The N Terminus of Saccharomyces cerevisiae Sst2p Plays an RGS-Domain-Independent, Mpt5p-Dependent Role in Recovery From Pheromone Arrest

Corresponding author: Janet Kurjan, 5400 NE 200th Pl., Lake Forest Park, WA 98155., jkurjan2{at}attbi.com (E-mail)

Communicating editor: S. SANDMEYER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Saccharomyces cerevisiae RGS protein Sst2p is involved in desensitization to pheromone and acts as a GTPase-activating protein for the G subunit Gpa1p. Other results indicate that Sst2p acts through Mpt5p and that this action occurs downstream of Fus3p and through Cln3p/Cdc28p. Our results indicate that the interaction of Sst2p with Mpt5p requires the N-terminal MPI (Mpt5p-interacting) domain of Sst2p and is independent of the C-terminal RGS domain. Overexpression of the MPI domain results in an Mpt5p-dependent increase in recovery from pheromone arrest. Overexpression of either intact Sst2p or the MPI domain leads to partial suppression of a gpa1 growth defect, and this suppression is dependent on Mpt5p, indicating that MPI function occurs downstream of Gpa1p and through Mpt5p. Combination of an mpt5 mutation with the GPA1^G302S mutation, which uncouples Gpa1p from Sst2p, results in pheromone supersensitivity similar to the sst2 mutant, and promotion of recovery by overexpression of Sst2p is dependent on both Mpt5p and the Gpa1p interaction. These results indicate that Sst2p is a bifunctional protein and that the MPI domain acts through Mpt5p independently of the RGS domain. RGS family members from other fungi contain N-terminal domains with sequence similarity to the Sst2p MPI domain, suggesting that MPI function may be conserved.

MATING between a and cells of Saccharomyces cerevisiae involves the secretion of the peptide pheromones, a- and -factor, respectively. Response to pheromone results in changes in cellular morphology, transcriptional activation of genes involved in this process, and arrest in the G₁ phase of the cell cycle (reviewed in SPRAGUE and THORNER 1992 ; KURJAN 1993 ). The pheromones act through G-protein-linked receptors (Ste2p in a cells and Ste3p in cells) and a heterotrimeric G protein, consisting of the Gpa1p (), Ste4p (ß), and Ste18p () subunits. In the yeast pathway, Ste4p/Ste18p (ß) plays a positive role in activation of the pathway, and Gpa1p () acts upstream of Ste4p/Ste18p to keep the pathway inactive in the absence of pheromone (DIETZEL and KURJAN 1987B ; MIYAJIMA et al. 1987 ; WHITEWAY et al. 1989 ). Therefore, ste4 or ste18 mutations result in defects in pheromone response and mating, whereas gpa1 mutations result in constitutive activation of the pathway and cell cycle arrest. Downstream components of the pathway include a mitogen-activated protein (MAP) kinase cascade and the Ste12p transcription factor. Cell cycle arrest involves Far1p, which is phosphorylated by the MAP kinase homolog Fus3p, and then acts to inhibit G1 cyclin (Cln1p or Cln2p)-Cdc28p kinase activity.

After a period of exposure to pheromone, cells are able to recover from pheromone-induced arrest. Inactivation of a- and -factor by secreted activities plays a role in recovery (MANNEY 1983 ; MACKAY et al. 1988 ; MARCUS et al. 1991 ). An additional desensitization process was identified by the characterization of sst2 mutants (CHAN and OTTE 1982A , CHAN and OTTE 1982B ; MOORE 1984 ). sst2 mutations result in a 200-fold increase in sensitivity to pheromone and a defect in recovery from pheromone arrest. Sst2p plays a similar role in a and cells, and this role is cell intrinsic and independent of pheromone degradation. Sst2p expression is increased by exposure to pheromone (DIETZEL and KURJAN 1987A ; DOHLMAN et al. 1995 , DOHLMAN et al. 1996 ).

The identification of a family of Sst2p homologs, called the RGS family, has provided additional information on Sst2p function. The discovery that the mammalian RGS protein GAIP could interact with G_i3 in the two-hybrid system (DE VRIES et al. 1995 ) suggested that RGS proteins could regulate G subunits. Subsequently, many RGS proteins have been shown to act as GTPase-activating proteins (GAPs) for G subunits (BERMAN and GILMAN 1998 ). Similarly, Sst2p has been shown to act as a GAP for Gpa1p (APANOVITCH et al. 1998 ; KALLAL and FISHEL 2000 ).

In a separate approach to investigating Sst2p function, Mpt5p was identified in a two-hybrid screen for Sst2p-interacting proteins (CHEN and KURJAN 1997 ). mpt5 mutations result in an increase in sensitivity to pheromone and a defect in recovery from pheromone arrest, although the phenotypes are less severe than those observed for sst2 mutants (KIKUCHI et al. 1994 ; CHEN and KURJAN 1997 ). Overexpression of either Mpt5p or Sst2p results in increased recovery from pheromone arrest (CHEN and KURJAN 1997 ). Increased recovery resulting from Mpt5p overexpression requires Sst2p, but increased recovery resulting from Sst2p overexpression is only partially dependent on Mpt5p. The ability of Mpt5p to interact with Fus3p and the ability of either mpt5 or cln3 mutations to partially suppress the pheromone arrest and mating defects of fus3 mutants suggested that Mpt5p might act downstream of Fus3p and in the same pathway as Cln3p. On the basis of the similar effects of the mpt5 and cln3 mutations and the ability of Mpt5p to interact with Cdc28p, we proposed that the role of Mpt5p in pheromone sensitivity and recovery occurs through G₁ cyclin/Cdc28p activity, although the mechanism of this action is unknown.

The suggestion that Sst2p-dependent Mpt5p function in pheromone sensitivity and recovery occurs at the level of the cell cycle machinery (CHEN and KURJAN 1997 ) seemed to contradict results indicating that Sst2p acts as a GAP for Gpa1p (APANOVITCH et al. 1998 ; KALLAL and FISHEL 2000 ). Characterization of Sst2p domains indicates that the interaction of Sst2p with Mpt5p is independent of the RGS domain and that the N-terminal Mpt5p-interacting domain plays a role that is independent of the RGS domain.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains:
Escherichia coli strain TG1 was used for plasmid analysis and construction. S. cerevisiae strains are shown in Table 1. N435-1A (MATa) and N435-2A (MAT) were used as mating testers. Y190 and Y187 (provided by S. Elledge) were used for the two-hybrid assays. All other yeast strains used in this study were isogenic to W303-1A (MATa ade2-1 can1-100 ura3-1 leu2-3, 112 trp1-1 his3-11,15).

The a gpa1::LEU2 [pTRP-MAT] and a gpa1::LEU2 [pRS316-GPA1] strains were obtained by transforming D111 with the indicated plasmids, followed by sporulation and dissection. Strains BXK41-BXK46, KSK1, and KSK32-7C were made by crosses, followed by sporulation and dissection to obtain the haploid strains of the opposite mating type or double mutants.

To generate a genomic GPA1^G302S replacement, the XhoI-PstI fragment from pGPA1^G302S (described below) was cotransformed with a LEU2-containing plasmid (pRS425; CHRISTIANSON et al. 1992 ) into SDK1. Leu⁺ transformants were selected and replicated to 5-fluoroorotic acid (5-FOA) medium to identify replacements of gpa1::URA3. The ability of the strain to lose pTRP-MAT was tested to confirm the replacement of gpa1::URA3 with a functional allele and pheromone sensitivity of the GPA1^G302S replacement was tested for the phenotype previously observed (JACKSON and HARTWELL 1990 ; DIBELLO et al. 1998 ).

Site-directed mutagenesis and other plasmid constructions:
The SST2^P20L mutation was constructed by using the Quikchange site-directed mutagenesis kit (Stratagene, La Jolla, CA) with the oligonucleotides P20L-1 and P20L-2 (Table 2) and pRS316-SST2 as the template. The mutation was confirmed by automated sequencing, and several independent mutant constructs were tested for phenotypes.

The GPA1^G302S allele was obtained by oligonucleotide-directed mutagenic PCR (KALLAL and KURJAN 1997 ) using the mutagenic primer G302ser-KS1 and PCR3'-1 (Table 2) for a first round of PCR, and primers 002 and PCR3'-1 (primer 002 is toward the 5' end of the gene) for a second round of PCR using the first-round PCR product and the overlapping EcoRI-SphI fragment of wild-type GPA1 as the template. The resulting product was cleaved with HindIII-SphI and subcloned into a pSK(+) plasmid containing 1.9 and 1.5 kb of 5' and 3' flanking sequences, respectively (DESIMONE and KURJAN 1998 ), to produce pGPA1^G302S. The mutant sequence was confirmed by sequencing of the GPA1 open reading frame.

To construct pPGK-SST2^P20L, the SST2^P20L fragment was amplified from pRS316-SST2^P20L by PCR using oligonucleotides SST2-H (33 nucleotides upstream of the SST2 initiation codon; Table 2) and SST2-3, cleaved with BamHI-HindIII and ligated into the pPGK vector.

To express N-terminal portions of SST2 under the control of the PGK promoter, the sst2[1-271] and sst2[1-401] HindIII-BamHI fragments were amplified using the upstream oligonucleotide SST2-H and the downstream oligonucleotides SST2-4 and SST2-5, respectively (Table 2), cleaved, and subcloned into pPGK. pPGK-sst2^P20L[1-271] and pPGK-sst2^P20L[1-401] were constructed by the same approach using the mutant templates.

To make SST2 constructs expressed from the PGK promoter in a HIS3 vector, the PGK-SST2, PGK-SST2^P20L, PGK-sst2[1-401], and PGK-sst2^P20L[1-401] fragments from the respective pPGK-SST2 constructs were amplified by PCR using oligonucleotides PGK-5 and PGK-3 (Table 2) and subcloned into pRS423.

The 4.5-kb SST2 HindIII fragment from pHSS6-B3 (DIETZEL and KURJAN 1987A ) was subcloned into pRS315 and pRS316 to construct YCp-SST2 plasmids. The 1.9-kb GPA1 EcoRI fragment was subcloned into pRS316 and pRS313 to produce YCp-GPA1 plasmids and into pRS423 to produce a YEp-GPA1 plasmid.

Two-hybrid assays:
To test interactions of various portions of Sst2p with Mpt5p, regions of SST2 were amplified by PCR using the oligonucleotides indicated in Table 2 and subcloned into pAS2 to make fusion to the Gal4p-binding domain (BD). A BD-Sst2p^P20L fusion was constructed by the same method.

Two-hybrid interactions were assayed using the colony lift assay for ß-galactosidase activity (BREEDEN and NASMYTH 1985 ). Blue color appeared after incubation in 1 mg/ml X-gal in Z buffer for times ranging from 30 min to overnight. To quantitate ß-galactosidase activity, strains were grown in medium selective for plasmids to an OD₆₀₀ of 0.2–0.5 and harvested. Cells were prepared and permeabilized, o-nitrophenyl-ß-D-galactopyranoside was used as a substrate, and ß-galactosidase units were calculated as described (AUSUBEL et al. 1989 ).

Plasmid shuffle experiments:
Plasmid shuffle experiments utilized MATa gpa1::LEU2 strains containing either the pTRP-MAT or the YCp-GPA1 (URA3) plasmid; these strains are unable to lose the plasmid due to the growth defect of haploid gpa1 cells. For the pTRP-MAT experiments, the strain also contained a URA3 plasmid with an SST2 construct expressed under the control of the PGK promoter. Cells were grown in YEPD broth for 24 hr, plated for single colonies, and replica plated onto -trp medium to determine the percentage of colonies among 200–400 total colonies that had lost the pTRP-MAT plasmid. For the YCp-GPA1 (URA3) experiments, the PGK-SST2 constructs were present on HIS3 plasmids, and the ability to lose the YCp-GPA1 plasmid was scored by growth on 5-FOA medium.

Pheromone response and mating assays:
Pheromone spotting assays to test pheromone response and recovery were performed as described previously (DIETZEL and KURJAN 1987A , DIETZEL and KURJAN 1987B ); 1 x 10⁶ log-phase cells were spread onto YEPD or dropout plates, 4 µg -factor (Sigma, St. Louis) was spotted on the nascent lawn or onto a paper filter on the lawn, and the zone of growth inhibition was observed after incubation at 30° for 24–48 hr.

Qualitative mating assays were done by replica plating master plates onto thick lawns of tester cells (N435-2A) spread in 0.3 ml of YEPD medium on minimal plates and incubating at 30° for 48 hr (SPRAGUE 1991 ). Quantitative mating assays were done as described previously (XU and KURJAN 1997 ) using 3 x 10⁶ W303-1A cells containing various plasmids and 1.5 x 10⁷ W303-1B (Trp⁺) cells and incubating the mating mixtures on YEPD plates for 4–5 hr. Mating efficiencies were calculated by dividing the number of diploid colonies (-ura -trp double dropout medium) by the total number of diploid plus W303-1A colonies (-ura dropout medium).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Sst2p RGS domain is not necessary for the Mpt5p interaction:
To identify the region of Sst2p involved in the Mpt5p interaction, BD constructs containing various portions of Sst2p were tested for the ability to interact with activation domain (AD)-Mpt5p using the two-hybrid system (Fig 1). Western blots using anti-HA antibody indicated that all of the BD-Sst2p constructs produced proteins of the predicted M_r at relatively similar levels (data not shown). The RGS-domain-containing constructs BD-Sst2p[273-698] and BD-Sst2[410-698] did not show detectable interactions with AD-Mpt5p. The N-terminal BD-Sst2p[1-401] construct, however, was able to interact with AD-Mpt5p, indicating that the RGS domain was not necessary for this interaction. The interaction of AD-Mpt5p with BD-Sst2p[1-401] was 10-fold stronger than its interaction with BD-Sst2p, suggesting that the RGS domain inhibits the Sst2p-Mpt5p interaction. We refer to the N-terminal, Mpt5p-interacting domain of Sst2p (amino acids 1–401) as the MPI domain. Further truncations of the MPI domain indicated that residues within the first 135 amino acids of Sst2p and residues between 333 and 401 of Sst2p were required for the Mpt5p interaction.

View larger version (13K): In this window In a new window Download PPT slide   Figure 1. Identification of the Mpt5-interacting domain of Sst2p. Fusions of intact Sst2p or various portions of Sst2p in the Gal4p DNA-BD vector were tested for interactions with the Gal4p transcriptional AD fusion, AD-Mpt5p in Y187. +++, dark blue colonies in the colony lift assay (43 units ß-galactosidase activity); +, blue colonies (4 units ß-galactosidase activity); -, white colonies (undetectable ß-galactosidase activity).

A dominant gain-of-function SST2 mutant, SST2^P20L, that inhibits transcriptional induction and cell cycle arrest in response to pheromone has been identified (DOHLMAN et al. 1995 ). Because this mutation is within the MPI domain and potentially could act by increasing the affinity of the interaction with an Sst2p target, we tested whether this mutation affected the Mpt5p interaction. However, the P20L mutation did not alter the interaction of either Sst2p or Sst2p[1-401] in the two-hybrid assays (data not shown).

The Sst2p MPI domain plays an Mpt5p-dependent role in recovery:
To determine whether the Sst2p MPI domain plays a role in pheromone sensitivity or recovery, we tested whether overexpression of this domain had a phenotypic effect. Whereas the sst2 mutant shows a large, clear zone of growth arrest by pheromone (Fig 2, pPGK control), overexpression of intact Sst2p under the control of the high-expression, constitutive PGK promotor results in a small zone of growth arrest, and cells at the edge of the zone rapidly resume growth due to promotion of recovery (CHEN and KURJAN 1997 ). The PGK-sst2[1-401] construct reduced pheromone sensitivity, as indicated by a zone of growth arrest smaller than the PGK control (Fig 2). It also resulted in a mild increase in recovery, as indicated by the slight growth of cells within the zone of growth arrest. However, both effects of PGK-sst2[1-401] were mild in comparison to PGK-SST2.

View larger version (102K): In this window In a new window Download PPT slide   Figure 2. Assays for MPI domain activity. Various SST2 constructs expressed under the control of the PGK promotor were tested in the MATa sst2::LEU2 MPT5 and MATa sst2::LEU2 mpt5::ADE2 strains using the pheromone spotting assay for cell cycle arrest; 4 µg of -factor on the disk diffuses into the surrounding medium and results in growth arrest.

We tested whether the dominant P20L mutation had any effect on the phenotype resulting from overexpression of Sst2p[1-401] (Fig 2). The diameters of the zones of growth arrest of the sst2 mutant containing the PGK-SST[1-401] and PGK-SST2^P20L[1-401] constructs were similar; however, the P20L mutation greatly increased the turbidity of the zone, suggesting that Sst2p^P20L[1-401] promotes recovery from pheromone arrest.

Two results indicated that the role of the N-terminal MPI domain of Sst2p is dependent on Mpt5p. Overexpression of Sst2p[1-271], with or without the P20L mutation, did not affect the sst2 phenotype (Fig 2). Because Sst2p[1-401] can interact with Mpt5p, but Sst2p[1-271] cannot (Fig 1), this result suggests that the Mpt5p interaction is necessary for Sst2p[1-401] function. In addition, the effect of the PGK-SST2^P20L[1-401] construct on recovery observed in the sst2 mutant was not observed in the sst2 mpt5 double mutant (Fig 2). These results indicate that Sst2p[1-401] plays an RGS-domain-independent, Mpt5p-dependent role in recovery from pheromone arrest.

Overexpression of Sst2p inhibits mating:
Although overexpression of SST2 results in increased recovery from pheromone-induced cell cycle arrest and reduced induction of a pheromone-inducible reporter gene (DIETZEL and KURJAN 1987A ; DOHLMAN et al. 1995 ), overexpression of SST2 under a galactose-inducible promoter was not sufficient to inhibit mating (DOHLMAN et al. 1995 ). Because the PGK promoter results in a higher level of expression than the GAL promoter (KANG et al. 1990 ), we tested the effect of the PGK constructs on the mating of a wild-type strain. PGK-SST2 resulted in only a mild effect, a 4-fold decrease in mating (Table 3). PGK-SST2^P20L, however, resulted in a 100-fold decrease in mating. In contrast, neither PGK-sst2[1-401] nor PGK-sst2^P20L[1-401] resulted in a significant decrease in mating, indicating that the RGS domain is necessary for this effect.

Overexpression of Sst2p can partially suppress the gpa1 growth defect:
GAL-SST2 and GAL-SST2^P20L constructs were previously shown to be insufficient to allow suppression of the growth defect resulting from a gpa1 mutation or activating STE4 mutations, consistent with Sst2p action occurring at the level of Gpa1p (DOHLMAN et al. 1995 ). Because inhibition of mating was observed with PGK-SST2^P20L (Table 3) but not with GAL-SST2^P20L (DOHLMAN et al. 1995 ), we tested whether the higher constitutive expression of the various forms of Sst2p under PGK control could suppress the gpa1 growth defect.

Whereas haploid gpa1 mutants are unable to grow due to constitutive activation of the pheromone response pathway (DIETZEL and KURJAN 1987B ; MIYAJIMA et al. 1987 ; WHITEWAY et al. 1989 ), a/ gpa1/gpa1 diploids grow normally due to lack of expression of downstream components of the pheromone response pathway. A haploid a gpa1 strain containing a MAT plasmid mimics an a/ gpa1/gpa1 diploid and therefore is able to grow. However, the a gpa1[pTRP-MAT] strain is unable to lose the MAT plasmid (Table 4), thus providing a plasmid shuffle assay (MANN et al. 1987 ) for suppression of the a gpa1 growth defect. Addition of the complementing pGPA1 plasmid allowed loss of pTRP-MAT from almost half of the colonies from both the a gpa1 MPT5 [pTRP-MAT] and a gpa1 mpt5 [pTRP-MAT] strains. pPGK-SST2 allowed a reproducible, low level of loss of pTRP-MAT from the gpa1 strain, suggesting that overexpression of Sst2p under PGK control can weakly suppress the gpa1 growth defect. This effect was dependent on Mpt5p; no loss of pTRP-MAT was observed in a gpa1 mpt5 mutant. pPGK-sst2[1-401] allowed a similar level of Mpt5p-dependent pTRP-MAT loss as pPGK-SST2, indicating that the Sst2p MPI domain is sufficient for suppression. However, pPGK-sst2[1-271] did not allow pTRP-MAT loss, consistent with a requirement for the Sst2p-Mpt5p interaction for this effect. The P20L mutation in either intact Sst2p or Sst2p[1-401] led to at most a mild increase in the ability to lose pTRP-MAT.

Because the level of pTRP-MAT loss from the gpa1 strain containing the PGK-SST2 constructs was low in the above assays, we tested suppression of the gpa1 growth defect by additional assays. The various PGK-SST2 constructs were introduced into an a/ GPA1/gpa1::LEU2 diploid and the strains were characterized by tetrad analysis. Dissection of an a/ GPA1/gpa1 diploid produces two viable GPA1 spores and two gpa1 spores that form barely discernible colonies (Fig 3A, pPGK control; DIETZEL and KURJAN 1987B ; MIYAJIMA et al. 1987 ). Due to plasmid loss, dissection of the a/ GPA1/gpa1 diploid containing the complementing pGPA1 plasmid resulted in two to four viable spores per tetrad. Dissection of the a/ GPA1/gpa1 diploid containing the PGK-SST2 or PGK-sst2[1-401] constructs, with or without the P20L mutation, also resulted in tetrads with two to four viable spores. In each tetrad, two colonies showed wild-type growth, consistent with a 2 GPA1:2 gpa1 segregation pattern. The remaining viable spores, which are predicted to be the gpa1 spores containing a PGK-SST2 construct, showed quite variable growth. These results are also consistent with the ability of overexpression of Sst2p or Sst2p[1-401] to partially suppress the gpa1 growth defect. Similar experiments using an a/ GPA1/gpa1::LEU2 mpt5::ADE2/mpt5::ADE2 diploid indicated that suppression of the gpa1 growth defect by overexpression of Sst2p or Sst2p[1-401] was dependent on Mpt5p (data not shown).

View larger version (91K): In this window In a new window Download PPT slide   Figure 3. Assays for suppression of the gpa1 growth defect. The ability of various SST2 constructs expressed under the PGK promoter to suppress the gpa1 growth defect was tested using alternative approaches. (A) The URA3-containing plasmids containing the different SST2 constructs, as well as the vector (pPGK) and YCp-GPA1 (pGPA1) controls, were transformed into the a/ GPA1/gpa1::LEU2 (D111) diploid, the diploids were sporulated and dissected, and the growth of spore colonies was observed after growth at 30° for 3 days. (B) Plasmid shuffle assays were done in MATa gpa1::LEU2 MPT5 and MATa gpa1::LEU2 mpt5::ADE2 strains containing pGPA1-URA3, with the SST2 constructs and controls present on a HIS3 plasmid. The ability of the strains to lose pGPA1-URA3 was monitored by the ability to grow on 5-FOA plates.

A final approach involved plasmid shuffle experiments using a gpa1::LEU2 [YCp-GPA1-URA3] strain, allowing loss of the GPA1 plasmid to be assayed by growth on 5-FOA plates. Addition of the complementing GPA1(HIS3) plasmid allowed Mpt5p-independent growth on 5-FOA medium, whereas the control vector did not allow growth (Fig 3B). The YEp-PGK-SST2 and YEp-PGK-sst2[1-401] HIS3-containing plasmids allowed loss of the GPA1(URA3) plasmid. Addition of the P20L mutation in either the YEp-PGK-SST2 or the YEp-PGK-sst2[1-401] plasmids allowed slightly better growth on 5-FOA plates, suggesting that this mutation may have a mild effect on the ability to suppress the gpa1 growth defect. The PGK-SST2 constructs did not allow growth of the gpa1 mpt5 [pGPA1-URA3] strain on 5-FOA medium, indicating that the ability of overexpression of Sst2p or Sst2p[1-401] to suppress the gpa1 growth defect was dependent on Mpt5p.

In all of the above suppression assays, the growth of the gpa1 mutant containing the various PGK-SST2 constructs was slower than the growth of the gpa1 [pGPA1] strain, indicating that suppression of the gpa1 growth defect was partial. In addition, the gpa1 [PGK-SST2] cells showed aberrant morphology similar to cells with mutations that lead to partial activation of the pheromone response pathway (MIYAJIMA et al. 1989 ; KURJAN et al. 1991 ; XU and KURJAN 1997 ).

Because mutations of downstream components of the pheromone response pathway (ste mutations) can suppress the gpa1 growth defect (NAKAYAMA et al. 1988 ; BLINDER et al. 1989 ), it was possible that the growth of the gpa1 mutant resulted from a selection for ste mutations rather than from suppression by the PGK-SST2 constructs. In such a case, addition of a GPA1 plasmid to the resulting gpa1 ste [pPGK-SST2] strains should not allow mating. However, addition of a pGPA1 plasmid to the viable gpa1 [pPGK-SST2] strains allowed efficient mating (data not shown).

Another test for the presence of ste mutations that suppress the gpa1 growth defect involved an assay for the ability of the gpa1::LEU2 strains containing the various PGK-SST2 (URA3) constructs to lose the plasmid. A gpa1 ste strain should be able to lose the PGK-SST2 plasmid, whereas the strain should be unable to lose the PGK-SST2 plasmid if growth requires overexpression of Sst2p. The gpa1 [pPGK-SST2-URA3] strains were unable to grow on 5-FOA medium (Fig 4, top), indicating a selection for maintenance of the plasmid. In a control experiment, addition of a complementing YCp-GPA1 plasmid allowed loss of pPGK-SST2-URA3 from the gpa1 strains, as indicated by growth on 5-FOA medium (Fig 4, bottom). Therefore, the growth of the gpa1 strains was dependent on the PGK-SST2 plasmids and the suppression of the gpa1 growth defect results from overexpression of Sst2p or Sst2p[1-401] and not from the isolation of downstream ste mutations that suppress the gpa1 growth defect.

View larger version (39K): In this window In a new window Download PPT slide   Figure 4. Dependence of gpa1 growth on Sst2p overexpression. In this assay, the pPGK-SST2 plasmids contain a URA3 marker to allow assays for the dependence of the growth of the gpa1::LEU2 strain on the overexpression of Sst2p, as monitored by growth on 5-FOA plates. The YCp-GPA1 plasmid complements the gpa1::LEU2 mutation, providing a control condition under which loss of pPGK-SST2 can occur.

The GPA1^G302S mpt5 double mutant shows a phenotype similar to an sst2 mutant:
The results described above suggest that the Sst2p MPI domain plays a role that acts through Mpt5p that is independent of the role of the RGS domain, which acts as a GAP for Gpa1p. To examine these proposed roles further, we made use of a GPA1 mutation that uncouples RGS proteins from G subunits. The GPA1^G302S mutant was first identified by a defect in discrimination (JACKSON and HARTWELL 1990 ), which is a process by which cells are able to choose mating partners that secrete high levels of pheromone; sst2 mutants also show discrimination defects. The GPA1^G302S mutation also results in supersensitivity to pheromone, although both the discrimination and supersensitivity defects are less severe than those exhibited by sst2 mutants. More recent results indicate that the GPA1^G302S mutation results in a greatly decreased Sst2p-Gpa1p interaction and undetectable GAP activity of Sst2p on Gpa1p (DIBELLO et al. 1998 ). Therefore, this mutation would be predicted to eliminate the role of the Sst2p RGS domain, but to maintain any role of the MPI domain. Because the MPI role is dependent on Mpt5p, a GPA1^G302S mpt5 mutant should be unable to carry out either the MPI or the RGS roles of Sst2p and therefore should show a phenotype similar to an sst2 null mutant.

To test this hypothesis, we constructed the GPA1^G302S mutant. The GPA1^G302S mutant exhibited an increase in supersensitivity to pheromone that was less severe than the sst2 null mutant; the zone of growth inhibition by pheromone is somewhat smaller in the GPA1^G302S mutant than in an sst2 mutant, as shown previously (Fig 5; JACKSON and HARTWELL 1990 ). Whereas the increase in sensitivity of either the GPA1^G302S or the mpt5 mutant is less than that in the sst2 mutant, the GPA1^G302S mpt5 double mutant showed an increase in sensitivity similar to that of the sst2 mutant.

View larger version (58K): In this window In a new window Download PPT slide   Figure 5. The mpt5 and GPA1G302S mutations are synergistic. The wild-type and mutant strains were tested for pheromone sensitivity using the pheromone spotting assay for cell cycle arrest. The mutants were mpt5::HIS3, sst2::HIS3, and GPA1G302S mpt5::HIS3.

We previously showed that overexpression of Sst2p results in promotion of recovery from pheromone arrest, resulting in a small turbid zone of growth inhibition in the pheromone spotting assay, and that this effect was only partially dependent on Mpt5p (Fig 6; CHEN and KURJAN 1997 ). On the basis of our hypothesis that the MPI and RGS domains of Sst2p play separate roles, overexpression of Sst2p in the mpt5 mutant would lose the effect of the MPI domain but maintain the effect of the RGS domain, explaining the partial dependence on Mpt5p. Overexpression of Sst2p in the GPA1^G302S mutant also results in a distinct zone of growth arrest, indicating that the effect of overexpression of Sst2p is also partially dependent on the Gpa1p interaction (Fig 6). In contrast, the effect of overexpression of Sst2p^P20L [1-401] (Fig 2) was not affected by the GPA1^G302S mutation (data not shown). Therefore, the strong effect of overexpression of intact Sst2p is dependent on both Mpt5p and the interaction with Gpa1p, whereas the effect of overexpression of Sst2p[1-401] is dependent only on Mpt5p.

View larger version (72K): In this window In a new window Download PPT slide   Figure 6. Promotion of recovery by PGK-SST2 is dependent on both Mpt5p and the interaction with Gpa1p. The ability of PGK-SST2 to promote recovery from pheromone arrest was tested using the pheromone spotting assay. The mutants were sst2::HIS3, mpt5::HIS3, and GPA1G302S mpt5::HIS3.

If Sst2p activity occurs solely through MPI domain action on Mpt5p and RGS domain action on Gpa1p, overexpression of Sst2p would be predicted to have no effect in a GPA1^G302S mpt5 double mutant. In the GPA1^G302S mpt5 double mutant, the effect of overexpression of Sst2p was greatly reduced in comparison to either the GPA1^G302S or the mpt5 single mutants (Fig 6); the zone of growth arrest was clear, consistent with a defect in promotion of recovery by Sst2p. However, the diameter of the zone of growth arrest was reduced by overexpression of Sst2p, indicating that Sst2p retained some ability to reduce pheromone sensitivity when overexpressed. A possible explanation for this result is that Sst2p action occurs through a component in addition to Gpa1p and Mpt5p. However, we favor the hypothesis that the slight decrease in pheromone sensitivity when Sst2p is overexpressed in the GPA1^G302S mpt5 double mutant results from a residual interaction between Sst2p and Gpa1p^G302S, on the basis of evidence that Sst2p retains some ability to interact with Gpa1p^G302S while losing detectable GAP activity (DIBELLO et al. 1998 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sst2p was initially implicated as playing an important role in desensitization to pheromone because sst2 mutations result in greatly increased sensitivity to pheromone and a defect in recovery from cell cycle arrest in response to pheromone (CHAN and OTTE 1982A , CHAN and OTTE 1982B ). Although the sequence of Sst2p did not initially provide any clues to Sst2p function (DIETZEL and KURJAN 1987A ), genetic epistasis results were consistent with Sst2p acting at the level of the G subunit of the pheromone response pathway, Gpa1p (KURJAN et al. 1991 ; DOHLMAN et al. 1995 ). The discovery that mammalian Sst2p homologs, called RGS proteins, act as GAPs for G subunits (DOHLMAN and THORNER 1997 ; IYENGAR 1997 ; KOELLE 1997 ) led to evidence indicating that Sst2p acts as a GAP for Gpa1p (Fig 7; APANOVITCH et al. 1998 ; KALLAL and FISHEL 2000 ). Although this result seemed to be sufficient to explain the role of Sst2p in pheromone sensitivity and recovery, other results suggested that the role of the Sst2p-interacting protein Mpt5p in pheromone sensitivity and recovery occurs at the level of G₁ cyclin/Cdc28p activity (CHEN and KURJAN 1997 ). In this article, we show that the Sst2p RGS domain does not not play a role in the Mpt5p interaction and that the N-terminal MPI domain plays a separate role in recovery that is dependent on Mpt5p.

View larger version (19K): In this window In a new window Download PPT slide   Figure 7. Model for Sst2p function. In the pheromone response pathway, Ste4p/Ste18p (ß) activates the downstream pathway (dotted arrow). The MAP kinase homolog, Fus3p, acts through the Ste12p transcription factor to induce pheromone-inducible genes and through Far1p to inhibit Cln1p/Cdc28p and Cln2p/Cdc28p kinase activity, resulting in G1 arrest. On the basis of protein-protein interactions and genetic epistasis results, we previously proposed that Mpt5p acts downstream of Fus3p and upstream of Cln3p/Cdc28p (CHEN and KURJAN 1997 ). Sst2p is a member of the RGS protein family of G-GAPs (APANOVITCH et al. 1998 ; KALLAL and FISHEL 2000 ); the C-terminal half of Sst2p contains the sequence similarity to RGS proteins. The identification of an RGS-domain-independent, Mpt5p-dependent role for the N-terminal MPI of Sst2p indicates that Sst2p is a bifunctional protein, with the MPI domain acting through Mpt5p and Cln3p. R*, activated Ste2p or Ste3p.

Sst2p contains two functional domains:
The N-terminal, MPI domain of Sst2p is sufficient for the Mpt5p interaction (Fig 1). On the basis of the two-hybrid interactions, the C-terminal RGS domain may inhibit this interaction. The observation that high-level expression of the Sst2p MPI domain, particularly with the dominant P20L mutation, led to decreased sensitivity to pheromone and increased recovery from pheromone-induced arrest (Fig 2) indicates that the MPI domain plays a role in pheromone sensitivity and recovery that is independent of the RGS domain. Promotion of recovery by overexpression of the MPI domain function was dependent on Mpt5p, consistent with the MPI domain acting through Mpt5p.

Previous results in which Sst2p or Sst2p^P20L expressed under the GAL promoter was unable to suppress the growth defect of a gpa1 mutant or activating alleles of the Ste4p Gß subunit indicated that Sst2p acts at the level of Gpa1p (DOHLMAN et al. 1995 ). However, higher level expression of Sst2p or Sst2p[1-401], with or without the P20L mutation, allowed weak suppression of the gpa1 growth defect (Fig 3 and Table 4). Suppression was independent of the Sst2p RGS domain but was dependent on both Mpt5p (Fig 3 and Table 4) and Cln3p (data not shown). These results indicate that the Sst2p MPI domain plays an RGS-domain-independent role that acts through Mpt5p and Cln3p (Fig 7). The ability of this domain to partially suppress the gpa1 growth defect indicates that Sst2p MPI domain function acts downstream of Gpa1p.

These results suggest that Sst2p contains two functional domains, the MPI domain, which acts through Mpt5p, and the RGS domain, which acts as a GAP for Gpa1p. To test this hypothesis, we combined an mpt5 mutation with the GPA1^G302S mutation. The GPA1^G302S mutation allows activation by pheromones, but results in a supersensitive phenotype that is less severe than an sst2 null mutation (JACKSON and HARTWELL 1990 ; DIBELLO et al. 1998 ). This mutation and the analogous mutation in mammalian G_q subunits greatly reduces the interaction of the RGS protein with G and eliminates the GAP activity of the RGS protein (DIBELLO et al. 1998 ). In the GPA1^G302S mpt5 double mutant, the action of both proposed Sst2p domains should be eliminated; the mpt5 mutation prevents the action of the MPI domain and the GPA1^G302S mutation eliminates the action of the RGS domain. As predicted, the GPA1^G302S mpt5 double mutant showed a similar phenotype as the sst2 null mutant (Fig 5).

Previous experiments showed that the increase in recovery resulting from overexpression of Sst2p is only partially dependent on Mpt5p (CHEN and KURJAN 1997 ). In addition, the effect of overexpression of Sst2p is only partially eliminated by the GPA1^G302S mutation (Fig 6), consistent with Sst2p containing two functional domains, one of which acts through Mpt5p and the other through Gpa1p (Fig 7). Both the ability of overexpression of Sst2p or the Sst2p MPI domain to partially suppress the gpa1 growth defect and the increase in recovery resulting from overexpression of Sst2p were dependent on Cln3p as well as Mpt5p (data not shown), consistent with Mpt5p acting through Cln3-Cdc28p (Fig 7).

A report of sequence similarity between the catalytic domain of Ras-GAP and a region of the N-terminal MPI domain of Sst2p (amino acids 3–307) has been presented (DOHLMAN et al. 1996 ), leading to the suggestion that the role of the C-terminal Sst2p RGS domain is to provide an interaction with Gpa1p, but that the N terminus acts as the GAP (DOHLMAN and THORNER 1997 ). We do not think that the role of the Sst2p MPI domain can be explained by this Ras-GAP similarity for several reasons. The sequence identity of the Sst2p N terminus to Ras-GAP is <20% (DOHLMAN et al. 1996 ), and many highly conserved residues of Ras-GAPs (SCHEFFZEK et al. 1996 ) are not present in the Sst2p MPI domain. Mammalian RGS proteins that consist almost completely of the RGS domain have been shown to act as G-GAPs (BERMAN et al. 1996B ; WATSON et al. 1996 ; HEPLER et al. 1997 ) and to provide partial Sst2p function in yeast (DRUEY et al. 1996 ; SIDEROVSKI et al. 1996 ), suggesting that the C-terminal RGS domain of Sst2p contains the GAP activity. In addition, the mechanisms of action of G-GAP and Ras-GAP have important differences (MITTAL et al. 1996 ; SCHEFFZEK et al. 1996 ; TESMER et al. 1997 ). In particular, Ras-GAP is proposed to contribute a catalytic Arg residue that plays a role similar to a conserved Arg residue of G subunits (R178 in G_i1). In contrast, RGS GAPs are proposed to not contribute residues involved directly in catalysis, but instead to act by stabilizing the transition state for the G GTPase activity (BERMAN et al. 1996A ; HUNT et al. 1996 ; WATSON et al. 1996 ). Finally, our experiments indicate that the Sst2p MPI domain plays a role that is independent of Gpa1p, which cannot be explained by N-terminal GAP activity for Gpa1p (Fig 4 and Table 3).

Recent results indicate that endogenous Sst2p is proteolytically cleaved into two fragments (HOFFMAN et al. 2000 ). The sites of cleavage are after residues Ser 414 and Ser 416; therefore, the C-terminal fragment contains the RGS domain and the N-terminal fragment is almost identical to the Sst2p[1-401] MPI domain. Expression of the separate N- and C-terminal domains partially complemented an sst2 mutant, and intact Sst2p and the N-terminal and C-terminal fragments showed different localization patterns (HOFFMAN et al. 2000 ). This cleavage event and the differential localization patterns may be critical for the separate roles we have identified for the Sst2p MPI and RGS domains, which act at very different points of the pathway (Fig 7). However, in our assays for inhibition of mating (Table 3), the N-terminal P20L mutation greatly increased inhibition of mating and this strong inhibition was observed only with intact Sst2p^P20L but not with Sst2p^P20L[1-401] (Table 3), indicating that the RGS domain is required for this inhibition. This effect of the P20L mutation suggests that the N terminus of Sst2p is important for some aspects of Sst2p RGS domain function. Therefore, intact Sst2p as well as the proteolytic fragments corresponding to the MPI and RGS domains may contribute to in vivo Sst2p function, and regions of the Sst2p N terminus may contribute to Sst2p RGS function through effects on localization or protein-protein interactions.

Fungal Sst2p homologs contain MPI domains:
Many RGS proteins are small, consisting primarily of the C-terminal RGS domain, whereas other RGS family members, including Sst2p, contain large domains to either the N or the C terminus of the RGS domain (DE VRIES and FARQUHAR 1999 ). Although the N-terminal MPI domain of Sst2p does not show significant sequence similarity to most RGS proteins, three Sst2p homologs identified in fungi contain N-terminal domains with significant sequence similarity to the Sst2p MPI domain (Fig 8). Aspergillus flbA has been implicated in developmental signaling processes, and a recessive flbA mutation results in phenotypes opposite to the recessive mutations in the G gene (fadA) involved in this process (YU et al. 1996 ). Interestingly, FlbA has been suggested to have a function independent of the G that of subunit, suggesting that the FlbA N terminus could play a role similar to that of the Sst2p MPI domain. The Schizophyllum commune thn1 mutant shows a change in hyphal morphology, but the pathway in which it acts has not been determined (FOWLER and MITTON 2000 ). The function of the Schizosaccharomyces pombe RGS homolog (Y8c) has not been reported. The observation that the N termini of fungal RGS proteins show sequence similarity to the Sst2p MPI domain suggests that they may also be bifunctional proteins that act through Mpt5p homologs.

View larger version (66K): In this window In a new window Download PPT slide   Figure 8. Putative fungal MPI domains. The sequence similarity among the N-terminal domains of Sst2p (amino acids 3–366; accession no. P11972; DIETZEL and KURJAN 1987A ), Aspergillus FlbA (amino acids 207–519; accession no. P38093; LEE and ADAMS 1994 ), S. commune Thn1 (amino acids 3–338, accession no. AF267870; FOWLER and MITTON 2000 ), and S. pombe 8c (amino acids 29–319) is shown, with identical amino acids indicated by solid boxes and conserved residues (K,R; D,E; N,Q; S,T; Y,F; L,V,I,M; D,N; Q,E) indicated by shaded boxes.

The discovery of the role of the RGS family of G-GAPs provided an explanation for the long-elusive role of Sst2p in desensitization to pheromone. However, the discovery of a separate role for the Sst2p MPI domain and the sequence similarity between this domain and the N-terminal domains of other fungal RGS proteins suggests an additional level of complexity to the function of Sst2p and its homologs.

	FOOTNOTES

¹ Present address: Bing-E Xu, Department of Pharmacology, University of Texas, Southwestern Medical Center, Dallas, TX 75390-9041.
² Present address: Karlheinz R. Skowronek, Department of Physiology and Biophysics, SUNY, Stony Brook, NY 11794-8661.

	ACKNOWLEDGMENTS

We thank S. Elledge and T. Chen for yeast strains, S. DeSimone for plasmids, C. Staats and L. Hill-Eubanks for technical assistance, and H. de Nobel, S. DeSimone, and T. Fowler for comments on the manuscript. This work was supported by National Institutes of Health grant GM-40585.

Manuscript received August 8, 2000; Accepted for publication October 1, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

APANOVITCH, D. M., P. B. SLEP, P. B. SIGLER, and H. G. DOHLMAN, 1998  Sst2 is a GTPase-activating protein for Gpa1: purification and characterization of a cognate-RGS-G protein pair in yeast. Biochemistry 37:4815-4822[Medline].

AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN et al., 1989 Current Protocols in Molecular Biology, pp.13.6.1–13.6.4. John Wiley & Sons, New York.

BERMAN, D. M. and A. G. GILMAN, 1998  Mammalian RGS Proteins: barbarians at the gate. J. Biol. Chem. 273:1269-1272[Free Full Text].

BERMAN, D. M., T. KOZASA, and A. G. GILMAN, 1996a  The GTPase-activating protein RGS4 stabilizes the transition state for nucleotide hydrolysis. J. Biol. Chem. 271:27209-27212[Abstract/Free Full Text].

BERMAN, D. M., T. M. WILKIE, and A. G. GILMAN, 1996b  GAIP and RGS4 are GTPase-activating proteins for the G_i subfamily of G protein subunits. Cell 86:445-452[Medline].

BLINDER, D., S. BOUVIER, and D. D. JENNESS, 1989  Constitutive mutants in the yeast pheromone response: ordered function of the gene products. Cell 56:479-486[Medline].

BREEDEN, L. and K. NASMYTH, 1985  Regulation of the yeast HO gene. Cold Spring Harbor Symp. Quant. Biol. 50:643-650[Medline].

CHAN, R. K. and C. A. OTTE, 1982a  Isolation and genetic analysis of Saccharomyces cerevisiae mutants supersensitive to G1 arrest by a factor and factor pheromones. Mol. Cell. Biol. 2:11-20[Abstract/Free Full Text].

CHAN, R. K. and C. A. OTTE, 1982b  Physiological characterization of Saccharomyces cerevisiae mutants supersensitive to G1 arrest by a factor and factor pheromones. Mol. Cell. Biol. 2:21-29[Abstract/Free Full Text].

CHEN, T. and J. KURJAN, 1997  Saccharomyces cerevisiae Mpt5p interacts with Sst2p and plays roles in pheromone sensitivity and recovery from pheromone arrest. Mol. Cell. Biol. 17:3429-3439[Abstract].

CHRISTIANSON, T. W., R. S. SIKORSKI, M. DANTE, J. H. SHERO, and P. HIETER, 1992  Multifunctional yeast high-copy-number shuttle vectors. Gene 110:119-122[Medline].

DESIMONE, S. M. and J. KURJAN, 1998  Switch domain mutations of the Saccharomyces cerevisiae Gpa1p G subunit identify a receptor subtype-biased mating defect. Mol. Gen. Genet. 257:662-671[Medline].

DE VRIES, L. and M. G. FARQUHAR, 1999  RGS proteins: more than just GAPs for heterotrimeric G proteins. Trends Cell Biol. 9:138-144[Medline].

DE VRIES, L., M. MOUSLI, A. WURMSER, and M. G. FARQUHAR, 1995  GAIP, a protein that specifically interacts with the trimeric G protein G_i3, is a member of a protein family with a highly conserved core domain. Proc. Natl. Acad. Sci. USA 92:11916-11920[Abstract/Free Full Text].

DIBELLO, P. R., T. R. GARRISON, D. M. APANOVITCH, G. HOFFMAN, and D. J. SHUEY et al., 1998  Selective uncoupling of RGS action by a single point mutation in the G protein -subunit. J. Biol. Chem. 273:5780-5784[Abstract/Free Full Text].

DIETZEL, C. and J. KURJAN, 1987a  Pheromonal regulation and sequence of the Saccharomyces cerevisiae SST2 gene: a model for desensitization to pheromone. Mol. Cell. Biol. 7:4169-4177[Abstract/Free Full Text].

DIETZEL, C. and J. KURJAN, 1987b  The yeast SCG1 gene: a G-like protein implicated in the a- and -factor response pathway. Cell 50:1001-1010[Medline].

DOHLMAN, H. G. and J. THORNER, 1997  RGS proteins and signaling by heterotrimeric G proteins. J. Biol. Chem. 272:3871-3874[Free Full Text].

DOHLMAN, H. G., D. APANIESK, Y. CHEN, J. SONG, and D. NUSSKERN, 1995  Inhibition of G-protein signaling by dominant gain-of-function mutations in Sst2p, a pheromone desensitization factor in Saccharomyces cerevisiae. Mol. Cell. Biol. 15:3635-3643[Abstract].

DOHLMAN, H. G., J. SONG, D. MA, W. E. COURCHESNE, and J. THORNER, 1996  Sst2, a negative regulator of pheromone signaling in the yeast Saccharomyces cerevisiae: expression, localization, and genetic interaction and physical association with Gpa1 (the G-protein subunit). Mol. Cell. Biol. 16:5194-5209[Abstract].

DRUEY, K. M., K. J. BLUMER, V. H. KANG, and J. H. KEHRL, 1996  Inhibition of G-protein-mediated MAP kinase activation by a new mammalian gene family. Nature 379:742-746[Medline].

FOWLER, T. J. and M. F. MITTON, 2000  Scooter, a new active transposon in Schizophyllum commune, has disrupted two genes regulating signal transduction. Genetics 154:1585-1594.

HEPLER, J. R., D. M. BERMAN, A. G. GILMAN, and T. KOZASA, 1997  RGS4 and GAIP are GTPase-activating proteins for G_q and block activation of phospholipase Cß by -thio-GTP-G_q. Proc. Natl. Acad. Sci. USA 94:428-432[Abstract/Free Full Text].

HOFFMAN, G. A., T. R. GARRISON, and H. G. DOHLMAN, 2000  Endoproteolytic processing of Sst2, a multidomain regulator of G protein signaling in yeast. J. Biol. Chem. 275:37533-37541[Abstract/Free Full Text].

HUNT, T. W., T. A. FIELDS, P. J. CASEY, and E. G. PERALTA, 1996  RGS10 is a selective activator of G_i GTPase activity. Nature 383:175-177[Medline].

IYENGAR, R., 1997  There are GAPS and there are GAPS. Science 275:42-43[Free Full Text].

JACKSON, C. L. and L. H. HARTWELL, 1990  Courtship in S. cerevisiae: both cell types choose mating partners by responding to the strongest pheromone signal. Cell 63:1039-1051[Medline].

KALLAL, L. and R. FISHEL, 2000  The GTP hydrolysis defect of the Saccharomyces cerevisiae G-protein Gpa1(G50V). Yeast 16:387-400[Medline].

KALLAL, L. and J. KURJAN, 1997  Analysis of the receptor binding domain of Gpa1p, the G subunit involved in the yeast pheromone response pathway. Mol. Cell. Biol. 17:2897-2907[Abstract].

KANG, Y.-S., J. KANE, J. KURJAN, J. M. STADEL, and D. J. TIPPER, 1990  Effects of expression of mammalian G and hybrid mammalian-yeast G proteins on the yeast pheromone response signal transduction pathway. Mol. Cell. Biol. 10:2582-2590[Abstract/Free Full Text].

KIKUCHI, Y., Y. OKA, M. KOBAYASHI, Y. UESONO, and A. TOH-E et al., 1994  A new yeast gene, HTR1, required for growth at high temperature, is needed for recovery from mating pheromone-induced G1 arrest. Mol. Gen. Genet. 245:107-116[Medline].

KOELLE, M. R., 1997  A new family of G-protein regulators—the RGS proteins. Curr. Opin. Cell Biol. 9:143-147[Medline].

KURJAN, J., 1993  The pheromone response pathway in Saccharomyces cerevisiae. Annu. Rev. Genet. 27:147-179[Medline].

KURJAN, J., J. P. HIRSCH, and C. DIETZEL, 1991  Mutations in the guanine nucleotide-binding domains of a yeast G protein confer a constitutive or uninducible state to the pheromone response pathway. Genes Dev. 5:475-483[Abstract/Free Full Text].

LEE, B. N. and T. H. ADAMS, 1994  Overexpression of flbA, an early regulator of Aspergillus asexual sporulation, leads to activation of brlA and premature initiation of development. Mol. Microbiol. 14:323-324[Medline].

MACKAY, V. L., S. K. WELCH, M. Y. INSLEY, T. R. MANNEY, and J. HOLLY et al., 1988  The Saccharomyces cerevisiae BAR1 gene encodes an exported protein with homology to pepsin. Proc. Natl. Acad. Sci. USA 85:55-59[Abstract/Free Full Text].

MANN, C., J. BUHLER, I. TREICH, and A. SENTENAC, 1987  RPC40, a unique gene for a subunit shared between yeast RNA polymerases A and C. Cell 48:627-637[Medline].

MANNEY, T., 1983  Expression of the BAR1 gene in Saccharomyces cerevisiae: induction by the mating pheromone of an activity associated with a secreted protein. J. Bacteriol. 155:291-301[Abstract/Free Full Text].

MARCUS, S., C.-B. XUE, F. NAIDER, and J. M. BECKER, 1991  Degradation of a-factor by a Saccharomyces cerevisiae -mating-type-specific endopeptidase: evidence for a role in recovery of cells from G1 arrest. Mol. Cell. Biol. 11:1030-1039[Abstract/Free Full Text].

MITTAL, R., M. R. AHMADIAN, R. S. GOODY, and A. WITTINGHOFER, 1996  Formation of a transition-state analog of the Ras GTPase reaction by Ras GDP, tetrafluoroaluminate, and GTPase-activating proteins. Science 273:115-117[Abstract].

MIYAJIMA, I., M. NAKAFUKU, N. NAKAYAMA, C. BRENNER, and A. MIYAJIMA et al., 1987  GPA1, a haploid-specific essential gene, encodes a yeast homolog of mammalian G protein which may be involved in mating factor signal transduction. Cell 50:1011-1019[Medline].

MIYAJIMA, I., K.-I. ARAI, and K. MATSUMOTO, 1989  GPA1^Val-50 mutation in the mating-factor signaling pathway in Saccharomyces cerevisiae. Mol. Cell. Biol. 9:2289-2297[Abstract/Free Full Text].

MOORE, S. A., 1984  Yeast cells recover from mating pheromone factor-induced division arrest by desensitization in the absence of factor destruction. J. Biol. Chem. 259:1004-1010[Abstract/Free Full Text].

NAKAYAMA, N., Y. KAZIRO, K.-I. ARAI, and K. MATSUMOTO, 1988  Role of STE genes in the mating-factor signaling pathway mediated by GPA1 in Saccharomyces cerevisiae. Mol. Cell. Biol. 8:3777-3783[Abstract/Free Full Text].

SCHEFFZEK, K., A. LAUTWEIN, W. KABSCH, M. R. AHMADIAN, and A. WITTINGHOFER, 1996  Crystal structure of the GTPase-activating domain of human p120GAP and implications for the interaction with Ras. Nature 384:591-595[Medline].

SIDEROVSKI, D. P., A. HESSEL, S. CHUNG, T. W. MAK, and M. TYERS, 1996  A new family of regulators of G-protein coupled receptors?