Journ

alof

Cell

Scie

nce

Control of Sty1 MAPK activity through stabilisation ofthe Pyp2 MAPK phosphatase

Katarzyna M. Kowalczyk*, Sonya Hartmuth*, David Perera`, Peter Stansfield and Janni Petersen§

University of Manchester, Faculty of Life Sciences, C.4255 Michael Smith Building, Oxford Road, Manchester M13 9PT, UK

*These authors contributed equally to this work`Present address: Department of Oncology, University of Cambridge, Hutchison/MRC Research Centre, Hills Road, Cambridge CB2 0XZ, UK§Author for correspondence (Janni.Petersen@manchester.ac.uk)

Accepted 29 April 2013Journal of Cell Science 126, 3324–3332� 2013. Published by The Company of Biologists Ltddoi: 10.1242/jcs.122531

SummaryIn all eukaryotes tight control of mitogen-activated protein kinase (MAPK) activity plays an important role in modulating intracellularsignalling in response to changing environments. The fission yeast MAPK Sty1 (also known as Spc1 or Phh1) is highly activated inresponse to a variety of external stresses. To avoid segregation of damaged organelles or chromosomes, strong Sty1 activation

transiently blocks mitosis and cell division until such stresses have been dealt with. MAPK phosphatases dephosphorylate Sty1 to reducekinase activity. Therefore, tight control of MAPK phosphatases is central for stress adaptation and for cell division to resume. In contrastto Pyp1, the fission yeast Pyp2 MAPK phosphatase is under environmental control. Pyp2 has a unique sequence (the linker region)

between the catalytic domain and the N-terminal MAPK-binding site. Here we show that the Pyp2 linker region is a destabilisationdomain. Furthermore, the linker region is highly phosphorylated to increase Pyp2 protein stability and this phosphorylation is Sty1dependent. Our data suggests that Sty1 activation promotes Pyp2 phosphorylation to increase the stability of the phosphatase. ThisMAPK-dependent Pyp2 stabilisation allows cells to attenuate MAPK signalling and resume cell division, once stresses have been dealt

with.

Key words: Schizosaccharomyces pombe, Sty1, MAPK, Pyp2, Phosphatase, DUSP6

IntroductionIn all eukaryotes mitogen-activated protein (MAP) kinase cascades

regulate cell growth and cell homeostasis. Following changes in the

cell environment MAPKs play an important role in modulating

intracellular signalling to instigate changes in a range of processes,

including transcriptional control. Higher eukaryotes and budding

yeast have several such MAPK pathways, each of which responds

to a particular type of stress (Brewster et al., 1993; Schuller,

et al., 1994; Han et al., 1994). In contrast, in fission yeast

Schizosaccharomyces pombe one main MAPK, Sty1 (also known as

Spc1 or Phh1), is activated in response to a variety of extracellular

stimuli (Millar et al., 1995; Shiozaki and Russell, 1995a; Degols

et al., 1996; Shiozaki and Russell, 1996). Strong Sty1 activation

transiently blocks cell division to prevent segregation of damaged

organelles or chromosomes (Degols et al., 1996; Hartmuth and

Petersen, 2009). To allow for cells to adapt following stress and for

cell division to resume, Sty1 is negatively regulated by deactivating

phosphatases (Millar et al., 1995; Shiozaki and Russell, 1995a;

Shiozaki and Russell, 1995b). The phosphatases Pyp1 and Pyp2

dephosphorylate tyrosine 173 in the activation site (Dal Santo et al.,

1996; Millar et al., 1992). Levels of Pyp1 remain constant upon

exposure to stress (Chen et al., 2003). In contrast, Pyp2 protein

expression is enhanced by the Sty1-activated transcription factor

Atf1 following cell exposure to most environmental stresses

(Wilkinson et al., 1996). However, nutrient stress induces a very

rapid decline in Pyp2 levels. This is regulated through proteasome

mediated Pyp2 degradation (Petersen and Nurse, 2007). Thus, Pyp2

protein levels are under tight control.

In this report we provide evidence that the stability of the Pyp2

MAPK phosphatase is regulated through a destabilisation domain

that links the N-terminal MAPK-binding site with the C-terminal

phosphatase domain. We designate this domain the ‘linker region’.

The Pyp2 linker region is highly phosphorylated to enhance

protein stability. The absence of Sty1 activity severely impaired

Pyp2 phosphorylation and significantly reduced Pyp2 protein

stability. Thus, Sty1-dependent Pyp2 phosphorylation stabilises

the phosphatase, which in turn reduces Sty1 activity. Such a Sty1

self-regulatory mechanism allows the cell to efficiently attenuate

MAPK signalling to promote stress adaptation and cell cycle

progression.

ResultsThe Pyp2 linker region is a destabilisation domain

In order to identify potential Pyp2-specific regulatory domains,

the protein sequences of Pyp1 and Pyp2 were compared. A high

degree of homology within the N-terminal MAPK-binding site

(26% identity) and the C-terminal catalytic domains (40%

identity) was observed. Interestingly, Pyp2 has an additional

region of ,270 amino acids linking the N- and C-termini. This

sequence is absent in Pyp1 (Fig. 1A). We named this Pyp2-

specific domain the ‘linker region’. To investigate its role in

Pyp2 regulation, the linker region (a.a. 130–313; deletion of XhoI

fragment see Fig. 1A) was deleted from the genomic pyp2 locus

(pyp2.linker-free or pyp2.LF). Importantly, the shorter linker-free

Pyp2 retains phosphatase activity since we were able to generate

a pyp2.LF pyp1D double mutant (data not shown). It has

3324 Research Article

Journ

alof

Cell

Scie

nce

previously been shown that simultaneous deletion of both pyp1

and pyp2 is lethal due to high levels of Sty1 activity (Dal Santo

et al., 1996; Ottilie et al., 1992).

To allow characterisation of Pyp2 we integrated a myc-tag at the

C-termini of pyp2 and pyp2.LF with the 59 and 39 UTRs both left

intact. This C-terminal tag does not alter Pyp2 function because the

Pyp2.myc strain was phenotypically indistinguishable from wild-

type cells (data not shown). Removal of the majority of the linker

region (XhoI 130–XhoI 313; Fig. 1A) leads to a fourfold increase

in Pyp2.LF protein levels (Fig. 1B). Pyp2 protein levels in both

wild type and pyp2.LF mutants were lower than when pyp2 was

expressed from the fission yeast nmt81 promoter integrated into

the genome (Fig. 1B). This indicates that the steady state Pyp2

level in wild-type cells is very low. Ponceau S staining of total

protein is linear with protein concentration, see supplementary

material Fig. S1, and has therefore been used as a loading control.

The fourfold increase in Pyp2.LF protein levels compared to wild

type may be explained by increased Pyp2.LF protein stability.

Thus, to study the protein stability further all new protein synthesis

was blocked by the addition of cycloheximide. Time course

analysis of Pyp2 and Pyp2.LF protein levels following

cycloheximide addition, showed that Pyp2.LF levels do not drop

below 60% of the initial amount, whereas Pyp2 is reduced to less

than one third of the initial level (Fig. 1C,D). We therefore

conclude that sequences within the linker region appear to

destabilise the Pyp2 protein.

Sty1 controls Pyp2 phosphorylation and protein levels

As mentioned previously, increased Pyp2 expression following

stress is controlled by the Sty1-activated transcription factor

Atf1. In contrast, Pyp2 transcription in unstressed conditions is

Atf1 independent (Chen et al., 2003). We next explored whether

the Sty1 MAPK pathway also regulates Pyp2 protein stability.

Interestingly, Pyp2 levels were dramatically reduced in strains

compromised for Sty1 MAPK signalling (Fig. 2A). Similarly,

Pyp2 was hardly detected in wis1D mutants (Fig. 2A), which is

consistent with the role of Wis1 as the Sty1 activating kinase

(Millar et al., 1995; Shiozaki and Russell, 1995a; Degols et al.,

1996). In contrast, Pyp2 levels in wild-type and atf1D strains

were very similar. This is in agreement with previously published

micro-array studies, which showed that pyp2 mRNA levels in

unstressed cells are similar in wild type, atf1D and sty1D mutants

(Chen et al., 2003).

However, to uncouple Sty1-dependent control of protein

expression from the potential control of protein stability, pyp2

was expressed from the weak nmt81 promoter integrated into the

genome. Importantly, micro-array studies have shown that

regulation of the nmt promoter is Sty1 independent (Chen et al.,

2003). When expressed from the nmt81 promoter, Pyp2 could be

detected in the MAPK deficient mutants. However, Pyp2 protein

levels in these mutants were reduced and the protein migrated at a

faster rate (Fig. 2B). We therefore conclude that Sty1 MAPK

activity increases Pyp2 protein levels. Furthermore, Sty1 appears

Fig. 1. The Pyp2 linker region is a destabilisation domain. (A) A

schematic of the Pyp1 and Pyp2 protein structure. (B,C) Western

blot analysis of TCA-extracted total protein from early exponential

cell cultures. The arrows indicate hyperphosphorylated Pyp2 and the

arrowheads indicate hypophosphorylated Pyp2. (B, right)

quantification of relative protein levels. (C) Cells expressing

endogenous pyp2.myc (upper blot) or pyp2-LF.myc (lower blot) were

grown exponentially and treated with 100 mg/ml cycloheximide.

Samples were harvested at the times (in minutes) indicated at the

top. (D) Quantification of relative protein levels from C.

**P,0.009, ***P,0.0007.

MAPK control of MAPK phosphatase stability 3325

Journ

alof

Cell

Scie

nce

to regulate Pyp2 post-translational modifications. To study this

further, nmt81-expressed Pyp2 was immunoprecipitated from

wild-type and sty1D cultures. Treatment with lambda

phosphatase demonstrated that Pyp2 purified from wild-type

cells is phosphorylated (Fig. 2C). Importantly, no change in

protein migration was observed when Pyp2 expressed in sty1Dmutants was treated with lambda phosphatase (Fig. 2C).

Furthermore, lambda-phosphatase-treated Pyp2 expressed in

wild-type cells and Pyp2 expressed in sty1D mutants had

identical mobility on the SDS-PAGE gels (Fig. 2C). These data

suggest that Pyp2 is phosphorylated and that most of the

phosphorylation that is visualised by the mobility shift is Sty1

dependent. Additionally, mass spectrometry analysis of Pyp2

immunoprecipitates revealed that the Sty1 kinase purifies with

Pyp2 (Fig. 2E), which is consistent with previous reports

demonstrating an interaction between Sty1 and Pyp2 (Wilkinson

et al., 1999).

As mentioned above, Pyp2 protein levels are vastly reduced in

cells deleted for sty1. Interestingly, in contrast to endogenously

expressed Pyp2 (Fig. 2A), endogenously expressed Pyp2.LF

could be detected in sty1D cultures (Fig. 2D). Thus, transcription

and translation of full length pyp2 mRNA is likely to occur in

unstressed sty1D cells, but the unphosphorylated protein is

unstable. Furthermore, the mobility of Pyp2.LF in wild-type and

in sty1D cells was identical (Fig. 2D). Therefore, the Sty1-

dependent control of Pyp2 protein levels appears to be regulated

through phosphorylation of the linker region.

To further consolidate our data, which indicates that Sty1

controls Pyp2 protein levels through a mechanism independent of

its role in transcription, cycloheximide was added to wild type

and sty1D mutants in which pyp2 was expressed from the nmt81

promoter. All new protein synthesis was blocked by the addition

of cycloheximide. Time course analysis of Pyp2 following

cycloheximide addition showed that Pyp2 levels do not drop

below 50% of the initial amount in wild-type cells, whereas Pyp2

is immediately reduced to 10% of the initial level in the sty1Dmutant (Fig. 3A). A similar dependency of Sty1 on Pyp2 protein

stability was seen following drug induced Sty1 inhibition

(Hartmuth and Petersen, 2009). When Sty1 was inhibited in

cells expressing Pyp2 from the nmt81 promoter, after all new

protein synthesis was blocked by the addition of cycloheximide,

Pyp2 levels were reduced to less than 40% of the initial levels. In

comparison, Pyp2 levels in cycloheximide- and DMSO-treated

control cells did not drop below 60% of the initial amount

(Fig. 3B).

Heat stress promotes Pyp2 phosphorylation and increases

protein stability

Cells expressing pyp2.LF have increased Pyp2 phosphatase levels

due to removal of the destabilising linker sequence (Figs 1, 2).

We next examined whether this increase in MAPK phosphatase

levels altered sensitivity to stress. Interestingly, pyp2.LF mutants

are sensitive only to heat stress (Fig. 4A). This indicates that

survival following heat stress is particularly dependent on

appropriate Pyp2 regulation. Thus, the potential control of

Pyp2 levels by heat stress was investigated further. Pyp2

expressed from the nmt81 promoter, in order to prevent the

Atf1-dependent increase in pyp2 expression, was exposed to heat

Fig. 2. Sty1 controls Pyp2 protein levels. (A,B,D) Western blot

analysis of TCA-extracted total protein from early exponential cell

cultures. Arrows indicate hyperphosphorylated Pyp2 and

arrowheads indicate hypophosphorylated Pyp2. (C) Western blot

analysis of immunoprecipitated and l-phosphatase-treated

Pyp2.myc expressed from the nmt81 promoter in either wild-type

or sty1D cells. (E) Native anti-myc Pyp2 immunoprecipitates were

analysed by mass spectrometry. Numbers of unweighted spectra

and unique peptides detected by mass spectrometry are listed.

Journal of Cell Science 126 (15)3326

Journ

alof

Cell

Scie

nce

stress. After 10 minutes, Pyp2 hyperphosphorylation was observed

in wild-type cultures (Fig. 4B,C). In contrast, no Pyp2 mobility

shifts were seen in sty1D cultures, suggesting that Pyp2

phosphorylation following heat stress is Sty1 dependent

(Fig. 4B). Sty1 may therefore phosphorylate the linker region,

which leads to an increase in Pyp2 stability (Fig. 2). We next

examined whether the heat-stress induced hyperphosphorylation

could enhance Pyp2 stability. Cells expressing nmt81:pyp2 were

exposed to heat stress for 10 minutes and then returned to 28 C, at

which point cycloheximide was added. Compared to the non-

stressed control culture, heat-stress induced hyperphosphorylation

and enhanced Pyp2 protein stability (Fig. 4C,D). We therefore

conclude that heat stress activates Sty1, which phosphorylates the

Pyp2 linker region to increase the stability of the phosphatase.

To identify potential MAPK phosphorylation sites within the

linker region bioinformatics analysis was used. Seven serine or

threonine residues of potential MAPK phosphorylation sites were

mutated to alanines and integrated into the pyp2 genomic locus.

Because Pyp2 regulation is particularly sensitive to heat stress

(Fig. 4A), we screened the potential MAPK phospho-site mutants

for sensitivity to heat stress. Two residues, serine 234 and

threonine 279, both of which had high scores in phosphorylation

prediction [NetPhos 2.0 prediction (Blom et al., 1999) values:

0.957 and 0.916 for Ser234 and Thr279, respectively], showed

increased sensitivity to heat stress when mutated to alanine

(Fig. 4E). This indicates that Pyp2 serine 234 and threonine 279

in the linker region could represent potential MAPK

phosphorylation sites important for control of Pyp2 stability.

Pyp2 stability is regulated through serine 234 and

threonine 279 in the linker region

If Pyp2 is phosphorylated on serine 234 and threonine 279, pyp2

point mutations are likely to affect Pyp2 stability. Consistent with

this prediction, a reduction in protein levels was observed in the

alanine mutants of both pyp2.S234A and pyp2.T279A (Fig. 5A).

In contrast, the phosphomimetic mutants, pyp2.S234E and

pyp2.T279D, had increased Pyp2 levels and hyperphosphorylation

was observed (Fig. 5A). Addition of cycloheximide to unstressed

pyp2.S234A,T279A.myc (AA) mutants demonstrated that Pyp2

stability was reduced. In contrast, Pyp2 stability in

Fig. 3. Pyp2 phosphorylation is controlled by Sty1 MAP kinase.

(A) Left: western blots of cycloheximide-treated cell expressing

Pyp2.myc from the nmt81 promoter in wild-type or sty1D cells.

Arrows indicate hyperphosphorylated Pyp2 and arrowheads indicate

hypophosphorylated Pyp2. Right: quantification of the western

blots. Levels shown are relative to Pyp2 in non-treated cultures

(time 0 minutes). (B) Left: western blots of cells expressing

Pyp2.myc from the nmt81 promoter. Cells were treated with

cycloheximide in combination with either a MAPK inhibitor or

DMSO as a control. Right: quantification of the western blots.

Levels were calculated relatively to time-point 0 minutes.

**P,0.0015, ***P,0.0003.

MAPK control of MAPK phosphatase stability 3327

Journ

alof

Cell

Scie

nce

pyp2.S234E,T279D.myc (ED) mutants was enhanced (Fig. 5D,E)

compared to wild type (Fig. 4D). Together, these findings highlight

important roles for S234 and T279 in the control of Pyp2 stability.

To investigate the potential Pyp2 phosphorylation further we raised

phospho-specific antibodies to Pyp2.S234 and Pyp2.T279. We were

unable to raise phospho-specific antibodies capable of detecting

Pyp2.T279 phosphorylation in unstressed conditions (data not

shown), however a faint Sty1-dependent signal could be observed

following heat stress (supplementary material Fig. S2E).

Conversely, the anti-phospho-Pyp2.S234 antibodies recognised

Pyp2 in protein extracts prepared from unstressed wild-type

cultures. Importantly, no signal was observed in either

pyp2.S234A or pyp2.S234E mutants, or when pyp2 was deleted

(Fig. 5B,C). In addition, when pyp2 was expressed from the nmt81

promoter no signal was seen in sty1 or wis1 mutants (Fig. 6A). Thus,

the MAPK consensus site Pyp2.S234 is phosphorylated in a Sty1-

dependent manner. Consistent with this observation, a Sty1 in vitro

kinase assay confirmed that Sty1 could phosphorylate Pyp2.S234

(Fig. 6B). Because heat stress stimulates Sty1-dependent Pyp2

phosphorylation (Fig. 2C; Fig. 4B), this stress may boost S234

phosphorylation. Indeed, a shift from 28 C to 37 C enhanced Pyp2

S234 phosphorylation (Fig. 5C). We therefore considered whether

heat-stress-induced Pyp2 stabilisation is regulated mainly through

Pyp2.S234 and T279 phosphorylation. Protein stabilisation

following heat stress was still observed in pyp2.S234A,T279A.myc

and pyp2.S234E,T279D.myc mutants (Fig. 5D,E), suggesting that

additional Pyp2 regulation occurs following heat stress. However,

the pyp2.S234A,T279A.myc mutant reduces Pyp2 stability in

unstressed conditions (Fig. 5D,E), whereas the phospho-mimetic

mutant, pyp2S234ET279D.myc enhances Pyp2 stability. Together

our data suggests that the Sty1-dependent Pyp2 S234

phosphorylation increases Pyp2 stability following exposure to

heat stress.

Since Pyp2 phosphatase levels are reduced in the pyp2.S234A

mutant (Fig. 5A), Sty1 activity may be altered in this mutant.

Phospho-specific antibodies that recognise the active form of

Sty1 kinase (Gaits et al., 1998) showed a slight increase in Sty1

activity in both the pyp2 deletion mutant and the pyp2.S234A

mutant (Fig. 6C). This is consistent with the reduced levels of

this MAPK inhibitor. In cells with higher levels of Sty1 activity

(e.g. wis1.DD) the block of cell division, to avoid segregation of

damaged chromosomes following stress, is prolonged (Hartmuth

and Petersen, 2009). When exposed to heat stress the pyp2.AA

mutant also blocked cell division for longer than the wild-type

control (Fig. 6D). In contrast, the phospho-mimetic mutants

resumed cell division shortly after the heat-induced block.

Together our data suggest, that Sty1-dependent phosphorylation

of Pyp2 S234 and T279 is important for proper regulation of Pyp2

stability, but they are unlikely to represent the only Sty1-dependent

regulation of Pyp2 stability.

DiscussionThe fission yeast MAPK Sty1 blocks or advances cell cycle

progression depending on the level of its activation (Hartmuth

and Petersen, 2009). Therefore, tight regulation of Sty1 activity is

essential to prevent the fatal effect of excessive Sty1 activity

(Millar et al., 1995; Shiozaki and Russell, 1995a). Sty1 is

activated by the MAPK kinase Wis1 and inhibited by MAPK

phosphatases including Pyp1 and Pyp2. In contrast to Pyp1, Pyp2

levels are relatively low in unstressed cells. Additionally, the

Pyp2 protein contains a domain sandwiched between the MAPK-

binding site and the phosphatase domain, which is absent in

Fig. 4. Heat stress promotes Pyp2 phosphorylation and increases

protein stability. (A,E) Early exponentially EMM-G-grown cells

were spotted on YES, YES plus 1.5 M sorbitol, YES plus 0.6 M KCl

or YES plus 1 mM H2O2 and incubated at 30 C or 37 C as indicated.

(B,C) Western blot analysis of TCA-extracted total protein from early

exponential cell cultures. Arrows indicate hyperphosphorylated Pyp2

and arrowheads indicate hypophosphorylated Pyp2. (B) Cultures

expressing Pyp2.myc from the nmt81 promoter were incubated at

42 C for 10 minutes. (C) Cultures expressing Pyp2.myc from the

nmt81 promoter were split in two. The control was kept at 28 C, while

the other half was incubated at 42 C for 10 minutes. Afterwards both

cultures were treated with cycloheximide. Samples were harvested at

the indicated time-points. (D) Quantification of relative protein levels

from C. *P50.015.

Journal of Cell Science 126 (15)3328

Journ

alof

Cell

Scie

nce

Pyp1. We called this domain the ‘linker region’. Deletion of the

Pyp2 linker region (pyp2.LF) stabilises the phosphatase,

suggesting that sequences within this domain are responsible forregulating protein turnover. Pyp2 is phosphorylated within the

linker region and this promotes protein stability. There are a

number of possible reasons why phosphorylation may enhanceprotein stability here. First, as Pyp2 is degraded by the proteasome

(Petersen and Nurse, 2007), phosphorylation within the linker

region may block its ubiquitylation, which would normally targetthe protein for degradation. Preliminary data (supplementary

material Fig. S4) indicates that the unphosphorylated pyp2.AA

mutant has increased levels of ubiquitylation, which therefore may

contribute to the reduced protein stability. Second, Pyp2

phosphorylation is likely to change the protein charge andinfluence its conformation. Increased negative charge and/or

altered conformation may prevent Pyp2 from interacting with the

degradation machinery and therefore promote its stability.

Cells appear to be particularly reliant on the Pyp2 phosphatase

in response to heat stress, as the pyp2.LF mutant is only sensitiveto heat stress. This might be explained by the observation that

Pyp1 becomes insoluble and unable to inactivate Sty1 in responseto heat stress (Nguyen and Shiozaki, 1999), which may then

explain why cells are sensitive to altered Pyp2 regulation.

Furthermore, enhanced Atf1-dependent transcription caused by

increased Sty1 activity results in elevation of Pyp2 but not Pyp1

levels (Chen et al., 2003). Further characterisation of the pyp2.LF

mutant following heat stress showed that the cells arrest growth

at 37 C, however, they do not die (supplementary material Fig.

S2).

Our data presented here suggests that an additional,

transcription-independent control of the Pyp2 phosphatase ispresent in cells and that this relies upon Sty1 activity. Deletion of

sty1 results in a major reduction in Pyp2 protein levels. However,this is unlikely to be exclusively due to reduced transcription,

because pyp2.LF can be detected in cells deleted for sty1.

Importantly, when transcribed from the nmt81 promoter, whichis unlikely to be regulated by Sty1 signalling (Chen et al., 2003),

Pyp2 levels are reduced. Deletion of Sty1 abolished the majority of

Pyp2 phosphorylation. Because phosphorylation of the linkerregion stabilises Pyp2, the unphosphorylated phosphatase in sty1Dcells is likely to be unstable. We therefore propose a model

whereby a Sty1 self-regulatory mechanism functionsindependently of transcription (Fig. 7B). Hence, elevated Sty1

activity will promote Pyp2 phosphorylation and consequently, anincrease in phosphatase stability. This will return Sty1

phosphorylation/activity back to steady state levels when the

Fig. 5. Pyp2 stability is regulated through serine 234

and threonine 279 in the linker region.

(A–C) Western blot analysis of TCA-extracted total

protein from early exponential cell cultures. Arrows

indicate hyperphosphorylated Pyp2 and arrowheads

indicate hypophosphorylated Pyp2. (A) Western blot of

Pyp2.myc from pyp2.S234 and pyp2.T279

unphosphorylatable or phosphomimetic mutants.

(B,C) TCA-extracted protein samples from early

exponential cultures of wt, pyp2.S234A and pyp2.S234E

myc-tagged strains. (C) The cells were heat stressed at

37 C for indicated times. (D,E) Pyp2 stability in

pyp2.S234 and pyp2.T279 unphosphorylatable or

phosphomimetic mutants. Cells expressing pyp2.myc,

pyp2.S234A. T279A.myc (pyp2.AA.myc) or

pyp2.S234E.T279D.myc (pyp2.ED.myc) from the nmt81

promoter were treated as in Fig. 4C. Following western

blot analysis, relative protein levels were quantified; the

graphs represent two independent experiments.

*P,0.045.

MAPK control of MAPK phosphatase stability 3329

Journ

alof

Cell

Scie

nce

activity of the MAPK kinase Wis1 subsides. In contrast, a sudden

reduction in Sty1 activity will destabilise Pyp2. Thus, elevated

Pyp2 degradation will in turn lead to increase Sty1

phosphorylation/activation. The latter can be observed following

mild nutrient stress, as under these conditions TOR-pathway-

dependent degradation of Pyp2 promotes Sty1 activity (Petersen

and Nurse, 2007). Interestingly, TOR control of Pyp2 is controlled

through the linker region as well (supplementary material Fig. S3),

because the pyp2.LF mutant is completely unable to respond to

nutrient stress. This suggests that the Pyp2 linker region is

modified in response to several environmental stresses. However,

because the pyp2.S234 and Pyp2.T279 mutants still advance

mitosis following nutrient stress, we propose that TOR control of

Pyp2 is regulated through sites other than the Sty1 controlled S234

and T279. Therefore, control of phosphatase stability by post-

translational modification gives cells the possibility to quickly and

efficiently fine-tune MAPK activity.

Because of their physiological role in the control of MAPK

activity, misregulation of MAPK phosphatases has been linked to

diseases such as cancer (reviewed in Haagenson and Wu, 2010;

Bermudez et al., 2010). For instance, in pancreatic cancer, levels

of the MAPK phosphatase DUSP6, which is responsible for

Fig. 6. Sty1 control of Pyp2.S234 phosphorylation.

(A,C) Western blot analysis of TCA-extracted total protein

from early exponential cell cultures. Arrows indicate

hyperphosphorylated Pyp2 and arrowheads indicate

hypophosphorylated Pyp2. (B) Sty1 in vitro kinase assay, using

Pyp2-GST or GST as substrates. (C) Extracts were probed with

anti-phospho-Sty1 or anti-Sty1 antibodies. Right: signal

quantification. (D) Left: cells exponentially grown in YES were

shifted from 28 C to 37 C. At the indicated time-points samples

were collected and fixed, and the number of dividing cells

assessed; 500 cells were counted for each time-point. Right: the

mean and standard deviation of three independent experiments

are shown. *P50.022; **P,0.005; ***P,0.0007.

Fig. 7. Self-regulatory control of Sty1 kinase activity.

(A) Potential conservations of the Sty1-controlled Pyp2

phosphorylation sites within the linker of human DUSP6.

(B) Diagram of the proposed Sty1-dependent regulation of Pyp2

stability. Arrows indicate direct interactions.

Journal of Cell Science 126 (15)3330

Journ

alof

Cell

Scie

nce

deactivation of the ERK1/2 MAPKs that drive proliferation, are

reduced or absent (Furukawa et al., 2003). Similarly, in breast

neoplasms, high levels of DUSP6 are associated with resistance

to tamoxifen, a drug commonly used to treat oestrogen receptor

positive breast cancer patients (Cui et al., 2006). Therefore, not

surprisingly, the potential for MAPK phosphatases as anti-cancer

targets is being exploited in preclinical trials (reviewed in Nunes-

Xavier et al., 2011). Our data suggests that the Pyp2 linker region

is a destabilisation domain important for correct control of

MAPK phosphatase levels/activity. Interestingly five of the six

DUSP6 mutations reported in the Sanger Institute Catalogue Of

Somatic Mutations In Cancer (COSMIC; http://www.sanger.ac.

uk/cosmic) are found in the DUSP6 linker region (Bamford et al.,

2004). Finally, the MAPK-controlled Pyp2 phosphorylation sites

that we report here may be conserved in DUSP6 (Fig. 7A).

Therefore, a similar linker-dependent control of DUSP6 stability,

to fine-tune MAPK activity, is likely to be conserved in

mammalian cells.

Materials and MethodsCell cultures and strains

Strains used in this study are listed in supplementary material Table S1. Cells wereexponentially grown for 48 hours at 28 C in EMM2-N (Formedium) minimalmedium supplemented with L-glutamic acid (EMM-G) as a nitrogen source.Where necessary, L-leucine was added to a final concentration of 150 mg/ml. Forspot tests, cells were grown to a density of 3.56106 cells/ml. For western blotanalysis cells were grown to a density of 26106 cells/ml. For nutritional-stress cellswere grown as described previously (Petersen and Nurse 2007). Before heat stressof liquid cultures, cells were grown as described previously (Hartmuth andPetersen, 2009).

Microscopy

In order to determine septation index, calcofluor white (Sigma Aldrich) staining ofsepta was performed as described previously (Moreno et al., 1991). 500 cells werecounted for each time-point.

Cycloheximide and MAPK inhibitor treatment

Cultures were exponentially grown in EMM-G to a density of 26106 cells/ml andtreated with cycloheximide (Sigma-Aldrich) dissolved in DMSO at a finalconcentration 100 mg/ml or DMSO as a control. The cells were incubated furtherat 28 C. The MAPK inhibitor was used as described previously (Hartmuth andPetersen, 2009). At the indicated time-points samples were collected.

Biochemistry

Total protein extracts were prepared by TCA precipitation (Caspari et al., 2000).Myc-dependent immunoprecipitation of Pyp2 was carried out using protein GDynabeads (Invitrogen). The proteins were extracted by TCA precipitation andresolubilised in 200 ml sample buffer (80 mM Tris-HCl pH 6.8, 5 mM DTT,5 mM EDTA) plus 2% SDS. Samples were boiled for 3 minutes diluted with900 ml of sample buffer plus 1% Triton X-100 (Sigma-Aldrich). For l-phosphatase treatment the immunocomplexes were washed with (50 mM Tris-HCl pH 7.5, 100 mM sodium chloride, 0.1 EGTA, 2 mM DTT, 0.01% Brij 35)and l-protein phosphatase was added for 30 minutes at 30 C. For co-immunoprecipitation proteins were extracted and washed with IP buffer(50 mM Hepes pH 8.0, 100 mM sodium chloride, 0.1% Tween20, 1 mMEDTA, 50 mM sodium fluoride, 1 mM DTT, 1 mM PMSF, 2 mM Na3V04,20 mM sodium b-glycerophosphate, and complete protease inhibitor; Roche).Sty1 kinase assay was carried out as described previously (Nguyen and Shiozaki,1999). As substrate a 30 amino acid long Pyp2 peptide (with Ser234 located inthe middle) fused to GST (pET-41a; Novagen) was used. This fusion protein andthe GST control were expressed in E. coli. Proteins were detected using thefollowing antibodies: 1:500 4A6 anti-myc (Millipore), 1:500 anti-phospho-Sty1(raised in rabbit by Eurogentec), 1:200 anti-Hog1 antibodies (Santa CruzBiotechnology INC), 1:500 anti-phospho-pyp2.S234 (Eurogentec), 1:1500 anti-phospho-pyp2.T279 (Eurogentec), anti-ubiquitin (Dako UK Limited). Alkaline-phosphatase-coupled secondary antibodies (Sigma Aldrich) were used for allblots, followed by direct detection with NBT/BCIP (VWR) substrates on PVDFmembranes (Millipore). Co-immune-precipitates were separated on NU-PAGEgels (Invitrogen) and analysed by mass spectrometry. Signal intensities werequantified using ImageJ software. Unless otherwise stated, the graphs representthe quantified levels from three individual experiments.

AcknowledgementsWe thank the Biological Mass Spectrometry facility at ManchesterUniversity for protein identification, members of the laboratory forstimulating discussions and Elizabeth Davie for valuable commentson the manuscript.

Author contributionsK.K., S.H. and D.P. performed the experiments. K.K., S.H., D.P. andJ.P. analysed the data. P.S. performed preliminary experiments. K.K.and J.P. wrote the manuscript.

FundingThis work was supported by a Cancer Research UK project grant[grant number C10888/A9015 to J.P.]; a Cancer Research UK SeniorFellowship [grant number C10888/A11178 to J.P.]; and TheUniversity of Manchester.

Supplementary material available online at

http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.122531/-/DC1

ReferencesBamford, S., Dawson, E., Forbes, S., Clements, J., Pettett, R., Dogan, A., Flanagan,

A., Teague, J., Futreal, P., Stratton, M. R. et al. (2004). The COSMIC (Catalogue

of Somatic Mutations in Cancer) database and website. Br. J. Can. 91, 355-358.

Bermudez, O., Pages, G. and Gimond, C. (2010). The dual-specificity MAP kinase

phosphatases: critical roles in development and cancer. Am. J. Physiol. 299, C189-

C202.

Blom, N., Gammeltoft, S. and Brunak, S. (1999). Sequence and structure-based

prediction of eukaryotic protein phosphorylation sites. J. Mol. Biol. 294, 1351-

1362.

Brewster, J. L., de Valoir, T., Dwyer, N. D., Winter, E. and Gustin, M. C. (1993). An

osmosensing signal transduction pathway in yeast. Science 259, 1760-1763.

Caspari, T., Dahlen, M., Kanter-Smoler, G., Lindsay, H. D., Hofmann, K.,

Papadimitriou, K., Sunnerhagen, P. and Carr, A. M. (2000). Characterization of

Schizosaccharomyces pombe Hus1: a PCNA-related protein that associates with Rad1

and Rad9. Mol. Cell. Biol. 20, 1254-1262.

Chen, D., Toone, W. M., Mata, J., Lyne, R., Burns, G., Kivinen, K., Brazma, A.,

Jones, N. and Bahler, J. (2003). Global transcriptional responses of fission yeast to

environmental stress. Mol. Biol. Cell 14, 214-229.

Cui, Y., Parra, I., Zhang, M., Hilsenbeck, S. G., Tsimelzon, A., Furukawa, T., Horii,

A., Zhang, Z. Y., Nicholson, R. I. and Fuqua, S. A. (2006). Elevated expression of

mitogen-activated protein kinase phosphatase 3 in breast tumors: a mechanism of

tamoxifen resistance. Cancer Res. 66, 5950-5959.

Dal Santo, P., Blanchard, B. and Hoffman, C. S. (1996). The Schizosaccharomyces

pombe pyp1 protein tyrosine phosphatase negatively regulates nutrient monitoring

pathways. J. Cell Sci. 109, 1919-1925.

Degols, G., Shiozaki, K. and Russell, P. (1996). Activation and regulation of the Spc1

stress-activated protein kinase in Schizosaccharomyces pombe. Mol. Cell. Biol. 16,

2870-2877.

Furukawa, T., Sunamura, M., Motoi, F., Matsuno, S. and Horii, A. (2003). Potential

tumor suppressive pathway involving DUSP6/MKP-3 in pancreatic cancer. Am.

J. Pathol. 162, 1807-1815.

Gaits, F., Degols, G., Shiozaki, K. and Russell, P. (1998). Phosphorylation and

association with the transcription factor Atf1 regulate localization of Spc1/Sty1 stress-

activated kinase in fission yeast. Genes Dev. 12, 1464-1473.

Haagenson, K. K. and Wu, G. S. (2010). Mitogen activated protein kinase

phosphatases and cancer. Cancer Biol. Ther. 9, 337-340.

Han, J., Lee, J. D., Bibbs, L. and Ulevitch, R. J. (1994). A MAP kinase targeted by

endotoxin and hyperosmolarity in mammalian cells. Science 265, 808-811.

Hartmuth, S. and Petersen, J. (2009). Fission yeast Tor1 functions as part of TORC1 to

control mitotic entry through the stress MAPK pathway following nutrient stress.

J. Cell Sci. 122, 1737-1746.

Millar, J. B., Russell, P., Dixon, J. E. and Guan, K. L. (1992). Negative regulation of

mitosis by two functionally overlapping PTPases in fission yeast. EMBO J. 11, 4943-

4952.

Millar, J. B., Buck, V. and Wilkinson, M. G. (1995). Pyp1 and Pyp2 PTPases

dephosphorylate an osmosensing MAP kinase controlling cell size at division in

fission yeast. Genes Dev. 9, 2117-2130.

Moreno, S., Klar, A. and Nurse, P. (1991). Molecular genetic analysis of fission yeast

Schizosaccharomyces pombe. Methods Enzymol. 194, 795-823.

Nguyen, A. N. and Shiozaki, K. (1999). Heat-shock-induced activation of stress MAP

kinase is regulated by threonine- and tyrosine-specific phosphatases. Genes Dev. 13,

1653-1663.

Nunes-Xavier, C., Roma-Mateo, C., Rıos, P., Tarrega, C., Cejudo-Marın, R.,

Tabernero, L. and Pulido, R. (2011). Dual-specificity MAP kinase phosphatases as

targets of cancer treatment. Anticancer Agents Med. Chem. 11, 109-132.

MAPK control of MAPK phosphatase stability 3331

Journ

alof

Cell

Scie

nce

Ottilie, S., Chernoff, J., Hannig, G., Hoffman, C. S. and Erikson, R. L. (1992). Thefission yeast genes pyp1+ and pyp2+ encode protein tyrosine phosphatases thatnegatively regulate mitosis. Mol. Cell. Biol. 12, 5571-5580.

Petersen, J. and Nurse, P. (2007). TOR signalling regulates mitotic commitmentthrough the stress MAP kinase pathway and the Polo and Cdc2 kinases. Nat. Cell Biol.

9, 1263-1272.Schuller, C., Brewster, J. L., Alexander, M. R., Gustin, M. C. and Ruis, H. (1994).

The HOG pathway controls osmotic regulation of transcription via the stress responseelement (STRE) of the Saccharomyces cerevisiae CTT1 gene. EMBO J. 13, 4382-4389.

Shiozaki, K. and Russell, P. (1995a). Cell-cycle control linked to extracellularenvironment by MAP kinase pathway in fission yeast. Nature 378, 739-743.

Shiozaki, K. and Russell, P. (1995b). Counteractive roles of protein phosphatase 2C(PP2C) and a MAP kinase kinase homolog in the osmoregulation of fission yeast.EMBO J. 14, 492-502.

Shiozaki, K. and Russell, P. (1996). Conjugation, meiosis, and the osmotic stressresponse are regulated by Spc1 kinase through Atf1 transcription factor in fissionyeast. Genes Dev. 10, 2276-2288.

Wilkinson, M. G., Samuels, M., Takeda, T., Toone, W. M., Shieh, J. C., Toda, T.,

Millar, J. B. and Jones, N. (1996). The Atf1 transcription factor is a target for the Sty1stress-activated MAP kinase pathway in fission yeast. Genes Dev. 10, 2289-2301.

Wilkinson, M. G., Pino, T. S., Tournier, S., Buck, V., Martin, H., Christiansen, J.,

Wilkinson, D. G. and Millar, J. B. (1999). Sin1: an evolutionarily conservedcomponent of the eukaryotic SAPK pathway. EMBO J. 18, 4210-4221.

Journal of Cell Science 126 (15)3332