Dissection of Upstream Regulatory Components of the Rho1p Effector, 1,3-ß-Glucan Synthase, in Saccharomyces cerevisiae

Corresponding author: Yoshikazu Ohya, Graduate School of Frontier Science, University of Tokyo, Bldg. FSB-101, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8562, Japan., ohya{at}k.u-tokyo.ac.jp (E-mail)

Communicating editor: M. D. ROSE

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In the budding yeast Saccharomyces cerevisiae, one of the main structural components of the cell wall is 1,3-ß-glucan produced by 1,3-ß-glucan synthase (GS). Yeast GS is composed of a putative catalytic subunit encoded by FKS1 and FKS2 and a regulatory subunit encoded by RHO1. A combination of amino acid alterations in the putative catalytic domain of Fks1p was found to result in a loss of the catalytic activity. To identify upstream regulators of 1,3-ß-glucan synthesis, we isolated multicopy suppressors of the GS mutation. We demonstrate that all of the multicopy suppressors obtained (WSC1, WSC3, MTL1, ROM2, LRE1, ZDS1, and MSB1) and the constitutively active RHO1 mutations tested restore 1,3-ß-glucan synthesis in the GS mutant. A deletion of either ROM2 or WSC1 leads to a significant defect of 1,3-ß-glucan synthesis. Analyses of the degree of Mpk1p phosphorylation revealed that among the multicopy suppressors, WSC1, ROM2, LRE1, MSB1, and MTL1 act positively on the Pkc1p-MAPK pathway, another signaling pathway regulated by Rho1p, while WSC3 and ZDS1 do not. We have also found that MID2 acts positively on Pkc1p without affecting 1,3-ß-glucan synthesis. These results suggest that distinct networks regulate the two effector proteins of Rho1p, Fks1p and Pkc1p.

THE cell wall is a fundamentally rigid structure that defines the shape of plant and fungal cells. During morphogenesis of vegetatively growing cells of the budding yeast Saccharomyces cerevisiae, cell wall materials are incorporated into the restricted cell surface domain active for cell wall remodeling (CID et al. 1995 ). Therefore, it is conceivable that numerous signals representing spatial and temporal information are integrated into cell wall synthesis during the budding process. Although a number of proteins essential for establishment of cell polarity have been identified (PRUYNE and BRETSCHER 2000A , PRUYNE and BRETSCHER 2000B ), little is known about how cell polarity and spatial activation of cell wall synthesis are related.

The structural components of the yeast cell wall are 1,3-ß-glucan polymer with some branches of 1,6-ß-linkages, chitin, and mannoproteins. Among them, 1,3-ß-glucan is the main component responsible for the rigidity of yeast cells. 1,3-ß-glucan synthase (GS) is composed of a putative catalytic subunit and a regulatory subunit. The putative catalytic subunit is an integral membrane protein encoded by the two closely related genes, Fks1p and Fks2p (DOUGLAS et al. 1994 ; INOUE et al. 1995 ; MAZUR et al. 1995 ). Double deletions of FKS1 and FKS2 are inviable (INOUE et al. 1995 ; MAZUR et al. 1995 ), indicating functional redundancy between Fks1p and Fks2p. The regulatory subunit is copurified with GS and encoded by RHO1 (DRGONOVA et al. 1996 ; MAZUR and BAGINSKY 1996 ; QADOTA et al. 1996 ). Rho1p acts as a molecular switch and specifically activates its effectors in its GTP-bound form (CABIB et al. 1998 ). Thus, Rho1p likely has the ability to monitor and to receive upstream signals that direct cell wall remodeling and cell morphogenesis.

In addition to Fks1p/Fks2p, four effector molecules of yeast Rho1p have thus far been known. Rho1p binds and activates Pkc1p, which in turn stimulates the Pkc1p-mitogen-activated protein kinase cascade (Pkc1p-Mpk1p pathway) to organize actin cytoskeleton and to maintain cell integrity (PARAVICINI et al. 1992 ; NONAKA et al. 1995 ; KAMADA et al. 1996 ; HELLIWELL et al. 1998 ). Rho1p also interacts with Bni1p, a yeast homolog of formin (KOHNO et al. 1996 ). Bni1p might regulate actin cytoskeleton through interaction with an actin-binding protein, profilin (EVANGELISTA et al. 1997 ; IMAMURA et al. 1997 ). Rho1p also binds to Skn7p (ALBERTS et al. 1998 ), which is thought to function in oxidative stress response (MORGAN et al. 1997 ), in transcription during G1-S phase (BOUQUIN et al. 1999 ), and in the two-component regulatory pathway (BROWN et al. 1994 ). Furthermore, the active form of Rho1p interacts with Sec3p, a spatial landmark for polarized exocytosis, and Rho1p is required for the establishment and maintenance of the Sec3p localization (GUO et al. 2001 ).

Screens of components that physically and genetically interact with Rho1p have led to the identification of various Rho1p activators in yeast: Multicopy suppressor screening using a conditional rho1 mutant identified Rom1p and Rom2p, GDP-GTP exchange factors (GEF) of Rho1p (OZAKI et al. 1996 ). Two-hybrid screening using an RHO1 fragment identified Bem4p, which positively affects Rho1p function by a mechanism other than promotion of GDP-to-GTP exchange of Rho1p (HIRANO et al. 1996 ). WSC family genes and the functionally related MID2 and MTL1 genes were genetically placed upstream of RHO1 and the Pkc1p-mitogen-activated protein kinase (MAPK) cascade (GRAY et al. 1997 ; VERNA et al. 1997 ; JACOBY et al. 1998 ; KETELA et al. 1999 ; RAJAVEL et al. 1999 ). Both Wsc1p and Mid2p interact with Rom2p and stimulate nucleotide exchange activity toward Rho1p (PHILIP and LEVIN 2001 ). Although these factors have been shown to act upstream of the Pkc1p-MAPK cascade (KAMADA et al. 1995 ; VERNA et al. 1997 ; MARTIN et al. 2000 ; PHILIP and LEVIN 2001 ), the roles of Rho1p regulators on 1,3-ß-glucan synthesis remain unknown.

In this study, we have taken a genetic approach to isolate factors involved in 1,3-ß-glucan synthesis. We performed a high-copy suppressor screen using fks1-1154 fks2, a temperature-sensitive mutant of GS (simply referred to as fks1-1154 below). By detecting in vivo glucan synthesis with aniline blue, a specific fluorescent dye for 1,3-ß-glucan, and by incorporating labeled glucose into cell wall 1,3-ß-glucan, we showed that all the suppressors act positively on 1,3-ß-glucan synthesis. In addition, we examined whether the suppressors act positively on the Pkc1p-MAPK cascade. On the basis of these results we present a model of the molecular network upstream of Fks1p and Pkc1p.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Media, growth conditions, and genetic manipulation:
Tetrad analysis and mating-type determination were performed as described previously (KAISER et al. 1994 ). Yeast transformation was carried out using the lithium acetate method (ITO et al. 1983 ). Yeast cells were grown in rich medium (YPD) [1% Bacto yeast extract (Difco, Detroit), 2% polypeptone (Nihon Seiyaku, Tokyo), and 2% glucose (Wako Chemicals, Osaka, Japan)], or in synthetic growth medium (SD) [0.67% yeast nitrogen base (Difco) and 2% glucose] appropriately supplemented. To examine a phenotype of the cells with a reduced GS activity, we grew yeast cells in a 0.2% glucose medium [1% Bacto yeast extract, 2% polypeptone, and 0.2% glucose]. Sensitivity to echinocandin B and Calcofluor white were examined as previously (RAM et al. 1994 ; INOUE et al. 1995 ). For -Ura or -Trp selection, 0.5% casamino acid (Difco, Detroit) was added to SD. Standard procedures were used for all DNA manipulations and Escherichia coli transformation (SAMBROOK et al. 1989 ).

Strains:
The yeast strains used are listed in Table 1. All the strains are isogenic derivatives of YPH499, YPH500, or YPH501. To construct YOC748 (msb1), we replaced the 3.3-kb EcoRI-to-SphI fragment inside the open reading frame (ORF) of MSB1 with a DNA fragment containing the HIS3 gene (from pJJ217). YOC2521 (wsc3), YOC2526 (mid2 mtl1), YOC2573 (wsc1), YOC2576 (rom2), YOC2579 (lre1), and YOC2581 (zds1) were constructed by PCR-mediated gene disruptions described previously (SAKUMOTO et al. 1999 ): Primers were used to amplify the HIS3, LEU2, or TRP1 gene of Candida glabrata together with flanking sequences derived from the upstream and downstream regions of the corresponding genes. Details of YOC1001 (FKS1 fks2) and YOC1087 (fks1-1154 fks2) constructions will be described elsewhere (M. ABE, M. MINEMURA-ASAKAWA, T. UTSUGI, M. SEKIYA-KAWASAKI, A. HIRATA, H. QADOTA, K. MORISHITA, T. WATANABE and Y. OHYA, unpublished results). Briefly, YOC1087 is a temperature-sensitive strain with the FKS1 and FKS2 genes deleted and a mutant allele of fks1 (fks1-1154) integrated at the ADE3 locus. YOC1001 is a wild-type control of YOC1087 with FKS1 and FKS2 deleted and the wild-type FKS1 gene integrated at the ADE3 locus.

Plasmids:
The plasmids used in this study are listed in Table 2. A YEp13-based genomic library was described previously (YOSHIHISA and ANRAKU 1989 ).

• pYO971: The KpnI-SacI fragment of pRS316 was replaced with a KpnI-NheI-SphI-SacI linker.

  
  

  View this table: In this window In a new window   Table 2. Plasmids used in this study

• pYO326NS: Using a linker, NheI and SphI sites were inserted adjacent to the SmaI site of pYO326.

• pYO964 and pYO965: RHO1 mutations (G19V and Q68L) were introduced by two-step PCR mutagenesis (HO et al. 1989 ). The mutated RHO1 open reading frames were inserted into pYO701, which consists of pRS314, the 0.2-kb RHO1 promoter, and the 0.1-kb RHO1 terminator (QADOTA et al. 1994 ). To construct pYO964 and pYO965, we cloned the mutant RHO1 genes into pRS316.

• pYO2358: 7.8-kb SphI-NheI fragment including the entire ORF of FKS1 from pGSYE1 (INOUE et al. 1995 ) was inserted into pYO326NS.

• pYO2325: A genomic fragment containing the MTL1 ORF, plus 1 kbp upstream and 500 bp downstream, was amplified from the S. cerevisiae genome by high-fidelity PCR using primers that generate a SalI site at the 5'-end. The PCR product was cloned into pYO326. The DNA sequence of the cloned MTL1 gene was verified.

• pYO2361: A 1.9-kb HpaI-XhoI fragment including MID2 (derived from the YEp13 genomic bank) was cloned into pYO326.

• pYO2364: A 2.7-kb BglII-BglII fragment including WSC1 (derived from the YEp13 genomic bank) was cloned into pYO326.

• pYO2366: A 6.5-kb HpaI-HpaI fragment including ROM2 (derived from the YEp13 genomic bank) was cloned into pYO326.

• pYO2368: A 2.5-kb EcoRV-EcoRV fragment including LRE1 (derived from the YEp13 genomic bank) was cloned into pYO326.

• pYO2369: A 3.4-kb EcoRI-SphI fragment including WSC3 ORF, plus 0.9 kb upstream and 0.8 kb downstream (derived from the YEp13 genomic bank), were cloned into pYO326. The SphI site is on the YEp13 vector.

• pYO2370: A 3.7-kb EcoRI-ClaI fragment including ZDS1 ORF, plus 0.3 kb upstream and 0.6 kb downstream (derived from the YEp13 genomic bank), were cloned into pYO326. The ClaI site is on the YEp13 vector.

• pYO2371: A 5.5-kb SalI-BamHI fragment including MSB1 (derived from the YEp13 genomic bank) was cloned into pYO326.

Multicopy suppressor screening:
YOC1087 (fks1-1154) was transformed with a 2µ-based genomic library with a 5- to 10-kb insert for each clone. Out of 40,000 colonies, transformants growing at 37° were isolated. Then we rescued the plasmids responsible for suppression and retransformed the parental strain with them to recheck suppression. The inserts were directly sequenced and the clones containing FKS1 (6 clones) and FKS2 (15 clones) were excluded from further investigation. We carried out subcloning using standard techniques to determine which ORFs are responsible for suppression. Finally, we identified six genes (WSC1, WSC3, ROM2, LRE1, ZDS1, and MSB1) that are responsible for suppression when expressed on a multicopy vector. We independently isolated 3 WSC1 clones, 8 WSC3 clones, 2 ROM2 clones, 2 LRE1 clones, 4 ZDS1 clones, and 3 MSB1 clones.

Morphological observations:
For staining of yeast cells with aniline blue, early log-phase cells (1 x 10⁷ cells) were harvested by low-speed centrifugation (2000 rpm, 3 min). The cells were washed twice with PBS and mildly sonicated for 10–20 sec. The washed and sonicated cells were incubated with 0.05% aniline blue (Wako) for 5 min and then observed under the fluorescence microscope using the Olympus U-MNV DM455 filter set (excitation wavelength, 400–410 nm; emission wavelength, 455 nm). Procedures for immunofluorescence microscopy were as described previously (PRINGLE et al. 1989 ). For Fks1p staining, a mouse monoclonal antibody to Fks1p (T2B8; INOUE et al. 1995 ) was selected as a primary antibody and a FITC-conjugated goat anti-mouse antibody was used as a secondary antibody at a 1:50 dilution. Cells were viewed on an Olympus BX-FLA microscope (Olympus, Tokyo). Images were captured using a CCD camera (Olympus MicroMax/OL) and Metamorph Imaging software (Universal Imaging, West Chester, PA). All the images presented were processed using Adobe PhotoShop software.

Incorporation of [¹⁴C]glucose into 1,3-ß-glucan:
Labeling of 1,3-ß-glucan was carried out as described previously (HONG et al. 1994 ) with some modifications. Cells were grown to early log phase at 25° and shifted to appropriate temperature for 2 hr. Cells were diluted to OD₆₀₀ of 0.5 with 1 ml of low-glucose medium [1% Bacto yeast extract, 2% polypeptone, and 0.5% glucose] or 0.67% yeast nitrogen base and 0.5% glucose containing 10 µCi of [¹⁴C]glucose and labeled for 2 hr. Labeled cells were harvested at 4° and extracted with 1 N NaOH at 75° for 1 hr. The insoluble fraction was collected and washed two times with distilled water (DW). The insoluble pellet was resuspended in 300 µl of 10 mM Tris/HCl, pH 7.5, containing 5 mg/ml zymolyase 100T (Seikagaku) and incubated at 37° for 20 hr. After digestion, the zymolyase-resistant material was removed by centrifugation (15,000 x g for 20 min). The supernatant was filtered through a microcon-10 membrane (Amicon Bioseparations, Beverly, MA). The flow-through fraction was dried with a vacuum evaporator, resuspended in 20 µl of DW, and applied to Unifilter-GF/C (Packard Instrument, Meriden, CT). The difference in the incorporation rate for the strain was normalized by OD₆₀₀ that was determined before and after the labeling period.

Quantitative 1,3-ß-glucan measurements:
Amount of 1,3-ß-glucan per cell was measured using aniline blue as described previously (WATANABE et al. 2001 ) with some modifications. Cells were grown to early log phase and 2.5 x 10⁶ cells were harvested. The cells were washed twice with TE and resuspended in 500 µl TE. To the cells 6 N NaOH was added to a final concentration of 1 N, floated in a water bath at 80° for 30 min followed by addition of 2.1 ml of AB mix [0.03% aniline blue (Wako), 0.18 N HCl, and 0.49 M glycine/NaOH, pH 9.5]. The tube was vortexed briefly and then incubated at 50° for 30 min and an additional 30 min at room temperature to allow reaction with the fluorochrome and decolorization of the aniline blue. Fluorescence of 1,3-ß-glucan was quantified using a spectrofluorophotometer (RF-5300PC, Shimadzu, Kyoto, Japan). Excitation wavelength was 400 nm/slit, width 3 nm, and emission wavelength was 460 nm/slit, width 3 nm.

Assay of in vitro GS activity:
The membrane fraction was prepared as described previously (ABE et al. 2001 ). In brief, log-phase cultures were resuspended in 1 mM EDTA and 500 mM NaCl containing 1 mM phenylmethylsulfonyl fluoride and lysed with glass beads. The crude lysate was centrifuged at 1500 x g for 5 min so that cell debris and unbroken cells are separated. After a centrifugation at 100,000 x g for 30 min, the pellet was suspended in a buffer containing 50 mM Tris-HCl (pH 7.5), 1 mM EDTA, and 33% glycerol, and the suspension was used as the membrane fraction. GS activity was measured according to the procedure described previously (INOUE et al. 1995 ). As described (INOUE et al. 1995 ), the assay buffer (composition of the assay buffer) contains an excess amount of GTPS.

Immunoblot analysis of Mpk1p:
Cells were grown to the log phase at 25° in SD-Ura medium and shifted to 39° for 2 hr. The cell extracts were prepared as described (MARTIN et al. 2000 ). Protein samples (50 µg) were loaded on SDS-polyacrylamide gels, transferred to PVDF membranes (Amersham Bioscience, catalog no. RPN2020F), and blotted with either an anti-phospho-p44/42 MAPK (Thr²⁰²/Tyr²⁰⁴) antibody (catalog no. 9101, New England Biolabs, Beverly, MA) or an anti-MAPK antibody (catalog no. sc-6802, Santa Cruz Biotechnology). After reacting with ECL-Plus (Amersham Bioscience), the amount of each protein was quantified with LAS-1000plus (Fuji Photo Film).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

1,3-ß-Glucan synthesis is defective in fks1-1154 fks2 cells:
A temperature-sensitive mutant, fks1-1154 (K877N, A899S, Q977P) fks2 (referred to as fks1-1154 below), was isolated in a genetic study of GS function in S. cerevisiae (M. ABE, M. MINEMURA-ASAKAWA, T. UTSUGI, M. SEKIYA-KAWASAKI, A. HIRATA, H. QADOTA, K. MORISHITA, T. WATANABE and Y. OHYA, unpublished results). The fks1-1154 mutant was able to grow at 25°, but failed to grow >35°. The mutation sites of fks1-1154 are located within the hydrophilic sequences representing the putative catalytic domain (KELLY et al. 1996 ). As expected, the membrane fraction of fks1-1154 cells was virtually devoid of GS activity (Fig 1A). A kinetic analysis of GS activity of the membrane fraction revealed that the fks1-1154 mutant has decreased V_max values, which are one-fifth that of wild-type cells at 25° and one-seventh that of wild-type cells at 37°, while the K_m value was not significantly changed (data not shown). Since GS activities of membrane fraction were measured in the presence of GTPS, Rho1p in the in vitro reaction was assumed to be in its active state while Rho1p in vivo is supposed to be not fully activated. Next, to examine the in vivo 1,3-ß-glucan synthesis, we measured incorporation of [¹⁴C]glucose into the 1,3-ß-glucan. Incorporation of glucose at 37° was also severely reduced in the fks1-1154 mutant (Fig 1B). Thus, in both physiological and Rho1p-activated conditions, GS activities of the fks1-1154 mutant are dramatically reduced at the restrictive temperature, suggesting that the activity of Fks1-1154p is below the required minimum value for viability at 37°.

View larger version (48K): In this window In a new window Download PPT slide   Figure 1. Defective in vivo glucan synthesis of fks1-1154 cells. (A) In vitro glucan synthase activity of the membrane fraction of fks1-1154. fks1-1154 cells were incubated at 25° or shifted to 37° for 2 hr and then assayed for GS activity. The data represent the means and standard deviations of at least four experiments. Strains used: YOC1001 (wild type) and YOC1087 (fks1-1154). (B) Incorporation of glucose into 1,3-ß-glucan of fks1-1154. Cells were incubated at 25° or shifted to 37° for 2 hr, labeled with [14C]glucose for 2 hr, and measured for incorporation of glucose into 1,3-ß-glucan. The data represent the means and standard deviations of at least three experiments. Strains used: YOC1001 (wild type) and YOC1087 (fks1-1154). (C) AB staining of yeast cells. Cells were shifted from 25° to 37°, cultured for 4 hr, and then stained with aniline blue. (a and b) YOC1001 (wild type); (c and d) YOC1087 (fks1-1154). (a and c) Glucan staining using aniline blue; (b and d) phase-contrast images. Bar, 10 µm. (D) Defective localization of Fks1p in fks1-1154 cells. The same cells as in C were processed for observation of Fks1p localization and aniline blue staining. Cells with a small bud or tiny projections were observed. The fractions of the cells without polarized Fks1p localization (shaded bars) and those with decreased glucan staining in the bud (solid bars) were determined (n > 200). Strains used: YOC1001 (wild type) and YOC1087 (fks1-1154).

We next examined the growth phenotypes of the fks1-1154 cells. We found that the fks1-1154 cells are hypersensitive to a specific inhibitor of 1,3-ß-glucan synthase, echinocandin B (HECTOR 1993 ; DEBONO and GORDEE 1994 ), and to a chitin-binding reagent, Calcofluor white (RONCERO and DURAN 1985 ; RAM et al. 1994 ). In addition, the fks1-1154 cells failed to grow in a low-glucose medium, in which synthesis of 1,3-ß-glucan is limited. These results are consistent with the idea that GS in fks1-1154 cells is devoid of catalytic activity of GS.

Next, to monitor in vivo 1,3-ß-glucan synthesis in fks1-1154 cells, we stained mutant cells with aniline blue, a fluorescent dye that preferentially interacts with 1,3-ß-glucan. It is known that fluorescent intensity derived from aniline blue increases with the amount of associated 1,3-ß-glucan (EVANS and HOYNE 1984 ; SHEDLETZKY et al. 1997 ). The wild-type cells exhibited uniform staining of the cell wall on the entire cell surface, but most of the fks1-1154 cells appear to lose staining signal specifically in the bud at the restrictive temperature (Fig 1C and Fig D). A slight defect of fks1-1154 was observed at the permissive temperature (Fig 1D, 25°, solid bars). On the basis of these facts, we suggest that the fks1-1154 strain is severely defective in the synthesis of new 1,3-ß-glucan into the bud. We stained other cell wall components, chitin and mannan in the fks1-1154 cells, using Calcofluor white and FITC-Con A, respectively. Staining patterns were indistinguishable from those of the wild-type cells, suggesting that fks1-1154 has a specific defect in 1,3-ß-glucan synthesis.

Growing wild-type cells display a polarized distribution of Fks1p (QADOTA et al. 1996 ). By way of comparison, we examined the localization of Fks1p in fks1-1154 cells and found that Fks1-1154p is delocalized slightly even at the permissive temperature and severely at the restrictive temperature (Fig 1D). Delocalization of Fks1-1154p in the fks1-1154 fks2 background was restored by introduction of FKS2, implying that the delocalization of Fks1-1154p in the fks1-1154 fks2 strain is due to a secondary effect of largely compromised 1,3-ß-glucan synthesis.

Isolation of multicopy suppressors of fks1-1154:
To identify genes that regulate 1,3-ß-glucan synthesis, we performed a multicopy suppressor screen using the fks1-1154 strain. We finally identified WSC1, WSC3, ROM2, LRE1, ZDS1, and MSB1 as multicopy suppressors (for details of the screening, see MATERIALS AND METHODS). Since all suppressor genes were isolated as independent clones more than once, the screening was judged to be nearly saturated. Suppression of the temperature-sensitive growth defect of fks1-1154 by several genes is shown in Fig 2. All genes isolated in the screening effectively suppressed the growth defect of the fks1-1154 mutant at 34° (Fig 2A). We also found that constitutively active RHO1 alleles [RHO1 (G19V) and RHO1 (Q68L)] expressed on a centromere plasmid were able to suppress the fks1-1154 mutation (Fig 2A). Fig 2B shows the suppression ability at various temperatures (33°–37°): Robust growth was observed in fks1-1154 cells transformed with multiple copies of WSC1, ROM2, ZDS1, or with a single copy of RHO1 (G19V) or RHO1 (Q68L). We also examined suppression of the low-glucose sensitive growth defect of fks1-1154 by these suppressors. All multicopy suppressors suppress the low-glucose sensitive growth defect as well as the temperature-sensitive growth defect of fks1-1154. In addition, the multicopy suppressors did not suppress the synthetic lethality of fks1 fks2 double mutants (data not shown), indicating that the suppression of the growth defect is dependent on Fks1-1154p.

View larger version (41K): In this window In a new window Download PPT slide   Figure 2. Growth of fks1-1154 cells transformed with multicopy suppressors. (A) Growth of transformants in YPD. YOC1087 (fks1-1154) was transformed with either a 2µ plasmid [a control vector (pYO326), YEpU-FKS1, YEpU-WSC1, YEpU-WSC3, YEpU-MID2, YEpU-MTL1, YEpU-ROM2, YEpU-LRE1, YEpU-ZDS1, or YEpU-MSB1] or centromere-based plasmids [YCpU-RHO1 (G19V) and YCpU-RHO1 (Q68L)], streaked onto YPD plates, and incubated for 4 days at 25° (top) or 34° (bottom). (B) Growth of the transformants at various temperatures. Growth in YPD of each of the transformants used in A was scored at several temperatures. Solid, shaded, and open regions indicate growth equivalent to wild-type cells, slower growth, and no growth, respectively.

Wsc1p and Wsc3p are proposed to be transmembrane proteins localized on the plasma membrane and are putative upstream regulators of Pkc1p (GRAY et al. 1997 ; VERNA et al. 1997 ; JACOBY et al. 1998 ; LODDER et al. 1999 ). Since Mid2p and Mtl1p were shown to be functionally related to Wsc1p (KETELA et al. 1999 ; RAJAVEL et al. 1999 ), we asked whether overexpression of MID2 or MTL1 suppresses the fks1-1154 mutation. We found that MTL1, but not MID2, weakly suppressed the growth defect of the fks1-1154 mutant (Fig 2).

Rom2p is a GEF of Rho1p. OZAKI et al. 1996 reported that the phenotypes of a rom2 strain include cell lysis and the terminal phenotypes are similar to those of a rho1 and a pkc1 mutant, implying that Rom2p acts upstream of Rho1p and Pkc1p. Since Wsc1p, Wsc3p, Mtl1p, and Rom2p are known as putative upstream regulators of Pkc1p, one possible suppression mechanism is through Pkc1p activation. Therefore, we investigated whether stimulation of the Pkc1p-MAPK cascade rescues the defect of fks1-1154. We transformed fks1-1154 cells with multiple copies of PKC1, an activated PKC1 (R398P) allele (NONAKA et al. 1995 ), BCK1 (a MAPKKK homolog), an activated BCK1 (BCK1-20) allele (LEE and LEVIN 1992 ), or MPK1 (a MAPK homolog). None of them had any effect on temperature-sensitive growth of fks1-1154. With the use of a specific antibody we confirmed that PKC1 overexpression indeed increased Mpk1p phosphorylation in fks1-1154 cells (data not shown). These results suggest that suppression of the fks1-1154 mutation is not through Pkc1p activation. Besides the genes known to interact with Rho1p physically or genetically, three multicopy suppressors (LRE1, ZDS1, and MSB1) were isolated. LRE1 was previously identified as the gene responsible for conferring resistance to laminarinase, an enzyme that specifically degrades 1,3-ß-glucan (LAI et al. 1997 ). ZDS1 was identified in numerous screens, including those designed to enhance replication origin function (YU et al. 1996 ), to increase transcriptional silencing (ROY and RUNGE 1999 ), and to enhance the defect of CDC42 (BI and PRINGLE 1996 ). MSB1 was isolated in a screen for multicopy suppressors of the cdc24-4 mutation and was later shown to suppress a cdc42 mutant defective in polarity establishment and cellular morphogenesis (BENDER and PRINGLE 1989 ), although its function and relationship to Cdc24p and Cdc42p remain unsolved.

Defects of in vivo 1,3-ß-glucan synthesis in fks1-1154 are rescued by multicopy suppressor genes:
Two possible mechanisms of multicopy suppression of the GS mutant are restoration of defective 1,3-ß-glucan synthesis and reparation of compromised cell wall structure. To distinguish these two possibilities, we tested whether high dosages of the suppressors restore the defective in vivo glucan synthesis in fks1-1154 cells. Using fluorescent microscopic analyses with aniline blue staining, we examined populations of cells with decreased 1,3-ß-glucan staining in the bud (Fig 3A). Four hours after a shift to 35° of fks1-1154 cells with vector alone, >70% exhibited a loss of glucan staining in the bud. In contrast, overexpression of WSC1, WSC3, MTL1, ROM2, LRE1, MSB1, ZDS1, and activated forms of RHO1 significantly recovered defective glucan staining in fks1-1154 cells (Fig 3A).

View larger version (33K): In this window In a new window Download PPT slide   Figure 3. Rescue of glucan synthesis of fks1-1154 by the suppressors of the fks1-1154 mutation. (A) Fractions of cells with 1,3-ß-glucan staining in the bud. For aniline blue staining, cells were precultured in SD-Ura, cultured in YPD at 25° for 5 hr, and then shifted to 35° for 4 hr. Cells with a small bud or a tiny projection were observed. The transformants described in Fig 2 and a wild-type strain (YOC1001) with a control vector (pYO326) were tested. The fractions of cells with normal staining at the bud were determined (n > 200). (B) Quantification of the total amount of 1,3-ß-glucan in fks1-1154 cells transformed with each of the multicopy suppressor genes. Cells were cultured in SD-Ura at 25°, and the 1,3-ß-glucan level per cell was measured according to MATERIALS AND METHODS and expressed as relative fluorescence intensity. Each datum represents the means and SD of at least three experiments.

In addition to the microscopic observations, we measured incorporation of [¹⁴C]glucose into 1,3-ß-glucan in fks1-1154 cells transformed with multicopy suppressor genes (Table 3). Measurements of 1,3-ß-glucan revealed that fks1-1154 cells with vector alone incubated at 34° possessed less 1,3-ß-glucan compared to wild-type cells. We found that expression of all the suppressors increased the incorporation into glucan in fks1-1154 cells. As expected from the result presented in Fig 2, overexpression of MID2 had little effect on glucan synthesis in fks1-1154 cells. Of the suppressors, RHO1 (G19V) and RHO1 (Q68L) had a relatively strong effect on recovery of glucan. Introduction of multiple copies of WSC1, WSC3, MTL1, ROM2, LRE1, ZDS1, or MSB1 resulted in a significant increase in 1,3-ß-glucan, but the level was lower than that in wild-type cells. These multicopy suppressors also rescued the reduced incorporation of [¹⁴C]glucose in fks1-1154 cells incubated at 25° in the same way (data not shown). We also quantified the total amount of 1,3-ß-glucan in fks1-1154 cells with multicopy suppressors using aniline blue (Fig 3B). Consistently, overexpression of all the suppressors increased the amount of glucan in fks1-1154 cells while overexpression of MID2 had little effect on glucan synthesis in fks1-1154 cells. Together with these results (Fig 3 and Table 3), we concluded that all the suppressors have a facilitative effect on in vivo 1,3-ß-glucan synthesis. GS activity in fks1-1154 cells in the presence of the suppressors seems to be above the required minimum value for viability.

To examine whether these suppressors act on GS through Rho1p activation, we determined whether suppressors facilitate GS activity in cells that already have the active form of Rho1p. If a suppressor activates GS by converting Rho1p to the active form, no additional effect on GS activity will be observed. In contrast, if the suppressor activates GS in some other way, an additional effect will be seen. First, as a control experiment, we measured glucose incorporation into 1,3-ß-glucan in cells expressing both the active forms of Rho1p and Rom2p. No additive effect was observed with ROM2 (Table 4), which is consistent with the previous report that Rom2p is a GEF (OZAKI et al. 1996 ), and Rom2p is required for Rho1p activation (BICKLE et al. 1998 ). Next, measurement of glucose incorporation with the active form of Rho1p and other suppressors revealed that no additive effect was observed with WSC3, MTL1, LRE1, or ZDS1 (Table 4). These results suggest that Rom2p, Wsc3p, Mtl1p, Lre1p, and Zds1p act positively on GS through activation of Rho1p. On the other hand, an additive effect was observed by expressing MSB1 (Table 4), suggesting that Msb1p acts positively on GS by mechanism(s) not involving Rho1p activation. The additive effect was phenotypically verified on the suppression of the low-glucose sensitive growth defect. Expression of both active forms of Rho1p and Msb1p caused additional suppression in the low-glucose medium, while no additive effect was observed with other multicopy suppressors. Fluorescent measurements using aniline blue also confirmed that the additive effect was observed by expressing MSB1 and the active form of Rho1p in fks1-1154 cells.

We also investigated how Msb1p activates GS using the membrane fractions. We measured the in vitro GS activity of the membrane fraction from the fks1-1154 cells overexpressing Msb1p shifted from 25° to 34° and cultured for 2 hr. Since in vitro GS activity was assayed in the presence of an excess amount of GTPS, a nonhydrolyzable analog of GTP (INOUE et al. 1995 ), virtually all Rho1p is assumed to be in its active state. If MSB1 activates GS in some way other than Rho1p activation, an additional effect on GS activity will be seen in the presence of GTPS. We found that MSB1 slightly increased in vitro GS activity of fks-1154 (Fig 4), supporting the idea that in vitro GS activity increases by a mechanism other than shifting the equilibrium of Rho1p to the GTP-bound state (see DISCUSSION). In contrast, the activities in the presence of a constitutive active RHO1 allele and the rest of the multicopy suppressors were as low as the vector alone in the presence of GTPS (Fig 4). These multicopy suppressors did not increase the reduced activity in fks1-1154 cells incubated at 25° (data not shown).

View larger version (19K): In this window In a new window Download PPT slide   Figure 4. In vitro glucan synthase activity assayed with the membrane fraction of fks1-1154 transformed with a suppressor. fks1-1154 cells transformed with one of the suppressors were shifted from 25° to 34° for 2 hr and then assayed for GS activity. The transformants described in Fig 2 and a wild-type strain (YOC1001) with a control vector (pYO326) were tested.

Analyses of deletion strains of the multicopy suppressor genes:
We constructed deletion mutants of the multicopy suppressor genes (wsc1, wsc3, mid2 mtl1, rom2, lre1, zds1, and msb1) to examine their growth and in vivo 1,3-ß-glucan synthesis. We also examined the phenotypes of a mid2 mtl1 strain because it was reported that mid2 and mtl1 showed a synthetic growth defect (KETELA et al. 1999 ; RAJAVEL et al. 1999 ). We found that among the deletion strains, wsc1 and rom2 had a significant defect in synthesis of 1,3-ß-glucan at the bud (Fig 5A). These defects were more severe at the restrictive temperature than at the permissive temperature (Fig 5B). The other deletion mutants showed no detectable defects when incubated either at 30° or at 37° for 2 hr probably due to the presence of their functional redundant genes. Measurement of glucose incorporation into 1,3-ß-glucan revealed that the cellular glucan synthesis in wsc1 and rom2 cells, but not those in suppressor deletion mutant cells, significantly decreased compared to that in wild-type cells (Fig 6A). The glucose incorporation did not decrease in either mid2 or mtl1 cells (data not shown). Quantitative analyses using aniline blue revealed that the amounts of 1,3-ß-glucan in wsc1 and rom2 cells were reduced while other suppressor deletion mutant cells were not. We also observed that wsc1 and rom2 cells exhibited slow growth in the low-glucose medium.

View larger version (37K): In this window In a new window Download PPT slide   Figure 5. Defects of in vivo glucan synthesis in wsc1 and rom2 cells. (A) In vivo glucan synthesis of wsc1 and rom2 cells. Cells were shifted from 25° to 37°, cultured for 2 hr, and then stained with aniline blue. (a and b) YPH499 (wild type); (c and d) YOC2573 (wsc1); (e and f) YOC2576 (rom2). (a, c, and e) Glucan staining with aniline blue; (b, d, and f) phase-contrast images. Bar, 10 µm. (B) Localization of Fks1p in wsc1, rom2 cells. Cells were cultured and observed as described in Fig 1C and Fig D. Strains used: YPH499 (wild type), YOC2573 (wsc1), and YOC2578 (rom2).

View larger version (24K): In this window In a new window Download PPT slide   Figure 6. 1,3-ß-glucan levels in deletion strains of the multicopy suppressor genes. (A) Cells were cultured to the log phase in YPD at 30° and incorporation of glucose into 1,3-ß-glucan was measured. Strains used: YPH500 (wild type), YOC1001 (FKS1), YOC1087 (fks1-1154), YOC2573 (wsc1), YOC2576 (rom2), YOC2521 (wsc3), YOC2581 (zds1), YOC2579 (lre1), YOC748 (msb1), and YOC2526 (mid2 mtl1). The data represent the means and standard deviations of at least three experiments. (B) Rescue of glucan synthesis of wsc1 and rom2 by the active form of RHO1. YPH500 (wild type), YOC2573 (wsc1), or YOC2576 (rom2) was transformed with the active form of RHO1. Cells were cultured in SD-Ura at 30° and incorporation of glucose into 1,3-ß-glucan was measured. The means and standard deviations of at least three experiments are presented.

Although comparison with wild-type cells showed that Fks1p is partially delocalized in wsc1 and rom2 mutant cells, most cells with defective glucan synthesis retained Fks1p at the bud at both 25° and 37° (Fig 5B). The results suggest that the defect of glucan synthesis is not simply due to a loss of Fks1p localization. The partial loss of Fks1p localization in wsc1 and rom2 cells might be interpreted as a secondary effect of a defective cell wall structure, as was explained above with fks1-1154 (Fig 1D). Together with these results, deletion analysis suggested that Wsc1p and Rom2p play a crucial role in glucan synthesis. Consistently, we observed a synthetic growth defect between fks1-1154 and wsc1, rom2. To confirm that Rom2p and Wsc1p act as upstream regulators of GS through the activation of Rho1p, we tested suppression of the defects in growth and glucan synthesis of wsc1 and rom2 cells with a single copy of the activated form of Rho1p. The temperature-sensitive growth defect of wsc1 and rom2 cells at 37° (OZAKI et al. 1996 ; VERNA et al. 1997 ) was suppressed by expressing the active form of Rho1p. Moreover, glucan synthesis was restored to the wild-type level by expressing the activated form of Rho1p (Fig 6B). Fluorescent measurements using aniline blue confirmed that expressing the activated form of Rho1p rescued the decrease in 1,3-ß-glucan content in wsc1 and rom2 cells. These results suggest that Rom2p and Wsc1p activate GS at point(s) upstream of Rho1p.

A temperature-sensitive rho1 mutant is suppressed by the multicopy suppressors of fks1-1154:
We examined genetic interactions between the multicopy suppressors and the regulatory subunit of GS. Temperature-sensitive rho1-2 mutant cells were employed in this experiment because they showed both decreased in vitro GS activity (QADOTA et al. 1996 ) and decreased incorporation of labeled glucose into 1,3-ß-glucan (Table 5). We found that all of the multicopy suppressors (WSC1, WSC3, MTL1, ROM2, LRE1, ZDS1, and MSB1) effectively suppressed the temperature-sensitive growth defect of the rho1-2 mutant (Fig 7A) and that the suppression was detectable even at 37°. The interpretation of this genetic data, however, is complex, since Rho1p is a multifunctional protein that affects five downstream effector proteins. Mutations of Rho1p result in actin delocalization (HELLIWELL et al. 1998 ) and cell lysis phenotypes (KAMADA et al. 1996 ), both of which are characteristic phenotypes caused by the defect of the Pkc1p-MAPK cascade. These results indicate that the rho1-2 mutation most likely impairs both 1,3-ß-glucan synthesis and the Pkc1p-MAPK cascade. Therefore, although overexpression of WSC1 and ROM2 indeed increased the incorporation level of glucose into 1,3-ß-glucan in rho1-2 cells (Table 5) and the amount of 1,3-ß-glucan, the possibility that the growth suppression of the rho1-2 mutation by WSC1 and ROM2 was achieved by activation of the Pkc1p-MAPK cascade cannot be excluded.

View larger version (29K): In this window In a new window Download PPT slide   Figure 7. Interactions between the suppressors and the Pkc1p-MAPK cascade regulated by Rho1p. (A) Growth of rho1-2 cells transformed with one of the multicopy suppressors. YOC752 (rho1-2) was transformed with a 2µ plasmid (YEpU-RHO1, YEpU-WSC1, YEpU-WSC3, YEpU-MTL1, YEpU-ROM2, YEpU-LRE1, YEpU-ZDS1, YEpU-MSB1, or YEpU-MID2). At the six indicated temperatures growth of the transformants on YPD plates was scored. The results are presented as in Fig 2B. (B) Mpk1p phosphorylation in rho1-2 transformed with one of the suppressors. The transformants in A were shifted to 39° for 2 hr and then blotted with either an anti-phospho-p44/42 MAPK antibody or an anti-MAPK antibody. (C) Quantification of Mpk1p phosphorylation. The amount of Mpk1p phosphorylation was quantified with LAS-1000plus (Fuji Photo Film). The means and standard deviations of at least three experiments are presented.

MID2 effectively suppressed the temperature-sensitive growth defect of the rho1-2 mutant (Fig 7A). Measurement of the incorporation revealed that overexpression of MID2 did not increase the amount of 1,3-ß-glucan in rho1-2 cells (Table 5). In addition, several genetic studies revealed that MID2 is placed upstream of the Pkc1p-MAPK cascade and Rho1p (KETELA et al. 1999 ; RAJAVEL et al. 1999 ; ANDREWS and STARK 2000 ; MARCOUX et al. 2000 ; PHILIP and LEVIN 2001 ). Therefore, the most probable explanation is that the growth suppression of the rho1-2 mutant by MID2 is through activation of the Pkc1p-MAPK cascade.

Pkc1p-MAPK cascade is activated by overexpression of WSC1, ROM2, LRE1, MSB1, MTL1, and MID2:
We directly examined whether the suppressors act upstream of the Pkc1p-MAPK cascade by monitoring the phosphorylation state of Mpk1p with a specific antibody upon heat shock in rho1-2 mutant cells transformed with suppressors. As shown in Fig 7B, we found that overexpression of WSC1, ROM2, and MID2 significantly rescued the defect of Mpk1p phosphorylation upon heat shock (39°) in rho1-2 cells. Quantitative analysis revealed that overexpression of LRE1, MSB1, and MTL1 also rescued the reduced level but the effect was relatively minor (Fig 7C). Overexpression of WSC3 and ZDS1 showed little effect on Mpk1p phosphorylation. To rule out the possibility that the 39° heat shock is too severe for rho1-2 mutant cells, we also monitored Mpk1p phosphorylation at a moderately high temperature (37°). At 37°, Mpk1p was slightly less phosphorylated in wild-type cells and overexpression of WSC3 and ZDS1 showed little effect on the degree of phosphorylation. These results suggest that among the multicopy suppressors, WSC1, ROM2, LRE1, MSB1, MTL1, and MID2 act positively on the Pkc1p-MAPK cascade. On the basis of these results and those shown in Fig 3 and Table 3, we conclude that WSC1, ROM2, LRE1, MSB1, and MTL1 positively regulate both 1,3-ß-glucan synthesis and the Pkc1p-MAPK cascade, while MID2 is involved mainly in the Pkc1p-MAPK cascade. ZDS1 mainly regulates 1,3-ß-glucan synthesis. WSC3 mainly regulates 1,3-ß-glucan synthesis but we could not rule out the possibility that WSC3 has a minor effect on both effectors.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of upstream regulators of Fks1p:
Although the main components of GS, Fks1p and Rho1p, play crucial roles in defining cell shape during the budding cycle, no regulatory factors other than Rho1p have yet been suggested. Using aniline blue and glucose incorporation, we monitored cell wall glucan synthesis and showed that Wsc1p, Wsc3p, Mtl1p, Rom2p, Lre1p, Zds1p, and Msb1p act positively on glucan synthesis. We also present the first in vivo evidence suggesting that the active form of Rho1p promotes glucan synthesis.

All the suppressor genes isolated act positively on 1,3-ß-glucan synthesis. Under the vegetative growth condition, these suppressors seem to act through Fks1p/Fks2p, but not other GS, since the suppressors did not suppress the synthetic lethality of fks1 fks2 double mutants. This is consistent with the observation that Fks3p, possibly another subunit of GS, is not expressed under the vegetative growth condition (our unpublished results). We did not isolate the factors involved in the synthesis of other cell wall components (e.g., mannan, chitin, and 1,6-ß-glucan). Recently, a mutation in FKS1 was reported to show a synthetic growth defect with chs3, which is involved in chitin synthesis (OSMOND et al. 1999 ), implying that different cell wall components act cooperatively on cell wall structure. However, our data suggest that any compensation of other cell wall materials does not suppress the defect of the 1,3-ß-glucan biogenesis pathway.

GS activities of the fks1-1154 mutant at the restrictive temperature are reduced more severely than those at the permissive temperature in vivo and in vitro. However, the difference in the GS activity between the permissive temperature and the restrictive temperature is small while the viability of fks1-1154 cells is decreased severely at the restrictive temperature. We suppose that a threshold of the required minimum value for viability between the small differences in the activity of Fks1-1154p may exist. Introduction of multicopy suppressors increased 1,3-ß-glucan in fks1-1154 cells, but the effect on GS activity was small. Possibly a small increase in GS activity by overexpressing multicopy suppressors crosses over the minimum value for viability and the multicopy suppressors effectively suppress growth defects in fks1-1154 cells.

Possible functions of the multicopy suppressors:
Of the multicopy suppressors of the fks1-1154 mutation, several are putative upstream regulators of Rho1p and Pkc1p, including Rom2p, which is a GEF of Rho1p (OZAKI et al. 1996 ). Wsc family proteins (including Wsc1p, Wsc2p, Wsc3p, and Wsc4p) and the Mid2p homolog, Mtl1p, are putative cell wall surface sensors essential for cell integrity signaling (GRAY et al. 1997 ; VERNA et al. 1997 ; JACOBY et al. 1998 ; RAJAVEL et al. 1999 ; KETELA et al. 1999 ; PHILIP and LEVIN 2001 ). Although the Pkc1p-MAPK cascade is involved in transcription of cell wall biogenesis genes, including FKS1 (IGUAL et al. 1996 ; JUNG and LEVIN 1999 ), activation of the Pkc1p-MAPK cascade by overexpressing Pkc1p or Mid2p in fks1-1154 had little effect on the activation of GS (see RESULTS). Furthermore, we found that overexpression of the other two Rho1p targets, BNI1 and SKN7, had little effect on growth of fks1-1154 cells (our unpublished results). Together with these results, we propose that activation of 1,3-ß-glucan synthesis occurs through direct activation of GS by Rho1p and that Fks1p and Pkc1p are activated independently by distinct upstream regulators. This idea is further supported by evidence that Mpk1p phosphorylation is defective in mid2 cells (KETELA et al. 1999 ; RAJAVEL et al. 1999 ) while mid2 cells have a normal content of 1,3-ß-glucan (see RESULTS).

How do Wsc family proteins and Rom2p regulate glucan synthesis? It has been suggested that Wsc1p and Rom2p interact with each other (PHILIP and LEVIN 2001 ) and that these factors are important for activation of the Pkc1p-MAPK cascade upon environmental stress (GRAY et al. 1997 ; VERNA et al. 1997 ; MARTIN et al. 2000 ) and for relocalization of Fks1p upon heat shock (DELLEY and HALL 1999 ). However, no evidence indicates that Wsc1p and Rom2p are required for activation of 1,3-ß-glucan synthesis. Here we show that wsc1 and rom2 cells exhibit a significant defect in 1,3-ß-glucan synthesis even at the permissive temperature (Fig 5A and Fig 6). This is the first evidence suggesting that Wsc1p and Rom2p play roles in 1,3-ß-glucan synthesis as well as in the Pkc1p-MAPK cascade during normal budding. As reported previously, Wsc1p localization is dependent on the polarized actin cytoskeleton (DELLEY and HALL 1999 ). Although more detailed analysis concerning the timing of Wsc1p localization is needed, it is conceivable that localization of Wsc1p might depend on polarization of the actin cytoskeleton achieved by Cdc42p.

Although Lre1p, Zds1p, and Msb1p were not known to be involved in cell wall synthesis, we found that LRE1, ZDS1, and MSB1 act positively on 1,3-ß-glucan synthesis (Fig 3 and Table 3). Of all the multicopy suppressors of the fks1-1154 mutation identified, only MSB1 showed a significant increase of in vitro glucan synthesis (Fig 4), suggesting that Msb1p does not act on the GDP-to-GTP exchange on Rho1p. At least two possibilities are consistent with this result. One possibility is that Msb1p is a component of the GS complex. However, Msb1p was apparently lost during preparation of the purified GS fraction, suggesting that Msb1p is not a tightly bound component of the GS complex. The other possibility is that Msb1p increases the GS complex. In fact, the amounts of Fks1p and Rho1p were slightly increased by expression of Msb1p in fks1-1154 or rho1-2 cells (our unpublished results). This result suggests that MSB1 acts on GS positively by increasing the expression of GS components or by stabilizing the GS complex. Since msb1 cells did not exhibit a clear effect on GS activity, there may be a factor(s) that has a redundant function of Msb1p. It should be noted that MSB1 suppresses a mutation in BEM4, which encodes a protein that physically interacts with multiple types of Rho-type GTPase without affecting their GDP-GTP exchange cycle (MACK et al. 1996 ). Msb1p might function through a mechanism shared with Bem4p. Lre1p and Zds1p are novel components related to Rho1p. Although more study will be necessary to know the precise molecular mechanism of their functions, we will discuss their involvement in Rho1p-regulated pathways below.

Relationship between the suppressors and Rho1p:
We showed that all suppressor genes examined are genetically related to Rho1p and summarize the results of their effects on the Pkc1p-MAPK cascade and GS in Fig 8. The suppressors were classified into three groups: WSC1, ROM2, LRE1, MSB1, and MTL1 regulate both 1,3-ß-glucan synthesis and the Pkc1p-MAPK pathway, and ZDS1 mainly regulates 1,3-ß-glucan synthesis, while MID2 mainly regulates the Pkc1p-MAPK pathway. WSC3 regulates 1,3-ß-glucan synthesis or both effectors of Rho1p. Thus, we show that the two essential Rho1p targets, Fks1p and Pkc1p, have distinct upstream regulators. Moreover, among the positive factors of 1,3-ß-glucan synthesis tested, WSC1, WSC3, MTL1, ROM2, LRE1, and ZDS1 act through activation of Rho1p, while MSB1 does not.

View larger version (30K): In this window In a new window Download PPT slide   Figure 8. Distinct effects of Rho1p-related components on activation of Fks1p and Pkc1p. We showed that different components genetically related to Rho1p (Wsc1p, Wsc3p, Mid2p, Mtl1p, Rom2p, Msb1p, Lre1p, and Zds1p) affect the Rho1p effectors, Fks1p and Pkc1p, differently.

In this article, we proposed a model for the function of the suppressors under vegetative growth conditions. It is quite interesting to test the possibility that the subset of upstream components for the Rho1p targets could be different under different physiological conditions. For example, MID2 is known to be important for sensing cell wall stress both during vegetative growth and pheromone-induced conditions (ONO et al. 1994 ; KETELA et al. 1999 ; RAJAVEL et al. 1999 ). Future study needs to address whether Mid2p has a different subset of downstream regulators under these conditions.

In mammalian cells, GTP-bound Ras triggers several cellular activities, including cell proliferation, malignant transformation, differentiation, and apoptosis, depending on different cell types (SATOH and KAZIRO 1992 ; CAMPBELL et al. 1998 ). Like Rho1p, Ras has various kinds of upstream regulators (receptor-tyrosine kinases) and downstream effectors. Although unproven in mammalian cells, it is assumed that various activities are caused by different combinations of effector molecules. In yeast, it is known that RHO1 may activate a subset of effectors during cytoskeletal organization (HELLIWELL et al. 1998 ). However, no explanation has yet been offered on how various pathways are regulated by small GTPases.

Here we demonstrated that putative upstream regulators of Rho1p have major effects on the different effectors, Fks1p and Pkc1p. What is the determinant of this different effect? It seems important that C-terminal amino acid sequences of Mid2p, which fall in the putative cytoplasmic domain, are different from the corresponding sequences of Mtl1p, Wsc1p, and Wsc3p. In consideration of the previous report (PHILIP and LEVIN 2001 ), all these Wsc family proteins may interact with Rom2p and activate Rho1p, which is consistent with our results. Possibly each cytoplasmic domain of Wsc family proteins acts as a scaffold that assembles Rom2p and different Rho1p effector complexes. It is important to identify molecules interacting with the cytoplasmic domains and also necessary to know which signals reside in Mid2p, Mtl1p, Wsc1p, and Wsc3p. It has already been suggested that stress signals responsible for Pkc1p-MAPK activation are different in different Wsc-related proteins (KETELA et al. 1999 ; DE NOBEL et al. 2000 ; MARTIN et al. 2000 ) and that different Wsc-related proteins are modified by different factors (PHILIP and LEVIN 2001 ).

In conclusion, we identified putative regulatory components of 1,3-ß-glucan synthesis using fks1-1154. We also suggested a network of regulatory components upstream of the two Rho1p effectors, Fks1p and Pkc1p. Future study will include determination of the functions of the regulators in cell cycle progression, screening of more mutants with defects in 1,3-ß-glucan synthesis, and investigation of interactions of the regulatory components at the molecular level. Further investigation of the specificity of signals upstream of Rho1p effectors will give us a key to answering how orchestration of small GTPase protein functions can be achieved.

	FOOTNOTES

¹ These authors contributed equally to this work.

	ACKNOWLEDGMENTS

We thank Kana Morishita and Hiroshi Qadota for initiating this project. We also thank Kazuma Tanaka for providing us the PKC1 (R398P) allele and David Levin for giving us the BCK1-20 allele. We are also grateful to Keiichi Homma for critically reading the manuscript and members of the Laboratory of Signal Transduction for helpful discussion. This work was supported by grants from the Ministry of Education, Science, and Sports and Culture of Japan to Y. Ohya and from the Japan Society for the Promotion of Science for Young Scientists to M. Abe.

Manuscript received June 24, 2002; Accepted for publication July 15, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABE, M., I. NISHIDA, M. MINEMURA, H. QADOTA, and S. SEYAMA et al., 2001  Yeast 1,3-ß-glucan synthase activity is inhibited by phytosphingosine localized to the endoplasmic reticulum. J. Biol. Chem. 276:26923-26930.[Abstract/Free Full Text]

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 ß subunits and the yeast response regulator protein Skn7. J. Biol. Chem. 273:8616-8622.[Abstract/Free Full Text]

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]

BENDER, A. and J. R. PRINGLE, 1989  Multicopy suppression of the cdc24 budding defect in yeast by CDC42 and three newly identified genes including the ras-related gene RSR1.. Proc. Natl. Acad. Sci. USA 86:9976-9980.