Point Mutations Identify a Conserved Region of the Saccharomyces cerevisiae AFR1 Gene That Is Essential for Both the Pheromone Signaling and Morphogenesis Functions

Corresponding author: James B. Konopka, Department of Molecular Genetics and Microbiology, State University of New York, Stony Brook, NY 11794-5222., konopka{at}asterix.bio.sunysb.edu (E-mail)

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mating pheromone receptors activate a G protein signal pathway that leads to the conjugation of the yeast Saccharomyces cerevisiae. This pathway also induces the production of Afr1p, a protein that negatively regulates pheromone receptor signaling and is required to form pointed projections of new growth that become the site of cell fusion during mating. Afr1p lacks strong similarity to any well-characterized proteins to help predict how it acts. Therefore, we investigated the relationship between the different functions of Afr1p by isolating and characterizing seven mutants that were defective in regulating pheromone signaling. The AFR1 mutants were also defective when expressed as fusions to STE2, the -factor receptor, indicating that the mutant Afr1 proteins are defective in function and not in co-localizing with receptors. The mutant genes contained four distinct point mutations that all occurred between codons 254 and 263, identifying a region that is critical for AFR1 function. Consistent with this, we found that the corresponding region is very highly conserved in the Afr1p homologs from the yeasts S. uvarum and S. douglasii. In contrast, there were no detectable effects on pheromone signaling caused by deletion or overexpression of YER158c, an open reading frame with overall sequence similarity to Afr1p that lacks this essential region. Interestingly, all of the AFR1 mutants showed a defect in their ability to form mating projections that was proportional to their defect in regulating pheromone signaling. This suggests that both functions may be due to the same action of Afr1p. Thus, these studies identify a specific region of Afr1p that is critical for its function in both signaling and morphogenesis.

THE conjugation of the yeast Saccharomyces cerevisiae is initiated when haploid cells of opposite mating type, MATa and MAT, secrete peptide pheromones that bind to receptors on the surface of the other mating type. Ste2p, the -factor receptor, binds -factor pheromone on the surface of MATa cells and Ste3p, the a-factor receptor, binds a-factor pheromone on the surface of MAT cells (KURJAN 1993 ; BARDWELL et al. 1994 ; FERGUSON et al. 1994 ; LEBERER et al. 1997 ). The pheromone receptors belong to a large family of G protein-coupled receptors (GPCRs) that are distinguished by containing seven membrane spanning domains (DOHLMAN et al. 1991 ; WATSON and ARKINSTALL 1994 ). This family of receptors initiates signaling by stimulating the G subunit of a heterotrimeric G protein to exchange a bound GDP for a GTP, which consequently promotes the release of the G_ß subunits (HEPLER and GILMAN 1992 ; BOURNE 1997 ). In the yeast pheromone pathway, the free G_ß subunits are thought to recruit to the plasma membrane the Ste5p scaffold protein and associated protein kinases consisting of Ste11p, Ste7p, and Fus3p that are homologs of MEKK, MEK, and MAP kinases, respectively (HERSKOWITZ 1995 ; PRYCIAK and HUNTRESS 1998 ; MAHANTY et al. 1999 ). Signaling through this MAP kinase module then appears to be stimulated by Ste20p, a PAK kinase homolog (LEBERER et al. 1992 ). Finally, transcription of genes essential for mating is induced by the pheromone-responsive transcription factor Ste12p (SPRAGUE and THORNER 1992 ; COOK et al. 1996 ; TEDFORD et al. 1997 ).

The yeast pheromone signal pathway induces multiple cellular changes required for mating. Regulation of the cyclin-dependent kinase, Cdc28p, by Far1p causes cell cycle arrest in G1 phase (CHANG and HERSKOWITZ 1990 ). Continued signaling promotes polarized cell growth in the direction of a gradient of pheromone emanating from a cell of opposite mating type (JACKSON et al. 1991 ; SEGALL 1993 ). This results in the formation of an acute projection of new growth that becomes the site at which the mating cells fuse together to form a zygote (ROEMER et al. 1996 ; MARSH and ROSE 1997 ). Cells exposed to high concentrations of pheromone in absence of a gradient also produce mating projections, but usually do so at a default site that is adjacent to the site at which the previous bud was formed (MADDEN and SNYDER 1992 ). The pheromone signal is thought to promote the polarized growth that forms mating projections by activating Cdc42p, a member of the rho family of GTPases (SIMON et al. 1995 ; ZHAO et al. 1995 ; BUTTY et al. 1998 ; NERN and ARKOWITZ 1998 ). Cdc42p apparently acts to organize cytoskeletal components such as actin that mediate polarized growth (READ et al. 1992 ; CHENEVERT 1994 ; DRUBIN and NELSON 1996 ). Cells that are completely defective in polarizing actin fail to mate, in part because they do not form a mating projection that acts as a conjugation tube to connect to the partner cell. Recent studies further indicate that actin-mediated cell polarity is also required for the pheromone signal to be transduced (SIMON et al. 1995 ; AYSCOUGH and DRUBIN 1998 ; OEHLEN and CROSS 1998 ; PRYCIAK and HUNTRESS 1998 ).

In addition to the actin cytoskeleton, other components such as Spa2p (GEHRUNG and SNYDER 1990 ), Pea2p (VALTZ and HERSKOWITZ 1996 ), and the septins (GIOT and KONOPKA 1997 ) are also required for cells to form normal projections. Mutation of any of these components does not block pheromone signaling or cell polarization, but does result in altered mating projections that are broader and less pointed. These mutants mate well to a wild-type mating partner, indicating that pointed mating projections are not essential for mating. However, the ability to form pointed projections may enhance the efficiency of mating under certain conditions by concentrating the signal transduction and cell fusion components into a more specific zone (GEHRUNG and SNYDER 1990 ; CHENEVERT et al. 1994 ; GIOT et al. 1999 ).

Interestingly, the proper regulation of pheromone receptor signaling is also important for pointed projection formation, suggesting that signaling and morphogenesis are coordinately regulated (KONOPKA et al. 1988 ; CHEN and KONOPKA 1996 ). In addition, the pheromone-induced protein Afr1p was identified as functioning in both regulation of signaling and in mating projection formation (KONOPKA 1993 ). Overexpression of Afr1p promoted more rapid adaptation to pheromone and deletion of AFR1 resulted in a defect in projection formation. Afr1p appears to regulate signaling by blocking the ability of pheromone receptors to signal (DAVIS et al. 1998 ). Afr1p also interacts with Cdc12p, a member of the septin family of filament-forming proteins (LONGTINE et al. 1996 ), and colocalizes with the septins to the neck of mating projections following pheromone induction (KONOPKA et al. 1995 ). This specific subcellular localization appears to be crucial for function; Afr1p mutants that mislocalized were defective in producing distinct mating projections (GIOT and KONOPKA 1997 ). Although interaction with Cdc12p promotes proper subcellular localization of Afr1p, it was not absolutely required for Afr1p function. The sequences of Afr1p that are essential for both pheromone receptor regulation and projection formation have been mapped to a region between amino acids 194 and 350 that is distinct from the Cdc12p interaction domain (GIOT and KONOPKA 1997 ). In this study, we dissect this essential region by identifying point mutations that disrupt AFR1 function. The results indicate that the abilities of Afr1p to regulate pheromone signaling and to induce projection formation are tightly linked and may, in fact, be due to a single mechanism of action.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and media:
Yeast strains are described in Table 1. Strain yCRD100-I-2 was the result of a cross between JKY26 and yLG108-7-3. Cells were grown in media as described (SHERMAN 1991 ). Plasmids were transformed into yeast strains using the lithium acetate method (SCHIESTL and GIETZ 1989 ). Plasmid-containing cells were grown in synthetic medium containing adenine, uracil, and amino acid additives but lacking either tryptophan, leucine, or uracil to select for plasmid maintenance.

Construction of plasmids:
Vector p112* was made by cutting YEplac112 (GIETZ and SUGINO 1988 ) with PstI and destroying the site using T4 DNA polymerase. p112*AFR1 was made by subcloning the BamHI-SalI fragment of pJK52 (KONOPKA 1993 ) containing AFR1 into the BamHI-SalI sites in the p112* polylinker. AFR1 mutants were generated by PCR amplification using Taq DNA polymerase with the concentration of one dNTP (dATP, dCTP, dGTP, or dTTP) adjusted to 25 µM (the remaining three dNTPs were used at 125 µM). Mutagenic PCR was performed between oligonucleotide primers OLG1 and OLG2 (GIOT and KONOPKA 1997 ). The fragments generated from the PCR reactions were cotransformed into yCRD-100-I-2 with p112*AFR1 cut at PstI and ClaI (base 394–1045 of the AFR1 coding region) to create a gap. The endogenous yeast DNA repair mechanisms were relied upon to perform homologous recombination between the gapped p112*AFR1 plasmid and the PCR fragments to generate mutant plasmids. To ensure that the mutations were contained within the AFR1 coding region, the PstI-ClaI fragment in the wild-type plasmid p112*AFR1 was replaced with the PstI-ClaI fragments from the mutant plasmids. Plasmids were recovered into Escherichia coli for DNA sequence analysis using a DNA sequencing kit from United States Biochemical (Cleveland) or a BigDye Terminator cycle sequencing reaction kit from Applied Biosystems (Foster City, CA) as described below. The mutant AFR1 genes were then inserted into centromeric plasmids by subcloning the SalI-BamHI fragment containing the entire mutant AFR1 gene from the second generation plasmids into the SalI-BamHI sites in the polylinker of YCplac111 (GIETZ and SUGINO 1988 ).

The AFR1-GFP fusion plasmid pCRD357 was constructed by using PCR to introduce an XhoI site in the 3' end of the AFR1 open reading frame and then ligating a SalI-XhoI AFR1 fragment to a XhoI-BamHI fragment containing green fluorescent protein (GFP; CORMACK et al. 1997 ) into YCplac111 (GIETZ and SUGINO 1988 ). To construct GFP fusions for the AFR1 mutants, a BspEI-BamHI fragment from pCRD357 containing GFP was used to replace the BspEI-BamHI fragment of the AFR1 mutants in YCplac111. To analyze the AFR1 mutants in the two-hybrid assay, PCR was used to generate a SalI-BamHI fragment that was then ligated in frame with the lexA sequences in pLG3 (GIOT and KONOPKA 1997 ), a version of pBTM116 with a modified polylinker region. Plasmids were introduced into L40 yeast carrying pGAD424 or pGAD-CDC12 for two-hybrid assays (KONOPKA et al. 1995 ). STE2-AFR1 fusion plasmids were constructed by cloning a SphI-SalI fragment of STE2 from pDB02 (DUBE and KONOPKA 1998 ) into YCplac33 to create pER1 and then ligating in PCR-generated SalI-BamHI fragments containing codons 1–350 of AFR1 into pER1.

The entire YER158c open reading frame (ORF) plus 1862 nucleotides 5' and 818 nucleotides 3' were PCR amplified from yeast genomic DNA using primers p1 (5'-TGTCGACTTCCCTAAGACTG-3') and p2 (5'-GCGGATCCCCTTTTATGG-3'). These primers also added either SalI or BamHI restriction sites on the ends to facilitate cloning of the PCR product into pBluescript II KS(+) (Stratagene, La Jolla, CA) between the SalI and BamHI sites. To create a deletion allele, the YER158c gene in pBluescript was disrupted by subcloning LYS2 between the HindIII and AvrII sites, leaving only the first 93 nucleotides of the coding region. This yer158c::LYS2 construct was then used to replace the endogenous YER158c in yeast by one-step recombination (ROTHSTEIN 1983 ). A SphI-HindIII fragment containing 696 nucleotides 5' and 16 nucleotides 3' of the open reading frame was subcloned to YEplac181 (GIETZ and SUGINO 1988 ) for overexpression assays.

Cloning and sequencing of AFR1-homologous sequences from S. uvarum and S. douglasii:
A PCR approach was used as the initial approach to identify AFR1-homologous sequences in other yeasts. A variety of PCR primer combinations were tested for ability to generate PCR products using genomic DNA from S. uvarum and S. douglasii as a template. The primers included those that were used in previous studies (GIOT and KONOPKA 1997 ) as well as primer CD11 that contains sequence redundancy so that it is complementary to a conserved sequence motif present between AFR1 and YER158c (corresponding to the coding region for residues 139–144 of Afr1p). Most primer combinations failed to give a product but the combination of primers CD11 (5'-CA[AG]TT[CT]CC[AGCT]AA[CT]GG[AGCT]GA-3') and OLG5 (5'-CGGGATCCTACTCGAGTCTGATATACTCCCAG-3') gave a 1083-base product using S. douglasii DNA as a template. This PCR product was then subcloned into pBluescript to facilitate DNA sequence analysis. We failed to detect AFR1-homologous sequences from S. uvarum DNA using this approach so Southern hybridization was used to detect AFR1-related sequences in this yeast. A digoxigenin-labeled DNA probe was generated by using PCR to amplify the entire coding region of AFR1 using primers OLG1 and OLG4 (GIOT and KONOPKA 1997 ). The digoxigenin-labeling and subsequent hybridizations were carried out using a Genius System kit from Boehringer Mannheim (Indianapolis). The AFR1 probe detected a 1.8-kb band in EcoRI-digested genomic DNA from S. uvarum. This AFR1-related band was then cloned into pBluescript for DNA sequence analysis. Basically, a preparative EcoRI digest of S. uvarum DNA was separated by electrophoresis on an agarose gel and then the DNA fragments corresponding to the 1.6 to 2.0-kb size range were purified and ligated into pBluescript. E. coli colonies containing AFR1-homologous sequences were detected by colony hybridization. DNA sequencing was performed by the dideoxy termination method using [³⁵S-]ATP (New England Nuclear, Boston) and a Sequenase kit from United States Biochemical. The DNA sequences were also confirmed using an automated sequencer and a BigDye Terminator cycle sequencing reaction kit from Applied Biosystems. The latter reactions were analyzed by the sequencing core facility at the State University of New York at Stony Brook.

-Factor response assays:
Sensitivity to -factor was measured in halo assays that were performed by adding 15 µl of -factor to a sterile filter disk (Difco, Detroit) and then placing it on the surface of an agar plate that had been spread with a lawn of 1.5 x 10⁵ logarithmic phase cells. Plates were incubated for 2 days at 30°. Alternatively, spot assays were performed by growing cells to stationary phase in synthetic medium, diluting cells 1:10, 1:100, 1:1000, and 1:10,000, and then spotting 10 µl of the undiluted culture and each of the serial dilutions onto either synthetic medium agar plates or the same medium containing 10^-7 M -factor. Plates were grown at 30° for 2 days. Morphological analysis was performed with strain JKY26 (afr1) carrying the indicated plasmid. Cells were grown overnight to midlogarithmic phase in synthetic medium, adjusted to a density of 2.5 x 10⁶ cells/ml in YEPD, grown for 1 hr, and then induced with 10^-7 M -factor. Aliquots were taken at 3 and 6 hr, fixed by addition of formaldehyde, and then 200 cells from each sample were examined microscopically to determine the percentage of cells with mating projections. Cells carrying AFR1-GFP plasmids were grown overnight to log phase, diluted to 3 x 10⁶ cells/ml in medium containing 10^-7 M -factor, incubated for 2 hr at 30°, and then examined microscopically. Cells were photographed with Kodak TMAX film using an Olympus BH2 microscope. Immunoblot analysis of Afr1p was carried out essentially as described previously (KONOPKA et al. 1995 ). Cells were grown to logarithmic phase and then were induced with 10^-7 M -factor for 2 hr. Approximately 10⁸ cells were extracted with lysis buffer (2% SDS, 50 mM Tris, pH 6.8), and then equal amounts of cell extract were resolved by electrophoresis on a SDS-polyacrylamide gel. The proteins were transferred to nitrocellulose. The blot was probed with rabbit anti-Afr1p antibody (KONOPKA et al. 1995 ), and then the immunoreactive proteins were detected by chemiluminescence using an Amersham (Arlington Heights, IL) ECL kit. To verify that equal amounts of protein were loaded, a parallel blot was probed with anti-glucose-6-phosphate dehydrogenase antibody (Sigma, St. Louis).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Screen for AFR1 mutants:
The sequences required for Afr1p to regulate pheromone signaling and to induce proper mating projections have been mapped to the N-terminal half of the 620-amino acid Afr1p. In particular, the region between amino acids 194 and 350 is essential for both functions (GIOT and KONOPKA 1997 ). This region of Afr1p does not show strong similarity to any other previously characterized proteins. Therefore, to better define the sequences required for Afr1p function and to investigate the relationship between the regulation of signaling and morphogenesis during mating, we carried out a screen to identify substitution mutations that disrupt Afr1p function. Mutagenesis was targeted to the essential region of AFR1 by performing PCR under suboptimal conditions (see MATERIALS AND METHODS). A genetic screen for AFR1 mutants was designed that took advantage of the fact that overexpression of wild-type AFR1 promotes hyperadaptation to -factor. Cells that carry AFR1 on a multicopy plasmid can be distinguished from wild-type cells by their ability to grow in the presence of high concentrations of -factor that lead to prolonged cell division arrest of wild-type cells. This made it possible to identify loss-of-function mutations in AFR1 by the inability of cells carrying a YEp-AFR1 plasmid to form colonies when replica-plated to medium containing a high concentration of -factor (10^-7 M). From ~3000 transformants of cells carrying mutagenized AFR1 genes on a multicopy plasmid vector, 152 mutants were identified that were reproducibly sensitive to pheromone.

The AFR1 plasmids were then recovered from the mutant cells for further analysis. To ensure that the increased sensitivity to -factor was due to a mutation in AFR1, a PstI-ClaI fragment (bases 394–1045) containing the essential region of AFR1 was subcloned back into a wild-type AFR1 plasmid. These resulting secondary clones were transformed into yeast and tested for pheromone sensitivity in a more quantitative fashion by spotting dilutions of cell cultures onto plates containing -factor (Fig 1 and data not shown). As expected, the cells containing the control vector displayed little or no growth in the presence of 10^-7 M -factor. In contrast, cells overexpressing wild-type AFR1 grew well, indicating that they were highly resistant to -factor. Of the 152 original mutants, 44 showed greatly increased sensitivity to -factor in the spotting assay.

View larger version (116K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Identification of AFR1 mutants defective in promoting resistance to -factor. The afr1 strain yCRD100-I-2 carrying the indicated AFR1 allele on the multicopy plasmid p112* was tested for resistance to -factor-induced cell division arrest. Tenfold serial dilutions were spotted on the surface of an agar medium plate with either no pheromone (minus -factor) or with 10-7 M -factor (plus -factor). The plates were incubated for 2 days at 30° and then photographed.

The mutants were then tested for their ability to produce Afr1p by Western blot analysis. The Western blots showed that the majority of the mutants either failed to produce detectable levels of Afr1p or they produced severely truncated Afr1p (not shown). These mutants were not studied further because of their defect in Afr1p production. However, seven mutants (A31, A46, A47, T164, T181, T182, and T211) were able to make full-length protein (Fig 2 and data not shown). These seven mutants were saved for further study because they were defective in Afr1p function.

View larger version (50K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Western blot analysis of Afr1p mutants. The afr1 strain yCRD100-I-2 carrying the multicopy plasmid vector p112*, or p112* containing the indicated AFR1 allele was assayed for ability to make Afr1p. Cells were treated with 10-7 M -factor for 2 hr at 30° to induce production of Afr1p. Equivalent amounts of cell extract were resolved by SDS-PAGE, transferred to nitrocellulose, and then the blot was probed with rabbit anti-Afr1p antibody. As a control to show that approximately equal amounts of protein were loaded, a separate gel was analyzed in parallel with control antibody against glucose 6-phosphate dehydrogenase (not shown). Positions of prestained size standards are indicated on the left.

The seven final mutants fell into two classes based on their sensitivity to -factor. Class I mutants (A31, A46, T164, and T211) appeared to be completely defective. The class II mutants (A47, T181, and T182) differed in that they retained weak ability to confer resistance to -factor. The ability of the class II mutants to promote resistance to -factor was more apparent when they were challenged to grow in the presence of lower concentrations of -factor (e.g., 10^-8 M) than at higher concentrations of -factor (e.g., 10^-7 M; Table 2 and data not shown).

View this table: [in this window] [in a new window]   Table 2. Ability of AFR1 mutants to form mating projections and promote resistance to -factor

DNA sequence analysis of AFR1 mutants:
DNA sequence analysis revealed that each of the mutant genes contained one or two nucleotide changes that resulted either in a single or double amino acid substitution (Fig 3). The class I mutants contained the following substitutions: Mutant A31 contained two mutations that resulted in a substitution of asparagine-254 with serine and tyrosine-261 with histidine (N254S, Y261H), mutant T164 contained a single change of valine-257 to alanine (V257A), and both A46 and T211 were caused by the same change of phenylalanine-263 to serine (F263S). The class II mutants (A47, T181, and T182) all contained the same substitution mutation that replaced isoleucine-259 with lysine (I259K). Mutant A47 also had an additional mutation of leucine-238 to proline (L238P). The additional change at amino acid 238 apparently did not contribute to the defective phenotype of A47 as it did not appear to be any more defective than the other class II mutants.

View larger version (10K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Protein sequence alignment of Afr1p mutants. The mutant AFR1 genes were subjected to DNA sequence analysis to identify the mutations and then the changes in the protein sequences were predicted. The amino acid sequence of the wild-type Afr1p in the region affected by the mutations is shown at the top (residues 230–280 as indicated above the Afr1p sequence). The changes in the amino acid sequence that are predicted to occur in the mutants are identified below the corresponding wild-type residue. Positions where there was no change are indicated by a period.

Analysis of the sites of the mutations gave the surprising result that the mutations for both the class I and class II mutants were all clustered in a small domain in the region coding for amino acids 254–263. Since the entire region coding for the N-terminal half of Afr1p was targeted for mutagenesis, this clustering is interesting because it indicates that this small region is critically important for Afr1p function. Analysis of the types of amino acid substitutions showed that, in most cases, the substitutions result in a rather significant change in amino acid character that could account for their effect on Afr1p function. For example, the substitution of a charged lysine residue in place of the hydrophobic isoleucine-259 (I259K), or the substitution of a relatively small serine residue in place of the bulkier phenylalanine-263 (F263S), could easily have profound effects on the structure and function of Afr1p. However, the conservative substitution of alanine for valine-257 indicates that this residue is critically important for Afr1p function.

Afr1p mutants are defective in promoting resistance to -factor when fused to Ste2p:
The defects of the Afr1p mutants were examined next by constructing chimeric STE2-AFR1 genes that result in the fusion of Afr1p onto the C terminus of the -factor receptor (Ste2p). For this analysis, only the first 350 residues of Afr1p were included in the Ste2-Afr1p fusions. The C-terminal sequences of Afr1p that mediate interaction with the Cdc12p septin protein were not included as they would be expected to have deleterious effects on septin function during vegetative growth (KONOPKA et al. 1995 ). Interestingly, ste2 cells carrying a wild-type STE2-AFR1 fusion gene on a YCp plasmid were highly resistant to pheromone, as indicated by their ability to grow on medium containing -factor (Fig 4A). The negative effects of Afr1p on signaling in this context were apparently limited to the receptors to which Afr1p was fused as the YCp-STE2-AFR1 plasmid had no detectable effects on signaling in a STE2 strain that also produced wild-type receptors (Fig 4B). This indicates that Ste2-Afr1p does not cause a nonspecific effect on signaling. These results provide further evidence that Afr1p acts to regulate -factor receptor signaling and does not act on a downstream component of the pathway.

View larger version (93K): [in this window] [in a new window] [Download PPT slide]   Figure 4. -Factor sensitivity of strains expressing STE2-AFR1 fusions. ste2 strain YLG123 (A) or STE2 strain JKY25 (B) carrying the indicated STE2-AFR1 allele on the vector YCplac33 were tested for resistance to -factor-induced cell division arrest. Tenfold serial dilutions were spotted on the surface of an agar medium plate with either no pheromone (minus -factor) or with 10-7 M -factor (plus -factor). The plates were incubated for 2 days at 30° and then photographed.

In contrast, ste2 cells carrying a YCp-STE2-AFR1 plasmid containing one of the AFR1 point mutations remained sensitive to pheromone (Fig 4A). Thus, fusion to Ste2p did not suppress the defects of the mutant Afr1 proteins. In a different assay for pheromone-induced cell division arrest (halo assay), it was possible to detect weak effects of the Afr1p mutants on signaling, especially for the class II mutant T181 (not shown). However, these weak effects did not appear to be enhanced relative to the effects observed for overexpression of nonfused AFR1 mutants (Fig 1 and data not shown). Altogether, these results indicate that the Afr1p mutants are defective even when brought in close proximity to the -factor receptors.

Correlation between pheromone resistance and morphogenesis:
The essential region of Afr1p between amino acids 194 and 350 also functions to promote the formation of mating projections. The seven resistance-defective AFR1 mutant genes were therefore subcloned into a low-copy centromeric plasmid vector and then introduced into an afr1 strain to test for their ability to complement the projection formation defect. The formation of mating projections was assayed in cells carrying the various AFR1 mutant plasmids that were induced with -factor for 3–6 hr and then photographed to record the morphology of their mating projections (Fig 5). The ability of the cells to form mating projections was also quantitated by microscopic examination (Table 2). As expected, the majority of the afr1 cells carrying a wild-type AFR1 plasmid formed a pointed projection at 3 hr, and by 6 hr many of the cells had formed multiple projections. In contrast, afr1 cells carrying the vector control plasmid rarely formed a single pointed projection and generally appeared as elongated, heterogeneously shaped cells.

View larger version (95K): [in this window] [in a new window] [Download PPT slide]   Figure 5. Pheromone-induced morphology of AFR1 mutants. To examine the effects of the AFR1 mutants on mating projection formation, the low-copy vector YCplac111, or this vector carrying the indicated AFR1 gene, was introduced into afr1 strain (JKY26). The cells were incubated with 10-7 M -factor for 3–6 hr at 30° as indicated to induce mating projections. The cells were fixed with formaldehyde and then photographed.

Interestingly, on the basis of their morphogenic ability the mutants could again be divided into two classes. The class I mutants (A31, A46, T164, and T211), which were most defective in promoting resistance to -factor when overexpressed, were also strongly defective in forming mating projections. These mutants were similar to the vector control cells in that they formed only a few projections and rarely, if ever, made multiple projections after prolonged exposure to pheromone. The class II mutants (A47, T181, and T182), which retained partial ability to confer resistance to -factor, also retained partial ability to make mating projections. Pointed projections were observed in 36, 28, and 19%, respectively, of the class II mutant cells. However, these mutants were still substantially less efficient in forming projections than wild-type AFR1 cells, which made projections in 58% of cells under these conditions. After prolonged exposure of the class II mutants to pheromone, some cells were able to make multiple projections (8, 11, and 3%, respectively); however, this ability was substantially reduced compared to wild type (39%). In addition, the projections that were made by the class II mutants appeared to be qualitatively different from wild type in that they were usually less pointed and exhibited broader neck regions (Fig 5). Thus, all of the mutants that were isolated on the basis of their inability to promote resistance to -factor also showed a corresponding defect in forming mating projections. The fact that single point mutations can cause proportional defects in the ability of Afr1p to promote both resistance and morphogenesis indicates that these phenotypes are tightly linked and may result from a single mechanism of Afr1p action.

Interaction with Cdc12p:
Interaction with the Cdc12p septin is important to help Afr1p localize properly in vivo and to promote normal morphogenesis (KONOPKA et al. 1995 ; GIOT and KONOPKA 1997 ). To examine the subcellular localization of the Afr1p mutants, AFR1 coding sequences were fused in frame to the sequences encoding the GFP. In afr1 cells carrying an AFR1-GFP fusion plasmid, wild-type Afr1-GFP localized to the neck of the mating projections in >95% of cells induced with -factor (Fig 6, left). This is essentially the same pattern of Afr1p localization that was detected in previous studies by immunofluorescence (KONOPKA et al. 1995 ; GIOT and KONOPKA 1997 ). In some cells, Afr1p appeared as a set of bars parallel to the long axis of the cell. This pattern was often easier to observe in the class II mutant T181, perhaps because the intermediate morphogenesis defect caused by this AFR1 mutant results in cells with long broad necks. Interestingly, a similar pattern of bars was detected for the septins in budding cells defective for the Gin4 protein kinase that colocalizes with septins in vivo (LONGTINE et al. 1998 ), suggesting that septin structure is differentially regulated during the formation of buds and mating projections.

View larger version (63K): [in this window] [in a new window] [Download PPT slide]   Figure 6. Localization of Afr1-GFP in -factor-induced cells. afr1 strain JKY26 (A) or AFR1 strain JKY25 (B) carried the indicated wild-type or mutant allele of AFR1 fused to the green fluorescent protein (GFP). The cells were induced with -factor (10-7 M) for 2 hr and then examined microscopically for Afr1-GFP localization. In A and B, a light microscope picture is shown to the left to demonstrate cell morphology, and a fluorescence microscope picture of Afr1-GFP is shown to the right.

The localization of the class I mutant proteins (A31, A46, and T164) was more difficult to observe because the GFP fluorescence was faint and the cells did not form a typical mating projection due to their AFR1 mutation. For the A46 and T164 mutant Afr1-GFPs, ~50% of cells produced a detectable GFP signal that was usually observed to form a ring of fluorescence or was concentrated at the neck region, indicating that these Afr1p mutants interact with the Cdc12p in vivo. In contrast, only ~5% of the A31 mutant cells showed a GFP pattern that resembled neck localization. The A31 mutant Afr1-GFP was most often observed in a diffuse subcellular localization with clusters of stronger fluorescence present in about half the cells. Since the mutants are mostly recessive to wild type, Afr1-GFP localization was examined next in AFR1⁺ cells that form pointed mating projections (Fig 6, right). Wild-type Afr1-GFP localized properly to the neck of mating projections as expected. Interestingly, the fluorescence of wild-type Afr1-GFP was significantly weaker in AFR1 cells than in afr1 cells, with only ~65% of cells showing detectable fluorescence, suggesting perhaps that Afr1-GFP may not compete as well as untagged Afr1p for a limited number of binding sites in vivo. Nonetheless, most of the class I and class II mutant proteins were observed to localize efficiently to the neck of mating projections in the AFR1 cells. Only the neck localization of the A31 mutant version of Afr1-GFP was still difficult to observe as >90% of the cells showed either diffuse GFP signal over the entire cell or in clusters within the cell. It is not clear whether this represents a defect in binding Cdc12p in vivo or a dominant effect of the A31 mutant on cell polarity.

The ability of the Afr1p mutants to interact with Cdc12p was examined further using the two-hybrid assay (BARTEL et al. 1993 ). For this analysis, one plasmid expressing the DNA-binding domain of lexA fused to AFR1 and a second plasmid expressing the GAL4 transcriptional activation domain fused to CDC12 were introduced into strain L40. As shown previously (KONOPKA et al. 1995 ), interaction between Afr1p and Cdc12p in this assay reconstitutes a transcriptional activator that induces a reporter gene consisting of lexA binding sites upstream of lacZ. Interaction between Afr1p and Cdc12p in this assay resulted in a 29-fold increase in ß-galactosidase activity for the lexA-AFR1 cells relative to control cells that expressed only the lexA DNA binding domain (Table 3). The four different Afr1p mutant proteins were also capable of interacting with Cdc12p. Interestingly, cells carrying the lexA-AFR1-A31 mutant (N254S, Y261H) showed significantly higher reporter gene activation than the wild-type or the other AFR1 mutants, as indicated by a 172-fold increase in ß-galactosidase activity relative to the control cells expressing lexA. The elevated reporter gene activity was dependent on the GAL4-CDC12 fusion plasmid as none of the cells carrying the control GAL4 activation domain plasmid showed significant levels of ß-galactosidase activity (not shown). Interestingly, the GFP-tagged version of this mutant also showed altered subcellular localization as described above. The improved signal detected for the A31 mutant in the two-hybrid assay could indicate a better interaction with Cdc12p. Alternatively, it is also possible that this mutant Afr1p functions better in the two-hybrid assay for other reasons such as improved protein stability or better nuclear localization. Altogether, these results indicate that the point mutations identified in this study do not prevent Afr1p from binding Cdc12p. This is consistent with previous results that the region of Afr1p that was required for the interaction with Cdc12p (residues 300–475; GIOT and KONOPKA 1997 ) is distinct from the region affected by the point mutations identified in this study (residues 254–263).

Analysis of YER158c: a S. cerevisiae ORF with similarity to AFR1:
To aid further in understanding the role of Afr1p, we analyzed the YER158c open reading frame found in the yeast genome that contains significant homology to Afr1p. DNA microarray analysis indicates that YER158c expression is not cell cycle regulated (SMITH et al. 1996 ; SPELLMAN et al. 1998 ), is induced during diauxic growth (DERISI et al. 1997 ), and is repressed during sporulation (CHU et al. 1998 ). The predicted 573-amino acid protein product of YER158c shows an overall identity to Afr1p of 24% and a similarity of 49% (Fig 7). However, the major regions of homology between YER158c and Afr1p lie in the N and C termini of the proteins, which are areas of unknown function. The sequences required for Afr1p function identified in the mutagenic screen are not conserved in the predicted Yer158c protein. To determine whether YER158c can regulate pheromone signaling, cells carrying the YER158c sequences cloned into a 2µ multicopy vector were assayed for their sensitivity to -factor in a halo assay (Fig 8). Cells carrying the multicopy YEp-YER158c plasmid arrested cell division similar to the cells carrying an empty plasmid vector (Fig 8A). In contrast, cells carrying a YEp-AFR1 plasmid adapted to the presence of -factor and did not display a zone of growth inhibition. These results demonstrate that multicopy YER158c does not promote hyperadaptation to pheromone as does AFR1.

View larger version (67K): [in this window] [in a new window] [Download PPT slide]   Figure 7. Alignment of Yer158cp and Afr1p. A BLAST alignment (ALTSCHUL et al. 1990 ) showing the protein sequence similarity between Yer158cp and Afr1p. Solid boxes, identical residues. Shaded boxes, similar residues.

View larger version (85K): [in this window] [in a new window] [Download PPT slide]   Figure 8. Analysis of YER158c. (A) Halo assays for -factor-induced cell division arrest. Strain DJ211-5-3 carrying multicopy vector YEplac181, or this vector carrying either YER158c or AFR1, was plated as a lawn of cells on solid agar medium. Filter discs containing 15 µl of 10-5 M, 5 x 10-6 M, and 10-6 M -factor were applied to the surface of the agar. The zone of growth inhibition caused by -factor was photographed after 2 days. (B) Pheromone-induced morphogenesis was analyzed in strains DJ211-5-3 (wild type, AFR1+ YER158C+), JKY26 (afr1), and yCRD105 (yer158c) that were induced with 10-7 M -factor for 3 hr and then photographed.

To determine whether YER158c is required for projection formation, a deletion allele was constructed using LYS2 sequences to replace YER158c sequences in the genome (see MATERIALS AND METHODS). The yer158c cells were viable, in agreement with previous results inferred from the positions of TY1 transposon insertions into chromosome V (SMITH et al. 1996 ). Examination of -factor-treated yer158c cells showed that they made pointed mating projections with the same efficiency as wild-type cells (Fig 8B). In contrast, afr1 cells assayed under the same conditions showed the expected defect in forming mating projections. Furthermore, in other experiments there were no detectable effects of the YEp-YER158c plasmid on projection formation and a yer158c afr1 double mutant was no worse in signaling or morphogenesis than the afr1 single mutant (data not shown). We also failed to detect an interaction between Yer158cp and Cdc12p in the two-hybrid assay. These results indicate that YER158c does not play an obvious role in projection formation. Thus, although Yer158cp contains some sequence similarity with Afr1p, it does not perform the same functions as Afr1p and does not appear to play a detectable role in pheromone signaling. These results are consistent with the lack of sequence homology between Yer158cp and the essential region of Afr1p. YER158c may therefore be involved in other cellular processes.

AFR1 genes from S. uvarum and S. douglasii:
As an additional approach to identify the sequences that are important for Afr1p function, we searched for AFR1 homologs in other species using low-stringency Southern blotting and PCR. AFR1-homologous gene fragments were recovered from the yeasts S. uvarum and S. douglasii. These species of yeast are closely related to S. cerevisiae as they undergo essentially identical conjugation events and can mate and form interspecies diploids (HAWTHORNE and PHILIPPSEN 1994 ). The predicted Afr1 proteins from S. uvarum and S. douglasii are 56 and 69% identical, respectively, to the essential region of S. cerevisiae Afr1p between residues 194 and 350. When all three proteins are aligned together, they are 27% identical in this region (Fig 9). Interestingly, the domain containing the substitution mutations we identified (amino acids 254–263) is exactly conserved in the homolog from S. douglasii and shows only a single conservative difference in S. uvarum (serine 262 of S. cerevisiae aligns with a cysteine in S. uvarum). Furthermore, this domain contains the largest contiguous block of identical residues in the essential regions of the three proteins. The high degree of identity in this domain further indicates an important role for residues 254–263 in Afr1p function.

View larger version (33K): [in this window] [in a new window] [Download PPT slide]   Figure 9. Alignment of the Afr1p sequences from S. cerevisiae, S. uvarum, and S. douglasii. The essential region of Afr1p (residues 194–350) from S. cerevisiae is shown aligned with the homologous protein sequences from S. douglasii and S. uvarum. The consensus line shows the residues that are identical in all three Afr1 proteins. The underlined sequence (residues 253–263) corresponds to a conserved region that contains the cluster of AFR1 point mutations identified in this study.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of residues critical for Afr1p function:
Afr1p acts to negatively regulate pheromone receptor signaling and to promote proper morphogenesis to form mating projections. Deletion mutagenesis localized the sequences required for these two functions to the region spanning amino acids 194–350 (GIOT and KONOPKA 1997 ). This region of Afr1p lacks any readily identifiable consensus sequences or functional motifs that are commonly present in other known proteins. To better define the sequences required for Afr1p function, we carried out a genetic screen that identified seven mutants that were defective in Afr1p function. Interestingly, all seven mutants contained point mutations that were clustered in a small region coding for amino acids 254–263 that is highly conserved in Afr1p homologs from other species. Thus, this clustering of mutations in a conserved region identifies this domain as essential for Afr1p's ability to promote negative regulation of pheromone receptor signaling.

Four different types of amino acid substitutions were identified within the seven AFR1 mutants we identified. The substitutions include three single mutants (V257A, I259K, and F263S) and one double mutant (N254S Y261H). Analysis of Afr1p by computer predictions of secondary protein structure suggests that the residues in the region between 254 and 263 may form a ß-sheet structure. Interestingly, all of the mutants were predicted to alter the propensity of this region to form a ß-sheet. These results suggest that this domain of Afr1p is a conformationally sensitive region that is critical for Afr1p function.

Relationship between signaling and morphogenesis:
The signaling and morphogenesis functions of Afr1p both require residues 194–350, but it is not clear if both functions are due to a single mechanism of action or to two independent activities. Interestingly, all of the mutants identified as defective in promoting resistance to -factor showed a proportional defect in their ability to produce projections (Fig 5, Table 2). These results suggest that the signaling and morphogenesis functions of Afr1p are interrelated. The ability of Afr1p to regulate both signaling and morphogenesis may stem from the subcellular localization of Afr1p at the neck of the mating projection. This contrasts with the localization of pheromone receptors that can be detected throughout the mating projection (JACKSON et al. 1991 ). Negative regulation by Afr1p could therefore act as a boundary to restrict the zone of receptor signaling to the tip of the mating projection. Since recent studies indicate that the G_ß subunits lead to activation of Cdc42p and other components involved in morphogenesis (BUTTY et al. 1998 ; NERN and ARKOWITZ 1998 ), spatially restricted receptor signaling would lead to a refined region of morphogenesis by activating the components required for projection formation only at the site of active receptor signaling. Therefore, the ability of Afr1p to spatially regulate pheromone receptor signaling may directly influence morphogenesis and promote the formation of mating projections.

AFR1 has also been reported recently to display synthetic-lethal and two-hybrid interactions with IQG1, a gene involved in bud morphogenesis and cytokinesis (OSMAN and CERIONE 1998 ). Interestingly, Iqgp is localized to the region of the bud neck in the same vicinity as the septins (EPP and CHANT 1997 ; LIPPINCOTT and LI 1998 ; OSMAN and CERIONE 1998 ). These results raise the interesting possibility that Afr1p may also participate in the regulation of the signals at the neck of the bud that controls bud morphogenesis and cytokinesis.

Coordination of morphogenesis by spatial regulation of signaling is also important in other GPCR pathways. For example, the leukocyte chemokine receptors and the Dictyostelium discoideum cAMP receptors induce oriented chemotaxis toward gradients of their ligands (DEVREOTES and ZIGMOND 1988 ; MURPHY 1994 ). Furthermore, postreceptor signaling events were observed predominantly at the leading edge of chemotaxis in D. discoideum (PARENT et al. 1998 ). Interestingly, many of the components that regulate pheromone signaling in yeast are homologous to the components that regulate signaling in other organisms (MACKAY et al. 1988 ; DOI et al. 1994 ; CHEN and KONOPKA 1996 ). For example, the SST2 gene, which plays a major role in regulating adaptation to the pheromone signal in yeast (CHAN and OTTE 1982 ; DIETZEL and KURJAN 1987 ), is now recognized to encode a member of a large family of regulators of G protein signaling (RGS) proteins that regulates GTP hydrolysis by the G subunits in a variety of organisms (KOELLE and HORVITZ 1996 ; DOHLMAN and THORNER 1997 ). Due to this conservation of G protein-coupled receptor signaling pathways across species, Afr1p homologs are also likely to exist in higher eukaryotes. The closest functional analogs of Afr1p are the arrestin proteins that regulate GPCR signaling in multicellular organisms in two ways: they bind to receptors to uncouple them from the G protein to block signaling and they also mediate receptor endocytosis (FERGUSON and CARON 1998 ). However, Afr1p differs from the arrestin proteins in that it does not require the C terminus of the -factor receptor to regulate signaling, does not appear to play a role in receptor endocytosis, and does not show significant sequence similarity to proteins of the arrestin family (KONOPKA 1993 ; DAVIS et al. 1998 ). Possibly, Afr1p homologs in other organisms have not been easily recognized on the basis of sequence similarity because they have evolved to regulate distinct receptors. In addition, Afr1p homologs may not be readily detected biochemically because Afr1p is a highly insoluble protein. Thus, the identification of conserved residues that are critical for Afr1p function in this study will help to identify potential Afr1p homologs in other species that can be examined for a role in the spatial regulation of GPCR signaling.

	ACKNOWLEDGMENTS

We thank the members of our lab for their helpful comments on the manuscript. We also thank Erica Roessner and Adelaide Asamoah for technical assistance. C.R.D. was supported in part by a predoctoral training grant from the National Cancer Institute (T32CAO9176). This work was supported by grants awarded to J.B.K. from the American Cancer Society (RPG9301406MBC) and the American Heart Association.

Manuscript received November 3, 1999; Accepted for publication January 19, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALTSCHUL, S. F., W. GISH, W. MILLER, E. W. MYERS, and D. J. LIPMAN, 1990  Basic local alignment search tool. J. Mol. Biol. 215:403-410[Medline].

AYSCOUGH, K. R. and D. G. DRUBIN, 1998  A role for the yeast actin cytoskeleton in pheromone receptor clustering and signalling. Curr. Biol. 8:927-930[Medline].

BARDWELL, L., J. G. COOK, C. J. INOUYE, and J. THORNER, 1994  Signal propagation and regulation in the mating pheromone response pathway of the yeast Saccharomyces cerevisiae.. Dev. Biol. 166:363-379[Medline].

BARTEL, P. L., C.-T. CHIEN, R. STERNGLANZ and S. FIELDS, 1993 Using the two-hybrid system to detect protein-protein interaction, pp. 153–179 in Cellular Interactions in Development: A Practical Approach, edited by D. A. HARTLEY. Oxford University Press, Oxford.

BOURNE, H. R., 1997  How receptors talk to trimeric G proteins. Curr. Opin. Cell Biol. 9:134-142[Medline].

BUTTY, A. C., P. M. PRYCIAK, L. S. HUANG, I. HERSKOWITZ, and M. PETER, 1998  The role of Far1p in linking the heterotrimeric G protein to polarity establishment proteins during yeast mating. Science 282:1511-1516[Abstract/Free Full Text].

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

CHANG, F. and I. HERSKOWITZ, 1990  Identification of a gene necessary for cell cycle arrest by a negative growth factor of yeast: FAR1 is an inhibitor of a G1 cyclin, CLN2. Cell 63:999-1011[Medline].

CHEN, Q. and J. B. KONOPKA, 1996  Regulation of the G protein-coupled -factor pheromone receptor by phosphorylation. Mol. Cell. Biol. 16:247-257[Abstract].

CHENEVERT, J., 1994  Cell polarization directed by extracellular cues in yeast. Mol. Biol. Cell 5:1169-1175[Medline].

CHENEVERT, J., N. VALTZ, and I. HERSKOWITZ, 1994  Identification of genes required for normal pheromone-induced cell polarization in Saccharomyces cerevisiae.. Genetics 136:1287-1297[Abstract].

CHU, S., J. DERISI, M. EISEN, J. MULHOLLAND, and D. BOTSTEIN et al., 1998  The transcriptional program of sporulation in budding yeast. Science 282:699-705[Abstract/Free Full Text].

COOK, J. G., L. BARDWELL, S. J. KRON, and J. THORNER, 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[Abstract/Free Full Text].

CORMACK, B. P., G. BERTRAM, M. EGERTON, N. A. GOW, and S. FALKOW et al., 1997  Yeast-enhanced green fluorescent protein (yEGFP) a reporter of gene expression in Candida albicans.. Microbiology 143:303-311[Abstract].

DAVIS, C., P. DUBE, and J. B. KONOPKA, 1998  Afr1p regulates the Saccharomyces cerevisiae -factor receptor by a mechanism that is distinct from receptor phosphorylation and endocytosis. Genetics 148:625-635[Abstract/Free Full Text].

DERISI, J. L., V. R. IYER, and P. O. BROWN, 1997  Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278:680-686[Abstract/Free Full Text].

DEVREOTES, P. N. and S. H. ZIGMOND, 1988  Chemotaxis in eukaryotic cells: a focus on leukocytes and Dictyostelium.. Annu. Rev. Cell Biol. 4:649-686.

DIETZEL, C. and J. KURJAN, 1987  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].

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., J. THORNER, M. G. CARON, and R. J. LEFKOWITZ, 1991  Model systems for the study of 7-transmembrane-segment receptors. Annu. Rev. Biochem. 60:653-688[Medline].

DOI, K., A. GARTNER, G. AMMERER, B. ERREDE, and H. SHINKAWA et al., 1994  MSG5, a novel protein phosphatase promotes adaptation to pheromone response in S. cerevisiae.. EMBO J. 13:61-70[Medline].

DRUBIN, D. G. and W. J. NELSON, 1996  Origins of cell polarity. Cell 84:335-344[Medline].

DUBE, P. and J. B. KONOPKA, 1998  Identification of a polar region in transmembrane domain 6 that regulates the function of the G protein-coupled -factor receptor. Mol. Cell. Biol. 18:7205-7215[Abstract/Free Full Text].

EPP, J. A. and J. CHANT, 1997  An IQGAP-related protein controls actin-ring formation and cytokinesis in yeast. Curr. Biol. 7:921-929[Medline].

FERGUSON, B., J. HORECKA, J. PRINTEN, J. SCHULTZ, and B. J. STEVENSON et al., 1994  The yeast pheromone response pathway: new insights into signal transmission. Cell. Mol. Biol. Res. 40:223-228[Medline].

FERGUSON, S. S. and M. G. CARON, 1998  G protein-coupled receptor adaptation mechanisms. Semin. Cell Dev. Biol. 9:119-127[Medline].

GEHRUNG, S. and M. SNYDER, 1990  The SPA2 gene of Saccharomyces cerevisiae is important for pheromone-induced morphogenesis and efficient mating. J. Cell Biol. 111:1451-1464[Abstract/Free Full Text].

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534[Medline].

GIOT, L. and J. B. KONOPKA, 1997  Functional analysis of the interaction between Afr1p and the Cdc12p septin, two proteins involved in pheromone-induced morphogenesis. Mol. Biol. Cell 8:987-998[Abstract].

GIOT, L., C. DEMATTEI, and J. B. KONOPKA, 1999  Combining mutations in the incoming and outgoing pheromone signal pathways causes a synergistic mating defect in Saccharomyces cerevisiae.. Yeast 15:765-780[Medline].

HAWTHORNE, D. and P. PHILIPPSEN, 1994  Genetic and molecular analysis of hybrids in the genus Saccharomyces involving S. cerevisiae, S. uvarum and a new species, S. douglasii.. Yeast 10:1285-1296[Medline].

HEPLER, J. R. and A. G. GILMAN, 1992  G proteins. Trends Biochem. Sci. 17:382-387.

HERSKOWITZ, I., 1995  MAP kinase pathways in yeast: for mating and more. Cell 80:187-197[Medline].

JACKSON, C. L., J. B. KONOPKA, and L. H. HARTWELL, 1991  S. cerevisiae -pheromone receptors activate a novel signal transduction pathway for mating partner discrimination. Cell 67:389-402[Medline].

KOELLE, M. R. and H. R. HORVITZ, 1996  EGL-10 regulates G protein signaling in the C. elegans nervous system and shares a conserved domain with many mammalian proteins. Cell 84:115-125[Medline].

KONOPKA, J. B., 1993  AFR1 acts in conjunction with the -factor receptor to promote morphogenesis and adaptation. Mol. Cell. Biol. 13:6876-6888[Abstract/Free Full Text].

KONOPKA, J. B., D. D. JENNESS, and L. H. HARTWELL, 1988  The C terminus of the Saccharomyces cerevisiae -pheromone receptor mediates an adaptive response to pheromone. Cell 54:609-620[Medline].

KONOPKA, J. B., C. DEMATTEI, and C. DAVIS, 1995  AFR1 promotes polarized apical morphogenesis in Saccharomyces cerevisiae.. Mol. Cell. Biol. 15:723-730[Abstract].

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

LEBERER, E., D. DIGNARD, D. HARCUS, D. Y. THOMAS, and M. WHITEWAY, 1992  The protein kinase homologue Ste20p is required to link the yeast pheromone response G-protein ß subunits to downstream signalling components. EMBO J. 11:4815-4824[Medline].

LEBERER, E., D. Y. THOMAS, and M. WHITEWAY, 1997  Pheromone signalling and polarized morphogenesis in yeast. Curr. Opin. Genet. Dev. 7:59-66[Medline].

LIPPINCOTT, J. and R. LI, 1998  Sequential assembly of myosin II, an IQGAP-like protein, and filamentous actin to a ring structure involved in budding yeast cytokinesis. J. Cell Biol. 140:355-366[Abstract/Free Full Text].

LONGTINE, M. S., D. J. DEMARINI, M. L. VALENCIK, O. S. AL-AWAR, and H. FARES et al., 1996  The septins: roles in cytokinesis and other processes. Curr. Opin. Cell Biol. 8:106-119[Medline].

LONGTINE, M. S., H. FARES, and J. R. PRINGLE, 1998  Role of the yeast Gin4p protein kinase in septin assembly and the relationship between septin assembly and septin function. J. Cell Biol. 143:719-736[Abstract/Free Full Text].

MACKAY, V. L., S. K. WELCH, M. Y. INSLEY, T. R. MANNEY, and J. H