Catalytic Roles of Yeast GSK3ß/Shaggy Homolog Rim11p in Meiotic Activation

Corresponding author: Aaron P. Mitchell, Department of Microbiology, Columbia University, 701 W. 168th St., New York, NY 10032., apm4{at}columbia.edu (E-mail)

Communicating editor: A. G. HINNEBUSCH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Saccharomyces cerevisiae, many meiotic genes are activated by a heteromeric transcription factor composed of Ime1p and Ume6p. Ime1p-Ume6p complex formation depends upon the protein kinase Rim11p, which interacts with and phosphorylates both Ime1p and Ume6p in vitro. Rim11p may promote complex formation through its phosphorylation of Ime1p and Ume6p or simply through its interaction with both proteins. Here, we characterize mutant Ime1p derivatives that interact with Rim11p but are not phosphorylated in vitro. These mutant proteins are also defective in interaction with Ume6p. These results argue that Ime1p must be phosphorylated to interact with Ume6p. Our genetic observations suggest that Ime1p tyrosine residues are among the Rim11p phosphoacceptors, and we find that Ime1p reacts with an anti-phosphotyrosine antibody. Ime1p and Rim11p have been thought to act only through Ume6p, but we find that Ime1p and Rim11p promote meiosis at a very low level in the absence of Ume6p. A nonphosphorylatable mutant Ime1p derivative promotes sporulation through this Ume6p-independent pathway, as does a mutant Rim11p derivative that fails to interact with Ime1p. Therefore, Ime1p and Rim11p have two genetically separable functions in the sporulation program. However, catalytic activity of Rim11p is required for sporulation in the presence or absence of Ume6p.

THE protein kinase Rim11p (also called Mds1p and ScGSK3) is required for meiosis in yeast (MITCHELL and BOWDISH 1992 ; BOWDISH et al. 1994 ; PUZISS et al. 1994 ). Rim11p promotes formation of the Ime1p-Ume6p complex (RUBIN-BEJERANO et al. 1996 ), which activates transcription of early meiotic genes (reviewed in KUPIEC et al. 1997 ). Fusion of a transcriptional activation domain to Ume6p permits meiosis in the absence of Rim11p and Ime1p (RUBIN-BEJERANO et al. 1996 ), and so the only essential role in meiosis of Rim11p and Ime1p is apparently to modify Ume6p.

Rim11p is similar to members of the eukaryotic glycogen synthase kinase-3ß (GSK3ß)/shaggy family in both structure and function (CADIGAN and NUSSE 1997 ). GSK3ß family members have catalytic regions with >80% amino acid sequence identity and share 55–60% identity with Rim11p. (For comparison, the GSK3ß and protein kinase A catalytic domains share <25% identity.) Several of the GSK3ß family members act in the Wnt/wingless signaling pathway to promote formation of a ß-catenin-adenomatous polyposis coli (APC) complex (CADIGAN and NUSSE 1997 ). The Ime1p-Ume6p and ß-catenin-APC complexes have no structural similarity, but their relationships to Rim11p and GSK3ß share a common feature: the protein kinases bind to and phosphorylate both subunits of their target complex (Rim11p, BOWDISH et al. 1994 ; MALATHI et al. 1997 ; GSK3ß, RUBINFELD et al. 1996 ; BEHRENS et al. 1998 ; IKEDA et al. 1998 ; YAMAMOTO et al. 1998 ). This finding is consistent with two distinct models for Rim11p and GSK3ß function. One is that the kinases have a catalytic role in complex formation; for example, Rim11p phosphorylates both Ime1p and Ume6p, and phospho-Ime1p then interacts with phospho-Ume6p (Figure 1A). Numerous other protein kinases act catalytically. The other model is that the kinases have a structural role; for example, a Rim11p-Ime1p complex interacts with Ume6p, and the functional transcriptional activator may be a ternary Ime1p-Rim11p-Ume6p complex (Figure 1B). Pbs2p and Kss1p are examples of protein kinases that have structural roles (as well as catalytic roles) in signal transduction (POSAS and SAITO 1997 ; BARDWELL et al. 1998 ).

View larger version (22K): In this window In a new window Download PPT slide   Figure 1. Possible catalytic and structural roles for Rim11p. (A) Catalytic model. Rim11p phosphorylates Ime1p and Ume6p, then phospho-Ime1p binds to phospho-Ume6p. (B) Structural model. Rim11p binds first to Ime1p, then the Rim11p-Ime1p complex binds to Ume6p. (It is equally possible that Rim11p binds first to Ume6p, then the Rim11p-Ume6p binds to Ime1p.)

Properties of rim11 and ume6 mutants are consistent with either catalytic or structural roles for Rim11p (MALATHI et al. 1997 ; K. MALATHI and A. P. MITCHELL, unpublished results). One observation is that two catalytically defective mutant Rim11p derivatives fail to promote Ime1p-Ume6p complex formation, as predicted by the catalytic model. However, both Rim11p derivatives are also defective in interaction with Ume6p (and with Ime1p). A second observation is that the mutant Ume6-T99Np, which has a substitution that reduces its phosphorylation in vitro, fails to interact with Ime1p, as predicted by the catalytic model. However, Ume6-T99Np is also defective in interaction with Rim11p. So, in both cases, the defects in Ime1p-Ume6p complex formation are predicted by the structural model as well. Thus, it has not been possible to use the Rim11p-Ume6p complex to distinguish between catalytic and structural roles for Rim11p, because the mutations we have examined affect both catalytic and structural attributes.

Here, we examine the effects of ime1 mutations on both Rim11p-Ime1p and Ime1p-Ume6p complexes. Their properties support a catalytic role for Rim11p in promoting Ime1p-Ume6p complex formation. In addition, our analysis indicates that Ime1p and Rim11p have a second role in promoting meiosis that is independent of Ume6p.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains:
Strains used in this study (Table 1) were constructed through genetic crosses and transformations, following standard procedures (KAISER et al. 1994 ). All strains derive from crosses among SK-1 strains (KANE and ROTH 1974 ; ALANI et al. 1987 ) except strains Y187 and Y190 (DURFEE et al. 1993 ), which were used for two-hybrid interaction cloning. The ume6::TRP1 mutation was created by PCR product-directed gene disruption (BAUDIN et al. 1993 ; LORENZ et al. 1995 ) through amplification of the TRP1 cassette from plasmid pRS314 with primers UME6-1050 and UME6-3'-3590. The mutation removes UME6 sequences between 18 bp upstream of the initiation codon and 102 bp upstream of the termination codon.

P_GAL1-IME168-182 mutants:
P_GAL1-IME168-182 internal deletions of previously isolated random P_GAL1-IME1 mutants (SMITH et al. 1993 ) were constructed as follows. Plasmid pHS136 (YCpP_GAL1-IME1) was digested with PvuII and SacI to release the 3' portion of the IME1 open reading frame (ORF) and then ligated to 1.0-kb EcoRV/SacI fragments carrying the various ime1 mutations.

Several oligonucleotides (see sequences below) were used to create IME1 mutations. To construct site-directed alanine substitutions, plasmid pKM153 (P_GAL1-IME168-182 in vector pRS316) was mutagenized with oligonucleotides S1 (to create S352A), S2 (to create S356A), and S3 (to create S360A). The multi-site "ala8" replacement mutant was generated by PCR using oligonucleotides IM1C and YX3 with plasmid pKM153 as a template. A corresponding wild-type plasmid was constructed in a PCR with oligonucleotides M127 and YX3. The PCR products were digested with HindIII and SalI and cloned into HindIII- and SalI-digested plasmid pSV150, which is vector pRS316 carrying the GAL1 promoter (VIDAN and MITCHELL 1997 ). The resulting plasmids carried P_GAL1-IME168-182, but lacked IME1 3' noncoding sequences. Ime1p was expressed poorly from these plasmids; expression was improved by insertion of IME1 3' noncoding sequences, cloned into the plasmids' SalI site from a PCR with oligonucleotides M128 and M129. The sequence of the entire P_GAL1-IME168-182-ala8 ORF was verified.

Two-hybrid interaction assays:
Overnight cultures in SC-Trp-Leu medium were diluted into yeast extract peptone acetate (YPAc) and harvested after two to three doublings (12–14 hr). ß-Galactosidase assays were performed on permeabilized cells (BOWDISH and MITCHELL 1993 ). Each determination is an average of three independent transformants. The range was <20% of the mean for values over 150 Miller units and 25% of the mean for values under 150 Miller units. Transformants of strains Y190 and Y187 were mated for use in these assays.

Plasmids specifying Gal4p DNA binding domain (GBD)-Rim11p, GBD-Ime1p(294–360), and Gal4p activation domain (GAD)-Ume6p(1–161) have been described previously (MALATHI et al. 1997 ). Plasmids specifying GAD-Ime1p and derivatives were constructed as follows. PCR amplification from P_GAL1-IME1 templates with oligonucleotides IM1N and IM1B created an NcoI site at codon 1 and a BamHI site at a location 200 bp downstream of the IME1 stop codon. PCR products digested with NcoI and BamHI were cloned into NcoI- and BamHI-digested plasmid pACTII (DURFEE et al. 1993 ).

Plasmids specifying GBD-Ime1p(294–360) mutant derivatives were generated by in vivo homologous recombination (MA et al. 1987 ). PCR amplification of IME1 plasmids with oligonucleotides M132 and M133 created products in which IME1 codons 294–360 and 200 bp of 3' noncoding sequences were flanked by pAS1-CYH2 polylinker sequences, arranged to create an in-frame fusion. Purified PCR products were cotransformed into strain Y187 along with plasmid pAS1-CYH2 digested with NcoI and BamHI. Trp⁺ transformants were tested for expression of GBD-fusion protein by immunoblotting with anti-Ime1p and plasmids were retrieved in Escherichia coli. Presence of the mutations was confirmed by sequencing.

Synthetic oligonucleotides:
Synthetic oligonucleotides used in these studies had 5' to 3' sequences indicated in parentheses: IM1C (AGT ACT TGT CGA CAA TTA AGC AGC GGC TTT AGC AAA CTT GGC GGC TAT TTC TTG AAA CCT GAC CTT GTC AGC AGC ATC TTG ATC ATT AGA ACT GCT G), M127 (AGT ACT TGT CGA CAA TTA AGA ATA GGT TTT ACT), M128 (CTT AAT TGT CGA CAA GTA CTA CAA TC), M129 (CGA TAA GGG TAC CAA CGC TAC GGT A), M132 (GGT CAA AGA CAG TTG ACT GTA TCG CCG GTA TTG CAA TAC CCA GCT TTG ACT CAT ATG GCC ATG GTA GAA ATT GCC TTC GAC GTT GAA), M133 (GTT ACT CAA GAA CAA GAA TTT TCG TTT TAA AAC CTA AGA GTC ACT TTA AAA TTT GTA TAC ACT TAT GGA TCC TAA TAA CGC TAC GGT ATT ATG), IM1N (GGC ATG CCA TGG AGC AAG CGG ATA TGC ATG G), IM1B (CGA TAA GGA TCC TAA TAA CGC TAC GG), UME6-1050 (GAA GCG CCC ACC TTC GCA CAG CGC ACA GGA ACT AGG ACA CTA CCG CAC TCA AAC CAT TTG GCA GAT TGT ACT GAG AGT GC), UME6-3'-3590 (GAT TTC CTC CAG TTT CAT CTG TTT TTT CTT TGG ATC AGA TAC AAA ATC TGG TTT GAA CGC CGC ATC TGT GCG GTA TTT CAC), YX3 (CAG CGA AGC GAT GAT TTT TGA TC), S1 (GGT TTC AAG AAA TAG CCT ACA AAG), S2 (CCT ACA AGT TTG CTA AAA CCT AT), S3 (TAA AAC CTA TGC TTA ATT CTC G).

Miscellaneous:
Procedures for Rim11-HAp immunoprecipitation, immunoblotting, protein kinase activity assays, and measurement of sporulation have been described (BOWDISH et al. 1994 ; MALATHI et al. 1997 ). For anti-phosphotyrosine immunoblots, filters were blocked for 2 hr with PyTBST buffer (100 mM Tris pH 8.0, 750 mM NaCl, 1% Tween 20, 10 mM EDTA) containing 3% bovine serum albumin and 1% ovalbumin. The filter was then probed for 16 hr at 4° with mouse monoclonal antibody 4G10 (Upstate Biotechnology) diluted 1/1000 in the blocking solution, washed three times with PyTBST buffer, then incubated with 1/3000 diluted secondary antibody and ECL reagents as described previously (MALATHI et al. 1997 ). For immunoprecipitation experiments with strains expressing the Ime168-182-ala8p, we used 4 mg of protein extract and 4.0 µl of 5.5 mg/ml 12CA5 monoclonal antibody followed by a 2-hr incubation with 40 µl of 50% protein A Sepharose beads. Levels of the mutant Ime168-182-ala8p were much lower than Ime168-182p in crude extracts. Therefore, we used plasmid pKM212, which specifies Ime168-182p but lacks IME1 3' noncoding sequences, to express Ime168-182p at reduced levels as a positive control. The multi-copy RIM11-HA plasmids pKM116 (RIM11-HA), pKM117 (RIM11-HA-K68R), pKM118 (RIM11-HA-Y199F), and pKM119 (RIM11-HA-K68A) resulted from transfer to vector pRS423 of the inserts from plasmids pKB166, pKB171, pKB201, and pKB199, respectively (BOWDISH et al. 1994 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Interaction of Rim11p with Ime1p mutant derivatives:
Rim11p and Ime1p form a complex in which Ime1p is phosphorylated. To determine whether interaction or phosphorylation is required for Ime1p function, we have characterized Rim11p interaction with and phosphorylation of Ime1p mutant derivatives.

Complex formation was assayed through co-immunoprecipitation of Ime1p with epitope-tagged Rim11-HAp (Figure 2A and Figure B) and through two-hybrid interaction assays (Table 2). For coimmunoprecipitation experiments, ime1 missense mutations were introduced into a functional Ime168-182p internal deletion derivative (SMITH et al. 1993 ) because Ime1p comigrates with Rim11p on SDS-PAGE. Ime168-182p was detectable in Rim11-HAp immune complexes (Figure 2, lane 2 compared to lane 1). Mutant products Ime168-182-L321Fp and Ime168-182-R347Kp were not detectable in Rim11-HAp immune complexes (Figure 2, lanes 3 and 5), although they were present in crude extracts (Figure 2D, lanes 3 and 5). Similarly, the L321F and R347K substitutions caused a 12-fold Ime1p-Rim11p interaction defect in two-hybrid assays (Table 2). The S360F second-site substitution restores functional activity to L321F and R347K mutants (SMITH et al. 1993 ) and restores considerable ability to bind to Rim11p (Figure 2, lanes 4 and 6, and Table 2). The nonsense mutant product Ime1-Q340*p was detectable in Rim11-HAp immune complexes (Figure 2, lane 7) and had 2.5-fold decreased interaction with Rim11p in two-hybrid assays (Table 2). Therefore, the L321F and R347K substitutions cause a severe defect in Ime1p-Rim11p interaction and the Q340* truncation causes a mild defect.

View larger version (64K): In this window In a new window Download PPT slide   Figure 2. Effects of ime1 mutations on Rim11p-Ime1p complex formation and phosphorylation. Anti-HA immune complexes were prepared from strain M1085 (genotype, RIM11-HA/RIM11-HA ime1/ime1) carrying plasmids YCpPGAL1 (lane 1), YCpPGAL1-IME168-182 (lane 2), YCpPGAL1-IME168-182-L321F (lane 3), YCpPGAL1-IME168-182-L321F,S360F (lane 4), YCpPGAL1-IME168-182-R347K (lane 5), YCpPGAL1-IME168-182-R347K,S360F (lane 6), or YCpPGAL1-IME1-Q340* (lane 7) after growth in YPAc medium. The immune complexes were split: half was used in an immunoblot with anti-HA monoclonal antibody (A) and anti-Ime1p antiserum (B); half was used in protein kinase assays (C). Presence of Ime1p derivatives in crude extracts was determined through an immunoblot with anti-Ime1p antiserum (D).

The ability of each mutant Ime1p derivative to serve as a phosphorylation substrate was assayed in Rim11-HAp immune complexes (Figure 2C). Phosphorylation of Ime168-182p was detected (lane 2, compared to lane 1). Phosphorylation of the L321F and R347K mutant proteins was not detected (lanes 3 and 5). This result was expected from their poor recovery in immune complexes, and the S360F second-site substitution restored phosphorylation of both mutant proteins (lanes 4 and 6). Phosphorylation of Ime1-Q340*p was not detected (lane 7). The Ime1-Q340*p extract did not contain an inhibitor of Rim11-HAp protein kinase activity, because autophosphorylation of Rim11-HAp was detectable at similar levels in all extracts (Figure 2C). Thus the Q340* truncation permits Rim11p-Ime1p binding, but prevents Ime1p from serving as a substrate for Rim11p phosphorylation in vitro.

Identification of putative phosphoacceptor residues in Ime1p:
The studies above suggest that amino acid residues beyond Q340 of Ime1p are required for Ime1p phosphorylation by Rim11p. Therefore, the eight serine, threonine, and tyrosine residues distal to Q340 in Ime168-182p are candidate sites of phosphorylation. To test this hypothesis, we examined the Ime168-182-ala8p mutant derivative, in which these eight residues were replaced with alanines. Ime168-182-ala8p accumulates to lower levels than wild-type Ime168-182p (data not shown). Therefore, we used wild-type Ime168-182p expressed at reduced levels, due to a deletion of 3' noncoding sequences, as a standard for comparison with Ime1-ala8p. These two proteins were present at comparable levels in crude extracts (Figure 3D, lanes 3 and 4) and were recovered at similar levels in Rim11-HAp immune complexes (Figure 3B). However, Rim11-HAp immune complexes phosphorylated Ime168-182-ala8p very poorly compared to Ime168-182p (Figure 3C). These results indicate that some or all of the serine, threonine, and tyrosine residues in Ime1p distal to Q340 are required for phosphoacceptor activity in vitro.

View larger version (47K): In this window In a new window Download PPT slide   Figure 3. Effects of the ala8 substitution on Rim11p-Ime1p complex formation and phosphorylation. Anti-HA immune complexes were prepared from strains KB267 (genotype, rim11/rim11; lane 1) and M1085 (genotype: RIM11-HA/RIM11-HA; lanes 2–4) carrying plasmids pKM212 (YCpPGAL1-IME168-182 lacking 3' noncoding sequences; lanes 1 and 3), YCpPGAL1 (lane 2), or YCpPGAL1-IME168-182-ala8 (lane 4) after growth in YPAc medium. As in Figure 2, the immune complexes were split: half was used in an immunoblot with anti-HA monoclonal antibody (A) and anti-Ime1p antiserum (B); half was used in protein kinase assays (C). Presence of Ime1p derivatives in crude extracts was determined through an immunoblot with anti-Ime1p antiserum (D).

The serine residues S352, S356, and S360 match the GSK3 substrate site consensus S-X₃-S-X₃-S (ROACH 1991 ). To determine whether these residues are the main Ime1p phosphoacceptors, we characterized mutants in which these serines were replaced with alanines. All single and multiple mutant Ime1p derivatives were present at similar levels in crude extracts and in Rim11-HAp immune complexes, and all mutant Ime1p derivatives were phosphorylated in immune complexes (Table 3 and data not shown). Notably, the triple substitution eliminating all three serine residues had little effect on Ime1p phosphorylation. Thus the Ime1p residues in this consensus GSK3 site are not required for phosphoacceptor activity in vitro.

We considered the hypothesis that Rim11p phosphorylates Ime1p on one of the tyrosine residues (Y342, Y343, Y353, and Y359) among residues 340–360. To test this model, we assayed reactivity of Ime168-182p, after its phosphorylation in Rim11-HAp immune complexes, with an anti-phosphotyrosine antibody. We detected a protein that reacts with the antibody (Figure 4A, lane 4), which has an electrophoretic mobility similar to that of Ime168-182p. Presence of the protein depends upon expression of both Ime168-182p and Rim11-HAp (Figure 4A, lanes 3 and 1, respectively). Therefore, we believe that the protein is Ime168-182p. We detected no reaction of Ime168-182-ala8p with the antibody (Figure 4A and Figure B, lane 5). These results support the hypothesis that Rim11p phosphorylates Ime1p on tyrosine residues.

View larger version (28K): In this window In a new window Download PPT slide   Figure 4. Reactivity of Ime1p with anti-phosphotyrosine monoclonal antibody. Anti-HA immune complexes were prepared from strains KB267 (genotype, rim11/rim11; lanes 1 and 2) and M1085 (genotype, RIM11-HA/RIM11-HA; lanes 3–5) carrying plasmids pKM212 (YCpPGAL1-IME168-182 lacking 3'-noncoding sequences; lanes 1 and 4), YCpPGAL1-IME168-182-ala8 (lanes 2 and 5), or YCpPGAL1 (lane 3) after growth in YPAc medium. The immune complexes were incubated for 20 min under immune complex kinase assay conditions in the presence of 0.02 mM unlabeled ATP (and without radiolabeled ATP). The immune complexes were then split to use in immunoblots with anti-phosphotyrosine monoclonal antibody (A) and with anti-Ime1p antiserum (B). Presence of Ime1p derivatives in crude extracts was determined through an immunoblot with anti-Ime1p antiserum (C).

Functional analysis of Ime1p mutant derivatives:
Functional activity of new Ime1p mutant derivatives was assessed through their abilities to activate ime2-lacZ expression and promote sporulation (Table 3). Ime168-182-ala8p was completely defective in both assays. Reduced accumulation of the mutant protein could not account for the defect, because a reduction in expression of Ime168-182p caused only a mild reduction in ime2-lacZ expression and sporulation (Table 3). Prior studies indicated that Ime1-Q340*p had reduced activity (SMITH et al. 1993 ), and we found a slightly more severe defect in assays of Ime168-182-Q340*p (Table 3). Properties of these mutants argue that interaction of Ime1p with Rim11p is not sufficient for full functional activity; phosphorylation of Ime1p is required as well.

Relationship between Ime1p phosphorylation and Ime1p-Ume6p interaction:
To determine whether Ime1p phosphorylation may be required for Ime1p-Ume6p interaction, we carried out two-hybrid interaction assays with GBD-Ime1p mutant derivatives and GAD-Ume6p (Table 4). All GBD-Ime1p fusion proteins were detected on immunoblots probed with anti-Ime1p antiserum, although the wild-type fusion protein accumulated to levels approximately fivefold higher than several of the mutant fusion proteins (data not shown). Ime1p substitutions that impair binding to Rim11p (L321F and R347K), as well as those that permit Rim11p binding but impair phosphorylation (Q340* and ala8), all caused reduced interaction with Ume6p. The second-site substitution that improves Rim11p binding and phosphorylation (S360F) improved interaction between mutant Ime1p derivatives and Ume6p. In summary, alterations of Ime1p that result in phosphorylation defects also result in Ume6p interaction defects. These findings support the hypothesis that Ime1p phosphorylation by Rim11p, not simply its association with Rim11p, is required for Ime1p-Ume6p interaction.

Genetic evidence for an additional function of Ime1p and Rim11p:
It is well established that Ime1p and Rim11p act in conjunction with Ume6p to activate early meiotic genes. However, ime1 and rim11 mutants are unable to sporulate, whereas ume6 mutants are able to sporulate weakly (BOWDISH and MITCHELL 1993 ; STRICH et al. 1994 ; STEBER and ESPOSITO 1995 ). Thus it seemed possible that Ime1p and Rim11p might have an additional role in sporulation. We tested this hypothesis through assays of sporulation in ume6 mutant strains (Table 5). A UME6/UME6 diploid yielded 93% sporulation; an isogenic ume6/ume6 diploid yielded 3.2% sporulation. Homozygous null ime1 or rim11 mutations abolished sporulation in UME6/UME6 and ume6/ume6 backgrounds. These observations indicate that Ime1p and Rim11p promote sporulation through a second mechanism that is independent of Ume6p.

View this table: In this window In a new window   Table 5. Sporulation of ume6 strains

To determine whether Ume6p-independent sporulation depends upon Ime1p phosphorylation, we assayed several Ime1p mutant derivatives for their ability to stimulate sporulation in an ime1 ume6 homozygous diploid (Table 5). Expression of Ime168-182p yielded 4.9% sporulation. Sporulation was reduced only twofold by the Q340* substitution. Sporulation was abolished by the ala8 substitution. Reduced protein accumulation may have contributed to the ala8 defect, because reduced expression of Ime168-182p caused a fourfold reduction in sporulation ability. Functional activity of Ime168-182-Q340*p in this assay argues that the phosphorylation of Ime1p assayed in vitro is not required for Ume6p-independent sporulation.

To determine whether Ume6p-independent sporulation depends upon Rim11p protein kinase activity, we assayed Rim11p mutant derivatives for ability to stimulate sporulation in a rim11 ume6 homozygous diploid (Table 5). Expression of Rim11-HAp yielded 4.8% sporulation. Sporulation was abolished by K68R and K68A substitutions, which reduce or abolish protein kinase activity (BOWDISH et al. 1994 ). Sporulation was reduced only twofold by the Y199F substitution, which does not abolish Rim11p autophosphorylation activity but causes a severe defect in phosphorylation of Ime1p and Ume6p in vitro (BOWDISH et al. 1994 ; MALATHI et al. 1997 ). These observations support the hypothesis that phosphorylation of Ime1p is not required for Ume6p-independent sporulation. However, they argue that Rim11p protein kinase activity is required for Ume6p-independent sporulation.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Prior studies have shown that Rim11p is required for Ime1p-Ume6p complex formation (RUBIN-BEJERANO et al. 1996 ), and several observations have suggested that Ime1p and Rim11p activate meiosis primarily through formation of an Ime1p-Ume6p complex. In this article, we provide correlative evidence that Rim11p must phosphorylate Ime1p to promote Ime1p-Ume6p complex formation. In addition, we provide genetic evidence that Ime1p and Rim11p each promote sporulation through a second mechanism that is independent of Ume6p.

Function of Rim11p in Ime1p-Ume6p interaction:
Rim11p interacts with both Ime1p and Ume6p, and so it has been uncertain whether Rim11p plays structural or catalytic roles in the Ime1p-Ume6p complex. Here, we have characterized two mutant Ime1p derivatives—Ime1-Q340*p and Ime1-ala8p—that bind to Rim11p but undergo no detectable phosphorylation in vitro. Both mutants are also defective for interaction with Ume6p. Therefore, our findings argue that interaction between Rim11p and Ime1p is not sufficient to promote Ime1p-Ume6p interaction. The evidence does not rule out the possibility that Rim11p remains associated with the Ime1p-Ume6p complex; in fact, the stability of Rim11p-Ime1p in vitro leads us to favor this idea (K. MALATHI and A. P. MITCHELL, unpublished results). However, our observations support the hypothesis that Rim11p must phosphorylate Ime1p to promote Ime1p-Ume6p complex formation (Figure 1A).

It is possible that Ime1p residues 340–360 include the Rim11p phosphorylation sites, because this region is required for phosphoacceptor activity in vitro. It includes a GSK3 consensus site (ROACH 1991 ) and a group of tyrosine residues. These residues are all plausible phosphoacceptors for Rim11p, because rat GSK3ß is a mixed specificity protein kinase in vitro (WANG et al. 1994 ). The consensus site is clearly not the only site of phosphorylation because Ime1-S352A,S356A,S360Ap has significant activity in vivo and is a Rim11p substrate in vitro. In fact, the consensus site residue S360 has the opposite properties from those expected for a Rim11p phosphoacceptor, because the S360F substitution suppresses Ime1p defects. Thus, if the 340–360 interval is the sole phosphorylated region, then phosphorylation must occur on nonconsensus residues. This conclusion is further substantiated by reactivity of Ime1p with an anti-phosphotyrosine antibody. We believe that the simplest interpretation of our results is that Rim11p phosphorylates Ime1p on one or more of the tyrosine residues Y342, Y343, Y353, and Y359.

It is also possible that Ime1p residues 340–360 act as a regulatory domain that stimulates Rim11p to phosphorylate Ime1p elsewhere. Indeed, suppression of Ime1p mutant defects by the S360F substitution is consistent with a regulatory role for this region.

Genetic characterization of Ume6p-independent sporulation:
Null ume6 mutations relieve repression of early meiotic genes, and so it has seemed likely that sporulation of ume6 null mutants would be independent of Rim11p and Ime1p (BOWDISH and MITCHELL 1993 ; STRICH et al. 1994 ; BOWDISH et al. 1995 ; STEBER and ESPOSITO 1995 ). However, we have shown in this report that Ime1p and Rim11p are required for sporulation in a ume6 homozygote. The ume6 allele cannot specify a partially functional product, because it removes all of the UME6 ORF except for the most 3' 34 codons. Therefore, Ime1p and Rim11p must have at least two functions in sporulation: they act in conjunction with Ume6p and independently of Ume6p.

Analysis of Ime1p and Rim11p mutant derivatives in Ume6p-independent sporulation supports the hypothesis that each protein has two distinct functions. For Ime1p, the Q340* truncation severely reduces Ume6p-dependent sporulation but not Ume6p-independent sporulation. The simplest interpretation is that phosphorylation of Ime1p by Rim11p is required only for Ime1p-Ume6p complex formation, and not for the second function of Ime1p. Similarly, for Rim11p, the Y199F substitution abolishes Ume6p-dependent sporulation but not Ume6p-independent sporulation. This Rim11p mutant derivative has a severe defect in interaction with Ime1p, as expected from the location of the substitution in a likely substrate-binding subdomain (BOWDISH et al. 1994 ; MALATHI et al. 1997 ). This observation suggests that Rim11p may promote Ume6p-independent sporulation without interacting with Ime1p, or perhaps through a different kind of interaction with Ime1p.

We do not know what additional roles Ime1p and Rim11p may have in meiosis. The idea that Ime1p acts later in meiosis has been proposed by SHEFER-VAIDA et al. 1995 , based upon temperature-shift studies of the ime1-3 mutant. We note that Ume6p is required later in meiosis to reestablish repression of the same early meiotic genes that it activates in conjunction with Ime1p and Rim11p (BOWDISH et al. 1995 ; STEBER and ESPOSITO 1995 ). Thus a simple model is that Ime1p, Rim11p, and Ume6p first act together to turn early meiotic genes on, then Ime1p and Rim11p carry out their additional functions while Ume6p acts to turn early meiotic genes off.

	ACKNOWLEDGMENTS

We are grateful to members of this laboratory for many helpful discussions and particularly to Teresa Lamb for comments on this manuscript. This work was supported by funds from the National Institutes of Health (GM-39531).

Manuscript received April 23, 1999; Accepted for publication July 14, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALANI, E., L. CAO, and N. KLECKNER, 1987  A method for gene disruption that allows repeated use of URA3 selection in the construction of multiply disrupted yeast strains. Genetics 116:541-545[Abstract/Free Full Text].

BARDWELL, L., J. G. COOK, D. VOORA, D. M. BAGGOTT, and A. MARTINEZ et al., 1998  Repression of yeast Ste12 transcription factor by direct binding of unphosphorylated Kss1 MAPK and its regulation by the Ste7 MEK. Genes Dev. 12:2887-2898[Abstract/Free Full Text].

BAUDIN, A., O. OZIER-KALOGEROPOULOS, A. DENOUEL, F. LACROUTE, and C. CULLIN, 1993  A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acids Res. 21:3329-3330[Free Full Text].

BEHRENS, J., B. A. JERCHOW, M. WURTELE, J. GRIMM, and C. ASBRAND et al., 1998  Functional interaction of an axin homolog, conductin, with beta-catenin, APC, and GSK3beta. Science 280:596-599[Abstract/Free Full Text].

BOWDISH, K. S. and A. P. MITCHELL, 1993  Bipartite structure of an early meiotic upstream activation sequence from Saccharomyces cerevisiae. Mol. Cell. Biol. 13:2172-2181[Abstract/Free Full Text].

BOWDISH, K. S., H. E. YUAN, and A. P. MITCHELL, 1994  Analysis of RIM11, a yeast protein kinase that phosphorylates the meiotic activator IME1. Mol. Cell. Biol. 14:7909-7919[Abstract/Free Full Text].

BOWDISH, K. S., H. E. YUAN, and A. P. MITCHELL, 1995  Positive control of yeast meiotic genes by the negative regulator UME6. Mol. Cell. Biol. 15:2955-2961[Abstract].

CADIGAN, K. M. and R. NUSSE, 1997  Wnt signaling: a common theme in animal development. Genes Dev. 11:3286-3305[Free Full Text].

DURFEE, T., K. BECHERER, R. CHEN, S. H. YEH, and Y. YANG et al., 1993  The retinoblastoma protein associates with protein the phosphatase type 1 catalytic subunit. Genes Dev. 7:555-569[Abstract/Free Full Text].

IKEDA, S., S. KISHIDA, H. YAMAMOTO, H. MURAI, and S. KOYAMA et al., 1998  Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3beta and beta-catenin and promotes GSK-3beta-dependent phosphorylation of beta-catenin. EMBO. J. 17:1371-1384[Medline].

KAISER, C., S. MICHAELIS and A. MITCHELL, 1994 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

KANE, S. and R. ROTH, 1974  Carbohydrate metabolism during ascospore development in yeast. J. Bacteriol. 118:8-14[Abstract/Free Full Text].

KUPIEC, M., B. BYERS, R. E. EXPOSITO and A. P. MITCHELL, 1997 Meiosis and sporulation in Saccharomyces cerevisiae, pp. 889–1036 in The Molecular and Cellular Biology of the Yeast Saccharomyces: Cell Cycle and Cell Biology, edited by J. R. PRINGLE, J. R. BROACH and E. W. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

LORENZ, M. C., R. S. MUIR, E. LIM, J. MCELVER, and S. C. WEBER et al., 1995  Gene disruption with PCR products in Saccharomyces cerevisiae. Gene 158:113-117[Medline].

MA, H., S. KUNES, P. J. SCHATZ, and D. BOTSTEIN, 1987  Plasmid construction by homologous recombination in yeast. Gene 58:201-216[Medline].

MALATHI, K., Y. XIAO, and A. P. MITCHELL, 1997  Interaction of yeast repressor-activator protein Ume6p with glycogen synthase kinase 3 homolog Rim11p. Mol. Cell. Biol. 17:7230-7236[Abstract].

MITCHELL, A. P. and K. S. BOWDISH, 1992  Selection for early meiotic mutants in yeast. Genetics 131:65-72[Abstract].

POSAS, F. and H. SAITO, 1997  Osmotic activation of the HOG MAPK pathway via Ste11p MAPKKK: scaffold role of Pbs2p MAPKK. Science 276:1702-1705[Abstract/Free Full Text].

PUZISS, J. W., T. A. HARDY, R. B. JOHNSON, P. J. ROACH, and P. HIETER, 1994  MDS1, a dosage suppressor of an mck1 mutant, encodes a putative yeast homolog of glycogen synthase kinase 3. Mol. Cell. Biol. 14:831-839[Abstract/Free Full Text].

ROACH, P. J., 1991  Multisite and hierarchal protein phosphorylation. J. Biol. Chem. 266:14139-14142[Abstract/Free Full Text].

RUBIN-BEJERANO, I., S. MANDEL, K. ROBZYK, and Y. KASSIR, 1996  Induction of meiosis in Saccharomyces cerevisiae depends on conversion of the transcriptional represssor Ume6 to a positive regulator by its regulated association with the transcriptional activator Ime1. Mol. Cell. Biol. 16:2518-2526[Abstract].

RUBINFELD, B., I. ALBERT, E. PORFIRI, C. FIOL, and S. MUNEMITSU et al., 1996  Binding of GSK3beta to the APC-beta-catenin complex and regulation of complex assembly. Science 272:1023-1026. (see comments).[Abstract].

SHEFER-VAIDA, M., A. SHERMAN, T. ASHKENAZI, K. ROBZYK, and Y. KASSIR, 1995  Positive and negative feedback loops affect the transcription of IME1, a positive regulator of meiosis in Saccharomyces cerevisiae. Dev. Genet. 16:219-228[Medline].

SMITH, H. E., S. E. DRISCOLL, R. A. SIA, H. E. YUAN, and A. P. MITCHELL, 1993  Genetic evidence for transcriptional activation by the yeast IME1 gene product. Genetics 133:775-784[Abstract].

STEBER, C. M. and R. E. ESPOSITO, 1995  UME6 is a central component of a developmental regulatory switch controlling meiosis-specific gene expression. Proc. Natl. Acad. Sci. USA 92:12490-12494[Abstract/Free Full Text].

STRICH, R., R. T. SUROSKY, C. STEBER, E. DUBOIS, and F. MESSENGUY et al., 1994  UME6 is a key regulator of nitrogen repression and meiotic development. Genes Dev. 8:796-810[Abstract/Free Full Text].

VIDAN, S. and A. P. MITCHELL, 1997  Stimulation of yeast meiotic gene expression by the glucose-repressible protein kinase Rim15p. Mol. Cell. Biol. 17:2688-2697[Abstract].

WANG, Q. M., C. J. FIOL, A. A. DEPAOLI-ROACH, and P. J. ROACH, 1994  Glycogen synthase kinase-3 beta is a dual specificity kinase differentially regulated by tyrosine and serine/threonine phosphorylation. J. Biol. Chem. 269:14566-14574[Abstract/Free Full Text].

YAMAMOTO, H., S. KISHIDA, T. UOCHI, S. IKEDA, and S. KOYAMA et al., 1998  Axil, a member of the Axin family, interacts with both glycogen synthase kinase 3beta and beta-catenin and inhibits axis formation of Xenopus embryos. Mol. Cell. Biol. 18:2867-2875[Abstract/Free Full Text].

This article has been cited by other articles:

				W. Salameh, J. P. Helliwell, G. Han, L. McPhaul, and O. Khorram Expression of endometrial glycogen synthase kinase-3{beta} protein throughout the menstrual cycle and its regulation by progesterone Mol. Hum. Reprod., September 1, 2006; 12(9): 543 - 549. [Abstract] [Full Text] [PDF]

				K. Purnapatre, M. Gray, S. Piccirillo, and S. M. Honigberg Glucose Inhibits Meiotic DNA Replication through SCFGrr1p-Dependent Destruction of Ime2p Kinase Mol. Cell. Biol., January 1, 2005; 25(1): 440 - 450. [Abstract] [Full Text] [PDF]

				I. Rubin-Bejerano, S. Sagee, O. Friedman, L. Pnueli, and Y. Kassir The In Vivo Activity of Ime1, the Key Transcriptional Activator of Meiosis-Specific Genes in Saccharomyces cerevisiae, Is Inhibited by the Cyclic AMP/Protein Kinase A Signal Pathway through the Glycogen Synthase Kinase 3-{beta} Homolog Rim11 Mol. Cell. Biol., August 15, 2004; 24(16): 6967 - 6979. [Abstract] [Full Text] [PDF]

				L. Pnueli, I. Edry, M. Cohen, and Y. Kassir Glucose and Nitrogen Regulate the Switch from Histone Deacetylation to Acetylation for Expression of Early Meiosis-Specific Genes in Budding Yeast Mol. Cell. Biol., June 15, 2004; 24(12): 5197 - 5208. [Abstract] [Full Text] [PDF]

				N. Colomina, Y. Liu, M. Aldea, and E. Gari TOR Regulates the Subcellular Localization of Ime1, a Transcriptional Activator of Meiotic Development in Budding Yeast Mol. Cell. Biol., October 15, 2003; 23(20): 7415 - 7424. [Abstract] [Full Text] [PDF]

				S. M. Honigberg and K. Purnapatre Signal pathway integration in the switch from the mitotic cell cycle to meiosis in yeast J. Cell Sci., June 1, 2003; 116(11): 2137 - 2147. [Abstract] [Full Text] [PDF]

				T. B. Guo, K. C. Chan, H. Hakovirta, Y. Xiao, J. Toppari, A. P. Mitchell, and W. A. Salameh Evidence for a Role of Glycogen Synthase Kinase-3{beta} in Rodent Spermatogenesis J Androl, May 1, 2003; 24(3): 332 - 342. [Abstract] [Full Text] [PDF]

				M. Schaber, A. Lindgren, K. Schindler, D. Bungard, P. Kaldis, and E. Winter CAK1 Promotes Meiosis and Spore Formation in Saccharomyces cerevisiae in a CDC28-Independent Fashion Mol. Cell. Biol., January 1, 2002; 22(1): 57 - 68. [Abstract] [Full Text] [PDF]

				Y. Xiao and A. P. Mitchell Shared Roles of Yeast Glycogen Synthase Kinase 3 Family Members in Nitrogen-Responsive Phosphorylation of Meiotic Regulator Ume6p Mol. Cell. Biol., August 1, 2000; 20(15): 5447 - 5453. [Abstract] [Full Text]