the daughter cell and (ii) the new daughter cells that grow to a large cell size (as determined by microscopic analysis of pedigrees [13]) before they form a bud and replicate their DNA. As the SAP genes were progressively removed, the cells were increasingly delayed in G1 (Fig. 5). Importantly, the loss of all four SAP genes gave rise to similar increases in 1n cells relative to 2n cells, similar reductions in the percentage of budded

VOL. 16, 1996 TABLE 2. Sensitivity of a sap155 sap185 sap190 SSD1-v1 strain to the loss of CLN3, SIS2, or BEM2a
Growth rate of strain with genotype: Mutation SAP SIT4 sap155 sap185 sap190 sap155 sap185 sap190 sit4

THE SIT4-ASSOCIATED PROTEIN FAMILY

2751

cln3 sis2 bem2

Wild type Wild type Moderatec

Slightly slower Slower Slightly slower

Slightly slower Slower Lethald

Lethalb Lethal Lethale

Lethalb Lethal Lethal

a The effects due to the loss of CLN3, SIS2, or BEM2 on the growth rates of strains without the indicated SAP genes were determined. Slightly slower indicates that the particular sap segregants grew very slightly but detectably more slowly if they also had a cln3 or bem2 mutation. The strains used for the crosses were CY4985 and CY4102 for the cln3 cross, ML58-4B and CY4987 for the sis2 cross, and CY4985 and ML59-15A for the bem2 cross. All strains were SSD1-v. b Of the 18 sap155 sap185 sap190 cln3 segregants, 13 were inviable and 5 grew extremely slowly. Of the 40 sit4 cln3 segregants, 22 were inviable and 18 grew extremely slowly. c The growth rate of a bem2 strain is dependent on SSD1. In ssd1-d backgrounds, bem2 causes a slow growth rate. In SSD1-v backgrounds, bem2 causes a small defect in the growth rate. d Of the ve sap185 sap190 bem2 segregants, four were inviable and one grew extremely slowly and papillated to faster growth. e Of the 12 sap155 sap185 sap190 bem2 segregants, 11 were inviable and 1 grew extremely slowly.

cells, and as many large 1n cells as seen for the sit4 SSD1-v1 cells (Fig. 5). In summary, loss of all SAP function causes a G1 delay very similar to that caused by the loss of SIT4. A strain without the SAP genes is sensitive to the loss of CLN3, SIS2, or BEM2. Like a strain without SIT4, a strain without the SAP genes is sensitive to the loss of CLN3, SIS2, or BEM2. We have previously shown that SIT4 is required for the normal expression of the CLN1 and CLN2 G1 cyclins (13). CLN3 also provides a pathway for activating the expression of CLN1 and CLN2. Cells containing a deletion of CLN3 have a normal growth rate, but they accumulate CLN1 and CLN2 RNAs at slower than normal rates and initiate bud formation and DNA synthesis at a larger than normal cell size (9, 41). CLN3 and SIT4 provide additive pathways for CLN1 and CLN2 expression (13), so that cln3 sit4 SSD1-v1 segregants from a cross of a cln3 strain to a sit4 strain either are inviable or grow extremely slowly (1 week to form a very small colony). Moreover, the synthetic phenotypes resulting from combining the loss of CLN3 with the loss of SIT4 is due to low G1 cyclin levels because it is cured by expressing CLN2 at near wild-type levels from a heterologous SIT4-independent promoter (Schizosaccharomyces pombe ADH:CLN2 [13]). From a cross of a sap155 sap185 sap190 SSD1-v1 strain to a cln3 SSD1-v1 strain, the sap155 sap185, the sap155 sap190, and the sap185 sap190 segregants were not very sensitive to the loss of CLN3. However, the sap155 sap185 sap190 cln3 SSD1-v1 segregants either were inviable or grew extremely slowly (Table 2). Moreover, the synthetic phenotypes resulting from combining the loss of CLN3 with the loss of the three major SAP genes is due to low G1 cyclin levels because it was cured by expressing CLN2 from a heterologous SIT4-independent promoter (S. pombe ADH:CLN2) (data not shown). Therefore, by this test of a requirement for G1 cyclin expression, the loss of the three major SAPs is similar to the loss of SIT4. In sit4 SSD1-v strains, overexpression of SIS2 stimulates the growth rate and increases the expression of SWI4, CLN1, and CLN2 (8). Importantly, high-copy-number SIS2 increased

the growth rate of a sap4 sap155 sap185 sap190 SSD1-v1 strain to the same extent as it did for a sit4 SSD1-v1 strain (data not shown). Moreover, sis2 sit4 SSD1-v1 cells are inviable (8). In a cross of a sap155 sap185 sap190 SSD1-v1 strain to a sis2 SSD1-v1 strain (which has a wild-type growth rate [8]), all seven of the predicted sis2 sap155 sap185 sap190 segregants were inviable (Table 2). Even the sap155 sap185, the sap155 sap190, and the sap190 strains grew signicantly more slowly in the absence of SIS2. Therefore, like cells without SIT4, cells without the SAPs are very sensitive to the loss of SIS2, and their growth rate defect is partially suppressed by overexpression of SIS2. In addition to G1 cyclin expression, SIT4 is also required during late G1 for bud formation (13). Moreover, sit4 SSD1-v1 cells are inviable in the absence of BEM2 (10, 17). BEM2 is involved in bud emergence, but cells are normally viable in the absence of BEM2 (2, 16). Signicantly, like cells without SIT4, cells without SAP155, SAP185, and SAP190 are inviable in the absence of BEM2 (Table 2). Therefore, the loss of SAP function is equivalent to the loss of SIT4 with respect to the genetic interactions with BEM2. The loss of BEM2 in combination with only certain deletions of the SAP genes is discussed below. In summary, these genetic interactions with CLN3, SIS2, and BEM2 further demonstrate that the SAPs function positively with SIT4. In vivo, SAP function requires SIT4, and conversely, SIT4 function requires the SAPs. The growth rate of a sap4 sap155 sap185 sap190 strain on YPD medium was not stimulated by overexpression of SIT4 (from its own promoter or the ADH1 promoter, both on a high-copy-number plasmid) (data not shown). In addition, overexpression of either SAP155, SAP185, or SAP190 (the given SAP gene on a highcopy-number plasmid) did not increase the growth rate of a sit4 SSD1-v1 strain on YPD medium (data not shown). When deletions of SAP155, SAP185, and SAP190 were combined with a deletion of SIT4 (when SAP4 was present), the strain grew at the same rate as a sit4 SSD1-v1 strain (on either YPD or YPGE medium) (data not shown). Therefore, by this test, SAP155, SAP185, and SAP190 do not have important functions in the absence of SIT4. In addition, this nding shows that SAP4 is not able to stimulate the growth rate and may be nonfunctional in the absence of SIT4. Further evidence suggesting that SIT4 combines with the SAPs to provide a function not provided by either SIT4 alone or the SAPs alone is from the effects due to the overexpression of both SIT4 and either SAP155 or SAP190 (all expressed from the ADH1 promoter on a 2 m plasmid). Overexpression of either SIT4 alone, SAP155 alone, or SAP190 alone caused either no effect or a very slight slow-growth-rate phenotype (data not shown). By contrast, overexpression of both SIT4 and SAP155 or of both SIT4 and SAP190 caused a very strong defect in the growth rate (data not shown). In summary, these ndings suggest that the function of SIT4 requires the SAPs and that, in turn, the functions of the SAPs require SIT4. The SAPs provide distinct functions. By sequence similarity, the SAPs fall into two groups: the SAP4/SAP155 group and the SAP185/SAP190 group (Fig. 1 and 2). The rst indication that the SAPs provide distinct functions was the effect due to the loss of certain SAP combinations on the growth rates of the cells (Fig. 4). Another indication that the SAPs provide distinct functions was the sensitivity of a bem2 strain to the loss of SAP function. In 55 tetrads from a cross of a sap155 sap185 sap190 SSD1-v1 strain to a bem2 SSD1-v1 strain, the sap155, the sap155 sap185 (SAP190 only, not considering SAP4), and the sap155 sap190 (SAP185 only) segregants grew only slightly more slowly if they also had a deletion of

2752

LUKE ET AL.

MOL. CELL. BIOL.

FIG. 6. SAP185 and SAP190 provide a function that is distinct from that provided by SAP155. Strain CY5001 ( sap185 sap190) (A) and strain CY5254 ( sap155) (B) were transformed with either YEp24 (control) or the indicated SAP gene on YEp24 (HC# SAP; see Materials and Methods for descriptions of plasmids). The transformants were streaked onto plates containing SC medium without uracil and with 2% glucose and grown for 2 days at 30 C.

FIG. 7. The SAPs are hyperphosphorylated in the absence of SIT4. (A) Western analysis of SAP155, SAP185:HA, and SAP190:HA in extracts prepared from isogenic wild-type SIT4 (WT) or sit4 ( ) SSD1-v1 strains grown in SC medium without uracil and with 2% glucose. The strains were CY4796 ( sit4) and CY4802 (SIT4) for the SAP190:HA samples and CY5067 (SIT4) and CY5077 ( sit4) for the SAP185:HA and SAP155 samples. Three lanes were included for the SAP190 samples to demonstrate the small mobility change. Interestingly, the levels of SAP155 were about threefold higher and the levels of SAP155 RNA (which do not vary in the cell cycle) were about two- to threefold higher in sit4 SSD1-v1 strains than in isogenic SIT4 strains (data not shown). Perhaps the cells detect the loss of SIT4 and induce the expression of SAP155. The SAP185:HA and SAP190:HA blots were probed with monoclonal antibody 12CA5 (anti-HA ascites uid), and the SAP155 blot was probed with an antiSAP155 monoclonal antibody. (B) SAP185:HA was immunoprecipitated from either a wild-type SIT4 strain (CY5067; lane 1) or a sit4 strain (CY5077; lanes 2 to 5). For lanes 1 and 2, the immune complexes were kept on ice without further treatment. For lanes 3 through 5, the immune complexes were treated for 15 min at 30 C with either calf alkaline phosphatase (lane 3), a phosphatase inhibitors mix plus the calf alkaline phosphatase (lane 4), or the phosphatase inhibitors mix only (lane 5) as described in Materials and Methods.

BEM2 (Table 2 and data not shown). By contrast, of the ve predicted sap185 sap190 bem2 segregants, four were inviable and one grew extremely slowly and papillated to faster growth (Table 2). Therefore, the function of SIT4 that is essential in the absence of BEM2 requires either SAP185 or SAP190. This function of SIT4 cannot be provided by SAP155. These effects are not simply due to the difference in the quantity of the remaining SAP activity. We determined if overexpression of a SAP from one group could cure the growth defects due to the absence of the other group. The growth rate of a sap185 sap190 strain was not stimulated by either the SAP155 gene or the SAP4 gene on a high-copy-number plasmid (Fig. 6A). By contrast, either the SAP185 or the SAP190 gene on a high-copy-number (Fig. 6A) or a low-copy-number CEN plasmid (data not shown) caused the sap185 sap190 strain to grow like wild-type cells. In the strains containing high-copy-number SAP155, SAP155 was overexpressed at least 10-fold (as determined by Western analysis of serial dilutions of extracts [data not shown]). Therefore, overexpression of SAP155 cannot compensate for the loss of SAP185/SAP190 function. Similarly, neither the SAP185 gene nor the SAP190 gene on a high-copy-number plasmid stimulated the growth rate of a sap155 strain (Fig. 6B). Although the sequence of SAP4 is more related to that of SAP155 and the loss of SAP4 caused a sap155 strain to grow more slowly on SC-GE medium, highcopy-number SAP4 was not able to stimulate the growth rate of

a sap155 strain (Fig. 6B). This nding suggests either that SAP4 cannot substitute for SAP155 or that high-copy-number SAP4 does not provide enough SAP155-like function to be detected by this assay. In summary, SAP155 provides a cellular function that cannot be provided by either SAP185 or SAP190 and the SAP185/SAP190 group provides a cellular function that cannot be provided by SAP155. The SAPs are hyperphosphorylated in the absence of SIT4. We had previously shown that the SAPs which coimmunoprecipitate with SIT4 are phosphoproteins (38). With reagents to detect the SAPs independently of SIT4, we can now address whether the presence or absence of SIT4 affects the phosphorylation state of the SAPs. The SAP155, SAP185:HA, and SAP190:HA in extracts from sit4 SSD1-v1 strains all have forms that migrate more slowly on SDS-polyacrylamide gels than the proteins from isogenic SIT4 strains (Fig. 7A). Moreover, the more slowly migrating forms of the SAP185:HA protein (Fig. 7B) and the SAP155:HA and SAP190:HA proteins (data not shown) were converted to a faster-migrating form after treatment with alkaline phosphatase, indicating that the reduced mobility of the SAPs from sit4 cells is due to increased phosphorylation. In summary, the loss of SIT4 causes SAP155, SAP185, and SAP190 to become hyperphosphorylated. In experiments in which we added SIT4 (from either a wild-type strain or a strain lacking SAP155, SAP185, and SAP190) to the hyperphosphorylated SAPs obtained from a sit4 strain, we were not able to detect dephosphorylation of the SAPs by SIT4 in vitro (data not shown). However, in these

VOL. 16, 1996

THE SIT4-ASSOCIATED PROTEIN FAMILY

2753

experiments, the hyperphosphorylated SAPs did not bind to SIT4, so that whether SIT4 can dephosphorylate the SAPs in vivo remains unknown. DISCUSSION The SIT4 protein phosphatase has a unique subunit composition. SIT4 is about 55% identical to the catalytic subunit of type 2A phosphatases and about 45% identical to the catalytic subunit of type 1 phosphatases (1). In side-by-side comparisons of immunoprecipitates of the S. cerevisiae SIT4, type 1 (GLC7), and type 2A (both PPH21 and PPH22) phosphatases, the in vitro activity (toward lysozyme [Sigma L 1526] phosphorylated by protein kinase A) of SIT4 was at least as sensitive to low [nanomolar] concentrations of okadaic acid as the PPH21 and PPH22 type 2A phosphatases (20a). Therefore, by protein sequence and okadaic acid sensitivity, SIT4 is a type 2A-like phosphatase. However, SIT4 is clearly distinct from the type 2A phosphatases. First, SIT4 provides an essential (in ssd1-d backgrounds) or very important (in SSD1-v1 backgrounds) cellular function in late G1 that is distinct from the essential functions of the budding yeast type 2A phosphatases (19, 31, 36). Second, SIT4 differs from the type 2A phosphatases in subunit composition. In S. cerevisiae, as in other organisms (4), the type 2A catalytic subunits (PPH21 and PPH22) associate with two regulatory subunits, an A subunit (a 65-kDa protein encoded by TPD3 [43]) and a B subunit (a 60-kDa protein encoded by CDC55 [16]). Although we are able to detect TPD3 and CDC55 as very intense bands in Western analysis of either PPH21:HA or PPH22:HA immunoprecipitates, we have not been able to detect TPD3 or CDC55 in SIT4:HA immunoprecipitates (7a). Instead, SIT4:HA associates with at least three SAPs with apparent molecular masses of 155, 185, and 190 kDa. Moreover, we have not been able to detect the SAPs in immunoprecipitates of HA-tagged type 2A (or type 1) catalytic subunits. Therefore, SIT4 associates with the SAPs whereas the type 2A catalytic subunits associate with the TPD3 and CDC55 subunits. SIT4 homologs have been found and characterized in S. pombe and in Drosophila melanogaster (21, 22, 35). These ndings suggest that, like the conserved type 1 and type 2A protein phosphatases, SIT4 and the SIT4 homologs dene a unique class of protein phosphatase catalytic subunit that are conserved during evolution. In this report, we identied the highmolecular-weight SAPs as a new protein family whose members uniquely associate with SIT4. We suggest that the SAPs are also a conserved class of proteins whose members bind to and function positively with the SIT4 homologs in other eukaryotic organisms. Physical association of the SAPs with SIT4. Previously, we showed that SAP155 and SAP190 associate with SIT4 in a cell cycle-dependent manner (38). Reexamination of the data of Sutton et al. (38) shows that SAP185, which we thought at that time was a nonspecic band, also associates with SIT4 in a cell cycle-dependent manner. Several pieces of evidence show that the SAPs are in separate complexes with SIT4: (i) the SAP190SIT4 complexes can be partially resolved from the SAP155SIT4 complexes on gel ltration columns (38), (ii) we cannot detect any SAP155 in SAP190:HA immunoprecipitates (data not shown), (iii) the binding of SAP155 and SAP185 to SIT4 does not require the presence of SAP190 (data not shown), and (iv) SAP155 and SAP190 compete with each other for binding to SIT4 (Fig. 3B). These ndings suggest that there is probably a single SAP binding site on SIT4, and because of the similarity of the SAPs to one another, there may be common sequences within the SAPs that specify binding to SIT4.

The SAPs function positively with SIT4. Any one of the four SAP genes in high copy number can suppress the temperaturesensitive phenotype of sit4-102 strains. In addition, the loss of SAP function is, by every test that we have done, equivalent to the loss of SIT4. These ndings suggest that the SAPs function positively with SIT4 and that the four SAPs together are involved in the same set of cellular functions as SIT4. Moreover, the SAPs and SIT4 require each other to function in vivo. Although overexpression of any SAP can suppress the temperature-sensitive phenotype of sit4-102 strains, it cannot increase the growth rate of a sit4 SSD1-v1 strain. Similarly, overexpression of SIT4 cannot suppress the defects due to loss of the SAPs (either the loss of SAP155 alone or loss of both SAP185 and SAP190, of SAP155, SAP185, and SAP190, or of all four SAPs). Therefore, overexpression of SIT4 cannot bypass the requirement for the SAPs and likewise, overexpression of the SAPs cannot bypass the requirement for SIT4. Furthermore, combining a deletion of SIT4 with deletions of the three major SAP genes (SAP155, SAP185, and SAP190) is equivalent to the deletion of SIT4 alone. By these criteria, the SAPs do not have important functions in addition to their functions with SIT4, and SIT4 does not have signicant functions, except possibly those with SAP4, in addition to its functions with SAP155, SAP185, and SAP190. The role of SAP4. We have not yet been able to detect SAP4 as a discrete band in SIT4 immunoprecipitates of 35S-labeled extracts. However, the genetic interactions of SAP4 suggest that it functions as a SAP. High-copy-number SAP4 partially suppresses the temperature-sensitive phenotype of a sit4-102 strain. Also, in the absence of SAP155, SAP185, and SAP190, SAP4 stimulates the growth rate slightly on YPD medium and moderately on YPGE medium (Fig. 4A). Moreover, this ability of SAP4 to stimulate the growth rate is dependent on SIT4 because a sit4 SAP4 sap155 sap185 sap190 SSD1-v1 strain grows at the same rate as either a sit4 SSD1-v1 strain or a sap4 sap155 sap185 sap190 SSD1-v1 strain. Together, these ndings suggest that SAP4 functions as a SIT4-associated protein. The SAPs provide at least two distinct cellular functions. As mentioned above, the SAPs together are involved in most or all of the functions that are provided by SIT4. The combined loss of the SAP genes, like the loss of SIT4, causes either lethality or a slow growth rate with a G1 delay. In addition, the loss of SAP155, SAP185, and SAP190 is very similar to the loss of SIT4 in terms of the genetic interactions with the loss of either CLN3, SIS2, or BEM2. The genetic interactions with CLN3 show that, like SIT4, the SAPs are required for normal G1 cyclin expression. However, some of the functions of SIT4 are provided uniquely via SAP155 (function B in Fig. 8A) because the loss of SAP155 causes a growth defect that is not cured by overexpressing either SAP185 or SAP190. SAP185 and SAP190 probably have signicant or total functional overlap: deletion of either SAP185 or SAP190 does not cause a detectable growth rate defect, but deletion of both SAP185 and SAP190 causes a signicant growth rate defect. Moreover, this growth rate defect of sap185 sap190 strains is not cured by the overexpression of SAP155. Therefore, SAP185 and SAP190 together provide a unique SIT4-dependent function (function A in Fig. 8A). This SAP185-SAP190 function is required for the SIT4 activity that is essential in the absence of BEM2 (Table 2), raising the possibility that the SAP185/SAP190-dependent SIT4 function is involved, directly or indirectly, in budding. The loss of SAP155 combined with the loss of SAP185 and SAP190 causes inviability in ssd1-d backgrounds and a very slow growth rate in SSD1-v backgrounds (an effect more severe

2754

LUKE ET AL.

MOL. CELL. BIOL.

FIG. 8. Models of SIT4 and SAP function. (A) Different functions provided by the various SAPs. (B) SIT4 as a regulator of the SAPs (model 1) versus the SAPs as regulators of SIT4 (model 2).

than either deletion of SAP155 alone or deletion of both SAP185 and SAP190). This effect can be explained by one of two possibilities. For one possibility, SAP155 has no functional overlap with SAP185 and SAP190. SAP155 provides only function B, while SAP185 and SAP190 together provide only function A. The synthetic growth defect or lethality of combining the loss of SAP185/SAP190 with the loss of SAP155 is due only to the loss of both functions A and B. In this case, if function A and function B are in separate processes, then the abnormal operation of both processes resulting from the absence of both functions A and B causes ssd1-d cells to be inviable. Alternatively, function A and function B may be required at two different steps within the same process, so that in the absence of both functions, this single process is so defective that ssd1-d cells are inviable. For the second possibility, which is difcult to exclude, SAP155 shares a common function (function C in Fig. 8A) with SAP185 and SAP190. This common function would be distinct from functions A and B. The nding that a sap155 strain (but not a SAP155 strain) grows slightly more slowly in the absence of either SAP185 or SAP190 could be either due to SAP155 having a slight functional overlap with SAP185 and SAP190 or due to a sap155 strain having a greater requirement than a SAP155 strain for a fully active A function. If such a common function exists, then the synthetic defects due to the absence of SAP155, SAP185, and SAP190 could be due to the loss of function C alone (and unrelated to the loss of functions A and B) or due to any combination of losses of functions A, B, and/or C. Do the SAPs activate SIT4 or does SIT4 activate the SAPs? We envision two models for the function of the SAPs relative to SIT4. In model 1 (Fig. 8B), SIT4 regulates the SAPs. In this model, the SAPs have a different activity, localization, or function when they are activated by SIT4, either via association into a complex with SIT4 or via dephosphorylation by SIT4. In model 2, the SAPs regulate SIT4. In this model, the SAPs

could either physically target SIT4 to the specic SIT4 substrates (similar to the glycogen-targeting subunit of type 1 phosphatases [4]) or regulate the phosphatase activity of SIT4 so that SIT4 is much more active for dephosphorylating the specic SIT4 substrates (which are not the SAPs themselves). For the latter case, SIT4 would be a SAP-dependent phosphatase for the in vivo substrates, similar to the cyclin-dependent kinases such as CDC28. Both SIT4 and CDC28 are present at relatively constant levels throughout the cell cycle, but their activities may require the association with other proteins, and these associations are regulated in the cell cycle. That the SAPs are hyperphosphorylated in the absence of SIT4 is consistent with either model. In model 1, the SAPs would be the relevant physiological substrates of SIT4 and become hyperphosphorylated and inactive in the absence of SIT4. In model 2, the SAPs regulate the activity of SIT4 toward other substrates. However, the SAPs might become hyperphosphorylated either because sit4 cells spend a longer time in G1 (when SAP kinases might be more active) or because they do not undergo their normal cell cycle-regulated association with SIT4. Distinguishing between these models will require the determination of the downstream processes or components that are either regulated via dephosphorylation by SIT4 or regulated by the SAPs. For either model, the activity of the SAPs or the activity of SIT4 would be regulated by the association of the SAPs with SIT4. We previously showed that while the levels of SIT4 are constant during the cell cycle, the association of SIT4 with the SAPs is regulated during the cell cycle (38). Moreover, our initial characterizations show that the amount of the SAPs that are associated with SIT4 is regulated by cellular growth signals: by carbon source (glycerol-ethanol versus glucose), by the presence or absence of amino acids in the growth medium, and during saturation of the cultures (20b). Possibly, the SAPs are involved in transducing nutrient growth signals via their association with SIT4, linking growth signals with G1 cyclin expression, bud formation, and other late G1 processes. The elucidation of such a signal transduction pathway will require the determination of how the association of the SAPs with SIT4 is regulated and what downstream processes are regulated by the SAPs and SIT4.
ACKNOWLEDGMENTS We thank P. Burfeind for ow cytometry analysis, C. Bautista, and M. Falkowski for monoclonal antibodies, M. Christman for the cell line expressing the anti-PY antibody, J. Thorner for the OCR roman font used in Fig. 1, B. Osmond for plasmid pCS3-1, and A. Doseff, B. Futcher, D. Shevell, and A. Sutton for comments on the manuscript. This work was supported by Human Frontier postdoctoral fellowship LT-268/92 to F.D.S. and NIH grant GM45179 to K.T.A. M.M.L. and F.D.S. contributed equally to this work.
REFERENCES 1. Arndt, K. T., C. A. Styles, and G. R. Fink. 1989. A suppressor of a HIS4 transcriptional defect encodes a protein with homology to the catalytic subunit of protein phosphatases. Cell 56:527537. 2. Bender, A., and J. R. Pringle. 1991. Use of a screen for synthetic lethal and multicopy suppressee mutants to identify two new genes involved in morphogenesis in Saccharomyces cerevisiae. Mol. Cell. Biol. 11:12951305. 3. Carlson, M., and D. Botstein. 1982. Two differentially regulated mRNAs with different 5 ends encode secreted and intracellular forms of yeast invertase. Cell 28:145154. 4. Cohen, P. 1989. The structure and regulation of protein phosphatases. Annu. Rev. Biochem. 58:453508. 5. Collins, K., R. Kobayashi, and C. W. Greider. 1995. Purication of Tetrahymena telomerase and cloning of genes encoding the two protein components of the enzyme. Cell 81:677686. 6. Cross, F. R. 1990. Cell cycle arrest caused by CLN gene deciency in Saccharomyces cerevisiae resembles START-I arrest and is independent of

VOL. 16, 1996

the mating-pheromone signalling pathway. Mol. Cell. Biol. 10:64826490. 7. Cross, F. R., M. Hoek, J. D. McKinney, and A. H. Tinkelenberg. 1994. Role of Swi4 in cell cycle regulation of CLN2 expression. Mol. Cell. Biol. 14:4779 4787. 7a.Di Como, C. J. Unpublished results. 8. Di Como, C. J., R. Bose, and K. T. Arndt. 1995. Overexpression of SIS2, which contains an extremely acidic region, increases the expression of SWI4, CLN1, and CLN2 in sit4 mutants. Genetics 139:95107. 9. Dirick, L., T. Bohm, and K. Nasmyth. 1995. Roles and regulation of ClnCdc28 kinases at the start of the cell cycle of Saccharomyces cerevisiae. EMBO J. 14:48034813. 10. Doseff, A. I., and K. T. Arndt. 1995. LAS1 is an essential nuclear protein involved in cell morphogenesis and cell surface growth. Genetics 141:857 871. 11. Dujon, B., D. Alexandraki, B. Andre, W. Ansorge, V. Baladron, J. P. Ballesta, A. Banrevi, P. A. Bolle, M. Bolotin-Fukuhara, P. Bossier, et al. 1994. Complete DNA sequence of yeast chromosome XI. Nature (London) 369:371 378. 12. Feng, Z. H., S. E. Wilson, Z. Y. Peng, K. K. Schlender, E. M. Reimann, and R. J. Trumbly. 1991. The yeast GLC7 gene required for glycogen accumulation encodes a type 1 protein phosphatase. J. Biol. Chem. 266:23796 23801. 13. Fernandez-Sarabia, M. J., A. Sutton, T. Zhong, and K. T. Arndt. 1992. SIT4 protein phosphatase is required for the normal accumulation of SWI4, CLN1, CLN2, and HCS26 RNAs during late G1. Genes Dev. 6:24172428. 14. Field, J., J. Nikawa, D. Broek, B. MacDonald, L. Rodgers, I. A. Wilson, R. A. Lerner, and M. Wigler. 1988. Purication of a RAS-responsive adenylyl cyclase complex from Saccharomyces cerevisiae by use of an epitope addition method. Mol. Cell. Biol. 8:21592165. 15. Grussenmeyer, T., K. H. Scheidtmann, M. A. Hutchinson, W. Eckhart, and G. Walter. 1985. Complexes of polyoma virus medium T antigen and cellular proteins. Proc. Natl. Acad. Sci. USA 82:79527954. 16. Healy, A. M., S. Zolnierowicz, A. E. Stapleton, M. Goebl, A. A. DePaoliRoach, and J. R. Pringle. 1991. CDC55, a Saccharomyces cerevisiae gene involved in cellular morphogenesis: identication, characterization, and homology to the B subunit of mammalian type 2A protein phosphatase. Mol. Cell. Biol. 11:57675780. 17. Kim, Y.-J., L. Francisco, G.-C. Chen, E. Marcotte, and C. S. M. Chan. 1994. Control of cellular morphogenesis by the Ipl2/Bem2 GTPase-activating protein: possible role of protein phosphorylation. J. Cell Biol. 127:13811394. 18. Kunkel, T. A. 1985. Rapid and efcient site-specic mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82:488492. 19. Lin, F. C., and K. T. Arndt. 1995. The role of Saccharomyces cerevisiae type 2A phosphatase in the actin cytoskeleton and in entry into mitosis. EMBO J. 14:27452759. 20. Lowndes, N. F., A. L. Johnson, L. Breeden, and L. H. Johnston. 1992. SWI6 protein is required for transcription of the periodically expressed DNA synthesis genes in budding yeast. Nature (London) 357:505508. 20a.Luke, M. M. Unpublished results. 20b.Luke, M. M., F. Della Seta, and K. T. Arndt. Unpublished results. 21. Mann, D. J., V. Dombradi, and P. T. Cohen. 1993. Drosophila protein phosphatase V functionally complements a SIT4 mutant in Saccharomyces cerevisiae and its amino-terminal region can confer this complementation to a heterologous phosphatase catalytic domain. EMBO J. 12:48334842. 22. Matsumoto, T., and D. Beach. 1993. Interaction of the pim1/spi1 mitotic checkpoint with a protein phosphatase. Mol. Biol. Cell 4:337345. 23. Miosga, T., E. Boles, I. Schaaff-Gerstenschlager, S. Schmitt, and F. K. Zimmermann. 1994. Sequence and function analysis of a 9.74 kb fragment of Saccharomyces cerevisiae chromosome X including the BCK1 gene. Yeast 10:14811488. 24. Nasmyth, K., and L. Dirick. 1991. The role of SWI4 and SWI6 in the activity of G1 cyclins in yeast. Cell 66:9951013. 25. Ogas, J., B. J. Andrews, and I. Herskowitz. 1991. Transcriptional activation

THE SIT4-ASSOCIATED PROTEIN FAMILY

2755

26. 27.

28. 29. 30.

31. 32. 33. 34. 35.

36. 37. 38. 39. 40. 41. 42. 43.

44. 45.

of CLN1, CLN2, and a putative new G1 cyclin (HCS26) by SWI4, a positive regulator of G1-specic transcription. Cell 66:10151026. Ohkura, H., N. Kinoshita, S. Miyatani, T. Toda, and M. Yanagida. 1989. The ssion yeast dis2 gene required for chromosome disjoining encodes one of two putative type 1 protein phosphatases. Cell 57:9971007. Pringle, J. R., and L. H. Hartwell. 1981. The Saccharomyces cerevisiae cell cycle, p. 97142. In J. N. Strathern, E. W. Jones, and J. R. Broach (ed.), The molecular biology of the yeast Saccharomyces, life cycle and inheritance. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Reid, B. J., and L. H. Hartwell. 1977. Regulation of mating in the cell cycle of Saccharomyces cerevisiae. J. Cell Biol. 75:355365. Richardson, H. E., C. Wittenberg, F. Cross, and S. I. Reed. 1989. An essential G1 function for cyclin-like proteins in yeast. Cell 59:11271133. Riles, L., J. E. Dutchik, A. Baktha, B. K. McCauley, E. C. Thayer, M. P. Leckie, V. V. Braden, J. E. Depke, and M. V. Olson. 1993. Physical maps of the six smallest chromosomes of Saccharomyces cerevisiae at a resolution of 2.6 kilobase pairs. Genetics 134:81150. Ronne, H., M. Carlberg, G. Z. Hu, and J. O. Nehlin. 1991. Protein phosphatase 2A in Saccharomyces cerevisiae: effects on cell growth and bud morphogenesis. Mol. Cell. Biol. 11:48764884. Rose, M. D., P. Novick, J. H. Thomas, D. Botstein, and G. R. Fink. 1987. A Saccharomyces cerevisiae genomic plasmid bank based on a centromerecontaining shuttle vector. Gene 60:237243. Rose, M. D., F. Winston, and P. Hieter. 1990. Laboratory course manual for methods in yeast genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Sharp, P. M., T. M. Tuohy, and K. R. Mosurski. 1986. Codon usage in yeast: cluster analysis clearly differentiates highly and lowly expressed genes. Nucleic Acids Res. 14:51255143. Shimanuki, M., N. Kinoshita, H. Ohkura, T. Yoshida, T. Toda, and M. Yanagida. 1993. Isolation and characterization of the ssion yeast protein phosphatase gene ppe1 involved in cell shape control and mitosis. Mol. Biol. Cell 4:303313. Sneddon, A. A., P. T. Cohen, and M. J. Stark. 1990. Saccharomyces cerevisiae protein phosphatase 2A performs an essential cellular function and is encoded by two genes. EMBO J. 9:43394346. Stuart, D., and C. Wittenberg. 1994. Cell cycle-dependent transcription of CLN2 is conferred by multiple distinct cis-acting regulatory elements. Mol. Cell. Biol. 14:47884801. Sutton, A., D. Immanuel, and K. T. Arndt. 1991. The SIT4 protein phosphatase functions in late G1 for progression into S phase. Mol. Cell. Biol. 11:21332148. Sutton, A., F. Lin, and K. T. Arndt. 1991. The SIT4 protein phosphatase is required in late G1 for progression into S phase. Cold Spring Harbor Symp. Quant. Biol. 56:7581. Thomas, B. J., and R. Rothstein. 1989. Elevated recombination rates in transcriptionally active DNA. Cell 56:619630. Tyers, M., G. Tokiwa, and B. Futcher. 1993. Comparison of the Saccharomyces cerevisiae G1 cyclins: Cln3 may be an upstream activator of Cln1, Cln2 and other cyclins. EMBO J. 12:19551968. Tyers, M., G. Tokiwa, R. Nash, and B. Futcher. 1992. The Cln3-Cdc28 kinase complex of S. cerevisiae is regulated by proteolysis and phosphorylation. EMBO J. 11:17731784. van Zyl, W., W. Huang, A. A. Sneddon, M. Stark, S. Camier, M. Werner, C. Marck, A. Sentenac, and J. R. Broach. 1992. Inactivation of the protein phosphatase 2A regulatory subunit A results in morphological and transcriptional defects in Saccharomyces cerevisiae. Mol. Cell. Biol. 12:49464959. Wada, K., Y. Wada, H. Doi, F. Ishibashi, T. Gojobori, and T. Ikemura. 1991. Codon usage tabulated from the GenBank genetic sequence data. Nucleic Acids Res. 19(Suppl.):19811986. Wittenberg, C., K. Sugimoto, and S. I. Reed. 1990. G1-specic cyclins of S. cerevisiae: cell cycle periodicity, regulation by mating pheromone, and association with the p34CDC28 protein kinase. Cell 62:225237.