the ste20-like kinase psk regulates microtubule organization and

The STE20-like kinase PSK regulates microtubule organization and stability

Costas Mitsopoulos,1 Ceniz Zihni,1 Ritu Garg,2 Anne J. Ridley,2 and Jonathan D. H.

Morris1

Addresses: 1Department of Academic Surgery, GKT School of Medicine and

Dentistry, Rayne Institute, 123 Coldharbour Lane, London SE5 9NU and 2The

Ludwig Institute for Cancer Research (University College London Branch), 91 Riding

House Street, London W1W 7BS, and the Department of Biochemistry and Molecular

Biology, University College London, Gower Street, London WC1E 6BT, United

Kingdom.

The abbreviations are: CRIB, Cdc42/Rac interacting domain; ERK, extracellular

signal-regulated kinase; GCK, germinal center kinase; GSK-3β, glycogen synthase

kinase-3β; JNK, Jun N-terminal kinase; MAPs, microtubule associated proteins;

MAPK, mitogen-activated protein kinase; MLK, mixed-lineage kinase; MTs,

microtubules; PAK, p21-activated kinase; PSK, prostate-derived STE20-like kinase;

STE20, sterile 20.

To whom correspondence should be addressed: Rayne Institute, 123 Coldharbour

Lane, London SE5 9NU, UK. E-mail : [email protected]

Running title: PSK regulates microtubules

1

Copyright 2003 by The American Society for Biochemistry and Molecular Biology, Inc.

JBC Papers in Press. Published on March 13, 2003 as Manuscript M213064200 by guest on M

arch 31, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

SUMMARY

Sterile 20 (STE20) protein kinases, which include germinal center kinases (GCKs)

and p21-activated protein kinases (PAKs), are known to activate mitogen-activated

protein kinase (MAPK) pathways (JNK, p38 or ERK), leading to changes in gene

transcription (reviewed in (1)). Some STE20s can also regulate the cytoskeleton and

we have shown that the GCK-like kinase prostate-derived STE20-like kinase (PSK)

affects actin cytoskeletal organization (2). Here, we demonstrate that PSK co-

localizes with microtubules; and that this localization is disrupted by the microtubule

depolymerizing agent nocodazole. The association of PSK with microtubules results

in the production of stabilized perinuclear microtubule cables that are nocodazole-

resistant and contain increased levels of acetylated α-tubulin. Kinase defective PSK

(K57A) or the C-terminus of PSK (amino acids 745-1235) lacking the kinase domain

are sufficient for microtubule binding and stabilization demonstrating that the

protein’s catalytic activity is not required. The localization of PSK to microtubules

occurs via its C-terminus and PSK binds and phosphorylates α- and β-tubulin in

vitro. The N-terminus of PSK (1-940) is unable to bind or stabilize microtubules

demonstrating that PSK must associate with microtubules for their re-organization to

occur. These results demonstrate that PSK interacts with microtubules and affects

their organization and stability independently of PSK kinase activity.

2

by guest on March 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

INTRODUCTION

The sterile 20 (STE20)-like kinases are a large group of proteins whose catalytic

domains have homology to the STE20 kinase of Saccharomyces cerevisiae involved

in mitogen-activated protein kinase (MAPK) signaling of the pheromone response in

yeast (3). Over thirty mammalian STE20-like kinases have been identified and these

proteins divide into two subfamilies according to their structure and regulation

(reviewed in (1)). The p21-activated kinases (PAKs) have a C-terminal catalytic

domain and an N-terminal Cdc42/Rac interacting domain (CRIB); whereas the

germinal center kinase (GCK)-like kinases possess an N-terminal catalytic domain

and no CRIB. Most STE20s can act as mitogen-activated protein kinase (MAPK)

kinase kinase kinases (MAP4Ks), upstream of the Jun N-terminal kinases (JNK), p38

and/or extracellular signal-regulated kinases (ERK) MAPK signaling pathways.

Recent work has demonstrated that some STE20s can also regulate the cell

cytoskeleton. PAKs interact with Rac or Cdc42 GTPases via their CRIB domain and

act as downstream effectors for these small GTP binding proteins to regulate the actin

cytoskeleton (reviewed in (4,5)). Rac and Cdc42 stimulated morphological re-

arrangements are blocked by mutated PAK and activated PAK down-regulates actin

stress fibers and focal complexes (6-8). Some of the effects of PAK1 can be

attributed to the phosphorylation of downstream kinases such as myosin light chain

kinase which reduces its activity towards myosin light chain; and LIM kinase 1 which

phosphorylates and inactivates the actin depolymerizing protein cofilin to generate

actin clusters (9,10). X-PAK5 co-localizes to actin and microtubules (MTs) and

produces stabilized MTs that are associated in bundles, and PAK1 accumulates at the

MT-organizing center (MTOC) and along mitotic spindles during mitosis (11,12).

3


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Although GCK-like STE20s lack a CRIB domain, recent work has demonstrated that

members of this subfamily of STE20s can also regulate the actin cytoskeleton.

Proline- and alanine-rich STE20-related kinase (PASK) associates with actin, and

prostate-derived STE20-like kinase (PSK), Traf2- and Nck-interacting kinase

(TNIK) and STE20-like kinase (SLK) decrease actin stress fibers and focal adhesions

and inhibit cell spreading (2,13-15). SLK has recently been shown to associate with

MTs at the cell periphery (16).

There is increasing evidence that actin filaments and MTs are co-ordinately regulated

during the establishment and maintenance of cell polarity, division and motility. The

ability of MTs to undergo transitions between growth, shrinkage and pause are crucial

for their function and the regulation of these processes. MT dynamics and stability are

controlled by multiple factors which include MT-associated proteins (MAPs), MT-

affinity regulated kinases (MARKs), severing factors (e.g. katanin) and catastrophe

proteins (e.g. stathmin/OP18 and XKCM1) (reviewed in (17)). MAPs provide a focal

point for MT regulation and act as structural proteins which lack enzymatic activity

but promote the assembly of tubulin and MT stability. The affinity of MAPs for MTs

is controlled by their phosphorylation which causes the detachment of MAPs from

MTs and results in MT instability. A number of kinases can phosphorylate MAPs and

thereby modulate MT stability (reviewed in (17)). In contrast, phosphorylation of the

MT-destabilizing protein stathmin inactivates the protein and stabilizes MTs

(reviewed in (18)). Several kinases including PAK modulate the phosphorylation of

stathmin and may therefore regulate tubulin polymerization and MT stability (19).

The signaling pathways and components that regulate MT dynamics remain to be

determined but small GTP binding proteins and their downstream targets, which

4


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

include STE20-like kinases, appear to play crucial roles in controlling MT function.

Here we show that a recently identified member of the GCK-like family of STE20s,

PSK, co-localizes with MTs and produces stabilized perinuclear MT cables that are

nocodazole-resistant and contain increased levels of acetylated α-tubulin. These

findings suggest that a major function for PSK is to transduce signals to the MT

network.

5


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

EXPERIMENTAL PROCEDURES

Plasmids-pRK5PSK and pRK5PSK (K57A) were made previously (2). PSK1 (1-

940) and PSK1 (745-1235) were prepared by PCR amplification using appropriate

oligonucleotides and pRK5PSK as substrate and DNA products were ligated into the

pRK5 vector which provides a 5’ MYC-epitope tag.

Microinjection-Cells were grown in 10% FCS/DMEM. For microinjection, cells were

seeded on glass coverslips and incubated for 2 days. Where appropriate cells were

serum starved for 16 h prior to microinjection. Endotoxin-free plasmid DNA (100

ng/µl in PBS, Qiagen) was microinjected into the nuclei of approximately 100 cells

per coverslip. After 3-4 h cultures were fixed for 20 min with 4% paraformaldehyde,

10% DMSO and 1% glucose in PBS (37°C) and permeabilized with 0.2% Triton X-

100 for 5 min.

Immunolocalization-To detect MYC-tagged PSK, cells were incubated with 1:400

rabbit anti-Myc tag antibody (Santa Cruz Biotechnology) followed by 1:400 FITC-

conjugated goat anti-rabbit IgG (Jackson Immunoresearch). To visualize tubulin,

cells were incubated with either 1:100 mouse Cy3-conjugated anti-β-tubulin

antibody (TUB2.1, Sigma) or 1:100 mouse anti-α-tubulin antibody (DM1A, Sigma),

followed by 1:400 TRITC-conjugated goat anti-mouse IgG (Jackson

Immunoresearch). To detect acetylated α-tubulin, cells were incubated with 1:100

mouse anti-acetylated α-tubulin antibody (6-11B-1, Sigma) followed by 1:400

Alexa633-conjugated goat anti-mouse IgG (Molecular Probes). All antibodies were

incubated in 20% FCS/PBS. Cells were imaged with a confocal laser scanning

6


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

microscope (BioRad 1024). Image processing was performed using Adobe Photoshop

5.5.

Transfection-HEK293 (1.3 x 106) or COS-1 (5 x 105) cells were seeded onto 60-

mm Petri dishes and grown for 16 h before the indicated plasmids (3µg) were

transfected into cells with Lipofectin (Life Technologies, Inc.). After 5 h, transfected

cultures were returned to growth medium and incubated for 36 h.

Immunoblotting-Cells were lysed in lysis and binding buffer (1% Nonidet P-40, 130

mM NaCl, 1 mM dithiothreitol, 2 µg/ml leupeptin, 2 µg/ml aprotinin, 10 mM NaF, 0.1

mM Na3VO4, 1 mM PMSF, 20 mM Tris, pH 7.4). 100 µg of each sample was

analyzed by SDS-PAGE before proteins were transferred to nitrocellulose (Schleicher

& Schuell). Samples were processed as described previously (2) except that rabbit

anti-PSK serum was used as the primary antibody (raised against the amino acid

sequences GTLAGRRSRTRQSRALPPWR and EEEGAPIGTPRDPGDGC for PSK,

Sigma Genosys).

Immunoprecipitation-Cells were lysed as described above and 400 µg of protein were

incubated with rabbit anti-PSK serum for 1 h at 4 °C. Protein A-Sepharose beads

(Sigma) were added to each sample and after 1 h beads were pelleted by

centrifugation and washed three times in lysis and binding buffer.

In vitro kinase assays-Beads were washed once in kinase buffer (20 mM Tris-HCl

pH7.4, 4 mM MgCl2) and placed in 30µl of kinase buffer containing 40 µM ATP, 5

µCi of γ-[32P]ATP (Amersham Pharmacia Biotech) and 1µg of bovine α- and β-

7


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

tubulin (Cytoskeleton) which predominantly form unpolymerised tubulin dimers

under these conditions. After 60 min at 37 °C, kinase assays were terminated in gel

sample buffer and processed as described previously (2).

RESULTS

PSK localizes to microtubules- We have previously reported that PSK, a GCK-like

kinase, activates the JNK MAPK pathway and regulates the actin cytoskeleton and

PSK showed a punctate or filamentous distribution when expressed in fibroblasts (2).

To investigate the subcellular distribution of PSK in more detail, expression vectors

encoding MYC-epitope tagged PSK were microinjected into the nuclei of Swiss 3T3

cells and PSK was found to localize to a filamentous network in the cytoplasm of

injected cells which closely resembled MTs (Fig. 1A). The localization of PSK to

MTs was confirmed by co-staining injected cells for PSK and the MT component β-

tubulin (Fig. 1B). The immunofluorescence from PSK was more intense from MTs

positioned around the nucleus than from MTs at the periphery of the cell which

appeared fragmented (Fig. 1A, B). The staining patterns for both PSK and β-tubulin

were disrupted by addition of the MT depolymerizing agent nocodazole (5 µM)

demonstrating that PSK and β-tubulin are closely associated (Fig. 1C, D). Moreover,

the fixation of cells in formaldehyde, not paraformaldehyde/DMSO, resulted in the

8


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

disruption of MTs and a tubulovesicular staining pattern for PSK (2). Co-staining of

injected cells for PSK and markers for other cytoskeletal structures (vimentin and

actin) demonstrated that PSK localized specifically to the MT network (data not

shown).

PSK localizes to MTs via its C-terminus and alters MT organization-To determine

whether the catalytic activity and/or specific sequences outside of the kinase domain

of PSK are involved in localizing PSK to MTs, we prepared expression vectors

encoding MYC-tagged full length kinase-defective PSK (K57A) and the C-terminus

of PSK (amino acids 745-1235) for comparison with wild type PSK and the N-

terminus of PSK (amino acids 1-940) containing the kinase domain (see Fig. 2). PSK

and kinase-defective PSK (K57A) both co-localized with β-tubulin (Fig. 3A-D)

showing that the catalytic activity of the protein is not required for its localization.

PSK (745-1235), which lacks the entire kinase domain, also localized to MTs (Fig.

3E, F) whereas PSK (1-940), containing the kinase domain, failed to localize to MTs

and was cytoplasmic (Fig. 3G, H). Taken together, these results demonstrate that the

C-terminus of PSK is both necessary and sufficient for PSK to localize to the MT

network and this process occurs independently of the protein’s catalytic activity.

The localization of PSK to MTs resulted in a significant change in MT organization.

PSK generated MT cables around the nucleus that were absent in uninjected

surrounding cells (Fig. 3A, B). PSK (K57A) and PSK (745-1235), which both

localize to MTs, also produced perinuclear MT cables (Fig. 3C-F) and PSK was

predominantly localized on these structures. Moreover, MTs appeared to be thicker in

the presence of PSK and co-staining for injected PSK and pericentrin, a marker for

the MT-organizing center (MTOC), demonstrated that the MTOC was retained in

9


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

cells expressing PSK (data not shown). PSK (1-940), which is catalytically active,

failed to produce these MT structures demonstrating that PSK must directly associate

with MTs via its C-terminus for their re-organization to occur (Fig. 3G, H).

PSK stabilizes MTs and increases acetylation of α-tubulin-The ability of PSK to re-

organize MTs into perinuclear cables raised the possibility that PSK might also alter

the stability of MTs. To determine whether PSK could affect MT stability, PSK was

microinjected into cells and nocodazole (0.5 µM) was added to the culture medium

for 30 min. PSK stabilized MTs in the presence of nocodazole and prevented the loss

of MT structures observed in surrounding uninjected cells (Fig. 4A, B). The two other

forms of PSK that localize to MTs, kinase-defective PSK (K57A) and PSK (745-

1235), also produced nocodazole-resistant MTs and PSK was found wherever MTs

remained (Fig. 4C-F). MT stabilization by PSK therefore requires protein localization

but not catalytic activity and this was again illustrated by the inability of PSK (1-940)

to protect MTs from nocodazole (Fig. 4G, H).

Tubulin undergoes several post-translational modifications including the acetylation

of α-tubulin on lysine residue 40 and although this modification is a consequence not

a cause of stabilization, the acetylation of α-tubulin is frequently used to distinguish

between stable and dynamic forms of MTs (reviewed in (20)). We therefore examined

the effects of PSK on the levels of acetylated α-tubulin using anti-acetylated α-

tubulin antibodies. The three forms of PSK that localize to MTs, PSK, kinase-

defective PSK (K57A) and PSK (745-1235), each increased levels of acetylated α-

tubulin above those in neighboring uninjected cells (Fig. 5A-F). The N-terminal

kinase domain of PSK (1-940), which was unable to associate with MTs or generate

nocodazole-resistant MTs, failed to stimulate the acetylation of tubulin (Fig. 5G, H).

10


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Serum starvation of Swiss 3T3 cells appeared to result in a more cytoplasmic

localization for full length PSK, whereas C-terminal PSK (745-1235) remained

tightly associated with the MTs under identical conditions (Fig. 5A, E). Nocodazole-

resistant and acetylated MTs are increased in Swiss 3T3 fibroblasts after serum

stimulation (21,22), however, the PSK effects observed here occur in the absence of

serum.

PSK binds tubulin in vitro-Comparison of the C-terminal regulatory domain of PSK

with other sequences present in the GenBank database has shown that PSK does not

share significant homology with any other proteins or with the MT-binding domains

reported for MAPs which contain a proline-rich sequence and three pseudorepeats

(reviewed in (17)). The ability of the C-terminus of PSK (745-1235) to localize to

MTs and produce stabilized perinuclear MT cables led us to investigate whether these

sequences were able to interact with α- or β-tubulin in vitro. A C-terminal fragment

of PSK fused to glutathione-S-transferase (GST) was prepared and recombinant PSK

(amino acids 1064-1235) was able to bind α- and β- tubulin in vitro suggesting that

PSK may interact directly with MTs (Fig. 6, lanes 1-3).

Taxol dissociates PSK from MTs-The ability of PSK to bind and stabilize MTs led us

to examine the effect of taxol-stabilized MTs on PSK. Swiss 3T3 cells expressing

PSK were incubated in the presence of taxol for 30 minutes and cultures were fixed

and co-stained for PSK and MTs. Figure 7 shows that the addition of taxol causes

PSK to dissociate from MTs and become cytoplasmic (Fig. 7A, B) while the MTs

formed a separate ring around the nucleus (Fig. 7C, D). These results demonstrate that

PSK can respond to taxol-induced changes in MT dynamics and suggest that PSK

might dissociate from MTs that are sufficiently stable.

11


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

PSK phosphorylates tubulin-The localization of PSK to MTs brings the kinase into

the local vicinity of a number of potential MT-associated substrates that are regulated

by phosphorylation and these proteins could lead to additional affects of PSK on MTs.

Since PSK binds tubulin we investigated the ability of PSK to phosphorylate tubulin.

Plasmids encoding PSK or PSK (K57A) were transiently transfected into cell cultures

and immune complexes of each protein assayed for in vitro kinase activity using α-

and β-tubulin as substrates. The isoforms of tubulin were separated from each other

using SDS-PAGE/6M Urea and we found that PSK, but not kinase-defective PSK

(K57A), phosphorylated α- and β-tubulin (Fig. 8i, lanes 2 and 3).

Interestingly, addition of the MT-stabilizing agent taxol rapidly downregulated the

ability of PSK, and to a lesser extent PSK (1-940), to phosphorylate tubulin

demonstrating that the catalytic activity of PSK responds to taxol-induced changes in

MT stability (Fig. 8ii, lanes 2, 3, 5 and 6). In addition, transfected PSK, which is

normally expressed in cells as a protein doublet of 185 kDa and 165 kDa (2), was

converted to the smaller form of the protein in the presence of taxol (Fig. 8ii, lanes 2

and 5). Since 9E10 antibody detects the N-terminal MYC-epitope tag on both forms

of PSK (2), it is likely that the 185 kDa form of the protein is either truncated at the

C-terminus to generate a protein of 165 kDa that potentially lacks MT localization

sequences or alternatively PSK might undergo a change in post-translational

modification. In contrast, nocodazole had no effect on either the kinase activity or

mobility of PSK (data not shown).

12


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

DISCUSSION

We have shown that the STE20-like kinase PSK localizes to MTs and produces

stabilized perinuclear MT cables that are nocodazole-resistant and contain increased

levels of acetylated α-tubulin. The N-terminal kinase domain of PSK (1-940) is

unable to bind MTs or regulate their organization or stability, demonstrating that the

association between PSK and MTs is required for these MT alterations to occur. The

C-terminus of PSK (745-1235) contains the MT localization and regulatory domain

and recombinant PSK (1064-1235) binds α- and β-tubulin in vitro. The catalytic

activity of PSK is not required for these effects on MTs since kinase-defective PSK

(K57A) associates with MTs and regulates their organization and stability. The MT-

stabilizing drug taxol dissociates PSK from MTs and downregulates the protein’s

catalytic activity suggesting that PSK is sensitive to taxol-induced changes in MT

stability.

The stabilized perinuclear MT cables generated by PSK are similar to the MT

structures produced by another STE20, X-PAK5, which binds MTs and induces

stabilized curly MTs that form whorls around the nucleus (11). X-PAK5, like PSK,

binds to MTs via its non-catalytic regulatory domain and neither the stabilization nor

the re-organization of MTs require the protein’s catalytic activity (11). The Rho

effector Rho kinase α and DCAMKL1 also generate MT clusters around the nucleus

independently of their kinase activity and another non-enzymatic protein, mDia, co-

localizes and stabilizes MTs orientated towards the wound edge (23,24). Interestingly,

the ability of X-PAK5 to bind MTs is negatively regulated by its kinase activity since

constitutively activated X-PAK5 is unable to bind MTs and relocates to the

13


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

cytoplasm where it causes dissolution of actin stress fibers and cell retraction (11)

PSK also downregulates actin stress fibers and induces cell rounding and another

STE20 SLK associates with MTs located at adhesion sites and reduces actin stress

fibers (2,16). Interestingly, the catalytic activity of PSK is required for its effects on

actin and it is plausible that PSK’s kinase activity might act in a similar manner to X-

PAK5 and regulate its localization and function (2). How PSK kinase activity is

regulated in vivo is not known but we have been unable to detect clear differences

between the localization of PSK and kinase-defective PSK (K57A) in injected cells.

PSK does however become more cytoplasmic when cells are starved of serum

whereas the C-terminus of PSK (745-1235) lacking the kinase domain remains

tightly bound to MTs. The identification of upstream regulators for PSK and its kinase

activity will be required to explore these possibilities further.

Catalytic activity is not required by PSK to re-organize and stabilize MTs but the

localization of PSK to MTs would be expected to bring the kinase into the vicinity of

a number of potential MT-associated substrates. PSK could therefore have additional

regulatory affects on MTs that require its catalytic activity. PSK phosphorylates α-

and β-tubulin, and others have shown that Syk and Src phosphorylate tubulin (25,26).

The physiological significance of this modification and the sites of phosphorylation

are unknown, but it has been suggested that the phosphorylation of tubulin could alter

its binding properties to signaling proteins via SH2 domain-mediated interactions

and/or regulate the accessibility of kinases to their substrates (25,26).

Although a number of kinases can regulate the activity and function of MT

components such as MT-associated proteins (MAPs) by phosphorylation, only a few

kinases have been shown to localize to MTs. Interestingly, MT-associated kinases

14


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

include several components of MAPK signaling pathways such as JNK and ERK

(MAPKs), MAP kinase/ERK kinase (MEK) (MAP2K), mixed-lineage kinase (MLK)

(MAP3K), X-PAK5 and SLK (MAP4Ks) (11,16,27-29). PSK can activate JNK (2)

as can its rat homolog Tao2 (30) and it is therefore possible that this occurs on MTs.

Whether JNK is involved in regulating MT stability remains to be determined, but

JNK, ERK, p38 and GSK3β can each phosphorylate MAPs such as tau potentially

regulating its affinity for MTs and their stability (31). Moreover, the MT-stabilizing

agent taxol activates JNK via MAP3Ks such as MEK kinase 1 and apoptosis-

regulating kinase (32,33) but our finding that taxol downregulates the activity,

expression and localization of PSK indicates that PSK is unlikely to contribute to

taxol-induced JNK activation. Interestingly, ERK kinase activity has been shown to

decrease the stability of MTs (34). In contrast, phosphorylation and inactivation of the

MT-destabilizing protein stathmin stabilizes MTs and the STE20 PAK can modulate

the phosphorylation of stathmin (19), however, we have found that immune

complexes of PSK were unable to phosphorylate stathmin on serine residues 16, 25 or

38 (data not shown), suggesting that PSK does not regulate the stability of MTs via

stathmin. Another MT-associated protein, dishevelled, generates stabilized and

acetylated MTs by inhibiting glycogen synthase kinase-3β (GSK3β) and reducing its

ability to phosphorylate MAPs (35). Dishevelled activates JNK (36), as does PSK (2),

but the effects of PSK on GSK3β are not known. PSK does bind directly to MTs in

vitro whereas GSK3β has been shown to bind MTs indirectly via its association with

the neuronal MAP tau, and JNK may interact with MTs via JNK-interacting proteins

(37,38).

The ability of PSK, SLK and X-PAK5 to regulate MTs and actin filaments suggests

15


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

that these STE20 proteins could function in the regulation of both cytoskeletal

networks. Similarly, there is growing evidence that Rho family members (Rho, Rac

and Cdc42) may co-ordinately regulate both networks (39). MT growth activates

Rac, itself a MT binding protein, and promotes actin polymerization and lamellipodial

protrusion; whereas MT disassembly stimulates Rho and the formation of actin stress

fibers and focal adhesions (40,41). The Rho guanine nucleotide exchange factors

(GEFs), p190 Rho GEF, GEF-H1 and Lfc bind MTs and may locally activate Rho

(42-44). Lfc activates JNK via STE20s such as PAK or MLK (43) and it is possible

that PSK, which activates JNK, might also interact with Lfc or another MT-

associated Rho GEF. In addition, X-PAK5, which contains a CRIB domain, is

targeted away from MTs towards actin-rich structures by Rac and Cdc42 and PSK

might be regulated in a similar manner by small GTP binding proteins that do not

require a CRIB domain (11).

In conclusion, MTs play important roles in the regulation of cell morphogenesis,

division, migration and vesicle trafficking, and the ability of MTs to undergo changes

in their stability and dynamics are crucial for their function. Here we show for the first

time that a member of the GCK family of protein kinases, which lack CRIB domains,

can localize to MTs and regulate their stability and organization. These findings imply

a functional role for PSK, and perhaps other GCKs, in the regulation of MT

dynamics.

16


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

REFERENCES1. Dan, I., Watanabe, N. M., and Kusumi, A. (2001) Trends Cell Biol 11, 220-230.2. Moore, T. M., Garg, R., Johnson, C., Coptcoat, M. J., Ridley, A. J., and Morris, J. D. (2000) J

Biol Chem 275, 4311-4322.3. Herskowitz, I. (1995) Cell 80, 187-197.4. Daniels, R. H., and Bokoch, G. M. (1999) Trends Biochem Sci 24, 350-355.5. Bagrodia, S., and Cerione, R. A. (1999) Trends Cell Biol 9, 350-355.6. Dharmawardhane, S., Sanders, L. C., Martin, S. S., Daniels, R. H., and Bokoch, G. M. (1997)

J Cell Biol 138, 1265-1278.7. Manser, E., Huang, H. Y., Loo, T. H., Chen, X. Q., Dong, J. M., Leung, T., and Lim, L.

(1997) Mol Cell Biol 17, 1129-1143.8. Kiosses, W. B., Daniels, R. H., Otey, C., Bokoch, G. M., and Schwartz, M. A. (1999) J Cell

Biol 147, 831-844.9. Edwards, D. C., Sanders, L. C., Bokoch, G. M., and Gill, G. N. (1999) Nat Cell Biol 1, 253-

259.10. Sanders, L. C., Matsumura, F., Bokoch, G. M., and de Lanerolle, P. (1999) Science 283,

2083-2085.11. Cau, J., Faure, S., Comps, M., Delsert, C., and Morin, N. (2001) J Cell Biol 155, 1029-1042.12. Banerjee, M., Worth, D., Prowse, D. M., and Nikolic, M. (2002) Curr Biol 12, 1233-1239.13. Fu, C. A., Shen, M., Huang, B. C., Lasaga, J., Payan, D. G., and Luo, Y. (1999) J Biol Chem

274, 30729-30737.14. Sabourin, L. A., Tamai, K., Seale, P., Wagner, J., and Rudnicki, M. A. (2000) Mol Cell Biol

20, 684-696.15. Tsutsumi, T., Ushiro, H., Kosaka, T., Kayahara, T., and Nakano, K. (2000) J Biol Chem 275,

9157-9162.16. Wagner, S., Flood, T. A., O’Reilly, P., Hume, K., and Sabourin, L. A. (2002) J Biol Chem

277, 37685-37692.17. Drewes, G., Ebneth, A., and Mandelkow, E. M. (1998) Trends Biochem Sci 23, 307-311.18. Cassimeris, L. (2002) Curr Opin Cell Biol 14, 18-24.19. Daub, H., Gevaert, K., Vandekerckhove, J., Sobel, A., and Hall, A. (2001) J Biol Chem 276,

1677-1680.20. MacRae, T. H. (1997) Eur J Biochem 244, 265-278.21. Li, W., Zhao, Y., and Chou, I. N. (1996) Toxicol Appl Pharmacol 140, 461-470.22. Danowski, B. A. (1998) Cell Motil Cytoskeleton 40, 1-12.23. Palazzo, A. F., Cook, T. A., Alberts, A. S., and Gundersen, G. G. (2001) Nat Cell Biol 3, 723-

729.24. Lin, P. T., Gleeson, J. G., Corbo, J. C., Flanagan, L., and Walsh, C. A. (2000) J Neurosci 20,

9152-9161.25. Simon, J. R., Graff, R. D., and Maness, P. F. (1998) J Neurocytol 27, 119-126.26. Faruki, S., Geahlen, R. L., and Asai, D. J. (2000) J Cell Sci 113, 2557-2565.27. Reszka, A. A., Seger, R., Diltz, C. D., Krebs, E. G., and Fischer, E. H. (1995) Proc Natl Acad

Sci U S A 92, 8881-8885.28. Nagata, K., Puls, A., Futter, C., Aspenstrom, P., Schaefer, E., Nakata, T., Hirokawa, N., and

Hall, A. (1998) Embo J 17, 149-158.29. Willard, F. S., and Crouch, M. F. (2001) Cell Signal 13, 653-664.30. Chen, Z., and Cobb, M. H. (2001) J Biol Chem 276, 16070-16075.31. Reynolds, C. H., Betts, J. C., Blackstock, W. P., Nebreda, A. R., and Anderton, B. H. (2000) J

17


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Neurochem 74, 1587-1595.32. Wang, T. H., Wang, H. S., Ichijo, H., Giannakakou, P., Foster, J. S., Fojo, T., and

Wimalasena, J. (1998) J Biol Chem 273, 4928-4936.33. Yujiri, T., Fanger, G. R., Garrington, T. P., Schlesinger, T. K., Gibson, S., and Johnson, G. L.

(1999) J Biol Chem 274, 12605-12610.34. Harrison, R. E., and Turley, E. A. (2001) Neoplasia 3, 385-394.35. Krylova, O., Messenger, M. J., and Salinas, P. C. (2000) J Cell Biol 151, 83-94.36. Boutros, M., Paricio, N., Strutt, D. I., and Mlodzik, M. (1998) Cell 94, 109-118.37. Verhey, K. J., Meyer, D., Deehan, R., Blenis, J., Schnapp, B. J., Rapoport, T. A., and

Margolis, B. (2001) J Cell Biol 152, 959-970.38. Sun, W., Qureshi, H. Y., Cafferty, P. W., Sobue, K., Agarwal-Mawal, A., Neufield, K. D.,

and Paudel, H. K. (2002) J Biol Chem 25, 2539. Enomoto, T. (1996) Cell Struct Funct 21, 317-326.40. Best, A., Ahmed, S., Kozma, R., and Lim, L. (1996) J Biol Chem 271, 3756-3762.41. Waterman-Storer, C. M., Worthylake, R. A., Liu, B. P., Burridge, K., and Salmon, E. D.

(1999) Nat Cell Biol 1, 45-50.42. Ren, Y., Li, R., Zheng, Y., and Busch, H. (1998) J Biol Chem 273, 34954-34960.43. Glaven, J. A., Whitehead, I., Bagrodia, S., Kay, R., and Cerione, R. A. (1999) J Biol Chem

274, 2279-2285.44. van Horck, F. P., Ahmadian, M. R., Haeusler, L. C., Moolenaar, W. H., and Kranenburg, O.

(2001) J Biol Chem 276, 4948-4956.

18


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

ACKNOWLEDGEMENTS

We thank Tanya Moore, Gareth Jones, Phillip Gordon-Weeks, Brian Anderton and

Ron Senkus for many helpful discussions and Andre Sobel for anti-phospho-

stathmin antibodies.

This research was supported by the Biotechnology and Biological Sciences Research

Council, UK (Grant 29/C14086), the Association for International Cancer Research,

St. Andrews, Scotland, the Ludwig Institute for Cancer Research and by a donation

from Laura Price.

19


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

FIGURE LEGENDS

Figure 1. PSK localizes to MTs. Growing Swiss 3T3 cells were microinjected with

pRK5-MYC-PSK (wt) and after 3.5 h cells were treated with (C and D) or without

(A and B) 5µM nocodazole for 30 min and then fixed and co-stained to detect MYC-

PSK (A and C) and MTs (B and D). Cells become vesiculated under these conditions.

Bar represents 20µm.

Figure 2. PSK constructs prepared for functional characterization. The kinase domain

is shown in dark gray and the amino acids are indicated by numbers.

Figure 3. PSK localizes to MTs via its C-terminus and generates perinuclear MT

cables. Growing Swiss 3T3 cells were microinjected with pRK5-MYC-PSK (A and

B), pRK5-MYC-PSK (K57A) (C and D), pRK5-MYC-PSK (745-1235) (E and F) or

pRK5-MYC-PSK (1-940) (G and H) and after 3.5 h cells were fixed and stained to

detect MYC-PSK (A, C, E and G) and MTs (B, D, F and H). Perinuclear MT cables

were observed in approximately 75% of cells injected with PSK or PSK (K57A),

100% of cells injected with PSK (745-1235) but not in uninjected cells or cells

containing PSK (1-940).

Figure 4. PSK stabilizes MTs against disruption by nocodazole. Growing Swiss 3T3

cells were microinjected with pRK5-MYC-PSK (A and B), pRK5-MYC-PSK

(K57A) (C and D), pRK5-MYC-PSK (745-1235) (E and F) or pRK5-MYC-PSK

(1-940) (G and H) and after 3.5 h cultures were treated for 30 min with 0.5µM

20


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

nocodazole. Cells were fixed and stained to detect MYC-PSK (A, C, E and G) and

MTs (B, D, F and H).

Figure 5. PSK stimulates the acetylation of α-tubulin. Quiescent Swiss 3T3 cells

were microinjected with pRK5-MYC-PSK (A and B), pRK5-MYC-PSK (K57A) (C

and D), pRK5-MYC-PSK (745-1235) (E and F) or pRK5-MYC-PSK (1-940) (G

and H). After 3.5 h cells were fixed and stained to detect MYC-PSK (A, C, E and G)

and acetylated MTs (B, D, F and H).

Figure 6. GST-PSK (1064-1235) binds α- and β-tubulin in vitro. 3 µg of tubulin

was mixed with glutathione beads (lane 1), glutathione beads coupled to GST (lane 2)

or glutathione beads coupled to GST- PSK (1064-1235) (lane 3), pre-blocked with

20% goat serum in PBS overnight. Samples were incubated in 1 ml of kinase buffer

(1 h, 4° C), washed three times in kinase buffer and separated using SDS-PAGE /6M

urea. α- and β-tubulin were detected by immunoblotting and ECL.

Figure 7. Taxol dissociates PSK from MTs. Growing Swiss 3T3 cells were

microinjected with pRK5-MYC-PSK and after 3 h cultures were treated for 30 min

with (B and D) or without (A and C) 5µM taxol. Cells were fixed and stained to detect

MYC-PSK (A and B) and MTs (C and D).

Figure 8. PSK phosphorylates α- and β-tubulin and is downregulated by taxol. (i)

Growing COS-1 cells were transfected with either PSK (lanes 1 and 2), kinase-

defective PSK (K57A) (lane 3) or pRK5 vector (lane 4). After 36 h cell lysates were

21


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

immunoprecipitated with anti-PSK antibodies and immune complexes were subjected

to an in vitro kinase assay using α- and β-tubulin as substrate. Tubulin was omitted

from lane 1 as a control. Tubulin subunits were separated by SDS-PAGE/6M urea

and the subunits identified by immunoblotting with α- or β-tubulin specific

antibodies (DM1A or TUB2.1 respectively). (ii) HEK 293 cells were transfected with

either pRK5-MYC vector alone (lanes 1 and 4), pRK5-MYC-PSK (lanes 2 and 5) or

pRK5-MYC-PSK (1-940) (lanes 3 and 6). After 36 h cells were treated with (lanes

4, 5, and 6) or without (lanes 1, 2 and 3) 10µM taxol for 30 min. PSK was

immunoprecipitated using anti-PSK antibodies and immune complexes were

subjected to an in vitro kinase assay using bovine α- and β-tubulin as substrate.

Immunoprecipitated PSK was detected by western blotting.

22


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

by guest on March 31, 2018 http://www.jbc.org/ Downloaded from

http://www.jbc.org/


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Costas Mitsopoulos, Ceniz Zihni, Ritu Garg, Anne J. Ridley and Jonathan D. H. MorrisThe STE20-like kinase PSK regulates microtubule organization and stability

published online March 13, 2003J. Biol. Chem.

10.1074/jbc.M213064200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M213064200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M213064200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/03/13/jbc.M213064200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2003/03/13/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2003/03/13/jbc.M213064200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

the ste20-like kinase psk regulates microtubule organization and

Documents