Fus3-Regulated Tec1 Degradation through SCFCdc4 Determines MAPK Signaling Specificity during Mating in Yeast

Cell, Vol. 119, 981–990, December 29, 2004, Copyright ©2004 by Cell Press

Fus3-Regulated Tec1 Degradation through SCFCdc4 Determines MAPK Signaling Specificity during Mating in Yeast Song Chou,1 Lan Huang,2 and Haoping Liu1,* 1 Department of Biological Chemistry University of California, Irvine Irvine, California 92697 2 Department of Physiology and Biophysics and Department of Developmental and Cell Biology University of California, Irvine Irvine, California 92687

Signaling specificity is fundamental for parallel mitogen-activated protein kinase (MAPK) cascades that control growth and differentiation in response to different stimuli. In Saccharomyces cerevisiae, components of the pheromone-responsive MAPK cascade activate Fus3 and Kss1 MAPKs to induce mating and Kss1 to promote filamentation. Active Fus3 is required to prevent the activation of the filamentation program during pheromone response. How Fus3 prevents the crossactivation is not clear. Here we show that Tec1, a cofactor of Ste12 for the expression of filamentation genes, is rapidly degraded during pheromone response. Fus3 but not Kss1 induces Tec1 ubiquination and degradation through the SCFCdc4 ubiquitin ligase. T273 in a predicted high-affinity Cdc4 binding motif is phosphorylated by Fus3 both in vitro and in vivo. Tec1T273V blocks Tec1 ubiquitination and degradation and allows the induction of filamentation genes in response to pheromone. Thus, Fus3 inhibits filamentous growth during mating by degrading Tec1.

et al., 1997). Mating genes are induced by Ste12 via its binding to the Ste12 binding sites in the promoters of mating genes (Roberts et al., 2000; Zeitlinger et al., 2003). Expression of filamentation genes requires both Ste12 and Tec1, a highly conserved transcription factor of the TEA/ATTS family (Madhani and Fink, 1997; Mosch and Fink, 1997). In response to pheromone, Ste12 activity and mating genes are highly induced, but filamentation genes are not (Madhani et al., 1997). In the absence of Fus3 kinase, however, the expression of filamentation genes is increased in response to pheromone (Madhani et al., 1997). It was thought that Kss1 could be activated in response to pheromone in the absence of Fus3 and that the active Kss1 would leak the signal to the filamentation program (Madhani et al., 1997). Later, it was shown that Kss1 is rapidly phosphorylated and potently activated by pheromone even in wild-type cells and that signaling specificity is likely maintained partially because active Fus3 limits the extent of Kss1 activation, thereby preventing inappropriate signal crossover (Sabbagh et al., 2001). Despite the differential activation of Fus3 and Kss1 in strength and duration by several upstream steps (Andersson et al., 2004; Sabbagh et al., 2001), how the two MAPKs generate two different developmental outputs is still not clear. Here we show a novel mechanism for how Fus3 retains MAPK signaling specificity. Fus3 phosphorylates Tec1, which triggers ubiquitin ligase SCFCdc4-mediated ubiquitination and degradation of Tec1. Stable Tec1 mimics the effect of the fus3 mutant and allows the induction of filamentation genes during pheromone response.

Introduction

Results

MAPK pathways play important roles in regulating the key cellular processes of proliferation, differentiation, and apoptosis. It is critical that these pathways maintain specificity in signaling to elicit the activation of a specific program of gene expression. Signaling specificity is particularly pertinent when components of parallel pathways regulate different developmental programs. In Saccharomyces cerevisiae, several elements of the MAPK pathway for the mating pheromone response are also required for invasive and filamentous growth (Liu et al., 1993; Roberts and Fink, 1994). These elements include Ste20 (a homolog of the mammalian P21-activated kinase, PAK), Ste11 (a MEKK homolog), and Ste7 (a MEK homolog). Both Fus3 and Kss1 MAP kinases are activated during pheromone response (Sabbagh et al., 2001) and have overlapping functions in mating (Elion et al., 1991). During filamentous growth, only Kss1 is activated to promote filamentation (Cook et al., 1997; Madhani et al., 1997; Sabbagh et al., 2001). The downstream targets of the active MAPKs include the Ste12 transcription factor and its associated repressors Dig1 and Dig2 (Cook et al., 1996; Elion et al., 1993; Tedford

Tec1 Protein Is Degraded during Pheromone Response in Mating In an experiment aimed at studying Ste12 and Tec1 interaction during mating, we noticed that the protein level of a C-terminal HA-tagged Tec1 decreased during the mating of haploid cells of opposite mating types (Figure 1A). We also observed that Tec1 protein disappeared in 15 min after exposing yeast cells to 5 ␮M pheromone, whereas the Ste12 protein level remained unchanged under the same conditions (Figure 1B). Since TEC1 transcription is induced by pheromone (Oehlen and Cross, 1998), the disappearance of Tec1 protein is likely due to pheromone-induced protein degradation. Pheromone-induced degradation of Tec1 protein was further confirmed by expressing a myc-tagged TEC1 under the GAL1 promoter. The Tec1-myc was expressed transiently and turned off before exposing the cells to pheromone. Tec1 protein disappeared at around 45 min after cells were exposed to either 5 ␮M or 100 nM pheromone, whereas it was relatively stable in cells not exposed to pheromone (Figure 1C). This confirmed that pheromone induction triggers the degradation of Tec1 protein. In the GAL shutdown assay, the difference in phero-

Summary

*Correspondence: [email protected]

Cell 982

Figure 1. Fus3 but Not Kss1 Is Required for Tec1 Degradation in Response to Pheromone Induction (A) HLY3334 (TEC1-HA STE12-myc) was grown to mid-log phase and mixed with an equal amount of wild-type MAT␣ cells (10560-6B). Cells were removed at the indicated times for immunoblotting with an anti-HA antibody (Roche 12C5). Immunoblotting with an anti-PSTAIRE (Santa Cruz) antibody was performed as a loading control. (B) HLY3320 (TEC1-myc) was grown to mid-log phase and treated with 2 ␮M or 5 ␮M ␣ factor for the indicated lengths of time. Tec1-myc was detected by immunoblotting with an anti-myc antibody. Ste12 protein level in pheromone treated cells was determined in Strain HLY3321 (STE12-myc). (C) GAL shutdown analysis of wild-type strain (10560-4A) carrying plasmid GAL1p-TEC1-myc. Cells were treated with either no pheromone, 100 nM ␣ factor, or 5 ␮M ␣ factor. (D) GAL shutdown analysis of Tec1 stability in fus3 (YM106), kss1 (YM105), and far1 (HLY1619) mutants carrying plasmid GAL1p-TEC1-myc in 5␮M ␣ factor.

mone concentrations (5 ␮M versus 100 nM) did not make much difference in Tec1 degradation. However, when TEC1 is under its endogenous promoter at its genomic locus, different pheromone concentrations made a noticeable difference in Tec1 protein level (Figure 1B, 2 ␮M versus 5 ␮M). Tec1 protein level did not decline as much at 2 ␮M pheromone, which probably represents a balance between transcriptional induction and protein degradation.

Pheromone-Induced Tec1 Degradation Is Blocked in fus3 but Not kss1 As Tec1 is required for the expression of filamentation genes and Fus3 activity prevents the upregulation of filamentation genes during pheromone induction, we speculated that Fus3 but not Kss1 was responsible for the degradation of Tec1. We examined Tec1 protein stability in fus3 and kss1 mutants. Pheromone-induced Tec1 degradation was blocked in the fus3 mutant, and the Tec1 level remained unchanged even after 120 min of pheromone treatment (Figure 1D). In contrast, the half-life of Tec1 in the kss1 mutant was similar to that in wild-type. The failure to degrade Tec1 in fus3 was not due to its defect in cell cycle arrest at G1, because Tec1 was not stabilized in a far1 mutant (Figure 1D).

Therefore, Tec1 degradation specifically requires Fus3 but not Kss1. Tec1 Degradation Requires the SCFCdc4 Ubiquitin Ligase To determine whether Tec1 degradation is regulated by ubiquitination, we examined Tec1 stability in yeast strains with mutations in genes encoding ubiquitin-conjugating enzymes that are available from the Yeast Deletion Collection as well as cdc34-3. We found that Tec1 degradation was not blocked in mutants with deletions in MMS2, RAD6, UBC4, UBC5, UBC7, UBC8, UBC10, UBC11, UBC12, UBC13, or UDF2 (data not shown) but was stabilized by the conditional loss of Cdc34 (Figure 2A), the E2 that normally functions together with an SCF complex type of ubiquitin ligase (Deshaies, 1999; Koepp et al., 1999). SCF complexes contain Skp1p, Rbx1p, Cdc53p, and an F-box protein, which is involved in specific substrate recognition. Therefore, we examined Tec1 stability in mutants of some known and predicted F-box proteins (Kipreos and Pagano, 2000), including Cdc4, Met30, Grr1, and Dia2. A cdc4 mutant led to a near-complete stabilization of the Tec1 protein (Figure 2A), whereas the other mutants, including grr1 and dia2 that are known to have enhanced filamentous

Fus3 Induces Tec1 Destruction via SCFCdc4 983

growth and FLO11 expression (Palecek et al., 2000), did not block Tec1 degradation (Figure 2A). Together, these data demonstrate that rapid Tec1 degradation during pheromone response is dependent upon the SCFCdc4 ubiquitin ligase. T273 of Tec1 Is Essential for Tec1 Degradation F-box proteins are known to bind phosphorylated substrates (Skowyra et al., 1997). For Cdc4, a high-affinity phosphopeptide binding motif termed the Cdc4 phospho-degron (CPD) has been defined based on the Cdc4 binding sequences from cyclin E1 and Gcn4 (Nash et al., 2001). The CPD bears the consensus I/L-I/L/P-pTP-⬍KR⬎4, where ⬍⬎ indicates a disfavored residue. A single CPD is sufficient for targeting proteins for degradation (Nash et al., 2001). Inspection of the Tec1 protein sequence identified a match to the consensus sequence centered at Thr273 (LLTP) (Figure 2B). LLTP is also an optimal sequence for Erk1 phosphorylation (Songyang et al., 1996) and therefore a likely phosphorylation site of Fus3. Interestingly, two mutations at T273 and P274 were previously isolated from a screen of random mutations of TEC1 for enhanced filamentous growth (Kohler et al., 2002). To determine whether the predicted site was essential for Tec1 degradation, we introduced a point mutation that replaced T273 with Val. Tec1T273V was completely stabilized in cells exposed to pheromone (Figure 2C). Thus, the motif at T273 is essential for Tec1 degradation in yeast. T273 Is Phosphorylated by Fus3 In Vitro and during Pheromone Induction in Yeast Because Fus3 is required for pheromone-induced Tec1 degradation, we wanted to determine whether Fus3 could directly phosphorylate Tec1. We performed an in vitro Fus3 kinase assay with MBP-Tec1-FLAG expressed in E. coli (Madhani and Fink, 1997). Fus3-myc and Fus3K42R-myc were affinity purified from pheromoneinduced cells. We found that Fus3 phosphorylated recombinant Tec1, whereas the kinase-dead mutant Fus3K42R did not (Figure 3A). Similarly, TAP-tagged Fus3 purified from pheromone-treated cells efficiently phosphorylated recombinant Tec1. Unphosphorylated Tec1 ran as a sharp band. After phosphorylation, Tec1 showed a shift in mobility and ran as a fuzzy doublet (Figure 3D, bottom panel). These data suggest that Tec1 is a direct target of Fus3. Since T273 is essential for Tec1 degradation, we wanted to determine whether T273 was phosphorylated by Fus3 in vitro. We analyzed Fus3-phosphorylated MBP-Tec1-FLAG by tandem mass spectrometry. A doubly protonated ion, MH22⫹ of mass to charge (m/z) 633.83, was observed in the LC MS run and was automatically selected for CID fragmentation. The resulting tandem mass spectrum is illustrated in Figure 3B. This peptide is a phosphorylated version of the peptide LLT 273PIT 276AS278NEK of the Tec1 protein. All the y ion series (y1–y10) are observed, among which only y9 and y10 ions are found in a phosphorylated state, whereas the other y ions (y1 to y8) are observed in an unphosphorylated state. In addition, the loss of H3PO4 from y9 and y10 ions is obtained, which is the characteristic loss of

Figure 2. Pheromone-Induced Tec1 Degradation Requires the SCFcdc4 Ubiquitin Ligase and T273 of Tec1 (A) GAL shutdown analysis of Tec1 stability in PY1 (bar1), PY23 (cdc34-3 bar1), PY187 (cdc4-3 bar1), dia2, and grr1 (HLY1377) carrying GAL1p-TEC1-myc in 5 ␮M ␣ factor. (B) Schematic diagram of Tec1 with a predicted high-affinity Cdc4 binding site. T273 is marked by asterisk. (C) GAL shutdown analysis of Tec1 stability in wild-type (10560-4A) carrying GAL1p-TEC1T273V-myc, induced in 5 ␮M ␣ factor.

serine or threonine phosphorylation in CID analysis. This determines the phosphorylation site at the ninth amino acid (T273) from the C terminus, which is further confirmed by the bn ions. In Figure 3B, the b ions (b3–b7), with the exception of the b2 ion, are observed as the phosphorylated forms with the loss of H3PO4. The strong b3 ion (408.20) and its corresponding b3* ion (310.23) with the loss of H3PO4 have demonstrated that T273 is phosphorylated at position 3 from the N terminus. Although there are three potential phosphorylation sites

Cell 984

Figure 3. T273 Is Phosphorylated by Fus3 In Vitro and Phospho-T273 Is Detected in ␣ Factor-Induced Cells (A) In vitro phosphorylation of MBP-Tec1 by Fus3. Autoradiograph of MBP-Tec1 phosphorylated by Fus3-myc or Fus3K42R-myc. The amounts of MBP-Tec1 used are the following: lanes 1 and 4, no Tec1; lanes 2 and 5, 100 ng; and lanes 3 and 6, 400 ng. (B and C) MBP-Tec1, phosphorylated by Fus3-TAP in vitro, and Tec1-myc from pheromone-induced cdc34 (PY187) cells were analyzed in nanoflow liquid chromatography tandem mass spectrometry. The NanoESI tandem mass spectra of the doubly charged phosphorylated peptide containing T273 at mass to charge (m/z) 633.83 was detected both in vitro (B) and in vivo (C). The peptide sequence was determined as LLpT273PITASNEK, where T273 was phosphorylated both in vitro and in vivo. The amino acid sequence coverage is provided by bn and yn fragment ions for the observed phosphopeptides. bn* or yn* refer to (bn ⫺ H3PO4)⫹ and (yn ⫺ H3PO4)⫹, yn*2⫹ refers to [(yn⫹H) ⫺ H3PO4)]2⫹, and MH22⫹* refers to [(M ⫹ 2H) ⫺ (H3PO4)]2⫹, respectively. MH22⫹ is the selected parent ion. (D) Cdc4 binds to phospho-Tec1 in vitro. In vitro-translated 35S-Cdc4 was incubated with Fus3-phosphorylated MBP-Tec1-FLAG. The recombinant Tec1 protein was precipitated with M2 anti-FLAG agarose. Bound proteins were eluted with FLAG peptide and analyzed by autoradiography (top panel). In vitro-translated Skp1 was included together with Cdc4 in the incubation. Lane 1, 35S-Cdc4 input. Lanes 2, 3, and 4 are immunoprecipitations with the following: no Tec1 (lane 2), Tec1 of mock phosphorylation from untagged cells (lane 3), or Tec1 phosphorylated by Fus3-TAP (lane 4). A portion of the immunoprecipitated proteins were analyzed by SDS-PAGE and stained with silver (bottom panel) to determine the amount of MBP-Tec1-FLAG precipitated in each immunoprecipitation.

(T273, T276, and S278) present in this peptide, T273 is identified unambiguously as the phosphorylated site based on the MS/MS spectrum. In addition to T273, several other SP or TP sites are also identified by MS/

MS to be phosphorylated. These include S86, T297, and S325. T476 is not phosphorylated. To further determine whether T273 is phosphorylated in vivo, Tec1-myc was affinity purified from pheromone-

Fus3 Induces Tec1 Destruction via SCFCdc4 985

treated cdc4-3 cells and analyzed by tandem mass spectrometry. The same peptide (MH22⫹ of m/z 633.83) was found in the LCMS run, and the resulting MS/MS spectrum is displayed in Figure 3C. Although the signal is much weaker compared to that in Figure 3B, the overall fragmentation pattern remains the same. The key fragment ions b3, b3*, b5*, b7*, y2, y4–y9, y9*, and y10* are observed in the MS/MS spectrum and are sufficient for the unambiguous identification of T273 as the phosphorylation site in vivo. Cdc4 Interacts with Fus3-Phosphorylated Tec1 In Vitro Cdc4 should interact with phospho-Tec1 if it is an F-box protein for Tec1 ubiquitination. To address this, Fus3phosphorylated MBP-Tec1-FLAG was incubated with in vitro-translated 35S-Cdc4 and Skp1. MBP-Tec1-FLAG and its associated proteins were affinity purified with M2 FLAG affinity beads. As shown in Figure 3D, Cdc4 preferentially bound to Fus3-phosphorylated Tec1 rather than the control Tec1, which was treated with mock purification from untagged yeast cells. The phospho-dependent interaction of Cdc4 with Tec1 in vitro suggests that Cdc4 is an F-box protein directly responsible for pheromone-induced Tec1 degradation. Tec1 Is Ubiquitinated in Pheromone-Treated Cells, and the Ubiquitination Needs Fus3, Cdc4, and T273 The results described above suggest that Fus3 phosphorylation of Tec1 may bring the SCFCdc4 ubiquitin ligase to Tec1 and trigger the ubiquitination and degradation of Tec1. To demonstrate whether Tec1 protein is ubiquitinated in vivo, we used cells carrying Tec1-myc and a His6-tagged ubiquitin variant UbiG76A, as described by Flick et al. (2004). UbiG76A can protect ubiquitinated proteins from deubiquitination in vivo and increase the fraction of a protein obtained in the ubiquitinated forms (Hodgins et al., 1992). Ubiquitin-conjugated proteins were purified by Ni-NTA affinity resin and analyzed by immunoblotting for Tec1-myc. A smear of high molecular weight Tec1 was detected in cells carrying His6UbiG76A but not in cells expressing untagged Ubi (Figure 4). This result suggests that Tec1 is ubiquitinated in vivo. Most of the ubiquitinated forms of Tec1 were gone in Tec1T273V, and only a small amount of ubiquitinated Tec1 was detectable in Tec1T273 after a longer exposure of the immunoblot (Figure 4A). This suggests that T273 is a major site required for Tec1 ubiquitination. Tec1 ubiquitination was blocked in either fus3 and cdc4-3 mutants (Figures 4B and 4C). Therefore, Fus3 and SCFCdc4 are required for Tec1 ubiquitination. Stable Tec1T273V Allows Crossactivation of Filamentation Genes during Pheromone Response To assess the effect of Fus3-regulated Tec1 degradation in maintaining signal specificity during pheromone response, we placed TEC1T273V under its endogenous promoter. A lacZ reporter with two Tec1 binding sites (TCS) was used as a reporter for the expression of filamentation genes. The reporter behaved similarly to other TCS reporters and the reporter of a filamentation response

Figure 4. Tec1 Ubiquitination in Pheromone-Treated Cells Requires Fus3, Cdc4, and T273 (A) In vivo ubiquitination assay of Tec1 and Tec1T273V. Strain PY555 (pre1-1 pre4-1) harboring either ADH1p-TEC1-myc or ADH1pTEC1T273V-myc plasmid was transformed with CUP1p-6xHIS-UBI or CUP1p-UBI (Flick et al., 2004). Cells were grown to OD 0.2–0.3 in selective minimal medium and induced with 100 ␮M CuSO4 for 3.5 hr and then with 2 ␮M ␣ factor for another hour. Ubiquitinated proteins were purified with Ni-NTA beads, separated by SDS-PAGE, and analyzed with a peroxidase conjugated anti-myc antibody (Sigma). (B) In vivo ubiquitination assay of Tec1 in wild-type (10560-4A) and fus3 (YM106) harboring ADH1p-TEC1-myc and CUP1p-6xHIS-UBI or CUP1p-UBI. (C) In vivo ubiquitination assay of Tec1 in wild-type (PY1) or cdc4-3 (PY187) with ADH1p-TEC1-myc and CUP1p-6xHIS-UBI or CUP1pUBI. Cells were grown at 25⬚C to OD 0.2–0.3 in selective minimal medium, induced with 100 ␮M CuSO4 for 3 hr, and then switched to 37⬚C for 20 min before induction with 2 ␮M ␣ factor for another hour at 37⬚C.

element (FRE), which contains a tandem Ste12 binding site and a Tec1 binding site (Kohler et al., 2002). As shown previously for FREs (Sabbagh et al., 2001), TCSlacZ was not induced by pheromone in wild-type cells but was induced by pheromone in a fus3 mutant, and the expression required Kss1 (Figure 5). TEC1T273V showed a similar pattern of TCS-lacZ expression to that of the fus3 mutant: it had higher expression than wild-type

Cell 986

Figure 5. Fus3 Retains Pathway Specificity during the Pheromone Response by Degrading Tec1 Yeast strains tec1, fus3tec1, kss1tec1, and fus3kss1tec1 carrying a TCS-lacZ plasmid were transformed with either TEC1 or TEC1T273V on CEN plasmids, and the transformants were grown in YEPD to mid-log phase. Half of the cells were induced with 5 ␮M ␣ factor for 2 hr, while the other half were not. Proteins were extracted for the ␤-galactosidase assay. ␤-galactosidase activities shown are averages from 5 to 7 transformants for each strain.

cells in the absence of pheromone and showed increased expression upon pheromone induction. Therefore, the stable Tec1 bypassed the need to delete FUS3 for crossactivation of the filamentation program during pheromone response. Interestingly, the TEC1T273V fus3 double mutant gave similar levels of reporter activity in the absence or presence of pheromone to those of the single fus3 or TEC1T273V mutants, and no additive effect was seen in the double mutant. These data suggest that Tec1 degradation is a major mechanism for how active Fus3 prevents crossactivation of the filamentation program during mating. Fus3 and Kss1 Play a Redundant Role in Transcriptional Activation of Filamentation Genes Kss1 is defined as the filamentation MAPK and is required for the expression of filamentation genes (Cook et al., 1997; Madhani et al., 1997; Sabbagh et al., 2001). However, we found that TEC1T273V in either kss1 or fus3 gave similar levels of reporter activity as in uninduced cells and gave additional induction of reporter activity in response to pheromone. This suggests that either Fus3 or Kss1 can activate transcription from TCSs (Figure 5). Thus, the stable Tec1T273V revealed that Fus3 and Kss1 play a redundant role in the transcriptional activation of Ste12/Tec1 for the expression of TCS-lacZ. This suggests that Fus3 has dual functions in the regulation of filamentous growth, activating Ste12/Tec1 and degrading Tec1. Because wild-type TEC1 failed to induce TCS-lacZ in the kss1 mutant, the function of Fus3 in Tec1 degradation is epistatic to its activation of Ste12/ Tec1 in filamentous growth.

Discussion Fus3-Activated Tec1 Degradation Determines Signaling Specificity during Differentiation The MAPK pathways that control mating and filamentation in S. cerevisiae provide a paradigm for studying how signaling specificity is obtained for parallel MAPK

Figure 6. Schematic Diagram of how Fus3 Prevents the Expression of the Filamentation Program during Mating Both Fus3 and Kss1 MAPKs are activated in response to pheromone. They have redundant functions in the transcriptional activation of Ste12/Tec1. Unlike Kss1, Fus3 phosphorylates Tec1, which induces SCFCdc4-mediated ubiquitination and degradation of Tec1. Tec1 degradation precludes the expression of filamentation genes.

Fus3 Induces Tec1 Destruction via SCFCdc4 987

Table 1. Strains Used in this Study Strain

Genotype

Source or Reference

Strains in ⌺1278 background 10560-4A MATa ura3–52 his3::hisG leu2::hisG trp1::hisG 10560-6B MAT␣ ura3–52 his3::hisG leu2::hisG trp1::hisG YM105 MATa kss1:hisG ura3–52 his3::hisG leu2::hisG trp1::hisG YM106 MATa fus3:TRP1 ura3–52 his3::hisG leu2::hisG trp1::hisG YM204 MATa fus3:TRP1 kss1:hisG ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3320 MATa TEC1-(13myc)::HIS3 ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3321 MATa STE12-(13myc)::HIS3 ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3334 MATa TEC1-(3HA)::TRP1STE12-(13myc)::HIS3 ura3–52 his3::hisG leu2::hisG trp1::hisG HLY1619 MATa far1::URA3 ura3–52 his3::hisG leu2::hisG HLY1377 MATa grr1::LEU2 ura3–52 leu2::hisG his3::hisG HLY3349 MATa tec1:KANR ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3351 MATa tec1:KANR fus3:TRP1 ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3363 MATa tec1:KANR kss1:hisG ura3–52 his3::hisG leu2::hisG trp1::hisG HLY3364 MATa tec1:KANR fus3:TRP1 kss1:hisG ura3–52 his3::hisG leu2::hisG trp1::hisG Strains in BF264-15Dau or S288C background PY1 MATa bar1 PY23 MATa cdc34-3 bar1 PY187 MATa cdc4-3 bar1 PY555 MATa pre1-1 pre4-1 YOR080wa MATa dia2:: KANR a

Liu et al., 1993 Liu et al., 1993 Madhani and Fink, 1997 Madhani and Fink, 1997 Madhani et al., 1997 This study This study This study This study This study This study This study This study This study Kaiser et al., 1998 Kaiser et al., 1998 Kaiser et al., 1998 Morris et al., 2003 Invitrogen yeast deletion

The dia2::KANR deletion was confirmed by PCR amplification of the dia2 locus.

pathways with shared components in regulating growth and differentiation. Initial models proposed that Fus3 and Kss1 carry out exclusive functions for mating and filamentation, respectively (Madhani et al., 1997). Later it was found that Kss1 is also activated as part of the wild-type pheromone response and contributes to mating (Sabbagh et al., 2001). Both models recognized Fus3 as the determining factor for pathway specificity. Here we show that an important mechanism that Fus3 utilizes to control pathway specificity is induction of Tec1 degradation (Figure 6). The degradation of Tec1 prevents the expression of filamentation genes, because a stable Tec1 behaves like a fus3 mutant in crossactivating the expression of filamentation genes during pheromone response. The lack of an additive effect in the double mutant of fus3 and stable Tec1 further suggests that Fus3-activated Tec1 degradation is the primary mechanism for inhibiting the expression of filamentation genes under the tested condition (5 ␮M ␣ factor). Tec1 degradation was also observed in mating cells of opposite mating types (Figure 1A), which suggests that Tec1 degradation is a robust and vital part of specificity regulation under physiological conditions. During pheromone response, TEC1 transcription is induced, whereas its protein is degraded. Therefore, Tec1 level in vivo is a balance of the two opposing effects. This explains why Tec1 degradation is induced by 100 nM pheromone in GAL shutdown analysis (Figure 1C), whereas Tec1 level drops less significantly in 2 ␮M pheromone than in 5 ␮M when TEC1 is expressed from its own promoter (Figure 1B, 2 ␮M versus 5 ␮M ␣ factor). The effect of Tec1 degradation on Tec1 protein level may be further amplified by a positive feedback loop, in which Tec1 and Ste12 positively regulate TEC1 expression. The higher the rate of Tec1 degradation, the less amount of active Tec1 is left to transcribe itself, thus leading to even less Tec1 production. Thus, in vivo Tec1 protein level is governed by the speed of its expres-

sion and degradation, both of which are affected by pheromone concentration. At a low pheromone concentration, Fus3-induced Tec1 degradation might not bring down Tec1 level adequately, and it might use additional means, such as attenuating the strength and duration of Kss1 activity (Sabbagh et al., 2001), to achieve optimal pathway specificity. MAPK Substrate Specificity Underlines the Mechanism for MAPK Signaling Specificity One of the important findings from this study is that the functional redundancy between Fus3 and Kss1 is not limited to transcriptional activation of mating genes but also applies to transcriptional activation of filamentation genes (Figure 6). Kss1 was suggested as the only MAPK that can activate filamentous growth through Ste12. But when the negative function of Fus3 toward filamentous growth was abolished in Tec1T273V, its ability to activate the filamentous program was shown to be as potent as Kss1. Thus, Fus3 is capable of all the functions that Kss1 is known of. This is consistent with the findings that both Ste12 and Dig1 are phosphorylated to a similar extent by Kss1 and Fus3 (Cook et al., 1996; Tedford et al., 1997). However, Fus3 also phosphorylates specific substrates that Kss1 is incapable of, all of which are involved in the regulation of mating. We show that Fus3 but not Kss1 induces Tec1 ubiquitination and degradation through phosphorylation. Two other Fus3-specific substrates have been previously reported. Far1 is phosphorylated much more efficiently by Fus3 than by Kss1, both in vivo and in vitro (Breitkreutz et al., 2001; Peter et al., 1993). Recently, Fus3 has been found to phosphorylate Bni1 to regulate cell polarization (Matheos et al., 2004). Interestingly, Kss1-specific substrates have not been identified. This is consistent with the fact that, beside the redundant role shared with Kss1 in the transcriptional activation of the mating program, Fus3 also has additional mating-specific functions, such as

Cell 988

pheromone-induced G1 arrest through Far1, polarity formation through Bni1 for shmooing, and maintaining pathway specificity via degrading Tec1. The extra mating-specific functions of Fus3 also explain why Fus3 activation has to be tightly controlled in vivo and is strictly dependent on the Ste5 scaffold (Andersson et al., 2004). All these findings raise an interesting and important question: what is the difference between Fus3 and Kss1 that can give rise to different substrate specificities? Since MAPK substrate specificity is determined by specific protein interactions between MAPKs and substrates through docking site motifs (Kusari et al., 2004), it will be interesting to determine what docking sites are in common among the three Fus3 substrates and what region of Fus3 specifically interacts with these docking sites. SCFCdc4 Links MAPK Activities with Protein Degradation to Define Signaling Specificity during Cell Differentiation Here we show that Fus3 phosphorylation of Tec1 triggers SCFCdc4-mediated degradation of Tec1 (Figure 6). A conditional cdc4 mutant (Table 1) abolishes pheromoneinduced ubiquitination and degradation of Tec1. Cdc4 specifically interacts with Fus3-phosphorylated Tec1 in vitro, suggesting that the effect of cdc4 on Tec1 degradation is likely direct. T273, which is centered in the only predicted high-affinity phosphopeptide binding sequence for Cdc4, is indeed phosphorylated by Fus3 both in vivo and in vitro. The importance of T273 is further highlighted by the ability of its mutation to block Tec1 ubiquitination and degradation. Therefore, T273 is likely the major site that mediates Cdc4 binding. In addition to T273, Fus3 also phosphorylates several other T(S)P sites in vitro. These sites could potentially contribute to the overall Cdc4 binding. Although Tec1 and Sic1 are both targets of SCFCdc4, they are very different in terms of Cdc4 binding sites. Instead of having a highaffinity Cdc4 binding motif, Sic1 uses multiple weak Cdc4 binding sites to capture Cdc4 (Nash et al., 2001). A single mutation in any of the low-affinity sites would not block Sic1 degradation. SCFCdc4-mediated Tec1 degradation is the first case in yeast that links MAPK signaling with protein degradation to control signaling specificity that defines differentiation versus growth. Similar cases of MAPK-regulated degradation of key transcription factors that are important for growth and differentiation also exist in mammalian systems. For an example, JNK activates c-Jun and promotes apoptosis in neurons, and the degradation of phosphorylated c-Jun through the SCFFbw7 (hCDC4) ubiquitin ligase inhibits apoptotic JNK signaling (Nateri et al., 2004). Cdc4 and its homologs are first known for their roles in degrading cell cycle regulators (Feldman et al., 1997; Koepp et al., 2001; Skowyra et al., 1997; Yamano et al., 2000). Recent studies on Cdc4 structure and its binding preference of phosphopeptide substrates find that MAPK phosphorylation consensus sequences are favored over CDK consensus sequences (Nash et al., 2001; Orlicky et al., 2003), suggesting that MAPK pathways may have a close relationship with SCFCdc4. Thus, coupling MAPK activity with SCFCdc4mediated degradation may be a general theme for signal transduction in differentiation.

Experimental Procedures Yeast Strains and General Methods Standard yeast manipulation methods were used. Yeast strains used in this study are listed in Table 1. HLY3349, HLY3351, HLY3363, and HLY3364 were constructed by PCR amplifying a tec1::KANR fragment from the tec1 deletion strain in the Yeast Deletion Collection (Invitogen) and transforming the PCR product into strains 10560-4A, YM106, YM105, and YM204, respectively.

Construction of myc- and HA-Tagged Strains and Plasmids TEC1-13myc (HLY3320), FUS3-13myc (HLY3387), and TEC1-3HA (HLY3324) strains were generated using the method introduced by Longtine et al. (1998). The TEC1-13myc module was then amplified by PCR and inserted into the EcoRI-XhoI site of pRS vectors (Sikorski and Hieter, 1989). Either an ADH1 or a GAL1-10 promoter was amplified and inserted into the NotI-EcoRI site to complete the construction of ADH1p-TEC1-myc and GAL1p-TEC1-myc. The ADH1p-FUS3-myc plasmid was generated in the same way, except FUS3 CDS was cloned into the ClaI-SalI site. The FUS3 K42R module was amplified from plasmid pJB278 (Brill et al., 1994) and used to replace FUS3 CDS to generate plasmid ADH1p-FUS3 K42R-myc. Highfidelity Pfx polymerase (Invitrogen) was used for all PCR amplification.

Construction of the Tec1T273V Mutant The TEC1T273V mutant was created by replacing nucleotide numbers 866–869 of TEC1 from TACG to GGTA. Two-step PCR amplification was used to generate the mutant with HLY3320 genomic DNA as the template. Two pair of primers, P511 (CCCATCGATGAGTCTT AAAGAAGACGACTTT) and P512 (CAGTGATTGGTACCAGCAGTTT TGAATGTACTGACAAT), and P513(CAAAACTGCTGGTACCAATCA CTGCTTCCAACGAG) and P507(CCGCTCGAGTGCCGGTAGAGGT GTGGTC), were used to PCR amplify overlapping Tec1-myc fragments with the mutation in the overlapping region. The resulting PCR products were purified and mixed as templates for another round of PCR amplification using P511 and P507 as the primers, which produced the full-length TEC1 T273V-myc sequence. The PCR product was digested with ClaI and XhoI and cloned into pRS vectors with either GAL1p or ADH1p. The insert was sequenced to confirm TEC1T273V-myc.

GAL Shutdown Analysis of Tec1 Stability Cells carrying a GAL1p-TEC1-myc plasmid were grown in synthetic growth media to mid-log phase, washed three times with water, and released into YEP ⫹ 2% raffinose media for several hours. Galactose (2%) was added for 30 min, and GAL induction was then shut down by changing the media to YEPD. ␣ factor (5 ␮M) was added 30 min later, and an equal amount of culture was collected at each time point for Western analysis of Tec1 level with an anti-myc antibody (Santa Cruz).

Immunoprecipitation Wild-type (10560-4A) carrying plasmid ADH1p-FUS3-myc or ADH1pFUS3 K42R-myc was treated with 2 ␮M ␣ factor for 30 min. Fus3myc and FusK42R-myc were immunoprecipitated with an anti-myc antibody (Santa Cruz) in IPP150 (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.1% NP40) plus ␣-Pr (protease inhibitors: 1 mM PMSF, 2 mM benzamidine, 2 ␮g/ml leupeptin, 2 ␮g/ml pepstatin, 2 ␮g/ml chymostatin, 5 ␮g/ml aprotinin), 0.5 mM orthovanadate, and 1⫻ phosphatase inhibitor cocktail (Sigma). Anti-myc antibody (2 ␮g) was added into each lysate from 50 ml of exponentially growing cells, and the immunocomplex was precipitated by protein A-Sepharose beads (Pharmacia) and used for an in vitro kinase assay. PY187 (cdc4-3) carrying plasmid GAL1-TEC1-myc was induced with galactose for 1 hr and then induced with 1 ␮M ␣ factor for 30 min. Tec1-myc protein was purified in the same way as above, except 300 mM NaCl was used. The beads were eluted with 1% SDS in TE and resolved by SDS-PAGE. The Tec1-myc band was excised and used for mass spectrometry analysis.

Fus3 Induces Tec1 Destruction via SCFCdc4 989

In Vitro Kinase Assay The Fus3-myc or FusK42R-myc beads were incubated with purified MBP-Tec1-FLAG (Madhani and Fink, 1997) in 20 mM HEPES (pH 7.5), 20 mM MgCl2, 0.5 mM orthovanadate, 1 mM DTT, and 1 ␮l ␥-P32 ATP at 30⬚C for 30 min. The phosphorylated MBP-Tec1-FLAG was separated by SDS-PAGE. The gels were dried and exposed to a phosphoimager screen.

A Modified Fus3 Kinase Reaction A modified kinase reaction was used to phosphorylate recombinant Tec1 for peptide sequencing with nanoflow liquid chromatography tandem mass spectrometry and for in vitro protein interaction assay with Cdc4. A MATa Fus3-TAP strain was purchased from Open Biosystems, and bar1 deletion was introduced to the strain. Cells were treated with 1 ␮M ␣ factor for 30 min and lysed in IPP150 plus 2 mM CaCl2, 0.5 mM orthovanadate, 0.25 mM NaF, 15 mM sodium pyrophosphate, and 15 mM pNPP (Zhan and Guan, 1999) plus ␣-Pr. Fus3-TAP was purified by calmodulin affinity beads (Stratagene) and used to phosphorylate MBP-Tec1-FLAG in 20 mM HEPES (pH 7.5), 20 mM MgCl2, 0.5 mM orthovanadate, 1 mM DTT, 15 mM pNPP, 5 mM EGTA, and 100 ␮M ATP at 30⬚C for 2 hr (Matheos et al., 2004).

In Vitro Interaction of Cdc4 and Phospho-Tec1 Phosphorylated MBP-Tec1-FLAG by Fus3-TAP was incubated with in vitro-translated 35S-Cdc4 in the same buffer used for Fus3-TAP purification, with the exception of CaCl2. TAP purification from a nontagged wild-type (bar1), treated with pheromone, was used for mock phosphorylation of MBP-Tec1-FLAG. Recombinant Tec1 was immunoprecipitated with M2 anti-FLAG agarose. Bound proteins were eluted with FLAG peptide and separated by SDS-PAGE. 35S-Cdc4 levels were determined by autoradiography. Skp1 was also in vitro translated and added together with Cdc4. CDC4 and SKP1 were cloned into pCITE 4b (⫹) vector (Novagen) and translated with STP3 in vitro transcription/translation coupled kit (Novagen).

Peptide Sequencing using Nanoflow Liquid Chromatography Tandem Mass Spectrometry The Tec1 protein gel bands were cut out, and in-gel digestion was performed using trypsin as described (Huang et al., 2001). The tryptic digest was extracted, concentrated, and subjected to nanoflow reverse phase liquid chromatography (RPLC) (Ultimate LC Packings, Dionex) coupled online to a quadrupole-orthogonal time-of-flight tandem mass spectrometer (QSTAR XL, Applied Biosystems/MDS Sciex). RPLC was performed using a PepMap column (75 ␮m ID ⫻ 150 mm long, LC Packings, Dionex), and the peptides were separated using a linear gradient of 2%–35% acetonitrile in 0.1% formic acid over 50 min. The eluted peptides were subsequently ionized by an applied voltage of 2 kV to a PicoTip emitter (New Objective), which was mounted onto a nanoelectrospray ion source (Applied Biosystems/MDS Sciex). The QSTAR MS was first operated in an information-dependent mode in which each full MS scan (1 s acquisition) was followed by three MS/MS scans (2 s acquisition for each MS/MS scan) where the three most-abundant peptide molecular ions were dynamically selected for low energy collision-induced dissociation (CID). The monoisotopic masses of both parent ions and their corresponding fragment ions, parent ion charge states and ion intensities from the acquired tandem mass spectra (MS/ MS), were automatically extracted using mascot script and directly submitted for automated database searching for protein identification and characterization using both MASCOT (www.matrixsciences. com) and development version of Protein Prospector installed locally. SwissProt protein database (9.28.2004) was used for database searching. The Tec1-myc sequence was also input into the userdefined database in Protein Prospector. To obtain the best quality MS/MS spectra of the desired phosphorylated peptides, information-independent data acquisition was carried out in which each full MS scan was followed by two MS/MS scans of the parent ions with known peptide masses at two different collision energies (30V and 36V) during the entire LC MS/MS run. The determined phosphorylation sites by automated database searching were confirmed by manual interpretation of the MS/MS spectra.

In Vivo Ubiquitination Assay An in vivo ubiquitination assay was carried out as previous described (Flick et al., 2004) except that ␣ factor was added 3.5 hr after the induction of the CUP1 promoter and cells were collected 1 hr after ␣ factor induction. Tec1-myc was detected with a peroxidase conjugated anti-myc antibody (Sigma). ␤-Galactosidase Assay A TCS-lacZ plasmid was constructed by inserting the AGAATGTG CATTATCGATTCATTCT sequence into the XhoI site in the pLG669Z plasmid (Guarente and Ptashne, 1981). A 3.4 kb fragment of TEC1, including its 1500 bp upstream and 500 bp downstream region, was amplified by PCR and inserted into a pRS vector to create plasmid TEC1p-TEC1/CEN. The XbaI-Stu1 region of TEC1 (including T273) was replaced with that from GAL1p-TEC1T273V-myc to generate TEC1p-TEC1T273V-CEN. A ␤-galactosidase assay was performed as previously described (Rose and Botstein, 1983), except that ␣-Pr was added to the breaking buffer. Acknowledgments We thank Peter Kaiser for strains and plasmids and for helpful suggestions and Peter Kaiser and members of the Liu lab for critical reading of the manuscript. This work is supported by National Institutes of Health Grant GM55155 (to H.L.). Received: September 8, 2004 Revised: October 28, 2004 Accepted: November 29, 2004 Published: December 28, 2004 References Andersson, J., Simpson, D.M., Qi, M., Wang, Y., and Elion, E.A. (2004). Differential input by Ste5 scaffold and Msg5 phosphatase route a MAPK cascade to multiple outcomes. EMBO J. 23, 2564– 2576. Breitkreutz, A., Boucher, L., and Tyers, M. (2001). MAPK specificity in the yeast pheromone response independent of transcriptional activation. Curr. Biol. 11, 1266–1271. Brill, J.A., Elion, E.A., and Fink, G.R. (1994). A role for autophosphorylation revealed by activated alleles of FUS3, the yeast MAP kinase homolog. Mol. Biol. Cell 5, 297–312. Cook, J.G., Bardwell, L., Kron, S.J., and Thorner, J. (1996). Two novel targets of the MAP kinase Kss1 are negative regulators of invasive growth in the yeast Saccharomyces cerevisiae. Genes Dev. 10, 2831–2848. Cook, J.G., Bardwell, L., and Thorner, J. (1997). Inhibitory and activating functions for MAPK Kss1 in the S. cerevisiae filamentousgrowth signalling pathway. Nature 390, 85–88. Deshaies, R.J. (1999). SCF and Cullin/Ring H2-based ubiquitin ligases. Annu. Rev. Cell Dev. Biol. 15, 435–467. Elion, E.A., Brill, J.A., and Fink, G.R. (1991). Functional redundancy in the yeast cell cycle: FUS3 and KSS1 have both overlapping and unique functions. Cold Spring Harb. Symp. Quant. Biol. 56, 41–49. Elion, E.A., Satterberg, B., and Kranz, J.E. (1993). FUS3 phosphorylates multiple components of the mating signal transduction cascade: evidence for STE12 and FAR1. Mol. Biol. Cell 4, 495–510. Feldman, R.M., Correll, C.C., Kaplan, K.B., and Deshaies, R.J. (1997). A complex of Cdc4p, Skp1p, and Cdc53p/cullin catalyzes ubiquitination of the phosphorylated CDK inhibitor Sic1p. Cell 91, 221–230. Flick, K., Ouni, I., Wohlschlegel, J.A., Capati, C., McDonald, W.H., Yates, J.R., and Kaiser, P. (2004). Proteolysis-independent regulation of the transcription factor Met4 by a single Lys 48-linked ubiquitin chain. Nat. Cell Biol. 6, 634–641. Guarente, L., and Ptashne, M. (1981). Fusion of Escherichia coli lacZ to the cytochrome c gene of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 78, 2199–2203. Hodgins, R.R., Ellison, K.S., and Ellison, M.J. (1992). Expression of a ubiquitin derivative that conjugates to protein irreversibly produces

Cell 990

phenotypes consistent with a ubiquitin deficiency. J. Biol. Chem. 267, 8807–8812.

(1993). FAR1 links the signal transduction pathway to the cell cycle machinery in yeast. Cell 73, 747–760.

Huang, L., Jacob, R.J., Pegg, S.C., Baldwin, M.A., Wang, C.C., Burlingame, A.L., and Babbitt, P.C. (2001). Functional assignment of the 20 S proteasome from Trypanosoma brucei using mass spectrometry and new bioinformatics approaches. J. Biol. Chem. 276, 28327– 28339.

Roberts, R.L., and Fink, G.R. (1994). Elements of a single MAP kinase cascade in Saccharomyces cerevisiae mediate two developmental programs in the same cell type: mating and invasive growth. Genes Dev. 8, 2974–2985.

Kaiser, P., Sia, R.A., Bardes, E.G., Lew, D.J., and Reed, S.I. (1998). Cdc34 and the F-box protein Met30 are required for degradation of the Cdk-inhibitory kinase Swe1. Genes Dev. 12, 2587–2597.

Roberts, C.J., Nelson, B., Marton, M.J., Stoughton, R., Meyer, M.R., Bennett, H.A., He, Y.D., Dai, H., Walker, W.L., Hughes, T.R., et al. (2000). Signaling and circuitry of multiple MAPK pathways revealed by a matrix of global gene expression profiles. Science 287, 873–880.

Kipreos, E.T., and Pagano, M. (2000). The F-box protein family. Genome Biol 1, REVIEWS3002. Published online November 10, 2000. 10.1186/gb-2000-1-5-reviews3002

Rose, M., and Botstein, D. (1983). Construction and use of gene fusions to lacZ (beta-galactosidase) that are expressed in yeast. Methods Enzymol. 101, 167–180.

Koepp, D.M., Harper, J.W., and Elledge, S.J. (1999). How the cyclin became a cyclin: regulated proteolysis in the cell cycle. Cell 97, 431–434.

Sabbagh, W., Jr., Flatauer, L.J., Bardwell, A.J., and Bardwell, L. (2001). Specificity of MAP kinase signaling in yeast differentiation involves transient versus sustained MAPK activation. Mol. Cell 8, 683–691.

Koepp, D.M., Schaefer, L.K., Ye, X., Keyomarsi, K., Chu, C., Harper, J.W., and Elledge, S.J. (2001). Phosphorylation-dependent ubiquitination of cyclin E by the SCFFbw7 ubiquitin ligase. Science 294, 173–177. Kohler, T., Wesche, S., Taheri, N., Braus, G.H., and Mosch, H.U. (2002). Dual role of the Saccharomyces cerevisiae TEA/ATTS family transcription factor Tec1p in regulation of gene expression and cellular development. Eukaryot. Cell 1, 673–686. Kusari, A.B., Molina, D.M., Sabbagh, W., Jr., Lau, C.S., and Bardwell, L. (2004). A conserved protein interaction network involving the yeast MAP kinases Fus3 and Kss1. J. Cell Biol. 164, 267–277. Liu, H., Styles, C.A., and Fink, G.R. (1993). Elements of the yeast pheromone response pathway required for filamentous growth of diploids. Science 262, 1741–1744. Longtine, M.S., McKenzie, A., III, Demarini, D.J., Shah, N.G., Wach, A., Brachat, A., Philippsen, P., and Pringle, J.R. (1998). Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953–961. Madhani, H.D., and Fink, G.R. (1997). Combinatorial control required for the specificity of yeast MAPK signaling. Science 275, 1314–1317.

Sikorski, R.S., and Hieter, P. (1989). A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122, 19–27. Skowyra, D., Craig, K.L., Tyers, M., Elledge, S.J., and Harper, J.W. (1997). F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex. Cell 91, 209–219. Songyang, Z., Lu, K.P., Kwon, Y.T., Tsai, L.H., Filhol, O., Cochet, C., Brickey, D.A., Soderling, T.R., Bartleson, C., Graves, D.J., et al. (1996). A structural basis for substrate specificities of protein Ser/ Thr kinases: primary sequence preference of casein kinases I and II, NIMA, phosphorylase kinase, calmodulin-dependent kinase II, CDK5, and Erk1. Mol. Cell. Biol. 16, 6486–6493. Tedford, K., Kim, S., Sa, D., Stevens, K., and Tyers, M. (1997). Regulation of the mating pheromone and invasive growth responses in yeast by two MAP kinase substrates. Curr. Biol. 7, 228–238. Yamano, H., Kitamura, K., Kominami, K., Lehmann, A., Katayama, S., Hunt, T., and Toda, T. (2000). The spike of S phase cyclin Cig2 expression at the G1-S border in fission yeast requires both APC and SCF ubiquitin ligases. Mol. Cell 6, 1377–1387.

Madhani, H.D., Styles, C.A., and Fink, G.R. (1997). MAP kinases with distinct inhibitory functions impart signaling specificity during yeast differentiation. Cell 91, 673–684.

Zeitlinger, J., Simon, I., Harbison, C.T., Hannett, N.M., Volkert, T.L., Fink, G.R., and Young, R.A. (2003). Program-specific distribution of a transcription factor dependent on partner transcription factor and MAPK signaling. Cell 113, 395–404.

Matheos, D., Metodiev, M., Muller, E., Stone, D., and Rose, M.D. (2004). Pheromone-induced polarization is dependent on the Fus3p MAPK acting through the formin Bni1p. J. Cell Biol. 165, 99–109.

Zhan, X.L., and Guan, K.L. (1999). A specific protein-protein interaction accounts for the in vivo substrate selectivity of Ptp3 towards the Fus3 MAP kinase. Genes Dev. 13, 2811–2827.

Morris, M.C., Kaiser, P., Rudyak, S., Baskerville, C., Watson, M.H., and Reed, S.I. (2003). Cks1-dependent proteasome recruitment and activation of CDC20 transcription in budding yeast. Nature 423, 1009–1013. Mosch, H.U., and Fink, G.R. (1997). Dissection of filamentous growth by transposon mutagenesis in Saccharomyces cerevisiae. Genetics 145, 671–684. Nash, P., Tang, X., Orlicky, S., Chen, Q., Gertler, F.B., Mendenhall, M.D., Sicheri, F., Pawson, T., and Tyers, M. (2001). Multisite phosphorylation of a CDK inhibitor sets a threshold for the onset of DNA replication. Nature 414, 514–521. Nateri, A.S., Riera-Sans, L., Da Costa, C., and Behrens, A. (2004). The ubiquitin ligase SCFFbw7 antagonizes apoptotic JNK signaling. Science 303, 1374–1378. Oehlen, L., and Cross, F.R. (1998). The mating factor response pathway regulates transcription of TEC1, a gene involved in pseudohyphal differentiation of Saccharomyces cerevisiae. FEBS Lett. 429, 83–88. Orlicky, S., Tang, X., Willems, A., Tyers, M., and Sicheri, F. (2003). Structural basis for phosphodependent substrate selection and orientation by the SCFCdc4 ubiquitin ligase. Cell 112, 243–256. Palecek, S.P., Parikh, A.S., and Kron, S.J. (2000). Genetic analysis reveals that FLO11 upregulation and cell polarization independently regulate invasive growth in Saccharomyces cerevisiae. Genetics 156, 1005–1023. Peter, M., Gartner, A., Horecka, J., Ammerer, G., and Herskowitz, I.