INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Sexual identity in the yeast Saccharomyces cerevisiae is controlled by the MAT locus. Transcriptional activators and repressors encoded by MATa or MAT alleles control the expression of a limited number of cell type-specific gene products that distinguish the two haploid mating types, the a cell and the  cell (24, 31). Chief among the cell type-specific products expressed is a receptor-ligand system which is used to direct the mating of the a and the  cells to form the a/ diploid. The a cell secretes the farnesylated peptide a-factor and expresses at its surface the -factor receptor (Ste2p), a G protein-coupled receptor which confers detection of the -factor peptide specifically produced and secreted by the  cell. Likewise, the  cell expresses a distinct G protein-coupled receptor, the a-factor receptor (Ste3p), which detects the a-factor secreted by the a cell.

This system of pheromones and receptors enables the communication of mating (for reviews, see references 18 and 19). Detection of pheromone alerts the cell to the proximity of a potential mating partner and prepares the cell for conjugation both through transcriptional induction of mating genes and through arrest of the cell cycle in G₁ at Start. Through polarized growth of the cell body, the two mating partners extend mating projections towards one another. The tips of the mating projections meet, cells adhere to one another, and, with the joining of the cell walls, the prezygote is formed. Finally, with dissolution of the intervening cell walls and subsequent fusion of cytoplasms and nuclei, the diploid zygote is formed: a cell with a characteristic dumbbell morphology, having two terminal bulbs derived from the cell bodies of the two haploid mating partners connected via a conjugation bridge derived from the tip-to-tip fusion of the two mating projections.

The mating process culminates with the transition of the new diploid cell to vegetative growth: the cells transition from G₁ to S, DNA replication is initiated, and a first mitotic bud begins to emerge from the midpoint of the zygotic conjugation bridge. Prior to zygote formation, during mating, pheromone signaling activates the cell cycle kinase inhibitor protein Far1p, which binds to and inhibits the Cdc28/Cln cell cycle kinase, and thus causes G₁ arrest. The reinitiation of the mitotic cycle for the new diploid cell requires that this inhibition be relieved.

The STE3^DAF mutation is a STE3 promoter mutation which confers cell type-independent expression of the a-factor receptor (Ste3p) (13). With this mutant it was found that the inappropriate expression of Ste3p in the a cell context leads to a striking inhibition of the pheromone signal transduction pathway, apparently exerted at the level of the G component of the heterotrimeric G protein, i.e., the first postreceptor signaling step (6, 13, 17). Inhibition depends on Ste3p and at least one MATa-specific gene product that is neither a-factor nor Ste2p (13). As Ste3p and the hypothetical a-specific interactor normally reside in different cell types, the biological relevance of this striking interaction has remained a mystery. Kim et al. (17) have suggested that Ste3p and its a cell interactor may gain access to one another in the zygote following fusion of the two mating partners and that the resulting inhibition of the pheromone response could serve to promote the transition of the zygote into a vegetative mode of growth. The present work identifies the a-specific Ste3p interactor as Asg7p and provides evidence which suggests that the inhibition of the pheromone response provided by Ste3p-Asg7p may indeed play a role in promoting the transition of the zygote to vegetative growth.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Plasmids. The LEU2 CEN ARS pRS315-derived plasmid pND541 expresses an N-terminally hemagglutinin (HA)-tagged Ura3p from the GAL1 promoter. Construction of pND541 first required fusion of an 820-bp GAL1,10 promoter fragment to the PstI site just upstream of the URA3 open reading frame (ORF). A restriction site for XhoI was then introduced by oligonucleotide-directed mutagenesis between codons 3 and 4 of the URA3 ORF (the introduced sequence translates to the dipeptide Leu-Glu). Finally, a DNA duplex encoding a single copy of the HA epitope, creating by annealing the two oligonucleotides 5'-pTCGAGTACCCATACGATGTTCCAGATTACGCTG-3' and 5'-pTCGACAGCGTAATCTGGAACATCGTATGGGTAC-3' was introduced into the XhoI site.

Strains. The strains used are listed in Table 1. The asg7::G418^R disruption allele, which replaces the entire ASG7 ORF with the G418^R marker from plasmid pFA-kanMX, was generated by PCR as described previously (33). The fus1::URA3 allele of plasmid construct pSL671 and the fus2::URA3 allele of p268 (complete descriptions of these plasmids are available from C. Boone upon request) were used to chromosomally transplace wild-type FUS1 and FUS2 alleles, respectively.

GAL1-STE3365

MAT GAL1-STE3365

Ste3p turnover. Ste3p turnover was monitored via a nonradioactive pulse-chase protocol as previously described (26). Briefly, a pulse of Ste3p synthesis was induced from GAL1-STE3 strains with the addition of galactose (2%) to cultures growing in yeast extract-peptone (YP)-raffinose (2%) medium. The chase period was initiated with the addition of glucose (3%), and, at various times thereafter, culture aliquots were removed for protein extract preparation (26). Extracts were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then Western analysis with affinity-purified rabbit polyclonal Ste3p-specific antibodies (28) or with the HA.11 monoclonal antibody (Berkeley Antibody Co., Berkeley, Calif.).

-Galactosidase assays. FUS1-LacZ bar1 MATa cells were cultured in YP-galactose (2%) medium and then treated with different concentrations of -factor for 1 h. Culture aliquots were collected by centrifugation, cells were permeabilized, and -galactosidase activity was determined as described previously (15).

Protease digestion of intact cells. The treatment of whole cells with pronase (Calbiochem-Novabiochem Corp., La Jolla, Calif.) and subsequent extract preparation were as described previously (7).

Matings. For quantitation of mating, aliquots containing approximately 10⁶ cells taken from log-phase cultures of MATa and MAT strains were mixed together and then applied, under gentle vacuum pressure, as an approximately 1-cm-diameter spot onto a 2.5-cm-diameter, 0.45-µm-pore-size nitrocellulose filter (Millipore, Inc.). The filter was then transferred to a 30°C yeast extract-peptone-dextrose (YPD) plate and incubated for 3 h. Cells were then eluted from filters with 1 ml of synthetic dextrose medium, subjected to a 3-s sonication at the lowest power setting of a Branson Sonifier 450, and then immediately diluted and plated onto medium selective for the diploid cells. For the microscopic inspection of zygotic and prezygotic morphology, matings were performed as described above except that 10⁷ cells of the two mating partners were applied to filters, filters were incubated on YPD plates at 30°C for various times, and then cells were eluted with 1 ml of 0.15 M NaCl-3.7% formaldehyde and stored at 4°C prior to analysis.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Impaired Ste3p turnover in MATa cells. The yeast a-factor receptor (Ste3p) is rapidly turned over via an endocytic mechanism that transports the receptor from the cell surface to the vacuole for degradation by the resident proteases (7). In its normal MAT cell context, the Ste3p turnover half-life is about 15 min whether the protein is expressed from its natural -cell-specific promoter (7) or from the GAL1 promoter (Fig. 1) (26). We find a different result for Ste3p ectopically expressed in the MATa context (Fig. 1). Quantification of the rate of this Ste3p loss indicates that turnover is indeed slowed in the a cell context relative to that in  cells (data not shown), indicating either that some -specific protein(s) acts to augment Ste3p turnover or, alternatively, that an a-specific gene product somehow interferes with the Ste3p turnover mechanism. In the a/ diploid cell context, we find rapid Ste3p turnover equivalent to that seen in  cells (data not shown), suggesting that the slowed turnover of a cells likely is the result of interference by some a-specific gene product.

View larger version (28K): [in this window] [in a new window]   FIG. 1.   Slowed Ste3p turnover in MATa cells responding to pheromone. Turnover of Ste3p was monitored in three isogenic yeast strains carrying a GAL1-STE3 construct in place of the wild-type STE3 locus: the wild-type MAT strain NDY1132, the wild-type MATa strain NDY1140, and the mfa1 mfa2 MATa strain SY2555. In addition, strains were transformed by pND541, a plasmid which expresses an HA-tagged Ura3p from the GAL1 promoter. A 3-h period of Ste3p and HA-Ura3p expression was induced from cultures growing in raffinose medium with the addition of galactose and subsequent termination by glucose addition. One of two SY2555 cultures was treated with two doses of -factor: a dose of 6 × 106 M administered 30 min prior to glucose addition and a dose of 15 × 105 M administered 30 min after the glucose addition. Protein extracts were prepared from culture aliquots, removed at the time of glucose addition (0-h time point) and at indicated times thereafter, and subjected to Western analysis both with Ste3p-specific antibodies (top) and with the HA.11 monoclonal antibody (bottom). The HA-Ura3p control provides the experiment with an internal standard of a protein not subject to rapid turnover.

GAL1-STE3 mfa1 mfa2

ASG7. Genome-wide analyses have identified ASG7 as an a-specific gene that is strongly induced by -factor (24, 36). ASG7 encodes a protein of 214 residues with two potential transmembrane domains (Saccharomyces Genome Database). Asg7p shows no significant homology to other yeast or mammalian proteins. Below, we test if ASG7 is the a-specific gene responsible for both the STE3^DAF phenotypes and slowed Ste3p turnover in MATa cells.

MAT GAL1-STE4 GAL1-STE3

View larger version (35K): [in this window] [in a new window]   FIG. 2.   ASG7 is the a-specific gene responsible for the Daf phenotype. Approximately 300 cells were spotted onto rich YP plates containing either 2% glucose or 2% galactose and allowed to grow for 2 days at 30°C. Cells from eight isogenic strains SY2150, SY2560, SY2561, NDY1050, and NDY1077 (see Table 1 for genotypes), as well as SY2560, NDY1050, and NDY1078, transformed by pND997 (pADH1-ASG7), a plasmid construct which constitutively expresses ASG7 from the ADH1 promoter, were tested. The relevant features of the strain genotypes are indicated to the left of growth spots.

MAT GAL1-STE4 GAL1-STE3

MAT GAL1-STE4 ste3 ADH1-ASG7

Coexpression of Asg7p and Ste3p blocks pheromone-induced transcription. We have examined the effects of Asg7p and Ste3p coexpression on the -factor-induced transcription of the pheromone-inducible gene FUS1 by assessing -galactosidase levels induced from a FUS1-LacZ reporter construct. To eliminate the potential for autocrine signaling through Ste3p, mfa1 mfa2 MATa cells were used. Responses were assessed over a 10,000-fold -factor concentration range. Consistent with the previous description of the Daf phenotype (13), ectopic expression of Ste3p in MATa strongly impairs the -factor response (Fig. 3; compare GAL1-STE3 ASG7⁺ to ste3 ASG7⁺). Wild-type responsiveness is restored to MATa GAL1-STE3 cells through ASG7 disruption (Fig. 3; see GAL1-STE3 asg7), indicating again that Ste3p and Asg7p together are required for the inhibition of pheromone signaling. We have also investigated the effects of constitutive expression of ASG7 from the ADH1-ASG7 construct. In terms of FUS1-LacZ induction, a much more pronounced inhibition was apparent: MATa GAL1-STE3 ADH1-ASG7 cells failed to show a detectable response to -factor (Fig. 3). Again, this inhibition wholly depended on Ste3p coexpression (Fig. 3; see ste3 ADH1-ASG7). The difference between the responsiveness of MATa GAL1-STE3 ADH1-ASG7 cells and that of MATa GAL1-STE3 ASG7^wt cells may be a reflection of the initial levels of Asg7p present in the two types of cells at the time of -factor addition. ASG7 is strongly inducible, and expression is virtually undetectable in MATa cells unstimulated by pheromone (16, 24, 36). Inhibition of signaling in the GAL1-STE3 ASG7^wt cells may await the new Asg7p that is synthesized following the pheromone challenge. Indeed, a previous kinetic analysis of the -factor response of MATa STE3^DAF cells showed that inhibition of signaling occurred only following an initial lag period: at early times following -factor addition, the response was found to be nearly wild type (17). With constitutive Asg7p expression from the ADH1 promoter, both Ste3p and Asg7p would be present in the cell at the time of the -factor challenge, possibly eliminating this lag in the institution of the Asg7p-Ste3p repression.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Ste3p and Asg7p coexpression impairs the response to -factor. -Galactosidase activity from a FUS1-LacZ reporter construct was used as a measure of the -factor-induced transcriptional response in six isogenic FUS1-LacZ bar1 STE2+ MATa strains: NDY1123, NDY1124, NDY1131, NDY1172, NDY1173, and NDY1174 (see Table 1 for strain genotypes). Cultures growing in YP-galactose (2%) medium were treated for 1 h with 109, 108, 107, 106, or 105 M -factor or were mock treated in parallel with no pheromone. Dose-response curves for induced -galactosidase activity versus -factor concentration are shown. As indicated to the right of each of the plots, the six strains differ both at the STE3 locus, which is ste3 or GAL1-STE3, and at the ASG7 locus, which is wild-type ASG7, asg7, or ADH1-ASG7.

Effects of Asg7p on Ste3p turnover and localization. Next, we examined if Asg7p is responsible for the slowed Ste3p turnover observed in a cells (Fig. 1). MATa mfa1 mfa2 GAL1-STE3 cells that were either ASG7⁺ or asg7 were treated with -factor to induce ASG7 expression, and Ste3p turnover was monitored (Fig. 4A). Disruption of ASG7 restores rapid turnover to these a cells, indicating that ASG7 is required for the slowed turnover previously observed in a cells (Fig. 1). Furthermore, with expression of ASG7 from the ADH1 promoter, slowed Ste3p turnover is apparent in both the a- and -cell contexts in the absence of added pheromone (Fig. 4B and C), indicating that for the slowed Ste3p turnover phenotype, pheromone is required simply for establishing elevated levels of Asg7p within the cell and not for activating Asg7p or Ste3p in some other way.

View larger version (15K): [in this window] [in a new window]   FIG. 4.   Asg7p is responsible for the slowed Ste3p turnover. Ste3p turnover was assessed by Western blotting as in Fig. 1; the loss of Ste3 antigen following the glucose repression of GAL1-STE3 constructs was monitored. (A) ASG7 encodes the pheromone-inducible factor responsible for slowed Ste3p turnover. The isogenic ASG7+ and asg7 MATa GAL1-STE3 mfa1 mfa2 strains SY2555 and NDY1086 were cultured and treated with -factor or were mock treated as described above for Fig. 1. (B) Constitutively expressed ASG7 suffices to retard Ste3p turnover in a cells. The isogenic asg7, and ADH1-ASG7 MATa GAL1-STE3 mfa1 mfa2 strains NDY1124 and NDY1131 were cultured as described above for Fig. 1, except -factor treatment was omitted. (C) Ectopic ASG7 expression in  cells retards Ste3p turnover. The GAL1-STE3 MAT SY2150 cells transformed by the ADH1-ASG7 plasmid pND997 or by the empty-vector plasmid (wt) were cultured as described for Fig. 1, except -factor treatment was omitted.

GAL1-STE3365

View larger version (35K): [in this window] [in a new window]   FIG. 5.   Asg7p delays Ste3p delivery to the cell surface. A 60-min period of receptor expression from GAL1-STE3365 was induced with galactose addition and terminated with the subsequent addition of glucose. Cell growth was continued for an additional 30 min, affording the newly synthesized receptor the opportunity to reach the cell surface. Culture aliquots at various time points were either treated with proteases (+) or were mock treated in parallel () (see Materials and Methods). Protein extracts prepared from these cells were subjected to Western analysis using Ste3p-specific antibodies. (A) Constitutive ASG7 expression delays the delivery of Ste3p to the cell surface in MATa cells. MATa GAL1-STE3365 mfa1 mfa2 cells, either asg7 (NDY1125) or ADH1-ASG7 (NDY1141), were cultured as described above. At 30 and 60 min following initiation of the galactose-induced pulse and 30 min after the subsequent addition of glucose (the 90-min time point), culture aliquots were removed for protease shaving. Indicated at right are the positions both of the undigested Ste3365p receptor and of a receptor digestion product corresponding to the protected cytoplasmic tail domain (CTD) plus the most C-terminal of the seven receptor transmembrane domains. (B) The Asg7p-mediated delay in Ste3p surface delivery occurs in a/ diploid cells which do not express the subunits of the heterotrimeric G protein. Localization of the Ste3365p expressed in MATa/ and MAT/ cells cultured as described above was assessed at the 90-min time point (following the 30-min glucose chase period) via the protease-shaving protocol. The MATa/ strains used, NDY1176 and NDY1178, and the MAT/ strains used, NDY1185 and NDY1186, were ASG7+/ASG7+ and ADH1-ASG7/ASG7+, respectively. (Wild-type ASG7 is expected to be transcriptionally silent in both a/ and / contexts [24]). For brevity, the portion of the gel displaying the lower-molecular-weight CTD fragment is not shown. (C) The Asg7p-mediated delay in Ste3p surface delivery occurs in cells with the G subunit-encoding gene STE4 deleted. MAT GAL1-STE3365 cells (NDY1200) as well as MAT GAL1-STE3365 ADH1-ASG7 (NDY1204) and MAT GAL1-STE3365 ADH1-ASG7 ste4::LEU2 cells (NDY1225) were cultured and treated with the protease-shaving protocol as described for panel B.

ADH1-ASG7 GAL1-STE3365

ASG7 and mating. The striking phenotypes noted above for ASG7 disruption or overproduction all involve the artificial coexpression of Asg7p and Ste3p in the same cell. In MATa cells not expressing STE3, both ASG7 deletion and overexpression were without discernible effect on the FUS1 transcriptional response (Fig. 3). Furthermore, we also find that asg7 and ADH1-ASG7 MATa cells show a morphogenetic response similar to that of wild-type MATa cells following 3 h of treatment with 10⁵ M -factor: normal mating projections were observed (data not shown).

MAT ADH1-ASG7

fus1 MAT

asg7 MAT

Deranged zygotic morphology in asg7 matings. We have performed microscopic analysis of zygotes formed from matings between wild-type MAT cells and asg7 MATa cells (Fig. 6). Here, we note several striking effects of the asg7 mutation. While zygotes are formed with wild-type efficiency in the mating of wild-type  cells to asg7 a cells (Table 3), these new diploid cells show a variety of morphologic abnormalities, often showing an unusual protuberant structure at the midregion of the conjugation bridge that connects the two cell bodies (Fig. 6). While approximately situated where the first diploid mitotic bud normally emerges, this protrusion does not show the neck-like constriction typical of emerging buds. These novel structures can be detected at very early mating time points (Fig. 6, 2.5-h time point) and often appear to be enlarged at later time points (Fig. 6, 5-h time point), suggesting that these may represent sites of aberrant polarized growth. Indeed, consistent with this being a locus of cell growth, mitotic buds are often found to emerge from the tips of these protuberances at late time points, (Fig. 6, 5-h time point). Zygotes with these protrusions generally manifest an overall bent morphology: instead of the linear dumbbell structure typical of wild-type zygotes, the two mating partners often have the aspect of having connected on an angle (Fig. 6). Finally, unlike the zygotes derived from wild-type matings, very few of the zygotes derived from early times of mating between wild-type MAT and the asg7 MATa cells are found with discernible emerging mitotic buds.

View larger version (70K): [in this window] [in a new window]   FIG. 6.   Zygotes from asg7 matings are morphologically deranged. Wild-type W303-1B MAT cells were mated to either wild-type W303-1A MATa cells or the isogenic asg7 strain NDY1089 for 2.5 or 5 h. Selected zygotes visualized by Nomarski optics are shown. White arrows, new mitotic buds; black arrows, stalk-like structures from the mitotic bud often seen to emerge for asg7 zygotes.

asg7 MAT

Epistasis of asg7 with fusion-defective mutations. The linear dumbbell presentation of the wild-type zygote results from tip-to-tip fusion of the two mating partners. The bent or angled morphology seen for many of the asg7-derived zygotes (Fig. 6) suggested the possibility that the asg7 a cells might be defective for this prezygotic connection to their  partners. Indeed, an angled connection between mating partners might also account for the midpoint protuberance seen for the asg7 zygotes, with the protuberance being derived from the angled fusion of the two mating projections. We were interested therefore to examine the morphology of prezygotes derived from asg7 pairings to see if the abnormalities seen for the zygotes are also present at this early step.

MAT fus1

fus1 asg7

fus2 asg7

MAT fus

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Asg7p-Ste3p coexpression phenotypes. We find that ASG7 is the a-specific gene responsible for the phenotype of the STE3^DAF allele. Though defined in a cells (13), the Daf phenotype, we also find, may be reproduced in  cells with the forced, inappropriate expression of Asg7p in this context. Thus, Asg7p is the only a-specific gene product and Ste3p is the only -specific gene product required for the inhibition. The key requirement for inhibition is coexpression of Asg7p and Ste3p in the same cell. We also report effects of ASG7 on Ste3p turnover and localization. Asg7p slows Ste3p turnover (Fig. 4). However, instead of being a direct result of Ste3p endocytosis, slowed turnover appears to be a secondary consequence of the Asg7p-mediated inhibition of the secretory delivery of Ste3p to the cell surface (Fig. 5).

Asg7p-Ste3p inhibition in the zygote. MATa asg7 cells respond to pheromone normally and mate with wild-type efficiency. However, while normal numbers of zygotes are formed, the zygotes are morphologically deranged and show a delay to the emergence of the first diploid mitotic bud. Since prezygotes examined from these same matings are morphologically normal, we concluded that perturbed zygotic morphology results from postfusion events and not from an initially misoriented connection of the two mating partners.

asg7 MAT

mat2

mat2

mat2

Mechanism of Asg7p-Ste3p inhibition. Little is understood regarding the mechanism of Asg7p-Ste3p inhibition. From the present analysis, it is clear that Ste3p is the only -specific protein required and Asg7p is the only a-specific protein required. It is not clear if these two proteins physically interact. Furthermore, it is not clear to what extent other proteins, present in both cell types, participate.