Suppression of the Profilin-Deficient Phenotype by the RHO2 Signaling Pathway in Saccharomyces cerevisiae

Corresponding author: Dominick Pallotta, Pavillon Marchand, Laval University, Ste-Foy, Quebec G1K 7P4, Canada., pallotta{at}rsvs.ulaval.ca (E-mail)

Communicating editor: P. G. YOUNG

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Profilin plays an important role in actin organization in all eukaryotic cells through mechanisms that are still poorly understood. We had previously shown that Mid2p, a transmembrane protein and a potential cell wall sensor, is an effective multicopy suppressor of the profilin-deficient phenotype in Saccharomyces cerevisiae. To better understand the role of Mid2p in the organization of the actin cytoskeleton, we isolated five additional multicopy suppressors of pfy1 cells that are Rom1p, Rom2p, Rho2p, Smy1p, and the previously uncharacterized protein Syp1p. The problems of caffeine and NaCl sensitivity, growth defects at 30° and 37°, the accumulation of intracellular vesicular structures, and a random budding pattern in pfy1 cells are corrected by all the suppressors tested. This is accompanied by a partial repolarization of the cortical actin patches without the formation of visible actin cables. The overexpression of Mid2p, Rom2p, and Syp1p, but not the overexpression of Rho2p and Smy1p, results in an abnormally thick cell wall in wild-type and pfy1 cells. Since none of the suppressors, except Rho2p, can correct the phenotype of the pfy1-111/rho2 strain, we propose a model in which the suppressors act through the Rho2p signaling pathway to repolarize cortical actin patches.

IN yeast, actin filaments form cortical patches and actin cables. Cortical patches are compact structures that relocalize during the cell cycle in accordance with changes in patterns of growth and secretion. During the isotropic growth in the G1 phase, the cortical patches are uniformly distributed within the cell. At the beginning of S phase they are located at the presumptive bud site, which becomes a site of active growth. As the cycle progresses, the cortical patches become localized almost exclusively in the growing bud. Electron microscopic observation showed that at this stage the cortical patches are associated with invaginations of the plasma membrane, presumably to maintain cell wall integrity during cell growth (MULHOLLAND et al. 1994 ). Finally, during cytokinesis the patches are found at the neck region that separates the mother and daughter cells. Any mutation that affects the location of the cortical patches also affects polarized growth. Cortical patches are thought to be essential for viability, since a viable cell without cortical patches has never been observed (KARPOVA et al. 1998 ).

Actin cables are long filamentous structures that traverse the cell. They have an uneven thickness along their length, suggesting that they are formed of short overlapping filaments (KARPOVA et al. 1998 ). Cables and patches are both present throughout the cell cycle and their distribution is coordinated. When patches are polarized the cables are oriented toward the patches and when patches are dispersed cables are randomly oriented. The frequent localization of patches at the ends of cables is further evidence for the association of these two structures.

The distribution and formation of actin cables and patches is controlled by many different proteins. Some, like Cdc42p, are signaling proteins that control actin distribution throughout the cell cycle. Others are structural proteins that bind directly to either filamentous or monomeric actin. Profilin is a small actin monomer-binding protein that influences actin organization. It is clearly involved in actin polymerization, although there is still controversy over whether profilin is a sequestering agent or whether it facilitates polymerization. All known profilins bind to poly-L-proline residues and to proline-rich sequences in proteins. In Saccharomyces cerevisiae, Pfy1p binds to the proline-rich formin-homology (FH) domains of Bni1p and Bnr1p (EVANGELISTA et al. 1997 ; IMAMURA et al. 1997 ). Bni1p is also a target of Rho1p, a Ras-like GTPase, involved in cell wall synthesis and organization of the actin cytoskeleton (KOHNO et al. 1996 ). Finally, profilin binds to the membrane phospholipid phosphatidylinositol-biphosphate (PIP₂), which in mammalian cells at least, inhibits its hydrolysis by phospholipase C (LASSING and LINDBERG 1985 ; GOLDSCHMIDT-CLERMONT et al. 1990 ). This suggests a role for profilin in cell signaling, possibly as an intermediate between signaling molecules and the organization of the actin cytoskeleton.

Profilin-deficient cells (pfy1) are viable but have an abnormal morphology. Actin cables are no longer visible, and cortical patches are present in the mother cell at all stages of the cell cycle (HAARER et al. 1990 , HAARER et al. 1993 ; MARCOUX et al. 1998 , MARCOUX et al. 1999 ). This results in large, round cells that grow slowly and undergo cell lysis at high temperatures. The polarity of the cells is also lost, and the normal axial budding of haploid cells becomes random. In addition, polarized secretion is compromised so that cells accumulate vesicular structures in their cytoplasm and show a lag in the secretion of the -factor. Maturation of the -factor is also reduced, which results in a decreased mating efficiency of the MAT strain, but not the MATa strain (MARCOUX et al. 1998 ).

We have previously shown that MID2 is an effective multicopy suppressor of the profilin-deficient phenotype (MARCOUX et al. 1998 ). The acronym MID stands for "mating pheromone-induced death" since MID mutants cannot resume the cell cycle after shmoo formation (IIDA et al. 1994 ). Mid2p is a small plasma membrane protein containing a serine/threonine-rich glycosylated extracellular domain, a single transmembrane domain, and a charged intracellular domain rich in aspartic acid residues. In addition to the suppression of the profilin-deficient phenotype, the MID2 gene was selected as a multicopy suppressor in several different genetic screens. Overexpression of MID2 suppresses the growth defect of the cik1 and kar3 strains, mutants defective in microtubule-based processes (MANNING et al. 1997 ), and it can correct the deleterious effects caused by overexpression of protein kinase A, Tpk1p (DANIEL 1993 ). Hcs77p, also called Slg1p or Wsc1p, is a transmembrane protein without any sequence similarity to Mid2p but with domains similar to those found in Mid2p (GRAY et al. 1997 ; VERNA et al. 1997 ; JACOBY et al. 1998 ; LODDER et al. 1999 ). The hcs77 cells undergo cell lysis at high temperatures, a phenotype that is corrected by the expression of MID2 (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). The phenotypes of temperature-sensitive growth defects and the inability to recover from mating pheromone-induced G1 arrest of the mpt5 mutation are also corrected by the overexpression of MID2. Finally, MID2 has also been selected as an activator of the Skn7p transcription factor (KETELA et al. 1999 ).

An explanation for these seemingly disparate results was provided by recent work identifying Mid2p as a potential cell wall stress sensor that activates the PKC1 pathway, probably through Rho1p, to ensure cell wall integrity (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). These results support a role for Mid2p in cell wall synthesis, but they do not indicate how it influences actin cytoskeleton organization. To obtain more information on this question, we used the caffeine and NaCl sensitivities of the pfy1 cells to identify five additional multicopy suppressors of the profilin-deficient phenotype. We propose a model in which Mid2p affects the actin cytoskeleton in the absence of profilin through a pathway leading to Rom1/Rom2 and Rho2p.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains, media, and transformations:
Strains used in this study are presented in Table 1. Cells were grown in either YPD (1% yeast extract, 2% peptone, 2% glucose) or SD (0.67% yeast nitrogen base w/o amino acids, 2% glucose) containing all amino acids except uracil (-Ura). Unless otherwise indicated, cells were transformed by a modified lithium acetate method (KAISER et al. 1994 ).

Deletions of MID2 and SYP1 were performed using the PCR-based method described by BAUDIN et al. 1993 . The coding sequence of MID2 was replaced with the HIS3 gene, amplified using oligos 5'-TTCGTTGAAGATTGGACATATAAAATACG CAAATCATAGTTTCCCTAGCATGTACGTGAG and 5'-GAAAAGTAGCCATAAGCACTAAATGATATGAATGGATATGGTGTCACTACATAAGAACAC. The PCR product containing the HIS3 gene flanked by 50 nucleotides of the 5' and 3' noncoding sequences of MID2 was used to transform the 22AB1-6A and DJP100 haploid strains by the standard lithium acetate method (KAISER et al. 1994 ). Deletion of MID2 was confirmed by PCR analysis.

The coding sequence of SYP1 was replaced with the kanamycin gene, amplified from the pFA6-kanMX4 plasmid (WACH et al. 1994 ) using oligos 5'-ATGACGGAACAAAGAACCAAA TATGCAGATAGCATATTGACTACTAAGAGTCAGCTGAAG CTTGTACGC and 5'-TCAAGCGAGACCATGGTAGTTACCAGTAGTCAAGTTCTTTTTGTGTTTACGCATAGGCCACTAG TGGATCTG. The resulting PCR product, containing the kan gene flanked by 51 bp homologous to the first and last nucleotides of the SYP1 coding sequence, was used to transform the BY4743 diploid strain by the standard lithium acetate method (KAISER et al. 1994 ). Transformants were selected on YPD medium containing 200 µg/ml geneticin, and the deletion of SYP1 was confirmed by PCR analysis. The haploid mutant syp1::kan strain was obtained by tetrad dissection from a heterozygous syp1::kan/SYP1 strain.

Plasmids:
Overexpression of Rho2p, Rho3p, and Rho4p was obtained by inserting the genes and their promoters into the multicopy pRS426 plasmid (CHRISTIANSON et al. 1992 ). The three genes were obtained by PCR amplification and ligated into the EcoRI-BamHI sites of pRS426. Sequences of oligos used are the following: RHO2, 5'-AAGGATCCATACCTCCA CAAGG and 5'-AAGAATTCGTAGGACATTA-ACGA; RHO3, 5'-ACTCGAATTCAGGCCACTTACC and 5'-AAGGGTACCTACAC-CTTCGACT; RHO4, 5'-GGAGAATTCTTTATACCCGT and 5'-GAGCGGTACCTCA-CACCAGATT. For overexpression of Mtl1p and Wsc11/Slg1/Hcs77, the plasmids pRS424 (MTL1) and pRS424 (HCS77) were used (RAJAVEL et al. 1999 ).

To construct the Syp1p-GFP fusion protein, the BssHII-NarI genomic fragment of SYP1, containing the complete coding sequence of SYP1 plus 544 bp upstream of the initiation codon and 746 bp downstream of the STOP codon, was inserted into the HincII/ClaI sites of pBluescript KS (+). The resulting plasmid, pBS-SYP1, was digested by NcoI and BamHI and the ends were filled in and religated, which eliminated the last three codons. The green fluorescent protein (GFP) sequence present in the plasmid p2312MAL (CORMACK et al. 1997 ) was amplified with oligos 5'-GAACGGATCCATGTCTAAAGGT GAAGAATT and 5'-GAACACTAGTTTATTTGTACAATTCATCCA. The PCR product was inserted into the BamHI-SpeI sites of pBS-SYP1, which places the GFP coding sequence in frame with the sequence coding for Syp1p at its C terminus, with the addition of a serine and a methionine between the two sequences. Finally, the XhoI-SpeI fragment of pBS-SYP1-GFP, containing the promoter of SYP1 plus the sequence coding for the Syp1p-GFP fusion protein was inserted into the centromeric plasmid pRS316 or the multicopy plasmid pRS416 (SIKORSKI and HIETER 1989 ).

Phalloidin staining, Calcofluor staining, and electron microscopy:
Staining of actin filaments was carried out according to the modified protocol of KARPOVA et al. 1998 . Cells from exponential phase cultures were fixed at room temperature for 30 min with 3.75% formaldehyde in the culture medium. Cells were washed with and resuspended in PBS. Staining was performed with 8.25 mM FITC-conjugated phalloidin for 90 min on ice in the dark. Cells were washed repeatedly with PBS before microscopic observation.

Bud scars were visualized by staining chitin rings with Calcofluor (PRINGLE 1991 ). Cells from overnight cultures were fixed for 30 min with 3.75% formaldehyde in the culture medium. Cells were pelleted and resuspended in Solution A (50 mM KPO₄ pH 6.5, 0.5 mM MgCl₂) containing 3.75% formaldehyde and further fixed for 1 hr (PRINGLE et al. 1991 ). Cells were then washed three times with and resuspended in Solution A. Staining was carried out by adding 1/500 volume of 1 mg/ml Calcofluor White and incubating for 1 hr at room temperature in the dark. Cells were finally washed repeatedly in PBS prior to microscopic observation. For electron microscopy, cells were fixed and prepared as previously described (MARCOUX et al. 1998 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of pfy1 suppressors:
To screen for multicopy suppressors, the pfy1 strain DJP102 was transformed with the YEp24 multicopy library constructed with DNA of the S288C strain (CARLSON and BOTSTEIN 1982 ). Approximately 15,000 transformants, representing about five genomes, were obtained and replicated onto -Ura plates containing 1.5 mg/ml caffeine or 1.5 M NaCl. The plasmids that allowed growth on both media were isolated and studied further. As expected from the number of genome equivalents screened, we obtained five plasmids containing PFY1. We also isolated five plasmids containing MID2. In addition, four other genes were isolated as suppressors: SMY1, ROM1, ROM2, and an open reading frame (ORF), YCR030C (Table 2). We named the ORF SYP1 for suppressor of yeast profilin deletion.

View this table: In this window In a new window   Table 2. Suppressors and nonsuppressors of the pfy1 phenotype

Given the established role for the Rho-type GTPases in controlling the actin cytoskeleton and the identification in our screen of Rom1p and Rom2p, two exchange factors for Rho proteins, we tested directly the effect of the five Rho-type GTPases, RHO1–RHO4 and CDC42, in the pfy1 strain DJP102. Among the RHO family members, only Rho2p is a suppressor of the growth defects of the pfy1 strain (Table 2). The pfy1 strain overexpressing Rho2p grows well in the presence of 1.25 mg/ml caffeine but grows slowly on 1.5 mg/ml caffeine, which may explain why it was not selected in the screen. A detailed analysis was carried out on Rho2p and the five suppressors isolated in the original screen.

Suppressors restore normal growth and an axial budding pattern to the pfy1 cells:
The suppressors were selected for their ability to confer growth in the presence of caffeine and NaCl. We also tested their ability to correct other phenotypic defects. The pfy1 cells are viable but grow slowly at 30°, with a doubling time of 6 hr in minimal medium, and do not grow at 37°. The pfy1 cells overexpressing any of the suppressors have a normal doubling time of 2 hr in minimal media and grow at 37°.

Normal haploid yeast cells have an axial budding pattern, meaning that the bud scars are formed on the same side of the cell and adjacent to the previous bud scar. Haploid pfy1 cells have a random distribution of bud scars and also a delocalized chitin deposition. We tested the bud site location in pfy1 cells expressing the different suppressors by staining for chitin with the fluorescent dye Calcofluor white. In mutant cells overexpressing either Smy1p (Fig 1) or Rho2p, Rom1p, Rom2p, and Syp1p (results not shown), the axial budding pattern is restored and chitin deposition is concentrated mainly in the bud scars, as it is in wild-type cells. In mutant cells overexpressing Mid2p, the Calcofluor white staining was delocalized and very intense, even more so than in the pfy1 cells. We were not able to identify the bud scars in these cells and thus could not determine whether they have an axial distribution. Similar results were obtained when Mid2p was overexpressed in wild-type cells. An explanation for this intense staining is found in the recent work showing that cells overexpressing Mid2p produce 250% more chitin than wild-type cells (KETELA et al. 1999 ). Since Calcofluor white intercalates into nascent chitin chains, cells overexpressing Mid2p are expected to show an intense staining reaction with this fluorescent dye. Thus, we were able to identify the bud scars in four of the suppressors, and they all showed the normal axial distribution in haploid cells.

View larger version (38K): In this window In a new window Download PPT slide   Figure 1. The normal axial budding pattern is restored in haploid pfy1 cells overexpressing Smy1p. Cells from exponential phase cultures were stained with Calcofluor and observed by fluorescence microscopy. Wild-type (WT) cells and pfy1 cells overexpressing Smy1p (pfy1/SMY1) have an axial distribution while pfy1 cells have a random distribution of bud scars.

Only some suppressors cause the formation of a thick cell wall:
We showed previously that pfy1 cells are large and accumulate numerous intracellular vesicular structures due to problems with intracellular trafficking. The overexpression of Mid2p decreases cell size to near normal, eliminates the intracellular vesicles, and results in an abnormally thick cell wall (MARCOUX et al. 1998 ). We reasoned that the phenotype of a thick cell wall could be informative for determining the signaling pathway involved in the suppression of the profilin-deficient phenotype. We therefore carried out electron microscopic examinations of pfy1 cells carrying different suppressors (Fig 2 and Table 3). It should be noted that the pfy1 cells have a wide range of size and morphology, and some of them from late log phase are much larger and contain more intracellular vesicular structures than the cell shown in Fig 2. The pfy1 cells overexpressing the suppressors are somewhat variable in size, but are always smaller than the pfy1 cells and are often about the same size as wild-type cells. No intracellular vesicles are seen in any of these cells.

View larger version (95K): In this window In a new window Download PPT slide   Figure 2. Electron microscopy of cell walls. Thick cell walls are seen in wild-type strains overexpressing Rho1p and Mid2p and pfy1 cells overexpressing Mid2p and Syp1p. All the other strains have normal cell walls.

View this table: In this window In a new window   Table 3. Production of a thick cell wall by the overexpression of Rho1p or the pfy1 suppressors in different strains as viewed by electron microscopy

The thickness of the cell wall in pfy1 and wild-type cells is similar. However, when Mid2p, Syp1p, Rom1p, or Rom2p are overexpressed in pfy1 cells, the cell walls in the mother cell are noticeably thicker. This is not accompanied by evident changes in the cell wall diameter of the bud. We also determined that the overexpression of Mid2p in wild-type cells leads to an increase in cell wall diameter of the mother cell, indicating that it is the suppressors alone that are responsible for the cell wall changes. In contrast, the cell walls in strains overexpressing Rho2p or Smy1p have normal dimensions. Therefore, not all suppressors cause the formation of a thick cell wall, indicating that a thick wall is not a prerequisite for the suppression of the profilin-deficient phenotype.

Nonetheless, we exploited this phenotype to determine whether Mid2p and Syp1p are in the same or in parallel signaling pathways. We reasoned that Mid2p, a plasma membrane and a putative sensor for cell wall integrity, should be the first step in a signaling pathway. If Syp1p is required to relay the signal from Mid2p to downstream components of the pathway, then overexpression of Mid2p in syp1 cells should give a normal cell wall. However, we observed that the syp1 cells overexpressing Mid2p have thick cell walls, suggesting that Mid2p does not signal cell wall biosynthesis through Syp1p (Table 3).

Rom2p is an exchange factor for Rho1p (OZAKI et al. 1996 ), and Mid2p activates the Rho1p signaling pathway (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). To investigate whether the increase in cell wall diameter implicates activation of the Rho1p signaling pathway by the multicopy suppressors, we overexpressed Rho1p in pfy1 and wild-type cells. Indeed, we found that an increased dosage of RHO1, in either pfy1 or wild-type cells, leads to a thick cell wall (Fig 2 and Table 3). We therefore conclude from these data that some of the pfy1 multicopy suppressors increase cell wall biosynthesis, at least in part, through the activation of the Rho1p signaling pathway. The facts that Rho1p is not a suppressor of the pfy1 mutant and that not all multicopy suppressors influence cell wall synthesis suggest that correction of the pfy1 phenotype implicates another cellular function. We thus turned our attention to the actin cytoskeleton.

Suppressors restore cortical patch polarity in the absence of visible actin cables:
Deletion of PFY1 results in large round cells that no longer have visible actin cables. These cells do have numerous, well-formed cortical patches, but their distribution is not normal. Although there is a concentration of cortical patches in small buds, numerous cortical patches are always found in the mother cell at all stages of the cell cycle (HAARER et al. 1990 ). Since the suppressors restored cell size to near normal in pfy1 cells, we verified whether this was accompanied by the formation of actin cables and by the normal, polarized distribution of cortical patches. We were able to visualize cables and patches by using the staining method of KARPOVA et al. 1998 . We made most of our observations on cells with small buds, since at this stage the polarized distribution of cortical patches is most evident (Fig 3). In wild-type cells, actin cables are easily visible in the mother cells and are oriented toward the small buds, where the cortical patches are concentrated; few cortical patches are seen in the mother cell at this stage. In pfy1 cells, actin cables are not visible and numerous patches are distributed throughout the mother cell, even when small buds are present. When Mid2p is overexpressed in these cells, actin patches become mostly concentrated in the small buds (Fig 3 and MARCOUX et al. 1998 ). The redistribution of cortical patches is not complete, since some cells still contain a few patches in the mother cell. In spite of the near-normal distribution of cortical patches, none of the pfy1 cells overexpressing Mid2p contain visible cables. Staining of actin filaments in pfy1 cells overexpressing any of the other suppressors gives similar results: cortical patches are localized almost exclusively in small buds and the cells have no visible actin cables. The results with Syp1p, Rho2p, and Smy1p are shown (Fig 3).

View larger version (66K): In this window In a new window Download PPT slide   Figure 3. Overexpression of the different suppressors in pfy1 cells repolarizes cortical patches without restoring visible actin cables. Cells from exponential phase cultures were stained with FITC-phalloidin to visualize actin filament distribution. Wild-type (WT), pfy1, and pfy1 cells overexpressing Mid2p (pfy1/MID2), Syp1p (pfy1/SYP1), Rho2p (pfy1/RHO2),and Smy1p (pfy1/SMY1) are shown.

In wild-type cells, cables are clearly visible in the mother cell. Although there are cables in the bud, the numerous cortical patches tend to mask their appearance. We do not believe, however, that the absence of cables in the pfy1 cells overexpressing the suppressors is due to a masking effect by the cortical patches. In most of these cells, few cortical patches are seen in the mother cell and in some instances none are present. However, we cannot rule out the presence of fewer, thinner, or much shorter actin cables. If present, however, they do not have the same organization as in wild-type cells.

Overexpression of Syp1p results in an abnormal phenotype:
Five of the proteins we identified as suppressors, Mid2p, Rom1p, Rom2p, Rho2p, and Smy1p, have been studied previously. They all play a role in cell wall biosynthesis or in the organization of the actin cytoskeleton. The other suppressor, Syp1p, is a previously uncharacterized protein. In an attempt to obtain more information about the role of SYP1, its entire coding sequence was deleted in a diploid strain by gene replacement using the kanamycin gene. Following sporulation, dissection of tetrads resulted in four haploid spores giving rise to colonies of approximately the same size. Haploid syp1 strains were identified by growth on geneticin-containing medium and confirmed by PCR analysis. Morphology of the haploid Syp1p-deficient cells is normal, as is actin distribution (results not shown). These cells have normal growth rates at 30° and 37° and in the presence of caffeine, NaCl, and benomyl, a microtubule depolymerizing drug. The budding patterns of haploid syp1 and diploid syp1/syp1 cells are also normal. Therefore, deletion of SYP1 does not result in an obvious morphological or growth defect, suggesting that Syp1p has a redundant function with at least one other protein. Sequence alignment with the S. cerevisiae and all other available databases does not reveal proteins with significant sequence homologies, identifying Syp1p as a novel protein.

Although we have not observed a phenotype associated with the deletion of SYP1, overproduction of Syp1p alters morphology of wild-type haploid and diploid cells. Overexpression of Syp1p causes the formation of abnormally long projections in 10% of exponentially growing haploid cells at 30° (Fig 4). Cortical patches are present in these projections, consistent with their localization in regions of cell growth. The cortical patches, however, are not located exclusively in the projections as they are also present in the mother cell. Staining of chitin rings with Calcofluor reveals that the budding pattern of these cells is normal. The long projection, however, does not respect the axial budding pattern as it always forms in a different area of the cell, separate from the bud scars (Fig 4). Occasionally, a projection and a bud form on the same cell, but the two are never contiguous. Thus, in these cells, the formation of the projection is not subject to the same spatial controls as bud formation. Chitin, in addition to its localization in bud scars, is also found throughout the projection, except at the rounded tip. The projections do not contain nuclei (results not shown). These projections are never formed by the overexpression of Syp1p in pfy1 cells.

View larger version (95K): In this window In a new window Download PPT slide   Figure 4. Effect of overexpression of Syp1p in wild-type cells. Morphology of haploid and diploid cells were visualized by Nomarski (N). Actin filament distribution was viewed by staining cells with FITC-conjugated phalloidin (P). Calcofluor staining of chitin rings was performed to observe budding patterns (C). The arrows indicate the position of protuberances formed in the diploid cells.

The effect of overproduction of Syp1p in diploid cells is less dramatic; no long projections are seen. In 10% of cells, however, a protuberance is seen. In cells with buds, the protuberance is found at the opposite end of the cell from the bud. This protuberance does not seem to be a site of active growth, since cortical patches are not present in this area. Chitin staining shows that the normal bipolar budding pattern of diploid cells is respected (Fig 4). This staining also shows that the small protuberance contains bud scars, suggesting that it may be formed by problems in chitin deposition or cytokinesis.

Syp1p localizes to the mother-bud neck and sites of active growth:
A cellular location has been established for five of the suppressor proteins we isolated. Mid2p is a plasma membrane protein and the four other proteins are localized in the cortical patch region of small buds. We reasoned that the cellular location of Syp1p could be informative for determining its role in the suppression of the profilin-deficient phenotype. It was localized using a Syp1p-GFP fusion protein. This fusion protein was expressed using the SYP1 promoter in a centromeric plasmid. Since syp1 cells have no observable phenotype, we could not test whether the protein was functional by suppression of the mutant phenotype. However, overexpression of the fusion protein in wild-type haploid cells resulted in the formation of long projections in 10% of the cells. In addition, overexpression of the fusion protein in pfy1 cells restored normal growth at 30° and 37° and eliminated the caffeine sensitivity phenotype. The pfy1 cells overexpressing the Syp1-GFP fusion protein or Syp1p are morphologically indistinguishable in the light microscope (results not shown). Therefore, Syp1p-GFP is functional, and its localization likely respects the normal distribution of Syp1p in cells.

The intracellular location of Syp1p was determined by fluorescence microscopy observations of exponentially growing cells expressing the Syp1p-GFP fusion protein from a centromeric plasmid. In cells without a bud, there is a low level of diffuse fluorescence staining as well as the presence of small dots located at the cell periphery (Fig 5A). The diffuse staining and the peripheral dots remain during the entire cell cycle, but an intense fluorescence develops in specific locations during budding. As the bud emerges, Syp1p becomes concentrated at the mother-bud neck and at the tip of the forming bud (Fig 5B). As the bud grows, the Syp1p fluorescence is abundant in the mother-bud neck region and in the bud (Fig 5C). At cytokinesis, the fusion protein is found predominantly at the junction between the mother cell and the bud (Fig 5D). In shmoos, Syp1p is present mainly at the base of the projection (Fig 5E). The distribution of Syp1p in growing buds, in the septum of dividing cells, and at the base of shmoo projections is consistent with a role for this protein in polarity establishment and/or cytokinesis.

View larger version (46K): In this window In a new window Download PPT slide   Figure 5. Distribution of Syp1p changes during the cell cycle. The Syp1p-GFP fusion protein was expressed from a centromeric plasmid in WT cells (a–e) and from a multicopy plasmid in the pfy1 strain DJP102 (f). The Syp1p-GFP was viewed by fluorescent microscopy in exponentially growing cells at different stages of the cell cycle (a–d and f) and during shmoo formation (e). To form shmoos, MATa wild-type cells (DJP100) expressing Syp1p-GFP were incubated with 6 µM -factor for 3 hr prior to observation by fluorescence microscopy.

The expression of the Syp1p-GFP fusion protein from a centromeric plasmid does not suppress the pfy1 phenotype. These cells, which are large and irregularly shaped, have such low levels of fluorescence that the fusion protein could not be localized. The expression of the Syp1p-GFP fusion protein from a multicopy plasmid in pfy1 cells gives an intense fluorescence. The location of the fusion protein is similar in these cells and in wild-type cells expressing the protein from a centromeric plasmid, although the fluorescence is much more intense in the pfy1 cells with the multicopy plasmid. A cell with a medium size bud is shown (Fig 5F). The mother cell has a peripheral fluorescence, usually in the form of small dots. The mother-bud neck junction and the contour of the bud are intensely stained, while a diffuse fluorescence is seen within the bud.

The expression of the Syp1-GFP fusion protein from a multicopy plasmid in wild-type cells gives an intense staining mainly in the 10% of the population with long projections. In G1 cells with projections, the fusion protein is found throughout the mother cell and the projection, except at the junction between the two where a pale region is seen. The most intense fluorescence is seen in the contour of the mother cell and the projection (results not shown).

Genetic interactions:
On the basis of previously published work, we hypothesized that Mid2p, Rom1p, Rom2p, and Rho2 act through the same signaling pathway to correct the profilin-deficient phenotype. To test this hypothesis and to obtain more information about all the suppressors, we constructed various double deletion strains. We first checked for possible synthetic lethal interactions among the suppressor genes. Haploid strains mid2/rho2, mid2/smy1, mid2/syp1, rho2/smy1, and rom2/syp1 are all viable. Also, as with cells carrying single deletions for any of these genes, no obvious morphological or growth defects were observed (Table 4). Thus, these suppressor genes do not show a synthetic lethality. We attempted to obtain double mutants with the pfy1 allele and the deleted suppressor genes, but our repeated attempts were unsuccessful. We assumed that this was due to the severely compromised growth of the pfy1 cells. We therefore constructed double mutants with the pfy1-111 allele (HAARER et al. 1993 ). This allele renders cells partly temperature sensitive; they grow slowly at an elevated temperature and a fraction of them undergo cell lysis. At the restrictive temperature, cells adopt a phenotype similar to but not as severe as the pfy1 cells. Phalloidin staining of pfy1-111 cells incubated at 37° shows a partial loss of actin patch polarization and also the absence of visible actin cables (Fig 6). Using this allele, we studied the double mutants pfy1-111/mid2, pfy1-111/rho2, pfy1-111/smy1, and pfy1-111/syp1.

View larger version (112K): In this window In a new window Download PPT slide   Figure 6. Actin filament distribution in different mutant strains. Strains were grown to exponential phase at 30° then shifted to 37° for 2 hr 30 min prior to staining with FITC-phalloidin. Cortical patches are seen in all the strains but are delocalized in the pfy1-111, pfy1-111/mid2, and pfy1-111/rho2 strains. Actin cables are seen in the WT, mid2, and rho2 strains only.

Deletion of SYP1 in the pfy1-111 background does not increase the morphological defects or the aberrant actin distribution of the pfy1-111 cells. In contrast, the pfy1-111/mid2 and pfy1-111/rho2 double mutants have a nearly completely depolarized actin distribution at the restrictive temperature. The pfy1-111/mid2 and pfy1-111/rho2 strains grow very slowly at 37° and few cells at this temperature have buds. Therefore, exponential phase cultures grown at 30° were shifted to 37° for 2 hr 30 min prior to staining with FITC-conjugated phalloidin. Under these conditions, both mid2 and rho2 cells contain abundant actin cables and show the normal polarized distribution of cortical patches (Fig 6). In contrast, the pfy1-111/mid2 and the pfy1-111/rho2 cells are larger than either single mutant, and actin patches are now completely depolarized compared to the pfy1-111 strain (Table 5, Fig 6). We attempted to obtain a pfy1-111/smy1 strain by sporulation and dissection of a pfy1-111, smy1 heterozygous diploid strain. A total of 80 tetrads originating from two different sporulations were dissected and incubated at 26° or 30°. No pfy1-111/smy1 double mutant was obtained, suggesting a synthetic lethal interaction between these two genes.

View this table: In this window In a new window   Table 5. Redistribution of cortical patches in the mutant strains pfy1-111/mid2 and pfy1-111/rho2 by Pfy1p and five of the suppressors

We overexpressed the suppressors in the pfy1-111/mid2 and pfy1-111/rho2 strains as a means for determining whether suppression by a given protein requires the presence of either Mid2p or Rho2p. In the pfy1-111/mid2 strain, overexpression of Pfy1p, Mid2p, Rho2p, or Rom2p all repolarize cortical patches to small buds (Table 5). However, overexpression of Smy1p does not correct the actin distribution or the growth defects of the pfy1-111/mid2 strain, indicating that Smy1p requires the presence of Mid2p for suppression. These results are compatible with a model placing Mid2p, Rom2, and Rho2 on the same signaling pathway. The results with Smy1p are more complex and do not allow us to include it in a model (see DISCUSSION).

In contrast, the results with the overexpression of Syp1p in the pfy1-111/mid2 cells are ambiguous. Approximately half the cells are large and round with nonpolarized cortical patches while the other half are small cells with polarized cortical patches. These two populations may be the result of varying numbers of plasmids and consequently the amount of Syp1p in the cells. These results suggest that Syp1p can function independently of Mid2p but that it is a more efficient suppressor in the presence of Mid2p.

In the pfy1-111/rho2 strain, the overexpression of Mid2p, Syp1p, or Smy1p does not repolarize cortical patches and only the overexpression of Pfy1p or Rho2p restores cortical patch polarity (Table 5). Overexpression of Rom2p partly corrects the distribution of cortical patches, but not as well as it does in cells with a functional Rho2p. Collectively these results again suggest a signaling pathway in which Mid2p signals through Rho2p to modify the actin cytoskeleton. They also indicate that Smy1p and Syp1p require Rho2p to repolarize cortical patches in pfy1 cells.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have identified a novel gene as a suppressor of pfy1 cells that we named SYP1. Overexpression of Syp1p in haploid wild-type cells causes the formation of a long projection in a fraction of the cells. In diploid cells, a small protuberance forms at the opposite end from the growing bud. However, no morphological or growth phenotype was observed following deletion of SYP1. In particular, there is no effect on the budding pattern of either haploid or diploid cells. Localization of Syp1p resembles that of the actin-binding protein Aip3p/Bud6p, which is also found in the bud tip and in the mother-bud neck (AMBERG et al. 1997 ). Both proteins have a putative coiled-coil domain and Aip3p has been shown to form homodimers. Apart from the coiled-coil domain, there is no obvious sequence homology between the two proteins. Unlike the deletion of SYP1, the deletion of AIP3 leads to a random budding pattern in diploid cells (AMBERG et al. 1997 ). Mutants of several other proteins of the actin cytoskeleton, including act1 mutants, specifically affect the budding pattern of diploid cells (ZAHNER et al. 1996 ; YANG et al. 1997 ). This implicates the actin cytoskeleton and proteins like Aip3p in localizing the positional signals for bipolar budding. Syp1p does not appear to be involved in this process.

All of the suppressors restore the polarized distribution of cortical patches in profilin-deficient cells, but none restore visible actin cables. This indicates that profilin is essential only for the polymerization of actin filaments that form actin cables. There is evidence that actin cables are responsible for polarization of cortical patches. Tropomyosin binds to and stabilizes actin cables. There are two tropomyosins in yeast, the major form, Tmp1p (LIU and BRETSCHER 1989 ), and the minor form, Tpm2p (DREES et al. 1995 ). Deletion of TPM1 causes the apparent loss of actin cables (LIU and BRETSCHER 1989 ) and deletion of both tropomyosins is lethal (DREES et al. 1995 ). A detailed study of the temperature-sensitive strain tpm1-2/tpm2 shows that at the permissive temperature these cells contain abundant actin cables (PRUYNE et al. 1998 ). Shifting cells to the restrictive temperature leads to a rapid disappearance of cables and subsequent depolarization of cortical patches. Reincubation at the permissive temperature causes the reappearance of cables oriented toward growing buds and a gradual repolarization of cortical patches. This suggests a role for actin cables in the establishment of polarity in S. cerevisiae cells.

Studies with the actin depolymerizing drug latrunculin-A (Lat-A) support this conclusion. Lat-A binds to actin monomers and inhibits their polymerization without directly causing depolymerization of actin filaments. Incubating yeast cells in the presence of Lat-A results in the disappearance of both actin cables and cortical patches (AYSCOUGH et al. 1997 ). Actin cables are the first to depolymerize, and, as a result, cortical patches become depolarized before they are also depolymerized. Therefore, delocalization of cortical patches is in part due to the absence of actin cables.

Most mutants of actin-binding proteins, such as cap2 and srv2, have depolarized cortical patches but still retain actin cables (KARPOVA et al. 1998 ). There must therefore exist another mechanism for establishing cell polarity and polarization of cortical patches distinct from actin cables. The results presented here suggest that the Rho2p pathway is implicated in this process. In the absence of profilin, actin cables disappear, leading to the depolarization of the cortical patches. The suppressors that we isolated can bypass the need for actin cables and somehow reestablish the positional signals needed to establish cell polarity.

The membrane protein Mid2p, which is a sensor for cell wall integrity, is a suppressor of the profilin-deficient phenotype. Overexpression of this protein increases chitin synthesis and probably other cell wall components, at least in part through the activation of Rho1p and the PKC1-MPK1 pathway (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). Our electron micrographic studies on the formation of a thick cell wall following overexpression of Mid2p and Rho1p are in agreement with a role for these proteins in cell wall biosynthesis.

Although they were not selected as suppressors in our screen, we also tested directly two additional plasma membrane proteins that are involved in cell wall integrity. One of the proteins is Mtl1p for MID2-like, which has 50% sequence identity to Mid2p. The high-copy-number expression of MTL1 from its own promoter did not allow pfy1 cells to grow in the presence of 1.25 mg/ml caffeine. These cells grow slowly at 30° and at 37°, with a doubling time of 6 hr, and they are much larger than pfy1 cells overexpressing Mid2p (results not shown). We conclude from these results that Mtl1p only partially suppresses the pfy1 phenotype.

Wsc1p, also called Slg1p or Hcs77p, is another plasma membrane protein that shows partial overlapping functions with Mid2p. The wsc1 cells undergo cell lysis at high temperatures, a phenotype that is corrected by the overexpression of Mid2p. The mating-induced death phenotype seen after the treatment of a MATa mid2 mutant with -factor is partially corrected by the overexpression of Wsc1p. Finally, cells carrying either the mid2 or the wsc1 single mutations are viable at room temperature while the double mutant is inviable (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). The pfy1 strain DJP102 overexpressing Wsc1p from its own promoter grows at 37° and in the presence of 1.25 mg/ml caffeine at 30°. They are smaller than the pfy1 cells and are morphologically similar in the light microscope to pfy1 cells overexpressing Mid2p (results not shown). On the basis of these findings, we conclude that Wsc1p is a multicopy suppressor of the profilin-deficient phenotype (Table 2).

Although Mid2p and Wsc1p signal through Rho1p, this protein is not a suppressor of the pfy1 phenotype. Rho1p mediates bud growth by controlling polarization of the actin cytoskeleton and cell wall synthesis (DRGONOVA et al. 1999 ). The downstream effectors of Rho1p are Skn7p (ALBERTS et al. 1998 ), a two-component signaling protein, Pkc1p (NONAKA et al. 1995 ; KAMADA et al. 1996 ) and Fks1p (ß-1,3-glucan synthase; DRGONOVA et al. 1996 ; QADOTA et al. 1996 ), and Bni1p, a member of the formin family (KOHNO et al. 1996 ). It is postulated that Rho1p controls the actin cytoskeleton via its interaction with Bni1p, which binds to profilin through its formin-homology (FH1) domain (EVANGELISTA et al. 1997 ; IMAMURA et al. 1997 ). Presumably, this locates the profilactin complex to the growing bud, where it can be used for polymerization of F-actin. In the pfy1 strain used in this study, the profilactin complex does not exist, and it cannot be recruited to regions of active growth by Bni1p. It is therefore reasonable the RHO1 is not a multicopy suppressor of the profilin-deficient phenotype.

Rho2p is the only yeast Rho family member that can repolarize the actin cortical patches in the pfy1 strain. The two exchange factors for Rho2p, Rom1p and Rom2p, are also suppressors of the profilin-deficient phenotype, further supporting the role for the Rho2p signaling pathway in actin cortical patch polarization. We therefore propose a model in which Mid2p, and possibly Wsc1p, act through the exchange factors Rom1p and Rom2p. These proteins would then activate the Rho1p signaling pathway to stimulate cell wall synthesis and would activate Rho2p to repolarize the actin cytoskeleton (Fig 7).

View larger version (14K): In this window In a new window Download PPT slide   Figure 7. Proposed model for the organization of the actin cytoskeleton by the Rho1p and Rho2p signaling pathways. Mid2p and Syp1p are postulated to activate Rho1p and Rho2p through their exchange factors Rom1p and Rom2p, either directly or through an intermediate.

The overexpression of Syp1p leads to a thick cell wall, suggesting that this protein is upstream of Rom1/Rom2p (Fig 7). This is further strengthened by the finding that Syp1p does not correct the phenotype of the pfy1-111/rho2 mutant phenotype. A thick cell wall is produced by the overexpression of Mid2p in syp1 cells suggesting that Mid2p does not signal through Syp1p and that Mid2p and Syp1p may be on parallel pathways. In support of this hypothesis, deletion of SYP1 in the pfy1-111 strain does not exaggerate the profilin mutant phenotype. In contrast, cortical patches of pfy1-111/mid2 cells are completely depolarized, and the cells are larger than the pfy1-111 cells.

The overexpression of Rho2p corrects the depolarized actin cortical patch phenotype, but it does not affect the diameter of the yeast cell wall. This is in agreement with Rho2p acting downstream of Rom1p and Rom2p. None of the suppressors, except for Rho2p, is able to fully correct the distribution of the actin cortical patches in a pfy1-111/rho2 strain, suggesting that all the suppressors tested act through the Rho2p signaling pathway or that they are dependent on the presence of Rho2p.

It is difficult, however, to position Smy1p precisely in the signaling pathway. Smy1p is localized in the cortical patch region of budding cells and is necessary for the proper localization of Myo2p, which is located in the same region (LILLIE and BROWN 1994 ). The amino acid sequence of Smy1p suggests that it is a kinesin-like protein, but microtubular organization in smy1 cells is normal and disruption of microtubules does not affect the localization of Smy1p. The smy1 strain shows synthetic lethality with the myo2-66 mutation but not with the yeast kinesin genes, suggesting a role for Smy1p in the organization of the actin cytoskeleton (LILLIE and BROWN 1992 , LILLIE and BROWN 1998 ).

The presence of a normal cell wall in strains overexpressing Smy1p argues that it is downstream of Rom1p/Rom2p or on a parallel pathway. The overexpression of Smy1p, however, does not correct the pfy1-111/rho2 mutant phenotype, suggesting that it is not downstream of Rho2p. Finally, the pfy1-111/rho2 double mutant is viable and the pfy1-111/smy1 is inviable. These results suggest a complex interaction between Smy1p and Rho2p that will require further study to resolve.

Evidence for a pathway leading from Mid2p to Rho2p via the exchange factor Rom2p was also obtained in a study of the microtubule cytoskeleton. The rho2 mutant shows no morphological or growth defects except for an increased sensitivity to benomyl, a microtubule depolymerizing drug (MANNING et al. 1997 ). Cik1p, for chromosome instability and karyogamy, is located in the spindle pole body and is essential for proper spindle organization. The cik1 mutant show defects in karyogamy at 30° and is inviable at 37° (PAGE and SNYDER 1992 ; PAGE et al. 1994 ). Cik1p interacts physically with Kar3p, which is a kinesin-like motor protein involved in nuclear migration during karyogamy (ROSE 1996 ). The kar3 mutant has a temperature-sensitive growth defect. Overexpression of Mid2p, Rho2p, or Rom2p allows both the cik1 and kar3 mutants to grow at the restrictive temperature (MANNING et al. 1997 ).

Genetic interactions have also implicated Rho2p in the organization of the actin cytoskeleton. Tor2p, a phosphatidylinositol kinase homologue, plays a role in cell cycle control and the organization of the actin cytoskeleton by activating Rom2p (SCHMIDT et al. 1997 ). Class A tor2-ts mutants, which have actin cytoskeleton defects, can be suppressed by the overexpression of several proteins, including Mss4p, a phosphatidylinositol-4-phosphate 5-kinase, Rom2p and Rho2p (HELLIWELL et al. 1998 ). The overexpression of Rho2p can also rescue the lethality and the actin cytoskeleton defects of a mss4-ts mutations (DESRIVIERES et al. 1998 ). In addition, the mss4-1 mutant is synthetically lethal with the profilin mutant cls5-1. It is noteworthy that the mss4-1 mutant, with its large cell size and the random distribution of cortical actin patches and random chitin localization, has a phenotype resembling that of the pfy1 cells (HOMMA et al. 1998 ).

Additional evidence implicating Rho2p in the organization of the actin cytoskeleton is seen in the work on GLC7, which encodes the catalytic subunit of a type 1 serine/threonine phosphatase. It controls glycogen accumulation by the dephosphorylation of glycogen synthase (CANNON et al. 1994 ; RAMASWAMY et al. 1998 ). A temperature-sensitive glc7 mutant has aberrant bud morphology, delocalized cortical actin patches, and few actin cables. These defects are suppressed by the over-expression of the protein kinase C homologue Pkc1p, but not by the components of the downstream mitogen-activated protein kinase cascade. Interestingly, Mid2p, Rho2p, Rom2p, and Wsc1p are also suppressors (ANDREWS and STARK 2000 ). We point out that these proteins are also suppressors of the pfy1 phenotype. Thus, Rho2p and other proteins affecting Rho2p signaling are involved in the control of the actin cytoskeleton. The eventual identification of downstream effectors of the Rho2p pathway should help determine how this is accomplished.

	ACKNOWLEDGMENTS

We thank Howard Bussey, Brian Haarer, Michael Hall, David Levin, and Michael Snyder for strains and plasmids. This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Fonds pour la formation des Chercheurs et l'Aide à la Recherche of Québec.

Manuscript received December 20, 1999; Accepted for publication June 26, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALBERTS, A. S., N. BOUQUIN, L. H. JOHNSTON, and R. TREISMAN, 1998  Analysis of RhoA-binding proteins reveals an interaction domain conserved in heterotrimeric G protein beta subunits and the yeast response regulator protein Skn7. J. Biol. Chem. 273:8616-8622[Abstract/Free Full Text].

AMBERG, D. C., J. E. ZAHNER, J. W. MULHOLLAND, J. R. PRINGLE, and D. BOTSTEIN, 1997  Aip3p/Bud6p, a yeast actin-interacting protein that is involved in morphogenesis and the selection of bipolar budding sites. Mol. Biol. Cell 8:729-753[Abstract].

ANDREWS, P. D. and M. J. STARK, 2000  Type 1 protein phosphatase is required for maintenance of cell wall integrity, morphogenesis and cell cycle progression in Saccharomyces cerevisiae.. J. Cell Sci. 113:507-520[Abstract].

AYSCOUGH, K. R., J. STRYKER, N. POKALA, M. SANDERS, and P. CREWS et al., 1997  High rates of actin filament turnover in budding yeast and roles for actin in establishment and maintenance of cell polarity revealed using the actin inhibitor latrunculin-A. J. Cell Biol. 137:399-416[Abstract/Free Full Text].

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

CANNON, J. F., J. R. PRINGLE, A. FIECHTER, and M. KHALIL, 1994  Characterization of glycogen-deficient glc mutants of Saccharomyces cerevisiae.. Genetics 136:485-503[Abstract].

CARLSON, M. and D. BOTSTEIN, 1982  Two differentially regulated mRNAs with different 5' ends encode secreted with intracellular forms of yeast invertase. Cell 28:145-154[Medline].

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].

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[Medline].

DANIEL, J., 1993  Potentially rapid walking in cellular regulatory networks using the gene-gene interference method in yeast. Mol. Gen. Genet. 240:245-257[Medline].

DESRIVIERES, S., F. T. COOKE, P. J. PARKER, and M. N. HALL, 1998  MSS4, a phosphatidylinositol-4-phosphate 5-kinase required for organization of the actin cytoskeleton in Saccharomyces cerevisiae.. J. Biol. Chem. 273:15787-15793[Abstract/Free Full Text].

DREES, B., C. BROWN, B. G. BARRELL, and A. BRETSCHER, 1995  Tropomyosin is essential in yeast, yet the TPM1 and TPM2 products perform distinct functions. J. Cell Biol. 128:383-392[Abstract/Free Full Text].

DRGONOVA, J., T. DRGON, K. TANAKA, R. KOLLAR, and G. C. CHEN et al., 1996  Rho1p, a yeast protein at the interface between cell polarization and morphogenesis. Science 272:277-279[Abstract].

DRGONOVA, J., T. DRGON, D. H. ROH, and E. CABIB, 1999  The GTP-binding protein Rho1p is required for cell cycle progression and polarization of the yeast cell. J. Cell Biol. 146:373-387[Abstract/Free Full Text].

EVANGELISTA, M., K. BLUNDELL, M. LONGTINE, C. CHOW, and N. ADAMES et al., 1997  Bni1p, a yeast formin linking Cdc42p and the actin cytoskeleton during polarized morphogenesis. Science 276:118-122[Abstract/Free Full Text].

GOLDSCHMIDT-CLERMONT, P. J., L. M. MACHESKY, J. J. BALDASSARE, and T. D. POLLARD, 1990  The actin-binding protein profilin binds to PIP2 and inhibits its hydrolysis by phospholipase C. Science 247:1575-1578[Abstract/Free Full Text].

GRAY, J. V., J. P. OGAS, Y. KAMADA, M. STONE, and D. E. LEVIN et al., 1997  A role for the Pkc1 MAP kinase pathway of Saccharomyces cerevisiae in bud emergence and identification of a putative upstream regulator. EMBO J. 16:4924-4937[Medline].

HAARER, B. K., S. H. LILLIE, A. E. ADAMS, V. MAGDOLEN, and W. BANDLOW et al., 1990  Purification of profilin from Saccharomyces cerevisiae and analysis of profilin-deficient cells. J. Cell Biol. 110:105-114[Abstract/Free Full Text].

HAARER, B. K., A. S. PETZOLD, and S. S. BROWN, 1993  Mutational analysis of yeast profilin. Mol. Cell. Biol. 13:7864-7873[Abstract/Free Full Text].

HELLIWELL, S. B., I. HOWALD, N. BARBET, and M. N. HALL, 1998  TOR2 is part of two related signaling pathways coordinating cell growth in Saccharomyces cerevisiae.. Genetics 148:99-112[Abstract/Free Full Text].

HOMMA, K., S. TERUI, M. MINEMURA, H. QADOTA, and Y. ANRAKU et al., 1998  Phosphatidylinositol-4-phosphate 5-kinase localized on the plasma membrane is essential for yeast cell morphogenesis. J. Biol. Chem. 273:15779-15786[Abstract/Free Full Text].

IIDA, H., H. NAKAMURA, T. ONO, M. S. OKUMURA, and Y. ANRAKU et al., 1994  MID1, a novel Saccharomyces cerevisiae gene encoding a plasma membrane protein, is required for Ca2+ influx and mating. Mol. Cell. Biol. 14:8259-8271[Abstract/Free Full Text].

IMAMURA, H., K. TANAKA, T. HIHARA, M. UMIKAWA, and T. KAMEI et al., 1997  Bni1p and Bnr1p: downstream targets of the Rho family small G-proteins which interact with profilin and regulate actin cytoskeleton in Saccharomyces cerevisiae.. EMBO J. 16:2745-2755[Medline].

JACOBY, J. J., S. M. NILIUS, and J. J. HEINISCH, 1998  A screen for upstream components of the yeast protein kinase C signal transduction pathway identifies the product of the SLG1 gene. Mol. Gen. Genet. 258:148-155[Medline].

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

KAMADA, Y., H. QADOTA, C. P. PYTHON, Y. ANRAKU, and Y. OHYA et al., 1996  Activation of yeast protein kinase C by Rho1 GTPase. J. Biol. Chem. 271:9193-9196[Abstract/Free Full Text].

KARPOVA, T. S., J. G. MCNALLY, S. L. MOLTZ, and J. A. COOPER, 1998  Assembly and function of the actin cytoskeleton of yeast: relationships between cables and patches. J. Cell Biol. 142:1501-1517[Abstract/Free Full Text].

KETELA, T., R. GREEN, and H. BUSSEY, 1999  Saccharomyces cerevisiae mid2p is a potential cell wall stress sensor and upstream activator of the PKC1-MPK1 cell integrity pathway. J. Bacteriol. 181:3330-03340