QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 18, Issue 11, 4405-4419, November 2007

This Article Abstract Full Text (PDF) All Versions of this Article: E07-05-0484v1 18/11/4405    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Madrid, M. Articles by Cansado, J. Search for Related Content PubMed PubMed Citation Articles by Madrid, M. Articles by Cansado, J.

Stress-activated Protein Kinase-mediated Down-Regulation of the Cell Integrity Pathway Mitogen-activated Protein Kinase Pmk1p by Protein Phosphatases

Yeast Physiology Group, Department of Genetics and Microbiology, Facultad de Biología, University of Murcia, 30071 Murcia, Spain

Submitted May 23, 2007; Revised July 11, 2007; Accepted August 15, 2007
Monitoring Editor: Fred Chang

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Fission yeast mitogen-activated protein kinase (MAPK) Pmk1p is involved in morphogenesis, cytokinesis, and ion homeostasis as part of the cell integrity pathway, and it becomes activated under multiple stresses, including hyper- or hypotonic conditions, glucose deprivation, cell wall-damaging compounds, and oxidative stress. The only protein phosphatase known to dephosphorylate and inactivate Pmk1p is Pmp1p. We show here that the stress-activated protein kinase (SAPK) pathway and its main effector, Sty1p MAPK, are essential for proper deactivation of Pmk1p under hypertonic stress in a process regulated by Atf1p transcription factor. We demonstrate that tyrosine phosphatases Pyp1p and Pyp2p, and serine/threonine phosphatase Ptc1p, that negatively regulate Sty1p activity and whose expression is dependent on Sty1p-Atf1p function, are involved in Pmk1p dephosphorylation under osmostress. Pyp1p and Ptc1p, in addition to Pmp1p, also control the basal level of MAPK Pmk1p activity in growing cells and associate with, and dephosphorylate Pmk1p both in vitro and in vivo. Our results with Ptc1p provide the first biochemical evidence for a PP2C-type phosphatase acting on more than one MAPK in yeast cells. Importantly, the SAPK-dependent down-regulation of Pmk1p through Pyp1p, Pyp2p, and Ptc1p was not complete, and Pyp1p and Ptc1p phosphatases are able to negatively regulate MAPK Pmk1p activity by an alternative regulatory mechanism. Our data also indicate that Pmk1p phosphorylation oscillates as a function of the cell cycle, peaking at cell separation during cytokinesis, and that Pmp1p phosphatase plays a main role in regulating this process.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In eukaryotic organisms, mitogen-activated protein kinase (MAPK) pathways transduce extracellular signals from hormones, growth factors, cytokines, or environmental stresses to trigger a diverse array of biological responses. Functional MAPK cascades comprise MAPK kinase kinases (MAPKKKs), which phosphorylate and activate MAPK kinases (MAPKKs), which in turn phosphorylate and activate MAPKs (Marshall, 1995; Waskiewicz and Cooper, 1995). Finally, active MAPKs phosphorylate different substrates to promote changes in gene expression that play a critical role in the adjustment of cells to environmental conditions. Three distinct MAPK signaling cascades have been identified so far in the fission yeast Schizosaccharomyces pombe. These include the stress-activated protein kinase (SAPK) pathway, the mating pheromone-responsive pathway, and the cell integrity pathway, whose central elements are MAPKs Sty1p/Spc1p, Spk1p, and Pmk1p/Spm1p, respectively (Toda et al., 1991, 1996; Millar et al., 1995; Shiozaki and Russell, 1995; Zaitsevskaya-Carter and Cooper, 1997).

The key player of the SAPK cascade in S. pombe is MAPK Sty1p, which shows high homology to mammalian p38 kinase and becomes activated by a wide range of stresses (Millar et al., 1995; Shiozaki and Russell, 1995; Degols et al., 1996; Soto et al., 2002) (Figure 1A). Sty1p is directly phosphorylated by MAPKK Wis1p, whereas Wis1p activation is dual, via either MAPKKKs Wak1p (also known as Wis4p or Wik1p) or Win1p (Shieh et al., 1997). Also, a response regulator protein, Mcs4p, associates with Wak1p and probably with Win1p to regulate MAPKKK activity in response to external stimuli (Shieh et al., 1997) (Figure 1A). Expression of many stress responsive genes in fission yeast is controlled by Sty1p through transcription factor Atf1p (Degols et al., 1996; Shiozaki and Russell, 1996; Wilkinson et al., 1996; Chen et al., 2003). In contrast, Pmk1p/Spm1p is a structural homologue to the cell integrity MAPK Mpk1p/Slt2p from Saccharomyces cerevisiae (Toda et al., 1996; Hohmann, 2002), and their activation is similar to the extracellular signal-regulated kinases ERK1/2p (p42/p44) from animal cells that become activated by growth factors, phorbol esters, cytokines, or osmotic stress (Roux and Blenis, 2004). In S. pombe, the involvement of MAPK Pmk1p in the maintenance of cell integrity, cytokinesis, and ion homeostasis derives from phenotypic analyses of mutant strains deleted in genes encoding the main components of the cascade, namely, mkh1⁺ (encoding MAPKKK Mkh1p) (Sengar et al., 1997), skh1⁺/pek1⁺ (encoding MAPKK Skh1p/Pek1p) (Sugiura et al., 1999; Loewith et al., 2000) and pmk1⁺/spm1⁺ (encoding MAPK Pmk1p/Spm1p) (Toda et al., 1996; Zaitsevskaya-Carter and Cooper, 1997). Deletion in any of these genes causes morphological changes, and cells display multiseptate phenotype, growth inhibition in response to potassium ions, hypersensitivity to -1,3-glucanases, and defective vacuole fusion under hypotonic stress (Toda et al., 1996; Sengar et al., 1997; Zaitsevskaya-Carter and Cooper, 1997; Bone et al., 1998; Sugiura et al., 1999; Loewith et al., 2000). MAPK Pmk1p is dually phosphorylated by Pek1p at two conserved threonine and tyrosine residues in positions 186 and 188, respectively (Sugiura et al., 1999; Loewith et al., 2000). Inactive (unphosphorylated) Pek1p binds Pmk1p in the absence of external stimulus and acts as a potent inhibitor of Pmk1p signaling (Sugiura et al., 1999). Although the Mkh1p–Pek1p-Pmk1p cascade was initially described as activatable by high temperatures or sodium chloride only (Toda et al., 1996; Zaitsevskaya-Carter and Cooper, 1997), recent work has shown that Pmk1p phosphorylation is induced by multiple stressing conditions (Madrid et al., 2006). In all cases, the stress-induced activation of Pmk1p was completely dependent on Mkh1p and Pek1p function, suggesting the existence of a linear, nonbranched MAPK cascade. Fluorescence microscopy studies detect Pmk1p into the cytoplasm and nucleus, whereas Mkh1p MAPKKK and Pek1p MAPKK are found exclusively into the cytoplasm. Notably, all three kinases also locate at the septum during cell separation (Madrid et al., 2006). Furthermore, stress treatments or the absence of Mkh1p or Pek1p do not alter the subcellular localization of MAPK Pmk1p, suggesting that its activation occurs at the cytoplasm or septum and that both its active and inactive forms cross the nuclear membrane. At present, the identity of transcription factor(s) regulated by Pmk1p is unknown.

View larger version (24K): [in this window] [in a new window]   Figure 1. The SAPK pathway controls Pmk1p deactivation in growing cells and under hypertonic stress. (A) Schematic arrangement of the main functional components of the SAPK pathway in the fission yeast. RR, response regulator; TF, transcription factor; , signal transmission; , enhanced transcription; , phosphatase inhibitory effect. See Introduction for details. (B) Strains MI200 (pmk1-HA6H, Control), MI208 (pmk1-HA6H, mcs4), MI 217 (pmk1-HA6H, wak1), MI210 (pmk1-HA6H, wis1), MI204 (pmk1-HA6H, sty1), and MI211 (pmk1-HA6H, atf1) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. Aliquots were harvested at the times indicated, and Pmk1-HA6H was purified by affinity chromatography. Activated Pmk1p was detected by immunoblotting with anti-phospho-p42/44 and total Pmk1p with anti-HA antibodies.

Currently, Pmp1p is the only phosphatase known to dephosphorylate and inactivate Pmk1p in fission yeast. The corresponding gene pmp1⁺ was originally isolated as a multicopy suppressor for the chloride hypersensitivity of one strain deleted in ppb1⁺, which encodes calcineurin (Sugiura et al., 1998). Further analysis indicated that Pmp1p is closely related to dual-specificity phosphatases and that it is able to bind and inactivate Pmk1p in vitro and in vivo, suggesting that it may negatively regulate the Pmk1p MAPK pathway (Sugiura et al., 1998). Microarray analyses have shown no significant changes in pmp1⁺ transcription during cell cycle or stress treatments (Chen et al., 2003). However, a recent work has demonstrated that the stability of pmp1⁺ mRNA relies on the function of the RNA binding protein Rnc1p, whose activity is up-regulated by Pmk1p-mediated phosphorylation (Sugiura et al., 2003). Thus, the negative-feedback loop that controls the posttranscriptional stabilization of Pmp1p transcripts by Rnc1p provides a regulatory mechanism for fine-tuning of the Pmk1p pathway.

We described previously the existence of cross-talk between Sty1p and Pmk1p MAPKs in S. pombe (Madrid et al., 2006). Sty1p function is required for correct deactivation of Pmk1p in cells subjected to osmotic upshifts by a mechanism dependent on Atf1p transcriptional activity and on de novo protein synthesis (Madrid et al., 2006). This result strongly suggests that one or several phosphatases whose expression is up-regulated by the SAPK pathway might be responsible for Pmk1p dephosphorylation. In this work, we have focused on this question to disclose that the SAPK-regulated tyrosine phosphatases Pyp1p and Pyp2p, as well as the threonine phosphatase Ptc1p, are able to down-regulate Pmk1p activity in S. pombe depending on the nature of the stimulus.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Strains and Growth Conditions
S. pombe strains (Table 1) were grown with shaking at 28°C in YES or EMM2 medium with 3% of glucose (Moreno et al., 1991), and supplemented with adenine, leucine, histidine, or uracil (100 mg/l; Sigma-Aldrich Co., St. Louis, MO) depending on their particular requirements. Mutant strains were obtained by standard transformation procedures or by mating and selecting diploids in EMM2 medium without supplements. Spores were obtained in MEL medium (Moreno et al., 1991), purified by glusulase treatment (Soto et al., 2002) and allowed to germinate in EMM2 plus the appropriate requirements. Transformation of yeast strains was performed by the lithium acetate method as described previously (Soto et al., 2002). In experiments performed with cdc25-22 thermosensitive mutant strains, the cells were grown in YES medium to an A₆₀₀ of 0.2 at 25°C (permissive temperature), shifted to 37°C for 3.5 h, and released from the growth arrest by transfer back to 25°C. In experiments for the expression of a particular gene driven by the thiamine repressible promoter, yeast cultures were grown in EMM2 with or without 5 mg/l thiamine for 12–24 h. Escherichia coli DH5F' was used as host to propagate plasmids by growth at 37°C in Luria-Bertani medium plus 50 µg/ml ampicillin.

Gene Disruption and Epitope Tagging
The pmk1⁺, pmp1⁺, atf1⁺, pyp1⁺, pyp2⁺, ptc1⁺, ptc2⁺, and ptc3⁺ null mutants were obtained by entire deletion of the corresponding coding sequences and their replacement with the KanMX6 (KanR) cassette by PCR-mediated strategy by using plasmid pFA6a-kanMX6 as template (Bahler et al., 1998). Alternatively, gene deletion was performed using plasmid pCR2.1.hph conferring hygromycin B resistance (Sato et al., 2005). Plasmid pFA6a-kanMX6-P41nmt1-GST (Bahler et al., 1998) was used to obtain strains expressing N-terminal, GST-tagged versions of Pyp1p and Pyp2p under the control of the medium-strength thiamine repressible promoter. To construct strains expressing C-terminal 13-myc or GST tagged versions of either pyp1⁺, pyp2⁺, or ptc1⁺, we used plasmids pFA6a-13Myc-kanMX6 and pFA6a-GST-kanMX6, respectively (Bahler et al., 1998). Primer sequences used in each case are available upon request. The Pmk1-HA6H–tagged strains were obtained after transformation with integrative plasmid pIH-pmk1-ura (Madrid et al., 2006) previously digested with BstxI at the unique site within the pmk1⁺ coding region. Strains expressing a C-terminal–tagged version of Pmk1p fused to the green fluorescent protein (GFP) were obtained by transforming with integrative plasmid pIL-pmk1-GFP (Madrid et al., 2006) digested at the unique NruI site within leu1⁺ and selecting for leucine prototrophy. In all cases, the correct construction of strains was verified by PCR, Western blot. or both (see below).

Stress Treatments
To investigate Pmk1p activation under hypertonic stress most experiments were made using log phase cell cultures (A₆₀₀ of 0.5) growing at 28°C in YES medium and the addition of 0.6 M KCl. At different times, the cells from 30 ml of culture were harvested by centrifugation at 4°C, washed with cold phosphate-buffered saline buffer, and the yeast pellets immediately frozen in liquid nitrogen for analysis.

Purification and Detection of Activated Pmk1-HA6H and Sty1-HA6H
Cell homogenates were prepared under native conditions by using chilled acid-washed glass beads and lysis buffer (10% glycerol, 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Nonidet NP-40, plus specific protease and phosphatase inhibitor cocktails for fungal and yeast extracts; Sigma Chemical). The lysates were cleared by centrifugation at 16,000 x g for 15 min and HA6H-tagged Pmk1p or Sty1p were purified by using nickel-nitrilotriacetic acid (Ni²⁺-NTA)-agarose beads (QIAGEN, Hilden, Germany) (Madrid et al., 2004). The purified proteins were resolved in 10% SDS-polyacrylamide gel electrophoresis (PAGE) gels, transferred to nitrocellulose filters (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom), and incubated with either monoclonal mouse anti-hemagglutinin (HA) (clone 12CA5; Roche Molecular Biochemicals, Basel, Switzerland), polyclonal rabbit anti-phospho-p42/44 antibodies (Cell Signaling Technology, Danvers, MA) (Madrid et al., 2006), or monoclonal mouse anti-phospho-p38 antibodies (Cell Signaling Technology) (Soto et al., 2002). Alternatively, we used a mouse monoclonal anti-phosphotyrosine antibody (PY99; Santa Cruz Biotechnology, Santa Cruz, CA). In this case, cell extracts were prepared as described by Shiozaki and Russell (1997) and incubated with Ni²⁺-NTA-agarose beads. After being processed as described above, immunoreactive bands were revealed with anti-mouse or anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies (Sigma Chemical) and the SuperSignal System (Pierce Chemical, Rockford, IL). Densitometric quantification of Western blot signals was performed using Molecular Analyst Software (Bio-Rad, Hercules, CA).

Detection of myc-tagged Fusions
Cell homogenates were prepared under native conditions as indicated above. The cleared lysates were resolved in 8% SDS-PAGE gels, transferred to nitrocellulose filters, and incubated with monoclonal mouse anti-c-myc antibody (clone 9E10; Roche Molecular Biochemicals). A polyclonal rabbit anti-Cdc2 antibody (PSTAIR; Upstate Biotechnology, Lake Placid, NY) was used as loading control. To detect immunoreactive bands, anti-mouse or anti-rabbit HRP-conjugated secondary antibodies (Sigma Chemical) and the SuperSignal System (Pierce Chemical) were used.

Purification and Detection of GST Fusions in E. coli and S. pombe
Purification of GST or GST-fused versions of phosphatases Pmp1p, Pyp1p, Pyp2p, and Ptc1p in E. coli DH5 was performed according to Millar et al. (1992). Briefly, the expression of the corresponding GST fusion was induced by adding 0.4 mM isopropyl -D-thiogalactoside to mid-log phase bacterial cultures. GST-fused proteins were precipitated from lysate supernatants by incubation at 4°C with glutathione-Sepharose (GE Healthcare). Sepharose-bound proteins were then eluted by incubating 1 h at 4°C with 10% glycerol, 100 mM KCl, 5 mM MgCl₂, 0.1 mM ZnCl₂, 0.1 mM EDTA, 2 mM dithiothreitol (DTT), 10 mM HEPES, and 10 mM reduced glutathione, pH 9.0. To purify GST or GST fusions in S. pombe, cells were grown in 1 l of EMM2 medium without thiamine to induce expression of GST, GST-Pyp1, or GST-Pyp2. The same medium was used to grow strains expressing Pyp1-GST or Ptc1-GST fusions under control of its own promoter. Yeast cells were recovered by filtration, washed, resuspended in 4 ml of lysis buffer (10% glycerol, 20 mM Tris-HCl, pH 8.2, 100 mM NaCl, 10 mM MgCl₂, 2 mM EDTA, 0.05% Nonidet NP-40, plus specific protease and phosphatase inhibitors), and disrupted with chilled acid-washed glass beads. The lysates were cleared by centrifugation at 15,000 x g for 25 min and incubated for 3 h with 1 ml of glutathione-Sepharose beads previously equilibrated in lysis buffer. After exhaustive washing with the same buffer, the bound proteins were eluted in the presence of reduced glutathione and concentrated by trichloroacetic acid (TCA) precipitation.

Coprecipitation Experiments
To detect interaction of GST-fusions with Pmk1-HA6H or Sty1-HA6H, the TCA precipitates were resolved by SDS-PAGE, transferred to nitrocellulose filters, and hybridized separately with anti-HA, anti-phospho-p44/42, or a polyclonal goat anti-GST antibody conjugated to HRP (GE Healthcare). The immunoreactive bands were revealed as indicated above.

Pmk1p Dephosphorylation In Vitro
Activated Pmk1-HA6H was purified with Ni²⁺-NTA-agarose beads (see above) from strain MI200 growing to mid-log phase in YES medium and subjected to stress with 0.6 M KCl for 15 min. The beads were washed twice and resuspended in phosphatase buffer containing 1 mM EDTA, 2 mM DTT, 50 mM imidazole, pH 7.2. Aliquots of Pmk1-HA6H were incubated at 30°C for 60 min in phosphatase buffer with 5 µg of either GST-Pmp1, GST-Pyp1, or GST-Pyp2, with or without 30 mM sodium orthovanadate. Phosphatase assays with GST-Ptc1 were performed by incubating activated Pmk1-HA6H in solution A (50 mM Tris-HCl, pH 7.0, 0.1 mM EGTA, 1 mg/ml bovine serum albumin, and 10 mM glutathione) with 20 mM MgCl₂ or 1 mM EDTA at 30°C for 45 min (Nguyen and Shiozaki, 1999). The reactions were stopped by adding sample buffer, subjected to SDS-PAGE, and analyzed by Western blot analysis with anti-p42/44, anti-HA, and anti-GST antibodies as described above.

Fluorescence Microscopy
Images were taken on a Leica DM 4000B fluorescence microscope with a 100x objective, captured with a cooled Leica DC 300F camera and IM50 software, and then imported and processed with Adobe PhotoShop 6.0 (Adobe Systems, Mountain View, CA). To localize Pmk1-GFP fusion, small aliquots (10 µl) from cells growing in YES medium were loaded onto poly-L-lysine–coated slides or fixed with formaldehyde as described previously (Alfa et al., 1993). Calcofluor white was used for cell wall/septum staining as described previously (Alfa et al., 1993). A minimum of 200 cells per strain was analyzed.

-1,3-Glucanase Sensitivity
Resistance to -1,3-glucanase was assayed in different mutants by the method of Loewith et al. (2000) with some modifications. Cells were grown in YES medium to an A₆₀₀ of 0.6, washed with 10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 1 mM -mercaptoethanol, and incubated with vigorous shaking at 30°C in the same buffer supplemented with 100 µg/ml Zymolyase 20-T (Seikagaku, Tokyo, Japan). Samples were taken every 15 min, and cell lysis was monitored by measuring A₆₀₀ decay.

Plate Assay of Stress Sensitivity for Growth
Wild-type and mutant strains of S. pombe were grown in YES liquid medium to log phase. Appropriate dilutions were spotted per duplicate on YES solid media or in the same medium supplemented with different concentrations of MgCl₂. The plates were incubated at 28°C for 3 d.

Reproducibility of Results
All experiments were repeated at least three times with similar results. Representative results are shown.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
SAPK Controls Pmk1p Deactivation during Growth and Hypertonic Stress
Sty1p MAPK is essential for proper Pmk1p deactivation after hypertonic stress in a process regulated by transcription factor Atf1p (Madrid et al., 2006). We performed a comparative analysis of Pmk1p activation/deactivation in salt-stressed S. pombe strains expressing an HA6H-tagged version of Pmk1p MAPK and deleted in different components of the SAPK pathway (Figure 1A) by using a p42/44 antibody, which detects dual phosphorylation at threonine and tyrosine. Pmk1p phosphorylation was evident in control cells after 20 min of hypertonic treatment, followed by a deactivation trend (Figure 1B). However, strains deleted in mcs4⁺ (response regulator), wak1⁺ (MAPKKK), wis1⁺ (MAPKK), sty1⁺ (MAPK), or atf1⁺ (transcription factor) showed a comparatively maintained stress-induced Pmk1p activation (Figure 1B). Identical results were obtained when these strains were subjected to nonsaline hypertonic stress (1 M sorbitol; data not shown). Also, a moderate but reproducible increase in basal Pmk1 phosphorylation was evident in the above-mentioned mutants compared with control cells (Figure 1B). These data confirm that the SAPK pathway transduces hypertonic stress signals via Sty1p-Atf1p to promote Pmk1p dephosphorylation, and they suggest that such pathway negatively regulates Pmk1p activity in growing cells.

Role of Tyrosine Phosphatases Pyp1p and Pyp2p as Negative Regulators of Pmk1p Activity
The above-mentioned results support that one or more protein phosphatases induced through the SAPK pathway might be responsible for Pmk1p deactivation in growing cells and under hypertonic stress. Thus, we first focused our attention on tyrosine phosphatases Pyp1p and Pyp2p, which account for Sty1p dephosphorylation during normal growth conditions (Pyp1p) and under osmostress (Pyp1p and Pyp2p) (Millar et al., 1995; Degols et al., 1996; Shiozaki and Russell, 1996). In fact, Sty1p–Atf1p function regulates during osmostress the basal expression of pyp1⁺ and the induced level of pyp1⁺ (moderate) and pyp2⁺ (strong) mRNAs (Degols et al., 1996; Shiozaki and Russell, 1996; Wilkinson et al., 1996, Chen et al., 2003). We determined Pmk1p activation/deactivation under hypertonic stress in control cells, and in cells lacking Pyp1p, Pyp2p, or Pmp1p, which is the only known protein phosphatase able to down-regulate Pmk1p activation in S. pombe (Sugiura et al., 1998). In pyp1 strain MI213, basal Pmk1p phosphorylation was higher than in control cells, whereas under osmostress the activation was significantly lower (Figure 2A). On the contrary, pyp2 cells showed unchanged basal level of Pmk1p activity, but the deletion prompted a marked phosphorylation and a kinetics of Pmk1p deactivation slower than in wild-type cells (Figure 2A). Except for a clear increase in the basal level of Pmk1p activity, pmp1⁺ deletion did not seem to affect significantly the rate of Pmk1p dephosphorylation under osmostress, although the overall level of Pmk1 phosphorylation was higher at all times (Figure 2A). We also analyzed Pmk1p phosphorylation under osmostress in control, pyp1, pyp2, and pmp1 strains by using specific anti-phosphotyrosine antibody (pY99) instead of anti-p42/44 antibody. The pattern of tyrosine phosphorylation of Pmk1p in these mutants was very similar to that observed with the antibody detecting dual phosphorylation at threonine and tyrosine (Figure 2B). To examine further the involvement of Pyp1p and Pmp1p phosphatases in regulating the basal level of activated Pmk1p, we analyzed this character in unstressed growing cells from different mutants. The p42/44 and pY99 levels in sty1, atf1, pyp1, and pmp1 strains were higher than in control or pyp2⁺-deleted cells (Figure 2C). Together, these results strongly suggest that both Pmp1p and Pyp1p phosphatases are responsible for Pmk1p tyrosine dephosphorylation under normal growth conditions, whereas Pyp2p is involved in Pmk1p down-regulation during hypertonic stress. However, additional phosphatases might be involved in this process, because Pmk1p dephosphorylation can be still recorded in the absence of Pyp2p. In contrast, the unexpected Pmk1 hypoactivation observed in pyp1 cells could result from Sty1 hyperactivation (see below).

View larger version (32K): [in this window] [in a new window]   Figure 2. Tyrosine phosphatases Pyp1p and Pyp2p are negative regulators of Pmk1p activation during growth and under hypertonic stress. (A) Strains MI200 (pmk1-HA6H, Control), MI213 (pmk1-HA6H, pyp1), MI214 (pmk1-HA6H, pyp2), and MI212 (pmk1-HA6H, pmp1) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. At timed intervals, Pmk1-HA6H was purified by affinity chromatography under native conditions. Activated and total Pmk1p was detected by immunoblotting with anti-phospho-p42/44 or anti-HA antibodies, respectively. (B) The same experiment described in A, except that Pmk1p phosphorylation was detected by immunoblotting with anti-phosphotyrosine (PY99) antibody. (C) Strains MI200 (pmk1-HA6H, Control), MI204 (pmk1-HA6H, sty1), MI211 (pmk1-HA6H, aft1), MI212 (pmk1-HA6H, pmp1), MI213 (pmk1-HA6H, pyp1), and MI214 (pmk1-HA6H, pyp2) were grown in YES medium to mid-log phase, and Pmk1p basal activity was determined with anti-phospho-p42/44 or anti-phosphotyrosine antibodies.

Tyrosine Phosphatases Pyp1p and Pyp2p Associate with, and Dephosphorylate Pmk1p
If Pmk1p were a substrate for Pyp1p and Pyp2p, these protein tyrosine phosphatases should be able to bind such MAPK to some extent. To test this, we expressed N-terminal GST-tagged versions of Pyp1p and Pyp2p under the control of the medium-strength thiamine repressible promoter nmt41 in strains MI502 and MI503, which contain genomic versions of Pmk1p fused to HA6H epitope at its C terminus. The Sty1p-tagged strains MI500 and MI501 were used as a positive control for binding to Pyp1p and Pyp2p, respectively. After growth of the cells in minimal medium with or without thiamine, GST-Pyp1p and GST-Pyp2p fusions were purified by affinity chromatography with glutathione-Sepharose beads. Finally, the bound proteins were subjected to Western blot analysis with anti-HA antibodies. Figure 3A shows that purification of both GST-Pyp1p and GST-Pyp2p fusions from cultures growing in the absence of thiamine allowed to detect Sty1p and Pmk1p with anti-HA antibodies, whereas in negative controls or in cultures grown in the presence of thiamine these MAPKs did not copurify with GST alone. These results suggest that Pyp1p and Pyp2p associate with Pmk1p. To ensure that the above-mentioned interactions were not due to overexpression of the phosphatases, we performed a coprecipitation experiment by using strain MI506, which expresses a C-terminal GST-tagged version of Pyp1p under the regulation of its own promoter in a Pmk1-HA background. Approximately equal amounts of GST (negative control) and Pyp1-GST fusions were purified from cell extracts with glutathione-Sepharose beads. As shown in Figure 3B, Pmk1-HA was exclusively detected after Pyp1-GST purification, indicating that most likely Pmk1 and Pyp1 interact in vivo.

View larger version (20K): [in this window] [in a new window]   Figure 3. Pyp1p and Pyp2p associate with, and dephosphorylate Pmk1p. (A) Strains MI500 (sty1-HA6H, nmt1:GST-pyp1, lanes 1 and 2), MI501 (sty1-HA6H, nmt1:GST-pyp2, lanes 3 and 4), MI502 (pmk1-HA6H, nmt1:GST-pyp1, lanes 5 and 6), MI503 (pmk1-HA6H, nmt1:GST-pyp2, lanes 7 and 8), and MI200 expressing unfused GST from plasmid pDS472a (lanes 9 and 10) were grown in EMM2 medium in the presence (+T) or absence (–T) of thiamine for 16 h. GST, GST-Pyp1, and GST-Pyp2 fusions were purified with glutathione-Sepharose beads, resolved by SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with anti-HA or anti-GST antibodies. (B) Pyp1p and Pmk1p associate in vivo. An MI200 transformant expressing unfused GST from plasmid pDS472a (lane 1), and strain MI502 (pmk1-HA6H, pyp1-GST) expressing a genomic version of Pyp1p fused to GST at its C terminus (lane 2) were grown in EMM2 medium in the absence of thiamine for 16 h. GST and Pyp1-GST fusions were purified with glutathione-Sepharose beads, resolved by SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with anti-HA or anti-GST antibodies. (C) Pyp1p and Pyp2p dephosphorylate Pmk1p in vitro. Activated Pmk1-HA6H was purified with Ni2+-NTA-agarose beads from strain MI200 after treatment with 0.6 M KCl for 15 min. The beads were incubated at 30°C for 60 min in phosphatase buffer with 5 µg of either GST (lane 1), GST-Pyp2 (lanes 2 and 3), GST-Pyp1 (lanes 4 and 5), or GST-Pmp1 (lanes 6 and 7) in the presence (even lanes) or absence (odd lanes) of 30 mM sodium orthovanadate. The samples were analyzed by SDS-PAGE and immunoblotted with anti-p42/44, anti-HA, and anti-GST antibodies.

Threonine Phosphatase Ptc1p Attenuates Basal and Activated Pmk1p under Osmostress
We focused on PP2C protein phosphatases Ptc1p, Ptc2p, and Ptc3p as potential candidates to negatively modulate Pmk1p activity because of the possible contribution of other phosphatases to Pmk1p deactivation (Figure 2, A and B). Threonine phosphatases Ptc1p and Ptc3p are involved in Sty1p down-regulation, and the induced expression of ptc1⁺ under stress is regulated by Sty1p-Atf1p (Gaits et al., 1997; Nguyen and Shiozaki, 1999). We found no significant change in the phosphorylation pattern of Pmk1p in ptc2 or ptc3 cells from either untreated or osmotic-stressed growing cultures compared with control cells (Figure 4A). However, deletion of ptc1⁺ resulted in increased basal level of Pmk1 activity and slower deactivation kinetics under osmostress than in wild-type cells (Figure 4A). This suggests that Ptc1p is involved in Pmk1p inactivation in both growing and osmostressed cells. Comparative analyses in control and mutant strains showed that deletion of either ptc1⁺ or pmp1⁺ induced a clear increase in Pmk1p phosphorylation in growing cells (Figure 4B).

View larger version (22K): [in this window] [in a new window]   Figure 4. Ptc1p associates with, and dephosphorylates, Pmk1p. (A) Strains MI200 (pmk1-HA6H, Control), MI216 (pmk1-HA6H, ptc1), MI218 (pmk1-HA6H, ptc2), and MI219 (pmk1-HA6H, ptc3) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. At different times Pmk1-HA6H was purified by affinity chromatography, and either activated or total Pmk1p was detected by immunoblotting with anti-phospho-p42/44 or anti-HA antibodies, respectively. (B) Strains MI200 (pmk1-HA6H, Control), MI212 (pmk1-HA6H, pmp1), MI216 (pmk1-HA6H, ptc1), MI218 (pmk1-HA6H, ptc2), and MI219 (pmk1-HA6H, ptc3) were grown in YES medium to mid-log phase, and Pmk1p basal activity was determined by using anti-phospho-p42/44 antibody as described above. (C) Pmk1p dephosphorylation in vitro. Activated Pmk1-HA6H or Sty1-HA6H was purified with Ni2+-NTA-agarose beads from strains MI200 (Pmk1p; lanes 1–3) or JM1521 (Sty1p; lanes 4–6) after treatment with 0.6 M KCl for 15 min. The beads were incubated at 30°C for 45 min in phosphatase buffer with 10 µg of GST-Ptc1 (lanes 2, 3, 5, and 6) in the presence of 10 mM MgCl2 (lanes 3 and 5) or 10 mM EDTA (lanes 2 and 6). The samples were analyzed by SDS-PAGE and immunoblotted with anti-p42/44, anti-HA, and anti-GST antibodies. (D) Pmk1p and Ptc1p interact in vitro. Cell lysates from exponentially growing strains JM1521 (Sty1-HA6H; lanes 1 and 3), MI200 (Pmk1-HA6H; lanes 2 and 4), and strain MI200 subjected to a 15-min treatment with 0.6 M KCl (lane 5) were incubated each with 15 µg of bacterially purified GST (lanes 1 and 2) or GST-Ptc1 (lanes 3–5) bound to glutathione-Sepharose beads. The beads were washed extensively, and the binding of Pmk1p or Sty1p fusions was detected by inmunoblotting with monoclonal anti-HA antibodies. (E) Pmk1p and Ptc1p interact in vivo. An MI200 transformant expressing unfused GST from plasmid pDS472a (lane 1) and strain MI505 (pmk1-HA6H, ptc1-GST) expressing a genomic version of Ptc1p fused to GST at its C terminus (lane 2) were grown in EMM2 medium in the absence of thiamine for 16 h. GST and Pyp1-GST fusions were purified with glutathione-Sepharose beads, resolved by SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with anti-HA or anti-GST antibodies.

Defective Pmk1p Activation under Saline Stress in pyp1 Cells Is Due to Sty1p Hyperactivation
The above-mentioned results strongly suggest that the SAPK-regulated phosphatases Pyp1p, Pyp2p, and Ptc1p deactivate Pmk1p, either in growing cells (Pyp1p and Ptc1p) or under osmostress (Pyp2p and Ptc1p). However, the defective Pmk1p activation in pyp1 cells subjected to osmostress was an intriguing result (Figure 2A). As indicated in Figure 5A, basal Sty1p phosphorylation measured by immunoblotting with anti-phospho-p38 antibody increased significantly in pyp1 cells, but not in the absence of Pyp2p, Ptc1p or Pmp1p. A reasonable interpretation for this is that, because the expression of pyp2⁺ and ptc1⁺ genes during osmostress depends on Sty1p-Atf1p (Degols et al., 1996; Shiozaki and Russell, 1996; Wilkinson et al., 1996, Chen et al., 2003), the hyperactivation of Sty1p in pyp1 cells should entail increased expression and synthesis of the two phosphatases, which would in turn lead to the decreased Pmk1p activation observed in the absence of Pyp1p. To examine this hypothesis, we undertook three different approaches. First, we constructed control and pyp1 strains expressing C-terminal 13myc-tagged versions of Pyp2p or Ptc1p phosphatases, and we determined Pyp2-13myc and Ptc1-13myc protein levels in cultures subjected to osmostress. As Figure 5B (top) shows, Pyp1p deletion prompted a clear increase in Pyp2-13myc levels in growing cells (0 min) and 15–30 min after the stress treatment with 0.6 M KCl compared with control cells. Pyp1p deletion also induced a modest but reproducible increase in Ptc1-13myc protein levels (Figure 5B, bottom). Second, we analyzed the osmostress-induced Pmk1p activation/deactivation in control cells and in strain MI709, expressing a hyperactive version of Wis1p MAPKK (wis1DD), which prompts the constitutive activation of Sty1p (Shiozaki et al., 1998). As shown in Figure 5C, Pmk1p activation in the wis1DD mutant was lower than in control cells. Moreover, Pyp1p and Ptc1p protein levels were enhanced in the mutant relative to control cells upon osmostress (Figure 5D), whereas Pyp2p protein levels were detected in growing and 15-min salt-stressed wis1DD cells (Figure 5D). Third, additional deletion of atf1⁺ gene in a wis1DD background rescued the defective Pmk1p activation observed in stressed cultures from the wis1DD mutant (Figure 5C). As a whole, these data support the notion that an increased synthesis of phosphatases Pyp2p and Ptc1p (due to up-regulation of the Sty1p-Atf1p loop) causes defective Pmk1p activation under osmostress in pyp1 cells. Also, they confirm that Pyp2p and Ptc1p negatively modulate Pmk1p activity in such conditions.

View larger version (35K): [in this window] [in a new window]   Figure 5. Sty1p hyperactivation promotes defective Pmk1p activation under saline stress. (A) Sty1p basal activity in different mutants. Strains JM1521 (sty1-HA6H, Control), MI1001 (sty1-HA6H, pyp1), MI1002 (sty1-HA6H, pyp2), MI1003 (sty1-HA6H, ptc1), MI1000 (sty1-HA6H, pmp1), and MI100 (sty1-HA6H, pmk1) were grown in YES medium to mid-log phase, and Sty1-HA6H was purified by affinity chromatography. Activated or total Sty1p was detected by immunoblotting with anti-phospho-p38 and anti-HA antibodies, respectively. (B) Increased Pyp2p and Ptc1p synthesis in pyp1 cells. Strains MI702 (pyp2-13myc, Control) and MI704 (pyp2-13myc, pyp1) (top), MI703 (ptc1-13myc, Control), and MI705 (ptc1-13myc, pyp1) (bottom) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. Aliquots were harvested at the times indicated, and total extracts were obtained. Pyp2-13myc and Ptc1p-myc fusions were detected by immunoblotting with anti-c-myc antibody. Anti-Cdc2 antibody was used as loading control. (C) Pmk1p activation in wis1DD cells. Strains MI200 (pmk1-HA6H, Control), MI709 (pmk1-HA6H, wis1DD), and MI713 (pmk1-HA6H, wis1DD, atf1) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. Aliquots were harvested at the times indicated and Pmk1-HA6H was purified by affinity chromatography. Activated Pmk1p was detected by immunoblotting with anti-phospho-p42/44 and total Pmk1p with anti-HA antibodies. (D) Increased Pyp1p, Pyp2p and Ptc1p synthesis in wis1DD cells. Strains MI701 (pyp1-13myc, Control) and MI710 (pyp1-13myc, wis1DD) (top), MI702 (pyp2-13myc, Control) and MI711 (pyp2-13myc, wis1DD) (middle), MI703 (ptc1-13myc, Control) and MI712 (ptc1-13myc, wis1DD) (bottom) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. Aliquots were harvested at the times indicated, and total extracts were prepared. Pyp1-myc, Pyp2-13myc, and Ptc1p-myc fusions were detected by immunoblotting with anti-c-myc antibody. Anti-Cdc2 antibody was used as loading control.

View larger version (12K): [in this window] [in a new window]   Figure 6. Synergistic down-regulation of Pmk1p activity by Pyp1p and Ptc1p phosphatases. (A) Pmk1p basal activity in phosphatase-null mutants. Pmk1-HA6H was purified by affinity chromatography from strains MI200 (pmk1-HA6H, Control), MI213 (pmk1-HA6H, pyp1), MI216 (pmk1-HA6H, ptc1), and MI223 (pmk1-HA6H, pyp1, ptc1). Activated or total Pmk1p was detected by immunoblotting with anti-phospho-p42/44 or anti-HA antibodies, respectively. (B) Pmk1p activation in phosphatase-null mutants subjected to a saline stress. Strains MI200 (pmk1-HA6H, Control), MI223 (pmk1-HA6H, pyp1, ptc1), and MI213 (pmk1-HA6H, pyp1) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. At different times Pmk1-HA6H was purified by affinity chromatography, and either activated or total Pmk1p was detected by immunoblotting as described above.

View larger version (28K): [in this window] [in a new window]   Figure 7. Pyp1p and Ptc1p phosphatases can negatively modulate Pmk1p phosphorylation under osmotic stress in a Sty1p-independent manner. (A) Strains MI200 (pmk1-HA6H, Control), MI204 (pmk1-HA6H, sty1), MI220 (pmk1-HA6H, sty1, pyp1), MI221 (pmk1-HA6H, sty1, pyp2), and MI222 (pmk1-HA6H, sty1, ptc1) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. At timed intervals, Pmk1-HA6H was purified by affinity chromatography under native conditions, and activated or total Pmk1p was detected by immunoblotting with anti-phospho-p42/44 or anti-HA antibodies, respectively. (B) Synthesis of Pyp1p and Ptc1p phosphatases in sty1 cells. Strains MI701 (pyp1-13myc, Control) and MI706 (pyp1-13myc, sty1) (left), MI702 (pyp2-13myc, Control) and MI707 (pyp2-13myc, sty1) (middle), MI703 (ptc1-13myc, Control) and MI708 (ptc1-13myc, sty1) (right) were grown in YES medium to mid-log phase and treated with 0.6 M KCl. Aliquots were harvested at the times indicated, and total extracts prepared. Pyp1-myc, Pyp2-13myc, and Ptc1p-myc fusions were detected by immunoblotting with anti-c-myc antibody. Anti-Cdc2 antibody was used as loading control.

View larger version (35K): [in this window] [in a new window]   Figure 8. Differential regulation of MAPK functions by Pyp1p, Ptc1p, and Pmp1p. (A) Chloride sensitivity assays. Strains MI200 (Control), TK107 (sty1), MI102 (pmk1), MI107 (pmk1 sty1), MI104 (pmp1), MI116 (pmp1 sty1), MI115 (pmp1 pmk1), MI105 (pyp1), MI109 (pyp1 sty1), MI108 (pyp1 pmk1), MI106 (pyp2), MI118 (pyp2 sty1), MI117 (pyp2 pmk1), MI110 (ptc1), MI112 (ptc1 sty1), MI111 (ptc1 pmk1), and 2119 (wis1DD) were grown in YES medium to an A600 of 0.5. Samples containing 105, 104, 103 or 102 cells were spotted onto YES plates supplemented with 0.2 M MgCl2 and incubated for 3 d at 28°C before being photographed. (B) Cell wall integrity assays. Strains MI200 (Control), TK107 (sty1), MI102 (pmk1), MI104 (pmp1), MI116 (pmp1 sty1), MI115 (pmp1 pmk1), MI105 (pyp1), MI109 (pyp1 sty1), MI108 (pyp1 pmk1), MI110 (ptc1), MI112 (ptc1 sty1), and MI111 (ptc1 pmk1) were grown in YES medium to an A600 of 0.5, and the cells treated at 30°C with 100 µg/ml Zymolyase 20-T. Cell lysis was monitored by measuring A600 decay at different incubation times. Results represent the mean value of three independent experiments.

Cell Cycle-dependent Activation of Pmk1p
Microarray analysis has shown that pmk1⁺ is periodically expressed during the cell cycle, with increased transcript levels during mitosis and G1 phase (Rustici et al., 2004). We explored Pmk1p levels and its phosphorylation state during the cell cycle by introducing the Pmk1-HA6H fusion into a cdc25-22 strain. Cells from this mutant (strain MI600) were grown at 25°C to log phase, shifted to 37°C for 3.5 h to synchronize the cells in G2, and then returned to 25°C. Figure 9A indicates that the level of HA6H-tagged Pmk1p varied slightly through the cell cycle, with a small increase during G2 transition followed by a very low decrease at M and G1-S phases (Figure 9A). However, Pmk1p phosphorylation clearly changed during the cell cycle, increasing during M phase and reaching its maximum during cytokinesis (Figure 9, A and B). Notably, deletion of phosphatase Pmp1p altered the activation state of Pmk1p along the cell cycle. As seen in Figure 9, A and B, phosphorylation level of p42/44 in G2-arrested cells from strain MI602 (cdc25-22 pmp1 Pmk1-HA6H) was higher than in wild-type cells, increased moderately during cell separation, and did not decrease thereafter. A quantitative analysis of the effect of pmk1⁺ or pmp1⁺ deletion in the completion of cytokinesis indicated that whereas the septation index reached a value of 0–2% in control cells at the point of maximal cell separation, the separation was partially defective in pmk1 and pmp1 strains, with 20% cells unable to complete cytokinesis (Figure 9A). The consequence of this defect in pmk1 and pmp1 cells is a cumulative multiseptate phenotype during the second division cycle (Figure 9C, multiseptate cells marked with arrows), which is coincident with the morphological feature previously described for both mutants (Toda et al., 1996; Zaitsevskaya-Carter and Cooper, 1997; Sugiura et al., 1998).

View larger version (58K): [in this window] [in a new window]   Figure 9. Influence of Pmp1p activity on the cell cycle-dependent activation of Pmk1p. (A) Pmk1p activates periodically during the cell cycle. Top, cells from strains MI600 (cdc25-22, pmk1-HA6H) and MI602 (cdc25-22, pmp1, pmk1-HA6H) were grown to an A600 of 0.3 at 25°C, shifted to 37°C for 3.5 h, and then released from the growth arrest by transfer back to 25°C. Aliquots were taken at different time intervals, and Pmk1-HA6H was purified by affinity chromatography. Activated or total Pmk1p were detected by immunoblotting with anti-phospho-p42/44 or anti-HA antibodies, respectively, and anti-Cdc2 antibody was used as loading control. Bottom, septation index of strains MI600 (filled circles), MI601 (cdc25-22, pmk1; open circles), and MI602 (filled triangles). (B) Quantitative analysis of Pmk1p activity during cell cycle in strains MI600 (filled bars) and MI602 (empty bars) obtained from data in A. (C) Defective cell separation in pmk1 and pmp1 cells. Cells from strains MI600 (Control), MI601, and MI602 were taken at the times shown during cell cycle experiments, stained with Calcofluor white, and observed by fluorescence microscopy. Arrows indicate cells with multiseptate phenotype and defective cell separation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

View larger version (21K): [in this window] [in a new window]   Figure 10. Proposed regulatory links between the SAPK and the cell integrity pathways by protein phosphatases (shaded figures) during growth and osmostress. Pyp1p, Pyp2p, and Ptc1p protein phosphatases are negative regulators of MAPK Sty1p and MAPK Pmk1 activities, whereas Pmp1 only dephosphorylates MAPK Pmk1p (, inhibition). The expression/synthesis of the shared phosphatases is in turn under Sty1p/Atf1p-dependent and Sty1p/Atf1p-independent (X-dependent) regulation. The size of the discontinuous arrows indicates the relative transcriptional reliance. The question mark shows that the corresponding transcription factor is unknown. For more details on abbreviations and symbols, see Figure 1.

It has been recently reported that the hyperosmotic shock induces in budding yeast a delayed transient phosphorylation of Slt2p, which is not observed in strains deleted in members of the HOG pathway (García-Rodriguez et al., 2005). In striking contrast, we show that deletion of members of the SAPK pathway in fission yeast raises both the basal and the osmostress-induced Pmk1 phosphorylation. In fact, a constitutive Sty1p hyperactivation prompted by either pyp1⁺ deletion or the presence of wis1DD hyperactive allele caused a significant decrease in Pmk1p phosphorylation. Previous work has shown that the basal expression of pyp1⁺ and the stress-induced expression of pyp1⁺, pyp2⁺, and ptc1 ⁺ is triggered by the Sty1p–Atf1p pathway through a negative feedback loop (Degols et al., 1996; Shiozaki and