Mutational analysis of the cytoplasmic domain of the Wsc1 cell wall stress sensor

The cell wall of the budding yeast Saccharomyces cerevisiae is required to maintain cell shape and integrity (Cid et al., 1995; Klis, 1994). The cell must remodel this rigid structure during vegetative growth and during pheromone-induced morphogenesis. Wall remodelling is monitored and regulated by the cell integrity signalling pathway controlled by the Rho1 GTPase. Two essential functions have been identified for Rho1. First, it serves as an integral regulatory subunit of the 1,3-β-glucan synthase complex (GS) that stimulates GS activity in a GTP-dependent manner (Drgonova et al., 1996; Qadota et al., 1996). A pair of closely related genes, FKS1 and FKS2, encode alternative catalytic subunits of the GS complex (Douglas et al., 1994; Inoue et al., 1995; Mazur et al., 1995; Ram et al., 1995) that are the presumed targets of Rho1 activity.

A second essential function of Rho1 is to bind and activate protein kinase C (Kamada et al., 1996; Nonaka et al., 1995), which is encoded by PKC1 (Levin et al., 1990). Loss of PKC1 function, or of any of the components of the MAP kinase cascade under its control, results in a cell lysis defect that is attributable to a deficiency in cell wall construction (Levin & Bartlett-Heubusch, 1992; Levin et al., 1994; Paravicini et al., 1992). The MAP kinase cascade is a linear pathway comprising a MEKK (Bck1; Costigan et al., 1992; Lee & Levin, 1992), a pair of redundant MEKs (Mkk1/2; Irie et al., 1993) and a MAPK (Mpk1/Slt2; Lee et al., 1993; Torres et al., 1991). One of the consequences of signalling through the MAP kinase cascade is the activation of the SRF-like transcription factor, Rlm1 (Watanabe et al., 1997; Jung et al., 2002). Signalling through Rlm1 regulates the expression of at least 25 genes, most of which have been implicated in cell wall biogenesis (Jung & Levin, 1999).

Cell wall integrity signalling is induced in response to several environmental stimuli. First, signalling is activated persistently in response to growth at elevated temperatures (e.g. 3739 °C; Kamada et al., 1995), consistent with the finding that null mutants in many of the pathway components display cell lysis defects only when cultivated at high temperature. Second, hypo-osmotic shock induces a rapid, but transient activation of signalling (Davenport et al., 1995; Kamada et al., 1995). Third, treatment with mating pheromone stimulates signalling at a time that is coincident with the onset of morphogenesis (Buehrer & Errede, 1997; Errede et al., 1995). Indeed, mutants defective in cell integrity signalling undergo cell lysis during pheromone-induced morphogenesis. Finally, agents that cause cell wall stress, such as caffeine and the chitin antagonist Calcofluor White, also activate signalling (Ketela et al., 1999; Martin et al., 2000).

The mechanisms by which cell wall stress is transmitted to Rho1 is an area of active investigation. Several regulators of Rho1 activity have been identified. Rom1 and Rom2 comprise a redundant pair of guanine nucleotide exchange factors (GEFs) for Rho1 (Ozaki et al., 1996). Bem2 and Sac7 are GTPase-activating proteins (GAPs) for Rho1 (Kim et al., 1994; Peterson et al., 1994; Schmidt et al., 1997). Finally, a family of cell surface sensors for the activation of cell integrity signalling has been described. These include Wsc1, Wsc2, Wsc3, Mid2 and Mtl1 (Gray et al., 1997; Verna et al., 1997; Jacoby et al., 1998; Ketela et al., 1999; Rajavel et al., 1999). Among these, Wsc1 and Mid2 are the major sensors dedicated to signalling wall stress during vegetative growth and pheromone-induced morphogenesis (Rajavel et al., 1999; Ketela et al., 1999). Indeed, a wsc1Δ mid2Δ double mutant displays a severe cell lysis defect, indicating that these genes have an overlapping function during vegetative growth. The cytoplasmic domains of both Wsc1 and Mid2 interact with the N-terminal domain of the Rom2 (and presumably Rom1) guanine nucleotide exchange factor to stimulate GTP loading of Rho1 (Philip & Levin, 2001).

All members of the Wsc1/Mid2 family are transmembrane proteins that reside in the plasma membrane (Ketela et al., 1999; Lodder et al., 1999; Rajavel et al., 1999; Verna et al., 1997). Their overall structures are similar in that they possess small cytoplasmic domains, each has a single transmembrane region and their extracellular domains are rich in Ser/Thr residues. These Ser/Thr-rich regions are highly O-mannosylated, probably resulting in extension and stiffening of the polypeptide. Therefore, the extracellular domains have been proposed to act as rigid probes of the extracellular matrix (Rajavel et al., 1999). Despite the broad similarity among the proteins, there is very limited sequence identity among their cytoplasmic domains. Here we present a mutational analysis of the cytoplasmic domain of Wsc1.

Strains, growth conditions and transformations.
Yeast strains used in this study were derived from EG123 (Kamada et al., 1995). Wild-type haploid (1783) or diploid (1788) yeast cultures were grown in YEPD (1 % Bacto Yeast Extract, 2 % Bacto Peptone, 2 % glucose) or YEPD supplemented with 10 % sorbitol to prevent lysis of strain DL2282 (wsc1Δ : : LEU2 mid2Δ : : URA3; Rajavel et al., 1999). Synthetic minimal (SD) medium (Rose et al., 1990) supplemented with the appropriate nutrients was used to select for plasmid maintenance. Yeast transformations were carried out by the lithium acetate method (Ito et al., 1983). wsc1 mutant plasmids were tested for function in yeast strain DL2282, for Wsc1^HA protein expression and phosphorylation in the haploid parental strain 1783 and for Mpk1 activation in the diploid parental strain 1788. Escherichia coli DH5α was used to propagate all plasmids. E. coli cells were cultured in Luria broth medium (1 % Bacto Tryptone, 0·5 % Bacto Yeast Extract, 1 % NaCl) and transformed by standard methods.

WSC1 plasmids
All plasmids used in this study are shown in Table 1.

Table 1. Plasmids

Expression plasmids.
To construct the C-terminal truncation mutants of WSC1, sequences (including 600 bp 5' to the translational start site) were amplified by PCR through the final amino acid desired in the truncation using YEp352-WSC1^HA (p1657) as template and inserted blunt-ended into the SmaI site of episomal plasmid YEp352 [3xHA] (p845; Rajavel et al., 1999). The resultant clones fused the 3xHA epitope, in-frame with an additional glycine residue, to the various C termini of Wsc1. These mutant alleles were subcloned as KpnIPstI fragments into centromeric plasmid pRS314 (Sikorski & Hieter, 1989).

For construction of point mutants, pRS314-WSC1^HA (p1672) was used as template for site-directed mutagenesis by the PCR overlap extension method (Ho et al., 1989). The T7 and T3 primers were used in primary PCR reactions with forward and reverse mutagenic primers, respectively. The products of these reactions were mixed in secondary PCR reactions using only the T7 and T3 primers. The full-length products derived from these reactions were subcloned either as KpnISacI fragments into pRS314 (K301A, Y303A and Q304A) or as ApaIPstI fragments into pRS314-WSC1^HA (L369A, V371A, V372A, N373A, P374A, D375A and D378A). YEp352-wsc1-S328/329/331A-T337/341A^HA (p1847) was constructed stepwise by first mutating the two Thr residues in pRS314-WSC1^HA, followed by mutagenesis of the three Ser residues and finally subcloning the pentuple mutant as an MfeISalI fragment into YEp352-WSC1^HA. YEp352-wsc1-S319/320A^HA (p1868), YEp352-wsc1-S322/323A^HA (p1869) and YEp352-wsc1-S319/320/322/323A^HA (p1850) were constructed by simultaneous mutation of two or all four Ser residues, followed by subcloning the double and quadruple mutants as a MfeISalI fragment into YEp352-WSC1^HA. pRS314-wsc1-(1-346)-S319/320/322/323A^HA (p1866) was constructed by subcloning the truncated C terminus of wsc1-(1-346)^HA (from p1513) into pRS314-wsc1-S319/320/322/323A^HA by an ApaIPstI fragment (the ApaI site spans residues 326/327). The complete DNA sequences of all inserts were determined. Sequence analysis was performed by the JHU Biosynthesis and Sequencing Facility. PCR was performed using Pfu polymerase (Stratagene). Primers are available upon request.

Two-hybrid plasmids and assays.
Sequences encoding the C-terminal tails (residues 291378) of wsc1 point mutants were amplified by PCR using primers that placed a BamHI site before the N-terminal residue and a PstI site after the C-terminal residue. These fragments were cloned into the BamHI and PstI sites of pGBT9 (Clontech Laboratories) so as to fuse the wsc1 sequences in-frame with the Gal4-DNA-binding domain. All fusions were confirmed by DNA sequence analysis. pGBT9 clones of mutant wsc1 tails were cotransformed with pGAD424-rom2-(1-660) (p1667) into yeast strain SFY526 (Clontech Laboratories) and transformants were tested as described in Philip & Levin (2001) for two-hybrid interactions.

Immunodetection and phosphatase treatment of Wsc1^HA.
Extracts of yeast strain EG123 expressing forms of Wsc1^HA from YEp352 were made and tested by immunoblotting with mouse mAb 12CA5, as described by Philip & Levin (2001) with the following modifications to the lysis buffer: phosphatase inhibitors (30 mM sodium pyrophosphate and 0·2 mM sodium vanadate) and 0·5 mM EGTA were added and 0·5 % NP40 was substituted for Triton X-100. The phosphorylation state of Wsc1^HA forms was determined by treatment with λ protein phosphatase (New England Biolabs). Extracts (1525 µg protein) were treated with 400 U λ protein phosphatase for 2 h at 30 °C in λ phosphatase buffer (50 mM Tris/HCl, pH 7·5, 0·1 mM EDTA, 5 mM DTT, 0·01 % Brij35 and 2 mM MnCl₂) with or without phosphatase inhibitors (45 mM KF, 23 mM sodium pyrophosphate and 1·5 mM sodium vanadate). Treated extracts were subjected to SDS-PAGE on 415 % polyacrylamide gradient Ready Gels (Bio-Rad).

Immunodetection of Mpk1^HA and activated Mpk1 after heat shock.
Yeast cells grown to an OD₆₀₀ of 0·51·0 at 23 °C in YEPD were exposed to a mild heat shock (39 °C) by 1 : 1 dilution with fresh medium prewarmed to 55 °C and maintained at 39 °C for the indicated times. The cell response was terminated by further dilution (1 : 1) with ice-cold stop mix (Kamada et al., 1995). Extracts were made and Mpk1^HA was detected by immunoblot after samples (5 µg protein) were subjected to SDS-PAGE using mouse mAb 12CA5 (BaBCo) and horseradish peroxidase-linked secondary antibody (Amersham) as described by Kamada et al. (1995). Activated Mpk1 was detected with rabbit polyclonal anti-phospho-p44/p42 MAPK (Thr²⁰²/Tyr²⁰⁴) antibody (New England Biolabs) essentially as described by Martin et al. (2000) except that 20 µg protein was fractionated and primary antibody was used at a 1 : 1000 dilution.

As a first step in the functional analysis of the cytoplasmic domain of Wsc1, we constructed a set of C-terminal truncation mutants of this sensor (Fig. 1a) in which the HA epitope was fused to the new C termini. C-terminal fusions have been shown previously not to impair Wsc1 function (Lodder et al., 1999; Rajavel et al., 1999). We tested the ability of these truncated forms to complement the cell lysis defect of a wsc1Δ mid2Δ double mutant when expressed from a centromeric plasmid. The wsc1Δ mid2Δ mutant, which lyses at all growth temperatures in the absence of osmotic support (Rajavel et al., 1999), provides a more sensitive test of Wsc1 function than does a wsc1Δ mutant, which only displays a cell lysis defect at high temperatures (e.g. 3739 °C). Deletion of the C-terminal 11 amino acid residues in wsc1(1-367) was sufficient to eliminate complementation at 30 °C (Fig. 1b), suggesting that residues at the extreme C terminus of this protein are important to its function. Similarly, deletion of the C-terminal 32 residues in wsc1(1-346) eliminated complementation. Interestingly, removal of an additional 20 or 30 residues [in wsc1(1-326) or wsc1(1-316), respectively] partially restored complementation function, suggesting the presence of sequences between residues 317 and 345 that inhibit Wsc1 function. Finally, removal of nearly the entire cytoplasmic domain in wsc1(1-300) eliminated complementation, indicating that residues between 301 and 316 are necessary and sufficient for partial function of the cytoplasmic domain. All of the truncated forms of Wsc1 were expressed at wild-type levels (data not shown).

(43K): Fig. 1. (a) C-terminal truncations of Wsc1. (b) Complementation of the cell lysis defect of a wsc1Δ mid2Δ mutant by some of the Wsc1 C-terminal truncations. Yeast strain DL2282 (wsc1Δ mid2Δ) was transformed with centromeric plasmid pRS314 (vector) or pRS314 bearing the indicated allele of WSC1HA (p1672, p1511, p1512, p1513, p1514 or p1518). Transformants were streaked onto YEPD and allowed to grow at 30 °C for 2 days.

Although the cytoplasmic domain of Wsc1 has diverged appreciably from those of Wsc2 and Wsc3, and is only about half the length of the other two, there are two conserved regions of sequence among all three (Fig. 2a). One of these, identified by Lodder et al. (1999), is the KxYQ box, which resides at residues 301304 in Wsc1. The other region of Wsc1 that is conserved with the other sensors comprises the 10 residues at its extreme C terminus (residues 369378; Fig. 2a). These conserved sequences correspond to the two regions of the cytoplasmic domain assessed to be important for Wsc1 function by our truncation analysis. Therefore, we mutated each of the conserved residues within both regions individually to alanyl residues to determine their contribution to function. Fig. 2(b) shows that among the conserved residues in the KxYQ box, only the Y303A mutation failed to complement the growth defect of a wsc1Δ mid2Δ mutant at 30 °C (or at 37 °C, data not shown). Consistent with this result, only the Y303A mutant failed to interact with Rom2 by two-hybrid analysis (Fig. 2c), suggesting that Tyr303 is important for Rom2 association.

(48K): Fig. 2. (a) Conserved residues within the cytoplasmic tail of Wsc1, 2 and 3. (b) Complementation of a wsc1Δ mid2Δ mutant by point mutants in the KxYQ motif of Wsc1. Yeast strain DL2282 (wsc1Δ mid2Δ) was transformed with centromeric plasmid pRS314 (vector) or pRS314 bearing the indicated allele of WSC1HA (p1672, p1785, p1786 or p1787). Transformants were streaked onto YEPD and allowed to grow at 30 °C for 2 days. (c) Interaction of point mutants in the Wsc1 KxYQ motif with the N-terminal domain of Rom2. Plasmids expressing fusions between the Gal4-DNA-binding domain (in pGBT9) and the cytoplasmic domain of the indicated Wsc1 forms (p1324, p1790, p1818 or p1819) were cotransformed with pGAD424-rom2-(1-660) (p1667; expressing a Gal4-activation domain fusion to Rom2) into yeast strain SFY526. Transformants were patched onto YEPD for 24 h and tested for two-hybrid interaction. Dark patches indicate interaction.

We next mutated the seven conserved residues at the C terminus of Wsc1 individually to alanyl residues. Among these mutants, only the L369A and N373A alleles failed to complement the wsc1Δ mid2Δ mutant at 30 °C (Fig. 3a). Two additional mutants, V371A and D375A, displayed some impairment at 37 °C (Fig. 3b). Here again, the two-hybrid interactions of these mutant forms with Rom2 were consistent with their ability to complement the wsc1Δ mid2Δ mutant (Fig. 3c). All four mutants at the extreme C terminus that displayed some deficiency in complementation were deficient in Rom2 interaction, suggesting that this region of Wsc1 also interacts with Rom2. Interestingly, the wsc1-L369A and wsc1-N373A alleles were less effective at complementation than either the wsc1(1-316) or wsc1(1-326) truncation mutants (Fig. 3d), supporting the notion that negative regulatory sequences reside between residues 317 and 345. As with the truncation mutants, epitope-tagged versions of all the point mutants in WSC1 were expressed at wild-type levels (data not shown).

(48K): Fig. 3. (a, b) Complementation of a wsc1Δ mid2Δ mutant by point mutants in the extreme C terminus of Wsc1. Yeast strain DL2282 (wsc1Δ mid2Δ) was transformed with centromeric plasmid pRS314 (vector) or pRS314 bearing the indicated allele of WSC1HA (p1672, p1880, p1881, p1882, p1883, p1884, p1885 or p1889). Transformants were streaked onto YEPD and allowed to grow at 30 (a) or 37 °C (b) for 2 days. (c) Interaction of point mutants in the extreme C terminus Wsc1 with the N-terminal domain of Rom2. Plasmids expressing fusions between the Gal4-DNA-binding domain (in pGBT9) and the cytoplasmic domain of the indicated Wsc1 forms (p1324, p1891, p1893, p1894, p1895, p1896, p1901 or p2018) were cotransformed with pGAD424-rom2-(1-660) (p1667; expressing a Gal4-activation domain fusion to Rom2) into yeast strain SFY526. Transformants were patched onto YEPD for 24 h and tested for two-hybrid interaction. Dark patches indicate interaction. (d) Point mutants at the extreme C terminus of Wsc1 are more severely compromised than truncation mutants that remove regions of the C terminus. Yeast strain DL2282 was transformed with pRS314 or pRS314 bearing the indicated allele of WSC1HA. Transformants were treated as in (a).

Lodder et al. (1999) demonstrated that Wsc1 is phosphorylated within the cytoplasmic domain. Phosphorylation of Wsc1 can be detected as a band shift with increased mobility after treatment with λ protein phosphatase (λPPase; Fig. 4a and Lodder et al., 1999). Treatment of Wsc1(1-367) or Wsc1(1-346) with λPPase resulted in a band shift (Fig. 4b). However, λPPase treatment of Wsc1(1-326) or Wsc1(1-316) failed to produce a band shift, indicating that the regions bounded by residues 326 and 346 are important for phosphorylation. This was also a region determined by deletion analysis to be important for negative regulation of Wsc1 (Fig. 1b). There are five potential phosphorylation sites (three seryl and two threonyl residues) within this region (Fig. 5a). To determine if any of these are real phosphorylation sites, we mutated all five seryl and threonyl residues collectively to alanyl residues (designated ST5A). To our surprise, the Wsc1-ST5A mutant form behaved like wild-type with respect to the λPPase-induced band shift (Fig. 5b), indicating that the phosphorylation site(s) do not reside between residues 326 and 346 and must be further N-terminal to 326. There may be a protein kinase docking site within the region between residues 326 and 346, which would explain the importance of this region for phosphorylation.

(35K): Fig. 4. Wsc1 is phosphorylated on its cytoplasmic domain. (a) Protein extract of yeast strain 1783 expressing Wsc1HA from YEp352 (p1657) was treated with λPPase in the presence or absence of phosphatase inhibitors. Samples were subjected to SDS-PAGE separation followed by immunoblot detection of Wsc1HA using the 12CA5 antibody. (b) Extracts of yeast strain 1783 expressing C-terminal truncations of Wsc1HA from YEp352 (p1449, p1459, p1460, p1516 or p1657) were treated with λPPase and the indicated form of Wsc1 detected as in (a).

(36K): Fig. 5. (a) Sequence of the Wsc1 cytoplasmic tail. Seryl and threonyl residues between 326 and 346 (boxed) were mutated in a cluster to alanyl residues (yielding wsc1-ST5A). Seryl residues 319, 320, 322 and 323, which were mutated in a cluster in wsc1-S4A, are the presumptive phosphorylation sites. Truncation sites are also shown by vertical lines. (b, c) Extracts of yeast strain 1783 expressing the indicated point mutants of Wsc1HA from YEp352 (p1847, p1850, p1868, p1869 or p1657) were treated with λPPase and the indicated form of Wsc1 detected as in Fig. 4. (d) The S4A mutation partially suppressed the defect of the wsc1(1-346) truncation mutant. Yeast strain DL2282 (wsc1Δ mid2Δ) was transformed with centromeric plasmid pRS314 (vector) or pRS314 bearing the indicated allele of WSC1HA (p1513, p1672 or p1866). Transformants were streaked onto YEPD and allowed to grow at 28 °C for 2 days. (e) The S4A mutation potentiates Mpk1 activation in response to mild heat shock. Wild-type yeast strain 1788 transformed with plasmids bearing MPK1HA (p672) and either WSC1 (p1672) or WSC1-S4A (p1854) were grown to mid-exponential phase at 23 °C in YEPD and shifted to 39 °C for the indicated times. Extracts were tested by immunoblotting for dual phosphorylation of Mpk1 and Mpk1HA (upper panel) and for Mpk1HA protein levels (lower panel).

We next mutated the four seryl residues that reside between 316 and 326 (Fig. 5a) collectively to alanyl residues (designated S4A). In this case, the Wsc1-S4A mutant protein failed to undergo a λPPase-induced band shift (Fig. 5b), migrating at the same position as λPPase-treated wild-type Wsc1. We separated these mutations into adjacent pairs (S319/320A and S322/323A). Both of these mutant proteins behaved as unphosphorylated forms (Fig. 5c), suggesting that all four seryl residues may be phosphorylated in an interdependent manner. Alternatively, some of these seryl residues may be important for substrate recognition by the protein kinase to phosphorylate the others.

To determine the contribution of the phosphorylated seryl residues to Wsc1 function, we introduced the S4A mutation into the wsc1(1-346) truncation allele. Fig. 5(d) shows that eliminating these phosphorylation sites partially suppressed the growth defect associated with wsc1(1-346), supporting the conclusion that phosphorylation of Wsc1 serves a negative regulatory role. Phosphorylation site mutants of Wsc1 did not display enhanced two-hybrid interaction with Rom2 (data not shown). However, because only the cytoplasmic domain of Wsc1 was expressed in the two-hybrid clone to allow nuclear localization, it may not be phosphorylated in this setting. Finally, we tested the effect of the S4A mutation on Mpk1 activity by following dual phosphorylation of the MAP kinase (Martin et al., 2000). We did not detect constitutive activity of Mpk1 in cells expressing the WSC1-S4A allele under non-inducing conditions (Fig. 5e). However, a shift from 23 to 39 °C reproducibly induced a more rapid activation of Mpk1 than observed in wild-type cells (Fig. 5e), suggesting that dephosphorylation of Wsc1 potentiates signal transduction. Full activation of Mpk1 in response to mild heat shock normally requires 30 min (Kamada et al., 1995; Fig. 5e). By contrast, Mpk1 was fully activated within 10 min after temperature upshift in the WSC1-S4A mutant.

These results, taken together, suggest a model for the regulated interaction of Wsc1 with Rom2. We propose that two regions of the Wsc1 cytoplasmic tail interact with Rom2 (Fig. 6). One of these includes tyrosine 303, which is proximal to the plasma membrane. Because a truncation that removes most of the cytoplasmic tail [wsc1(1-316)] was partially functional, we conclude that this region is sufficient for Rom2 stimulation in the absence of other cytoplasmic domain sequences. The other Rom2-interacting region is at the extreme C terminus of Wsc1, defined by residues 369375. Disruption of either of these interactions in the context of the full-length protein prevents Rom2 interaction. The presence of a negative regulatory region between residues 316 and 345 also influences function. We identified a cluster of seryl residues within this region of Wsc1 (S319, 320, 322 and 323) that appear to be phosphorylated. Seryl-to-alanyl mutation of this cluster of residues partially suppressed the growth defect of a Wsc1 mutant missing the C-terminal interaction region. Moreover, a WSC1 mutant missing these phosphorylation sites potentiated activation of Mpk1 by mild heat shock. Therefore, we propose that phosphorylation of the Wsc1 cytoplasmic tail interferes with Rom2 interaction (Fig. 6) and that dephosphorylation provides a means of regulating this interaction in response to cell wall stress. However, we have not been able to detect stress-induced dephosphorylation of Wsc1. Perhaps only a small fraction of the Wsc1 is dephosphorylated in response to wall stress. It will be interesting to identify the protein kinase and phosphatase that are responsible for the regulation of Wsc1.

(15K): Fig. 6. Model for the regulated interaction of the Wsc1 cytoplasmic tail with Rom2. Phosphorylation of seryl residues 319/320/322/323 interferes with the interaction of the two indicated regions of Wsc1 with Rom2.

This work was supported by a grant from the NIH (GM48533) to D. E. L.

Footnotes

†Present address: Wyeth Pharmaceuticals, Cambridge, MA 02140, USA.