The serine/threonine kinase Cmk2 is required for oxidative stress response in fission yeast.

Cmk2, a fission yeast Ser/Thr protein kinase homologous to mammalian calmodulin kinases, is essential for oxidative stress response. Cells lacking cmk2 gene were specifically sensitive to oxidative stress conditions. Upon stress, Cmk2 was phosphorylated in vivo, and this phosphorylation was dependent on the stress-activated MAPK Sty1/Spc1. Co-precipitation assays demonstrated that Cmk2 binds Sty1. Furthermore, in vivo or in vitro activated Sty1 was able to phosphorylate Cmk2, and the phosphorylation occurred at the C-terminal regulatory domain at Thr-411. Cell lethality caused by overexpression of Wis1 MAPK kinase was abolished by deletion of cmk2 or by mutation of Thr-411 of Cmk2. Taken together, our data suggest that Cmk2 acts downstream of Sty1 and is an essential kinase for oxidative stress responses.

Stress-activated protein kinases (SAPKs) 1 are a conserved subfamily of MAPKs responsive to diverse environmental stress stimuli rather than to growth factors or other mitogenic stimuli (1,2). SAPKs in mammals and the fission yeast Schizosaccharomyces pombe are activated by various forms of stress (for review, see Ref. 3). In S. pombe, the SAPK Sty1/Spc1/Phh1 is activated by high osmolarity, oxidative stress, and heat shock (4 -6). Sty1 is activated through phosphorylation by the MAPK kinase (MEK), Wis1. Osmostress and oxidative stress are transmitted to Wis1 MEK by two MEK kinases (MEKKs), Wis4/Wik1/Wak1 (7)(8)(9) and Win1 (10). Farther upstream, Mcs4, a homologue of the Saccharomyces cerevisiae Ssk1 response regulator protein (11,12), is believed to regulate the Wis4 MEKK (8,9,13). In budding yeast Ssk1 acts in a multistep phospho-relay system to control the activity of the Hog1 MAPK (11,12,14), which is structurally related to the SAPK family, but it is activated only by increases in external osmo-larity (15,16). The Sty1 MAPK cascade is also controlled by a multistep phospho-relay system in response to oxidative stress but not to other forms of stress. Fission yeast Mpr1 is homologous to the Ypd1 response regulator phosphotransferase in budding yeast. Mpr1 binds to the Mcs4 response regulator and transmits oxidative stress signals to the Sty1 MAPK cascade (17,18).
The activation of Sty1 in response to stress stimulates gene expression via the Atf1 and Pap1 transcription factors, homologues of human ATF2 and c-Jun, respectively (19 -25). Atf1 is phosphorylated by Sty1 in vivo and in vitro (22), and both ⌬sty1 and ⌬atf1 mutants are defective in osmotic stress (22,26). Pap1 and Atf1 are required for the induction of ctt1 and other genes in response to oxidative stress. Ctt1 encodes cytoplasmic catalase, which decomposes hydrogen peroxide (H 2 O 2 ) and protects cells from oxidative stress (27). Oxidative stress brings accumulation of Pap1 to the nucleus in a Sty1-dependent manner (24). However, the reason why Sty1 is required for nuclear translocation of Pap1 is not known, since Pap1 is not a substrate of Sty1 (26).
Although the Atf1 and Pap1 transcription factors are key components of the fission yeast SAPK pathway, they are not the only targets for Sty1. Cells lacking Sty1 are delayed in the timing of mitotic initiation, whereas cells lacking both Atf1 and Pap1 are not (28). Here, we describe Cmk2 as a component of the fission yeast SAPK pathway. cmk2 was isolated by its sequence similarity to the yeast and mammalian calmodulin kinases. 2 It has a high degree of homology to budding yeast RCK2, previously isolated by virtue of its sequence similarity to mammalian calmodulin kinases (30). It has also been described as a suppressor of fission yeast checkpoint mutants (31) and a substrate of Hog1, the MAPK in budding yeast responsive only to osmolarity stress (32).

MATERIALS AND METHODS
Fission Yeast Strains, Media, and General Techniques-The strains used in this study are listed in Table I. The rich medium used was YES, and the selective medium was Edinburgh synthetic minimal medium supplemented with 225 mg/liter of the required amino acids (33). Yeast growth was at 30°C. Standard techniques for fission yeast genetics were used following Moreno et al. (33). Plasmid DNA was transformed by lithium acetate as described elsewhere (33).
Standard molecular biology techniques were applied (34). Restriction enzymes were used as recommended by their suppliers (New England Biolabs or MBI Fermentas). Recovery of DNA fragments from agarose gels was performed with a CLONTECH Advantage PCR pure kit, following the manufacturer instructions.
Drug Sensitivity Assay-The S. pombe strains to be assayed for sensitivity to various toxic compounds were first grown on fresh YES * This work was financially supported by Comisión Internacional de Ciéncia y Tecnologia Grant SAF97-0014 and by European Network Grant ERBFMRXCT980212 at the European Commission. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  plates, after which the cells were streaked on YES plates containing the specific compound at the indicated concentration (sodium arsenite 0.4 mM, calcium chloride 300 mM, hydrogen peroxide 0.6 mM) and incubated at 30°C for 3 days.
cmk2 Gene Disruption-The cmk2::ura4 ϩ disruptant mutant was generated by inserting a 1.8-kilobase fragment encoding the ura4 ϩ gene between the BglII-HinDIII sites of cmk2 from plasmid pVA21 (plasmid pBluescript containing the chromosomal PstI-XhoI fragment from cmk2). 2 The ura4 ϩ gene was amplified from pURA4 plasmid using VA6 and T7 oligonucleotides. VA6 is essentially the standard T3 promoter oligonucleotide with an added BglII site (underlined), cgccaaagatctattaaccctcactaaag. The amplified fragment was digested with BglII and Hin-DIII and ligated to pVA21, creating plasmid pVA24. The fragment PstI-BamHI isolated from plasmid pVA24 was used to transform the wild-type strain. Stable ura ϩ transformants were confirmed by PCR and Southern blotting.
Chromosomal Integration-To tag genomic cmk2 with two copies of the HA epitope and hexahistidine, plasmid pREP1-cmk2 2 was digested with PstI and SacI, releasing a ϳ3-kilobase fragment that contained the full nmt1-cmk2 expression cassette and was cloned into pBluescript SK Ϫ (Stratagene) digested with the same enzymes. The resulting plasmid was digested with HinDIII, which released the full nmt1 promoter and the first 215 amino acids of Cmk2, and ligated to leu2 ϩ from pREP1 plasmid digested with HinDIII. The resulting construction was linearized with SnaB1, transformed into the appropriate S. pombe strains (See Table I), and selected for Leu ϩ transformants. Plates were replicaplated four times on rich YES media and finally selected on Edinburgh synthetic minimal medium plates lacking leucine. Colonies that grew on selective media were screened for HA integration by immunoblotting with a specific anti-hemagglutinin (anti-HA) epitope antibody.
Cells containing cmk2-His6Ha did not show any phenotype compared with wild-type cells (i.e. oxidative stress). To replace the endogenous cmk2 gene by a cmk2 with a point mutation on Thr-411 to Ala, the cmk2T411A from pGEX-KG-cmk2T411A plasmid (described in next paragraph) was digested with SnaBI and NotI, releasing a fragment that contained the cmk2T411A, and ligated to plasmid pBluescript SK (described in the previous paragraph) digested with the same enzymes, and the construction was integrated as described above.
In Vivo Coprecipitation Assay-Wild-type cells were transformed with pREP41-sty1-9myc or pREP42-Cmk2-HA6His or both plasmids. Cells were grown in minimal medium for 20 h in the absence of thiamine. Pelleted cells were lysed into lysis buffer (150 mM NaCl, 50 mM Tris-HCl pH 8.0, 5 mM EDTA, 0.1% Triton X-100, 10% glycerol, 50 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 1 mM NaVO 4 , 5 g/ml aprotinin, 5 g/ml leupeptin) and purified by immunoprecipitation with anti-HA monoclonal antibody 12CA5 (Roche Molecular Biochemicals) and protein A-Sepharose beads or with anti-Myc and protein G-Sepharose (Sigma). Beads were washed with lysis buffer three times and resuspended with 35 l of 4ϫ SDS loading buffer. Proteins were resolved by SDS-PAGE, and co-precipitation was monitored by Western blotting with anti-HA monoclonal antibody 12CA5 (Roche Molecular Biochemicals) or anti-Myc.
In Vivo Kinase Assay-Phosphorylation of Cmk2 protein was monitored by Western blot analysis of HA-tagged Cmk2. The MB260, MB269, and MB264 strains (Table I) were grown in the presence of 1 mM sodium arsenite for 0, 5, 15, and 30 min. Cells were lysed by vortexing with glass beads in lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 10 mM imidazole, 1 mM EDTA, 1 mM EGTA, 0.2 mM Na 3 VO 4 , 10 mM NaF, 5 g/ml aprotinin, 5 g/ml leupeptin). The samples (15 g protein/l) were dephosphorylated by treatment with phosphatase (400 units/l, Calbiochem) for 60 min at 30°C. Cell extracts containing 100 g of total protein were run on 7.5% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Immobilon, Micropore). Membranes were probed with a monoclonal antibody to the HA epitope (12CA5, Roche Molecular Biochemicals). Phosphorylated and activated Sty1 protein was detected by Western blotting with antiphospho-p38 MAPK antibody (New England Biolabs).
In Vitro Kinase Assay-Phosphorylation of Cmk2 by Sty1 activated in vivo. The Sty1-HA6his protein was purified by immunoprecipitation with anti-HA monoclonal antibody 12CA5 (Roche Molecular Biochemicals) from yeast cells treated or not with 1 mM H 2 O 2 for 10 min in wild type or ⌬wis1 background. 5 g of GST-Cmk2KA or 5 g of the different Cmk2 forms fused to GST protein purified from E. coli were added to the purified Sty1-HA6his protein activated in vivo together with 20 M ATP and [␥-32 P]ATP (0.1 Ci/l). The mixture was then incubated for 20 min at 30°C, and the reactions were terminated by addition of SDS loading buffer. Labeled proteins were resolved by SDS-PAGE and detected by autoradiography using dried gels.
Phosphorylation of Cmk2 by Sty1 Activated in Vitro-One microgram of recombinant GST-Sty1 from E. coli was activated by phosphorylation using 0.5 g of GST-Pbs2(EE) in the presence of kinase buffer and ATP for 15 min at 30°C. 5 g of the different Cmk2 forms fused to GST protein purified from E. coli were added to the previous mixture together with [␥-32 P]ATP (0.1 Ci/l). The mixture was then incubated for 5 min at 30°C, and the reactions were terminated by addition of SDS loading buffer. Labeled proteins were resolved by SDS-PAGE and detected by autoradiography using dried gels.

Cmk2 Is Homologous to Rck2, a Hog1 Substrate in Budding
Yeast-We identified cmk2 in the fission yeast genome-sequencing project (Sanger Centre) while searching for open reading frames with homology to calmodulin-dependent kinases. Cmk2 was included in chromosome 1 cosmid (C23A1, GenBank TM accession number AL021813), and it contained  Cmk2 and Oxidative Stress three exons encoding a putative Ser-Thr protein kinase of 504 amino acids with a predicted molecular mass of 57 kDa. A computer-based amino acid sequence homology search for known proteins revealed that the greatest degree of amino acid sequence identity was shared with budding yeast RCK1 and RCK2 (CLK1) kinases (42 and 43% identity, respectively) and to calmodulin-dependent kinases (CaMKs) (40% identity to rat CaMKI and 35% to rat CaMKII). RCK1 and RCK2 were first described as suppressors of radiation sensitivity of fission yeast G 2 arrest-deficient mutants (rad1, rad3, rad9, rad17, and chk1) (31). They have a long glycine-rich insert between consensus domains VIb and VII of protein kinases, which is also present in Cmk2. Rck2p/Clk1p has also been described in budding yeast as a calmodulin kinase-like protein (30), although it does not bind calmodulin. Cells Lacking cmk2 Are Sensitive to Oxidative Stress-To examine the cellular function of Cmk2, gene disruption of cmk2 was performed. A construct in which cmk2 was replaced by the S. pombe ura4 ϩ gene was generated (see "Materials and Methods"). This construct was used to replace the genomic copy of cmk2 in a ura4-D18 strain. The correct integration of the construct in the resulting strain was confirmed by Southern hybridization analysis (data not shown). The cmk2::ura4 ϩ strain was viable, and it presented no morphological abnormalities.
To examine the role of Cmk2 in the stress response, ⌬cmk2 cells or cells lacking various components of the Sty1 MAP kinase pathway were grown on rich medium in the presence of osmotic (300 mM CaCl 2 ) or oxidative stress (0.6 mM H 2 O 2 or 0.4 mM sodium arsenite). We would like to highlight the fact that, like cells lacking either sty1 or pap1, ⌬cmk2 cells did not grow in conditions of oxidative stress caused by 0.6 mM hydrogen peroxide or 0.4 mM sodium arsenite (Fig. 1, B and C). In contrast to ⌬sty1 and ⌬atf1 cells, ⌬cmk2 cells proliferated in the presence of 300 mM CaCl 2 (Fig. 1A) or 1 M KCl (data not shown). Thus, cmk2 is required for oxidative stress response.

Cmk2 Is Phosphorylated in Vivo after Oxidative Stress in a SAPK-dependent Manner-The previous result shows that
Cmk2 is a component of the oxidative stress response. Because Sty1 is rapidly phosphorylated and activated by Wis1 after oxidative stress, we next determined whether Cmk2 is also phosphorylated after oxidative stress. Wild-type and ⌬sty1 strains were subjected to a brief oxidative stress, and endogenous expression of HA-tagged Cmk2 protein was monitored by Western blotting using anti-HA antibodies.
Under oxidative stress, Cmk2 showed slower mobility bands in addition to the main Cmk2 band ( Fig. 2A, upper panel). The slow mobility bands of Cmk2 observed at 15 and 30 min after oxidative stress were paralleled by the activation of Sty1 MAPK, as shown by Western blotting of the same samples using monoclonal antibody against phosphorylated p38 SAPK (human Sty1 homologue) ( Fig. 2A, lower panel). The slower migrating bands of Cmk2 appeared due to phosphorylation, since the altered mobility pattern was reversed on treating extracts from stressed cells with phosphatase ( Fig. 2A, upper  panel). In addition, when Cmk2 phosphorylation was studied upon oxidative stress in a mutant deficient in the SAPK pathway, ⌬sty1 strain, no slow mobility bands of Cmk2 were observed after oxidative stress compared with the wild-type strain ( Fig. 2A). Therefore, Cmk2 is phosphorylated after oxidative stress in a Sty1-dependent manner.

FIG. 2. In vivo phosphorylation of Cmk2 during oxidative stress.
A, wild-type (wt) and ⌬sty1 cells containing cmk2 HA-tagged (cmk2-HA6His) were grown and exposed to 1 mM sodium arsenite. Cells were taken at various intervals of oxidative stress, and cell extracts were prepared to detect Cmk2 by immunoblot analysis using anti-HA monoclonal antibody. The Cmk2 phosphorylation state was monitored by the appearance of slow mobility bands of the protein in wild-type cells (upper panel, first three lanes, 0Ј, 15Ј, and 30Ј) and in sty1-deleted cells (upper panel, lane marked ⌬sty1). Cell extract from 15-min stressed cells was treated with phosphatase (upper panel, lane marked 15Ј ϩ ), and Cmk2 was detected as described before. Activation of Sty1 by phosphorylation was detected from the same extracts by immunoblot analysis using anti-phospho p38 antibody (lower panel). B, wild-type cells containing a point mutation in Thr-411 to Ala of Cmk2 were subjected to 1 mM sodium arsenite. Cells were taken at various intervals of stress treatment, and cell extracts were prepared to detect Cmk2 as in A.
Cmk2 Is Phosphorylated by in Vivo and in Vitro Activated Sty1 at Thr-411-We then tested whether activated Sty1 was able to phosphorylate Cmk2. Wild-type cells or wis1-deleted cells expressing HA-tagged Sty1 were exposed to oxidative stress for 10 min, and Sty1-HA was then immunoprecipitated by using monoclonal anti-HA antibodies and protein A-Sepharose beads. The activation of Sty1-HA was assessed by Western blotting using a monoclonal antibody against phosphorylated p38 SAPK (Fig. 4A). Immunoprecipitated Sty1 was incubated in the presence of [␥-32 P]ATP and a catalytically inactive GST-Cmk2, named GST-Cmk2KA, which contains Lys-94 mutated to Ala. The use of a kinase-deficient Cmk2 was necessary to avoid autophosphorylation. As shown in Fig. 4B, Cmk2KA was significantly phosphorylated when the protein was incubated with activated Sty1. In contrast, no Sty1-dependent phosphorylation was detected when the protein was incubated with inactive Sty1 from wis1-deleted cells.
To map the phosphorylation site for Sty1 in Cmk2, we created several truncated Cmk2 alleles. Cmk2 contains four putative MAP kinase phosphorylation sites at the C-terminal domain (Thr-370, Thr-393, Thr-411, Ser-436). We generated truncated versions of Cmk2 containing different domains, one containing Thr-370 and Thr-393 (Cmk2⌬6KA, Fig. 5A) and the other containing Ther-411 and Ser-436 (Cmk2⌬7, Fig. 5A). The two Cmk2⌬6KA and Cmk2⌬7 alleles together with the kinase domain (Cmk2⌬3, Fig. 5A) and full-length Cmk2 were expressed as GST-fused proteins in E. coli and subjected to in vitro phosphorylation by Sty1 activated in vivo. Cmk2-truncated forms of Cmk2 contained the Lys-94 mutated to Ala (referred as KA) to create catalytically-deficient enzymes. The C-terminal-truncated forms containing the catalytic domain of Cmk2, Cmk2⌬3KA (from amino acids 1 to 341), and Cmk2⌬6KA (from amino acids 1 to 402) were not phosphorylated by Sty1 compared with the full-length protein (Fig. 5B). In contrast, Cmk2⌬7, which contains the last 100-residues of the C terminus, was phosphorylated by Sty1 as efficiently as the full-length, suggesting that the C-terminal regulatory domain of Cmk2 is the target of Sty1 phosphorylation and that phos-phorylation is restricted to Thr-411 or Ser-436.
We created a point mutation version to replace Thr-411 by Ala of the Cmk2⌬7 truncation (Cmk2⌬7T411A, Fig. 5A) and tested it for phosphorylation by Sty1. As shown in Fig. 5B, phosphorylation of Cmk2 by Sty1 was mainly abolished in the mutated version.
We then attempted to determine whether Sty1 directly phosphorylates Cmk2 using purified proteins in an in vitro kinase assay. For this purpose, Sty1 was purified as a GST fusion protein from E. coli (see "Material and Methods"). Purified Sty1 was incubated with a constitutively activated version of the S. cerevisiae Wis1-related kinase Pbs2 (32). In the first step of the reaction, Sty1 was activated by phosphorylation in the presence of purified Pbs2(EE) and ATP. Thereafter, Cmk2 fragments, purified from E. coli as GST fusion proteins, were added to the reaction together with [␥-32 P]ATP. Pbs2 did not phosphorylate Cmk2 (data do not shown). As shown in Fig. 5C, lane 1, full-length Cmk2 was phosphorylated directly by the Sty1 kinase. Removal of the C terminus of the protein abolished Cmk2 phosphorylation (Fig. 5C, lanes 3 and 4). Moreover, when a C-terminal polypeptide Cmk2⌬7 (amino acids 402-505) was tested, it was phosphorylated by Sty1, suggesting that the C-terminal region was indeed the main target for Sty1 phosphorylation. Interestingly, mutation of Thr-411 to Ala completely abolished phosphorylation of the C-terminal region (Fig. 5C, lane 6). Mutation of T411A in the full-length protein dramatically reduced its phosphorylation by Sty1 (Fig. 5C, lane  2). All these results indicate that Cmk2 is directly phosphorylated by Sty1 and that phosphorylation occurs mainly in the regulatory domain of Cmk2.
We also attempted to determine whether the in vivo phosphorylation of Cmk2 in response to oxidative stress was abol-  HA6His (lane 2), and both plasmids (lane 3), and Sty1-9myc and Cmk2-HA6His were expressed in the absence of thiamine. Sty1-9myc and Cmk2-HA6His were detected by Western blot using specific anti-Myc (upper panels) and anti-HA (lower panels). Expression of Sty1-9myc and Cmk2-HA6His proteins of transformed cells was detected from whole extracts (Total extracts). Sty1 co-precipitates with Cmk2 are as follows. Cmk2-HA6His was immunoprecipitated from cells extracts (Cmk2 IP, lower panel) and the presence of Sty1-9myc in the precipitates was detected (Cmk2 IP, upper panel). Cmk2 co-precipitates with Sty1: Sty1-9myc was immunoprecipitated from cell extracts (Sty1 IP, upper panel), and the presence of Cmk2-HA6His in the precipitates was determined (Sty1 IP, lower panel).

FIG. 4. In vivo activated StyI phosphorylates Cmk2.
A, wild-type (wt) and ⌬wis1 cells with HA-tagged sty1 (sty1-HA) were used to immunoprecipitate Sty1-HA by using anti-HA monoclonal antibody before (0Ј) or after (10Ј) treatment with 1 mM H 2 O 2 . The presence of Sty1-HA in the precipitates was monitored by immunoblot analysis using the anti-HA monoclonal antibody (lower panel). Activation of Sty1 by phosphorylation was monitored from the same precipitates by immunoblot analysis using anti-phospho p38 (upper panel). B, after immunoprecipitation, Sty1 was incubated with purified GST-Cmk2KA in the presence of kinase buffer and radioactive ATP. Phosphorylated proteins were separated by SDS-PAGE and detected by autoradiography. Purified GST-Atf1 transcription factor was used as a positive control for StyI phosphorylation activity.
ished in the point mutation T411A of Cmk2. A plasmid containing cmk2T411A tagged with the epitope HA (cmk2T411A-HA) was generated to replace the endogenous cmk2 and create a yeast strain containing cmk2T411A-HA under the control of the cmk2 promoter. This strain was subjected to a brief oxidative stress, and endogenous Cmk2T411A-HA-tagged protein was monitored by Western blotting using anti-HA antibodies. As shown in Fig. 2B, the phosphorylated bands of Cmk2 were abolished when Thr-411 was replaced by Ala, indicating that Thr-411 is the unique phosphorylation site for Sty1-dependent modification of Cmk2 after oxidative stress.
Deletion of Cmk2 Suppresses Cell Lethality Caused by Hyperactivation of the MAPKK Wis1-To further confirm the genetic data of the involvement of Cmk2 in Sty1 signaling, we tested whether deletion of cmk2 alters the Sty1 response. Overexpression of Wis1 results in cell lethality (4,5,35). This lethality can be prevented by deletion of downstream elements like the Sty1 MAPK. We thus tried to determine whether cell lethality caused by overexpression of Wis1 was suppressed by deletion of cmk2 and the phosphorylation site mutant cmk2T411A.
Wis1 was overexpressed in wild-type cells, in cmk2-deleted cells, and in cells in which the endogenous cmk2 gene was replaced by cmk2T411A. As shown in Fig. 6, both deletion of the cmk2 gene (Fig. 6A) or mutation of the Sty1 phosphorylation site of Cmk2 (Cmk2T411A) (Fig. 6B) partially suppressed cell lethality caused by overexpression of Wis1, further supporting Cmk2 as a direct element of the Sty1 pathway that acts downstream of the Sty1 MAPK. DISCUSSION We have identified Cmk2 kinase as a new component of the fission yeast oxidative stress-activated Sty1 MAP kinase response. One central observation is that Cmk2 kinase is essential for oxidative stress responses. Cmk2-deleted cells are sensitive to oxidative stress but not to osmotic, pH, or temperature stress. Furthermore, Cmk2 is a substrate of Sty1 MAPK. Cmk2 binds Sty1, which phosphorylates it in vivo and in vitro after oxidative stress activation. In addition, the biochemical and physiological level of Cmk2 phosphorylation by Sty1 depends on a single phosphorylation site. Finally, cell lethality caused by hyperactivation of Wis1 MAPKK can be suppressed by deletion of cmk2 or by mutation of the Cmk2 site phosphorylated by Sty1. This suggests that Cmk2 and Cmk2 phosphorylation by Sty1 are necessary for the MAPK response.
Fission yeast Cmk2 is homologous to budding yeast Rck2, which is a direct substrate of Hog1 MAPK (32,36). Hog1 is specific for the cellular response of osmotic stress in budding yeast. Although Rck2 binds and is phosphorylated by Hog1 MAPK, Rck2-deleted cells do not show increased sensitivity to osmotic stress (32,36). In contrast to the Hog1 kinase in bud- FIG. 5. Sty1 phosphorylates Thr-411 at the C-terminal domain of Cmk2. A, graphical representation of various truncated and point mutation forms created and expressed in E. coli as GST fusion proteins. B, the full-length and truncated recombinant tagged Cmk2 proteins described in A were purified from E. coli as described under "Materials and Methods." Equal concentration of Cmk2 forms were incubated with immunoprecipitated Sty1 from 10-min-stressed cell extracts (10Ј, 1 mM H 2 O 2 ) or nonstressed extracts (0Ј) in the presence of radioactive ATP. Phosphorylated proteins were resolved by SDS-PAGE gels and detected by autoradiography. The position of the phosphorylated Cmk2 is indicated on the left. C, Cmk2 is phosphorylated by in vitro activated Sty1. Various Cmk2 fragments were tested for their ability to be phosphorylated in an in vitroactivated Sty1 (as described under "Materials and Methods"). After in vitro kinase assay, phosphorylated proteins were resolved by SDS-PAGE and detected by autoradiography. Positions of the Cmk2 full-length or Cmk2⌬7 are shown on the right. Proteins were GST-tagged and contained the KA mutation to prevent autophosphorylation. ding yeast, Sty1 is activated by multiple environmental stresses including osmotic stress, heat shock, H 2 O 2 , UV light, certain DNA-damaging agents, and the protein synthesis inhibitor anisomycin (5,8,37). Like Rck2, Cmk2 is a substrate of Sty1 MAPK and in fission yeast Cmk2 is required for the cellular response to oxidative stress, as illustrated by the fact that cells lacking Cmk2 proliferate under oxidative stress.
Which is the role of Cmk2 in the oxidative stress response? In fission yeast, the Pap1 transcription factor is a target of Sty1 MAPK in oxidative stress conditions (24,25). Pap1 is required for the induction of catalase (ctt1), thioredoxin reductase (trr1), and other genes in response of oxidative stress. In addition, oxidative stress brings about nuclear accumulation of Pap1 in a Sty1-dependent manner (24). However, Pap1 is not a substrate of the Sty1 MAPK (26). Thus, the regulation of Pap1 by Sty1 is not understood. We investigated whether Cmk2 is involved in the regulation of Pap1 transcription activation or nuclear localization, but these are not affected by Cmk2. Loss of Cmk2 did not block the nuclear accumulation of ectopically expressed green fluorescent protein-Pap1 fusion protein (data not shown). Furthermore, Pap1 was not phosphorylated in vitro by Cmk2, and we have also confirm that Pap1 is not phosphorylated by purified in vitro activated Sty1 (data not shown). Thus, Cmk2 is not required for the induction of Pap1dependent gene transcription or Pap1 cellular localization.
In addition to the phosphorylation of transcription factors, MAP kinases are known to activate downstream protein kinases involved in several cellular processes. These include kinases such as MAP kinase-activated protein kinases (MAP-KAP-K2 and MAPKAP-K3), MAP kinase signal-integrating kinase (MNK), p38 regulated-activated kinase (PRAK), and mitogen-and stress-activated kinase (MSK) (29,38,40). Their activation results in the phosphorylation of both the transcription factors CREB and ATF1 and the essential proteins Hsp27 and eIF1e (38,39). During the preparation of this manuscript, in vitro phosphorylation of EF-2 by Rck2 budding yeast kinase has been reported (36), which suggests a role for Rck2 in translation. Although the environmental agents that stimulate Rck2 and Cmk2 kinases differ, a similar scenario may be found in fission yeast, in which Sty1 may regulate translation through Cmk2 to display oxidative stress-induced responses. Current studies are under way to identify Cmk2 substrates and further understand the Cmk2-mediated oxidative stress responses.