(The FASEB Journal. 2000;14:6-16.)
© 2000 FASEB
Dual specificity phosphatases: a gene family for control of MAP kinase function
MONTSERRAT CAMPS,
ANTON NICHOLS and
STEVE ARKINSTALL1
Serono Pharmaceutical Research Institute, Ares-Serono International SA, 1228 Plan-les-Ouates, Geneva, Switzerland
1Correspondence: Serono Pharmaceutical Research Institute, Ares-Serono International SA, 1228 Plan-les-Ouates, Geneva, Switzerland. E mail. steve.arkinstall@serono.com
 |
ABSTRACT
|
|---|
Mitogen-activated protein (MAP) kinases are important players in signal
transduction pathways activated by a range of stimuli and mediate a
number of physiological and pathological changes in cell function. MAP
kinase activation requires phosphorylation on a threonine and tyrosine
residue located within the activation loop of kinase subdomain VIII.
This process is reversible even in the continued presence of activating
stimuli, indicating that protein phosphatases provide an important
mechanism for MAP kinase control. Dual specificity phosphatases (DSPs)
are an emerging subclass of the protein tyrosine phosphatase (PTP) gene
superfamily, which appears to be selective for dephosphorylating the
critical phosphothreonine and phosphotyrosine residues within MAP
kinases. Some DSPs are localized to different subcellular compartments
and moreover, certain family members appear highly selective for
inactivating distinct MAP kinase isoforms. This enzymatic specificity
is due in part to powerful catalytic activation of the DSP phosphatase
after tight binding of its amino-terminal to the target MAP kinase. DSP
gene expression is induced strongly by various growth factors and/or
cellular stresses, providing a sophisticated transcriptional mechanism
for targeted inactivation of selected MAP kinase activities.—Camps,
M., Nichols, A., Arkinstall, S. Dual specificity phosphatases: a gene
family for control of MAP kinase function.
Key Words: MAP kinases • protein phosphatases • dual • specificity phosphatases • signal transduction
|
INTRODUCTION
|
|---|
PROTEIN PHOSPHORYLATION ON serine, threonine, or
tyrosine residues has emerged as a critically important
posttranslational modification at the heart of regulatory mechanisms
controlling cell activities. Levels of protein phosphorylation are
dictated by the coordinated activities of protein kinases and protein
phosphatases, and the large number of genes encoding these enzymes
testifies to their fundamental importance in the control of normal cell
and body physiology. In Caenorhabditis elegans, for example,
2.2% of all genes encode protein kinases while ~0.5% appear to be
protein tyrosine phosphatases (PTPs) (1)
. A similar
proportion of genes are also committed to direct control of protein
phosphorylation in the budding yeast Saccharomyces
cerevisiae (1.9% of all genes are protein kinases and 0.23% are
PTPs) (2)
. Extrapolation from these organisms predicts
that the human genome may encode 2000 protein kinases and up to 500
PTPs, providing a complex diversity of enzymes for controlling protein
phosphorylation and orchestrating essential changes in cell function.
|
MAMMALIAN MITOGEN-ACTIVATED PROTEIN KINASES
|
|---|
Mammalian protein kinases fall into two major families that
phosphorylate either serine and/or threonine residues (serine/threonine
kinases) or tyrosine residues (tyrosine kinases). One major role for
cellular kinases is their participation in signal transduction pathways
through which cells respond functionally to external messages or to
extracellular stresses. Among these, a specialized family of
serine/threonine mitogen-activated protein (MAP) kinases are activated
by a range of diverse stimuli (3
4
5
6)
. These display a high
level of evolutionary conservation and are essential for many cell
functions throughout the animal kingdom. These include mating in yeast,
morphogenesis in slime mold, vulval development in worms, eye
development in flies, mesoderm induction in frogs, as well as
lymphocyte differentiation and neuronal death in mammals (5
, 7
8
9
10)
. Extracellular signal-regulated kinase-1 (ERK1) and ERK2,
also known as p44 and p42 MAP kinases, represent the prototypical MAP
kinases in mammalian cells. ERK MAP kinase catalytic activation
requires phosphorylation by an upstream dual specificity MAP kinase/ERK
kinase (MEK) exemplified by MEK-1, which in turn is phosphorylated and
activated by a MAP kinase kinase kinase of the Raf family. Of these,
c-Raf-1 has been studied extensively and found to undergo activation at
the plasma membrane after binding to the GTP-bound form of the low
molecular weight G-protein p21ras (3
4
5
6)
. In all, four
mammalian MAP kinase cascades are currently recognized that, in
addition to ERK, include c-Jun NH2-terminal
kinase/stress-activated protein kinase (JNK/SAPK), p38/RK/CSBP (p38),
and big mitogen-activated protein kinase-1/ERK5 (BMK-1/ERK5) pathways
(3
4
5
6
, 11)
. As with ERK, these MAP kinases are activated
by phosphorylation that occurs at a specific threonine and tyrosine
residue localized within the activation loop motif TxY of
kinase subdomain VIII (where x represents glutamate in ERK,
proline in JNK/SAPK, and glycine in p38 MAP kinases). Multiple genes
and splice variants for each MAP kinase as well as a number of specific
upstream activating kinases responsible for MAP kinase phosphorylation
and activation have been identified; these are illustrated in
Fig. 1
.
View larger version (57K):
[in this window]
[in a new window]
|
Figure 1. Mammalian MAP kinase cascades. Schematic representation of mammalian
MAP kinase signal transduction pathways. MAP kinases are boxed together
and include the extracellular signal-regulated kinase (ERK), c-Jun
NH2-terminal kinase (JNK)/stress-activated protein kinase
(SAPK), p38, and big MAP kinase-1 (BMK1)/ERK5 subfamilies. Each MAP
kinase is activated after phosphorylation by specific upstream MAP
kinase kinases as indicated by the solid arrows. MAP/ERK kinase-1
(MEK1) and MEK2 activate the ERKs, JNK kinase-1 (JNKK1) and JNKK2
activate JNKs, MAP kinase kinase-3 (MKK3) and MKK6 activate the p38’s,
whereas MEK5 activates BMK1/ERK5. For several MAP kinase and MAP kinase
kinase genes, several alternative splice variants have been reported.
MAP kinase activation by many diverse stimuli has been reported.
Generally, ERKs are activated strongly by polypeptide growth factors as
well as by tumor-promoting phorbol esters and some oncogenes. By
contrast, JNK and p38 MAP kinases are activated by cell stresses (e.g.,
UV light, DNA damaging agents, oxidants, osmotic and heat shock),
cytokines, tumor necrosis factor- , Fas, T cell receptor (TCR)/CD28
costimulation, CD40 ligation, and microbial lipopolysaccharide (LPS).
Factors stimulating BMK1/ERK5 are unclear although activation by
epidermal growth factor (EGF), serum, hyperosmolarity, and oxidative
stress has been reported. MAP kinases phosphorylate serine or threonine
residues preceding a proline, although additional docking sites outside
of the kinase active site also appear to be important in defining
substrate selectivity. Some established target substrates for MAP
kinases include the transcription factors Elk-1, SAP-1, c-Jun, MEF-2C,
CHOP and PPAR , the microtubule regulator stathmin, cytosolic
phospholipase A2 (cPLA2), as well as the downstream target kinases MAP
kinase-activated protein kinase-1 (MAPKAP-K1, also known as ribosomal
p90 S6 kinase or RSK), mitogen- and stress-activated protein kinase-1
(MSK-1), MAPKAP-K2, MAPKAP-K3, MAP kinase-integrating kinase-1 (Mnk1),
and Mnk2. MAP kinase kinase activation (broken arrows) is mediated by
numerous upstream kinases acting together with various GTP binding
proteins as well as complex-forming adaptor proteins
(3
4
5
6)
.
|
|
Different cell stimuli activate preferentially distinct MAP kinases.
Hence, many growth factor and G-protein linked receptors, cell
adhesion, phorbol esters, and some oncogenes are linked to activation
of the ERKs, whereas inflammatory cytokines, trophic factor
deprivation, and a number of cell stresses lead preferentially to
activation of JNK/SAPK and p38 MAP kinases (3
4
5
6)
. Less is
known about BMK-1/ERK5 responsiveness, although this MAP kinase has
been reported to respond to both growth factors and stressful stimuli
(11)
(Fig. 1)
. MAP kinases target several substrate
proteins for phosphorylation, including additional ‘downstream’
serine/threonine kinases (which themselves become activated),
cytoskeletal elements, regulators of cell death, as well as a number of
nuclear receptors and transcription factors (Fig. 1)
. Such an array of
substrates suggests a key role for MAP kinases in controlling diverse
cell functions; recent studies using chemical inhibitors and dominant
inhibitory or active mutant kinases, as well as analysis of mice after
gene deletion, support this. For example, ERK MAP kinases are involved
in cellular chemotaxis, cell cycle progression and mitogenesis,
oncogenic transformation and metastasis, neuronal differentiation and
survival, and in processes underlying memory and learning (3
4
5
6
, 12
13
14
15
16
17
18
19)
. One recent study suggests that BMK-1/ERK5 can also
participate in pathways leading to cell cycle progression and
proliferation (11)
. Conversely, JNK/SAPK and p38 MAP
kinases are important in pathways controlling T cell differentiation,
production of inflammatory cytokines and eicosanoids, and apoptotic
cell death (3
, 8
9
10
, 19
20
21
22
23
24)
. These observations indicate
that mechanisms controlling MAP kinases activation state are of
critical importance to several diverse aspects of normal and
pathological cell functions. In this regard, studies over recent years
have provided clear evidence for important and specific roles for
upstream kinases in phosphorylating and activating MAP kinases in
response to extracellular stimuli (3
4
5
6)
(Fig. 1)
. In
addition to stimulatory input from such MAP kinase kinases, however,
activation of MAP kinase is a reversible process even in the continued
presence of activating stimuli, indicating that protein phosphatases
are also likely to provide an important mechanism for control.
|
MAP KINASE INACTIVATION BY THREONINE OR TYROSINE DEPHOSPHORYLATION
|
|---|
Dephosphorylation of either the threonine or tyrosine residue
within the MAP kinase activation loop TxY motif alone can
result in their enzymatic inactivation. In intact cells,
dephosphorylation and inactivation of MAP kinases occur, with kinetics
ranging from minutes to several hours depending on the cell type and
activating stimulus. In endothelial cells, for example, exposure to
serum leads to ERK activation that is sustained at high levels for over
2 h. In contrast, different patterns can be observed in the
pheochromocytoma PC12 cell line where EGF-stimulated ERK activation is
transient, with inactivation initiated within 5 min and nearly complete
within 15–30 min, whereas this MAP kinase displays prolonged
activation over several hours on stimulation with NGF
(25)
. It is believed that different temporal patterns of
ERK MAP kinase activation elicited by EGF and NGF underlie their
differential effects to drive either cellular proliferation or
differentiation, respectively (6)
. Using PC12 cells as a
model system to identify key phosphatases suppressing ERK activation,
biochemical studies revealed that early rapid inactivation of these MAP
kinases reflects, in part, threonine dephosphorylation by the PPP
serine/threonine protein phosphatase PP2A (25)
(Fig. 2
). In addition to threonine dephosphorylation, these studies also
indicated that tyrosine-specific protein phosphatases (PTPs) also
contribute to ERK MAP kinase inactivation (25)
. Currently,
50 or more PTPs have been characterized (26
27
28)
, and
although the PC12 cell PTPs were not identified molecularly
(25)
, recent studies in other cell types have identified a
possible role for three related PTP gene family members. Hence, PTP-SL
(also known as PCPTP1) and STEP are two homologous neuronal PTPs
existing in both transmembrane and cytosolic forms that bind tightly as
well as dephosphorylate and inactivate ERKs (29)
. HePTP
(same as LC-PTP) is an additional cytosolic PTP present in lymphoid
tissue also reported recently to bind and inactivate this class of MAP
kinase (30)
(Fig. 2)
. PTP-SL, STEP, and HePTP appear to
act as functional homologues of the tyrosine-specific PTPs Ptp2 and
Ptp3 as well as Pyp1 and Pyp2, identified genetically as
down-regulators of MAP kinases in budding and fission yeast,
respectively (31
32
33
34)
.
Notwithstanding the importance of these early reports on PP2A and
tyrosine-specific PTPs inactivating ERKs, little is known of their
general importance in terminating MAP kinase signaling, of the
molecular mechanisms that may control their phosphatase catalytic
activity, or of their specificity for inactivating different MAP kinase
isoforms. In contrast to these protein phosphatase classes, there has
been significant and rapid progress in our understanding of the role
played by a subclass of PTP that possess activity for dephosphorylating
both phosphotyrosine and phosphothreonine residues, hereafter termed
dual specificity phosphatases (DSPs). It is our current understanding
of MAP kinase control by DSPs that is the focus of this review.
|
THE DSP FAMILY
|
|---|
The prototypical DSP is the VH1 gene identified in 1991 as an open
reading frame in vaccinia virus expressed in late-stage viral infection
(35)
. In 1992, the first mammalian DSP was identified as
the mouse immediate early gene 3CH134 (later also cloned as Erp) or its
human orthologue CL100, which is induced rapidly after exposure to
growth factors, heat shock, or oxidative stress (36
37
38)
.
Sequence homology with VH1 suggested that CL100/3CH134 was a tyrosine
phosphatase and this was confirmed after its expression in bacteria.
Recombinant CL100/3CH134 was also shown to dephosphorylate threonine
and tyrosine residues of ERK MAP kinases, and this was paralleled by
kinase inactivation. These studies showed that CL100/3CH134 was
specific for dephosphorylation of ERK MAP kinases when compared to a
number of other unrelated phosphoproteins (36
, 39
40
41)
. In
mammalian cells, a correlation between 3CH134 levels and ERK
inactivation suggested specificity for inactivation of MAP kinases,
leading to its renaming as MAP kinase phosphatase-1 (MKP-1)
(42)
.
Despite this important early work, it is now unclear whether
CL100/MKP-1 is a physiological regulator of the ERKs rather than other
subclasses of MAP kinase. First, we now recognize that MKP-1 is at
least as effective in inactivating JNK/SAPK and p38 when compared to
ERK MAP kinases (43
, 44)
. Second, ERK activity is
apparently normal after deletion of the CL100/MKP-1 gene in mice
(45)
. Third, newly identified members of the DSP gene
family appear highly selective for ERK and may represent the true
physiological regulators of this MAP kinase isoform (see below).
Since the initial cloning of CL100/MKP-1, eight additional mammalian
DSP gene family members have been identified and characterized. These
include PAC1 (46
, 47)
, hVH-2/MKP-2 (also cloned as TYP-1)
(48
49
50)
, hVH3/B23 (51
, 52)
, hVH-5 (human
orthologue of M3/6) (53
, 54)
, MKP-3/PYST1 (same as rVH6)
(55
56
57)
, B59 (full-length clone of PYST2 and human
orthologue of MKP-X) (56
57
58)
, MKP-4 (59)
,
and MKP-5 (60)
(Fig. 3
). These DSPs all appear to be effective in mediating inactivation of
MAP kinases. VHR is an additional mammalian homologue of VH1
(61)
, but it lacks an amino-terminal extension identified
in all other DSP gene products that is now known to be important for
targeting specific MAP kinase substrates (see below). VHR also
dephosphorylates growth factor receptor tyrosine kinases and
appears to be relatively inactivate against MAP kinases (58
, 61)
. For these reasons, we consider VHR to represent a distinct
DSP subclass and so it is not considered further here.
View larger version (29K):
[in this window]
[in a new window]
|
Figure 3. Schematic representation of human dual specificity phosphatase (DSP)
amino acid sequence homologies. Numbering indicates first and last
residue of the amino- and carboxyl-terminal halves of each DSP. The
predicted amino acid sequence of these domains was used for independent
calculation of percentage sequence identity with the corresponding
region of CL100/MKP-1, using the SIM pairwise sequence alignment
program. Human sequences used for comparisons were obtained from the
GeneBank/EMBL data bases, using the following accession numbers:
CL100/MKP1, X68277; hVH2, U21108; hVH3, U16996; hVH5, U27193; PAC1,
L11329; MKP-3/PYST1, X93920; B59, X93921; MKP-4, Y08302; MKP-5,
AB026436.
|
|
|
WHERE ARE DSPs EXPRESSED?
|
|---|
In studies of DSP expression, some gene family members have been
found to be restricted to a subset of tissue types. These include PAC1,
which is enriched in hematopoietic cells, hVH-5, which is expressed
predominantly in brain, heart and skeletal muscle, MKP-5, which is
detected only in liver and skeletal muscle, and MKP-4, which was found
only in the placenta, kidney, and embryonic liver (46
, 47
, 53
, 54
, 59
, 60)
(Table 1
). Other DSPs also reveal a restricted expression pattern when studied
by in situ analysis. For instance, although Northern
blotting reveals that CL100/MKP-1, B29 (MKP-X), MKP-3/PYST1, and
hVH-3/B23 are all present in adult rat brain, more detailed analysis in
brain slices demonstrates a highly localized and distinct expression
pattern for these DSP gene family members (62)
. In
addition, DSPs can also occupy distinct subcellular compartments.
Hence, whereas CL100/MKP-1, PAC-1, hVH-2/MKP-2, and hVH-3/B23 are all
localized within the nucleus, MKP-3/PYST1 appears to be exclusively
cytosolic (46
, 48
, 51
, 56
, 57)
. MKP-4, in contrast, is
present in cytosol as well as in punctate nuclear bodies colocalizing
with the promyelocytic protein (59)
. Other DSPs are
different, again as illustrated by M3/6, which may be either nuclear or
cytosolic depending on its cellular environment (54)
.
|
TRANSCRIPTIONAL CONTROL OF DSPs
|
|---|
Another interesting feature of DSPs is their tight and rapid
transcriptional induction by growth factors and/or cellular stresses.
Indeed, CL100/MKP-1 was identified originally as an immediate early
gene induced rapidly within 30–120 min in cultured cells by mitogens,
heat shock, or oxidative stress (36
, 37)
. Other DSP genes
also undergo transcriptional up-regulation in cell lines exposed to
growth factors and/or cellular stress (summarized in Table 1
). In some
instances, DSP gene induction appears specific for a given stimulus. In
PC12 cells, for instance, CL100/MKP-1, hVH-2/MKP-2, and hVH-5 all
undergo rapid induction as immediate early genes in response to
mitogenic factors whereas MKP-3/PYST1 expression is stimulated
exclusively by agents promoting neuronal differentiation, most
particularly by NGF (50
, 53
, 54
, 56
, 63)
. Tight
transcriptional control of DSP genes also occurs in vivo as
indicated by studies in rats undergoing kainic acid-induced seizure
activity where CL100/MKP-1, hVH-3/B23, PAC-1, and MKP-3/PYST1 are all
reversibly up-regulated in discrete brain regions. Under these
conditions of up-regulation, different DSPs map to distinct brain areas
displaying either neuronal plasticity, apoptotic death, or survival
(62
, 64)
. CL100/MKP-1 also displays transient and highly
regionalized induction within specific brain regions during normal
embryonic development (62)
. Mechanisms controlling DSP
induction are likely to be complex although expression of some gene
family members, including CL100/MKP-1, hVH-2/MKP-2, and PAC1, is
reported to be dependent at least in part on MAP kinase activation
(65
66
67)
. This provides clear scope for negative feedback
of the inducing MAP kinase or for regulatory cross talk between
parallel MAP kinase pathways.
|
CHROMOSOMAL LOCALIZATION OF DSP GENES
|
|---|
MAP kinases have been implicated in processes underlying cellular
transformation, tumor formation, and metastasis (14
15
16
, 68
, 69)
. Consistent with this, expression of CL100/MKP-1, hVH-3/B23,
or B59 inhibits oncogene-driven DNA synthesis and cell transformation
(58
, 70)
. With such a potential tumor suppressor role in
mind, it is of note that the chromosomal localization of all human DSP
genes have now been mapped and several of these occupy regions
implicated in genetic reorganizations associated with the incidence of
some cancers (38
, 54
, 59
, 71
, 72)
(Table 1)
. MKP-3/PYST1,
for instance, is localized to a region of frequent allelic loss in
pancreatic cancer (12q21–23) and also displays lower levels of
expression in several pancreatic tumor cell lines (73)
.
|
DSP CATALYTIC DOMAIN
|
|---|
Amino acid sequence comparison between DSPs reveals the highest
similarity within and immediately around a catalytic active site
sequence motif
Dx26(V/L)x(V/I)HCxAG(I/V)SRSxT(I/V)xxAY(L/I)M
(where x is any amino acid) localized within the
carboxyl-terminal half of these enzymes (Fig. 4A
). Screening the EST database reveals at least six
additional potential mammalian members of this gene family
displaying a conserved DSP catalytic active site (Fig. 3B
).
This motif contains at least two residues (underlined) known to be
critically important within a two-step catalytic reaction. According to
a model of catalysis by PTPs, an essential cysteine functions as an
active site nucleophile forming a covalent thio-phosphate intermediate,
whereas the aspartate acts as a general acid to donate a proton to the
leaving group. The same aspartate then probably acts as a general base
taking a proton from a water molecule, which then attacks the
phospho-enzyme intermediate to eliminate phosphate and regenerate
active enzyme (28)
(Fig. 3C
). Both threonine
and tyrosine residues are likely to be dephosphorylated by the same
catalytic mechanism, and the crystal structure of VHR suggests that a
shallow active site (as opposed to the deeper cleft of
tyrosine-specific PTPs) accounts for the less stringent phosphoamino
acid specificity of the DSPs (28)
.
View larger version (97K):
[in this window]
[in a new window]
|
Figure 4. The conserved DSP extended active site motif. A)
Alignment and amino acid conservation of human DSP extended active
sequence motifs was performed using Clustal W (1.5) and GeneDoc
software. At any one position, conservation between DSP gene family
members of 100%, >80% or >60% is indicated by shading in black,
dark gray, and light gray, respectively. B) Sequence
alignment of extended active site motif of human CL100/MKP1 with amino
acids from potential DSPs encoded by mammalian EST sequences
corresponding to the accession numbers shown. Arrows indicate the
positions of the active site cysteine and aspartate residues predicted
to play an essential role in the catalytic mechanism of DSP hydrolysis.
C) Schematic representation of the DSP catalytic
mechanism (28)
. The DSP active site is represented by the
diagonal gray box, which shows the catalytic aspartate (Asp) and
cysteine (Cys) residues. The phosphotyrosine shown is localized within
the MAP kinase activation loop motif TxY of kinase
subdomain VIII. For clarity, phosphothreonine is not shown. The
phosphatase reaction proceeds in two steps. First, the thiolate anion
of cysteine attacks the tyrosine phosphate (1.) and forms a
cysteinyl-phosphate intermediate (2.). Release of tyrosine follows
donation of a proton by the aspartic acid acting as a general acid
(2.). For the second step of the catalytic reaction, the same aspartate
acts as a general base to take a proton from a water molecule, which in
turn attacks the phospho-enzyme intermediate to eliminate phosphate and
regenerate active DSP enzyme (4.). The phosphothreonine within the MAP
kinase TxY motif is probably dephosphorylated by an
identical mechanism.
|
|
|
NATURALLY OCCURRING INACTIVE DSP ‘MUTANTS’
|
|---|
In contrast to the DSP active site motif described above, a
naturally occurring enzymatically inactive DSP (STYX) has been
identified in which the catalytic cysteine is replaced by glycine. The
inactive ‘catalytic site’ of STYX may bind phosphotyrosine in a
manner analogous to SH2 and PTB binding domains found within various
signal transduction effector targets and adaptor proteins. An
additional noncatalytic DSP homologue Sbf1 has also been described
recently, and database searching reveals additional sequences such as
MK-STYX, which is predicted to encode similarly inactive DSPs
(74)
. This indicates that STYX could be the prototype of a
new subfamily of catalytically inactive DSPs, though their functional
significance is currently unclear.
|
SEQUENCE DIVERGENCE WITHIN THE AMINO-TERMINAL OF DSPs
|
|---|
Whereas the carboxyl-terminal catalytic core of DSPs is well
conserved, considerably less sequence homology is evident within the
amino-terminal half of these enzymes (Fig. 3)
. One role of specific
sequences within the amino terminus region could be to define their
subcellular localization. Indeed, a potential nuclear localization
motif has been suggested within the amino terminus of hVH-3/B23
(52)
, and similar sequences can also be identified in
CL100/MKP-1, hVH-2/MKP-2, and PAC-1. In contrast, putative nuclear
export motifs have been proposed within the amino terminus of rVH-6
(MKP-3/PYST1) and B59 (55
, 58)
. The presence of these
motifs is consistent with the subcellular localization reported for
these DSPs (see above). Despite sequence divergence within the
amino-terminal, all DSPs possess two short stretches of homology
containing residues conserved with two segments flanking the active
site of the Cdc25 phosphatase. These motifs have been designated CH2
(Cdc25 homology regions) A and B or, alternatively, CH2-N and CH2-C
(75
, 76)
. The functional significance of these motifs is
currently unknown, although a role for the amino-terminal in binding
substrate MAP kinases (see below) could reflect properties associated
with one or both CH2 domains. Examination of the gene structures of
CL100/MKP-1 and PAC1 (38
, 76
, 77)
indicates that their
amino termini may have distinct evolutionary origins to the
carboxyl-terminal catalytic core. In fact, of four exons comprising the
murine CL100/MKP-1 gene, exons 1 and 2 encompass the CH2-N and CH2-C
domains, respectively, tempting speculation that these exons may have
converged with a smaller gene encoding simply a DSP catalytic core.
|
SELECTIVE MAP KINASE INACTIVATION BY DSPs
|
|---|
One striking development in our understanding of DSP action came
with the discovery that some gene family members display specificity
for inactivating different MAP kinases. Hence, studies using purified
proteins and transfected mammalian cells have revealed that low
concentrations of MKP-3/PYST1 completely inactivate ERK1 and ERK2 but
not JNK/SAPK or p38 MAP kinases (57
, 78)
. This contrasts
with another DSP, M3/6, which appears highly specific for inactivating
JNK/SAPK and p38 MAP kinases, as even high expression levels failed to
inhibit ERK activation by growth factor or oncogenic p21ras
(78)
. Other DSPs also display selectivity in that MKP-4 is
more effective at inactivating ERKs and MKP-1 acts preferentially on
JNK/SAPK and p38 MAP kinases, whereas PAC-1 and hVH-2/MKP-2 show
limited specificity for ERK and either p38 and JNK/SAPK, respectively
(43
, 44
, 59)
(Table 1)
. Although this selectivity could
reflect merely a substrate preference by the catalytic domains of these
different DSPs, it appears that an additional unexpected mechanism
accounts at least in part for specific MAP kinase inactivation.
|
MAP KINASE SELECTIVITY REFLECTS DSP BINDING AND CATALYTIC
ACTIVATION
|
|---|
In 1996 a report from Steve Keyse’s laboratory demonstrated
that MKP-3/PYST1 is able to form a stable complex with ERK MAP kinases
(57)
. More detailed analysis of this interaction soon
revealed that MKP-3/PYST1 binding to ERK MAP kinases was direct and
unexpectedly involved the noncatalytic amino terminus of the
MKP-3/PYST1 protein. Indeed, when expressed alone, the
carboxyl-terminal catalytic domain of MKP-3/PYST1 was devoid of ERK
binding activity (79)
. Strikingly, when interaction with
different MAP kinases was compared, the MKP-3/PYST1 amino terminus was
found to bind tightly to both ERK1 and ERK2 but not to JNK3 (SAPKß),
JNK2 (SAPK), or p38 MAP kinases (79
, 80)
. This binding
specificity mirrors the selectivity of MKP-3/PYST1 for inactivating
different MAP kinases and raised the question of whether amino-terminal
substrate binding alters the catalytic activity of this DSP. As
indicated by MKP-3/PYST1-dependent hydrolysis of the artificial
substrate p-nitrophenyl phosphate (p-NPP),
purified ERK2 was found to stimulate the catalytic activity of this DSP
by more than 40-fold. This effect was independent of protein kinase
activity and required binding of ERK2 to the noncatalytic amino
terminus of MKP-3/PYST1. Neither JNK/SAPK nor p38 MAP kinases bound or
stimulated significant MKP-3/PYST1 catalytic activity over basal levels
(80)
. Together, these observations indicate that
MKP-3/PYST1 selectivity for ERK MAP kinases may reflect tight substrate
binding through its amino terminus, leading to its catalytic activation
and consequent dephosphorylation and inactivation of the bound MAP
kinase (Fig. 5
). In support of this, an amino-terminally truncated version of
MKP-3/PYST1 devoid of binding activity loses its selectivity for
inactivating ERKs and displays similar but reduced activity against all
MAP kinases (79
, 80)
. In addition, a mutant form of ERK2
(D319N) analogous to the Drosophila gain-of-function ERK-A
mutation sevenmaker is disabled in its ability to either
bind or activate MKP-3/PYST1. This may account for its hypersensitivity
to activation in vivo as well as its resistance to
inactivation by DSPs (43
, 80
, 81)
. Purified MKP-4
(80)
and MKP-1 (S. Arkinstall, unpublished observations)
also bind and undergo catalytic activation by purified MAP kinases in
parallel with their enzymatic selectivity, suggesting that this is a
general mechanism for activating DSPs and targeted inactivation of MAP
kinases.
View larger version (32K):
[in this window]
[in a new window]
|
Figure 5. Schematic representation of ERK binding and catalytic activation of
MKP3/PYST1. This cartoon represents MKP3/PYST1 either free or bound to
an ERK MAP kinase. Two black bars indicate the position of
amino-terminal Cdc25 homology (CH2) domains whereas a black box the
position of the DSP catalytic domain within the carboxyl terminus. ERKs
bind specifically to the amino-terminal noncatalytic region of
MKP3/PYST1. Binding triggers powerful catalytic activation of
MKP3/PYST1 phosphatase activity, leading to rapid dephosphorylation and
inactivation of bound ERK1 or ERK2. Both binding and catalytic
activation of MKP3/PYST1 are independent of ERK enzymatic activity.
JNK/SAPK and p38 MAP kinases fail to bind or trigger catalytic
activation of MKP3/PYST1. The gain-of-function
sevenmaker mutant ERK2 D319N (not shown) is also
deficient in MKP3/PYST1 binding and catalytic activation and this may
account for its resistance to inactivation in vivo.
|
|
|
MODEL FOR MAP KINASE CONTROL BY DSPs
|
|---|
Together, these observations suggest the following model for MAP
kinase inactivation by DSPs (Fig. 6
). Stimulation by growth factors, cytokines, cell stresses, or some
active oncogenes leads to rapid transcription of one or a subset of DSP
genes. Increased DSP transcription may reflect activation of specific
MAP kinases, although alternative pathways are not excluded. After
translation of the DSP mRNA into protein, the catalytically inactive
DSP translocates to a specific subcellular compartment within either
the nucleus or cytosol. Upon encountering its target MAP kinase, the
DSP binds tightly through its amino terminus, which in turn triggers
activation of the phosphatase catalytic domain. If the bound MAP kinase
is already activated, then this will result in its rapid inactivation.
Conversely, if the MAP kinase is not active, then its tight interaction
with an active DSP is expected to block any possibility of kinase
activation by a subsequent stimulus. MAP kinases that fail to bind the
DSP within its amino terminus remain active or susceptible to
activation after extracellular stimulation. Depending on their cellular
localization, these regulatory effects allow for selected inhibition of
MAP kinase activities in specific subcellular compartments. Some DSPs
have been shown to possess short half-lives (38
, 39)
,
suggesting that in the absence of continued gene transcription and
protein synthesis, their rapid turnover limits their duration of action
in cells. Overall, tight control of DSP gene induction, combined with
their differential binding and catalytic activation by a specific
repertoire of MAP kinases, provides a sophisticated mechanism for rapid
and targeted inactivation of selected MAP kinase activities.
View larger version (36K):
[in this window]
[in a new window]
|
Figure 6. Model of MAP kinase inactivation by DSPs. Cell exposure to growth
factors, cytokines, cell stresses or activated oncogenes leads to
induction of a subset of DSP genes. Increased expression is likely to
reflect activation of transcription factors (black circles) via both
MAP kinase-dependent and independent pathways. Newly synthesized DSPs
translocate to specific subcellular compartments as dictated by
anchorage and/or localization motifs not yet identified. Specific
binding to target MAP kinases through regions within the DSP amino
terminus then triggers activation of the phosphatase catalytic domain.
Bound MAP kinases are in turn inactivated by dephosphorylation on
threonine and tyrosine residues localized within the ‘activation
loop’ motif TxY. Inactive MAP kinases then dissociate,
leaving the DSP free to bind and inactivate another MAP kinase
molecule. In the absence of continued DSP gene transcription and
protein synthesis, rapid degradation may limit their duration of
activity in cells.
|
|
From a broader perspective, tight binding between DSP family members
and MAP kinases represents a further example of an association
indicative of higher order complexes within cellular signal
transduction pathways. ERK, for example, binds to a number of proteins
involved in this MAP kinase cascade. These include the activating
kinase MEK-1, the target transcription factors Elk-1, c-Myc and
STAT1, downstream kinases including the MAPKAP-K1 (p90 S6 kinase,
RSK), Mnk1, and Mnk2, as well as the potential scaffold proteins MP1
and Ksr (82
83
84
85
86
87
88
89
90)
. Together with the DSP MKP-3/PYST1, these
proteins may form stable or dynamic multimolecular complexes similar to
those underlying pheromone-stimulated MAP kinase activation in budding
yeast (7)
. Ordered protein complexes probably play a major
role defining specificity of communication within MAP kinase signaling
cascades and is an emerging theme also seen in other signal
transduction pathways, as exemplified in the multimeric ‘signalsome’
comprising several essential components required for activation of the
transcription factor NF-
B (91)
.
|
REFERENCES
|
|---|
-
. The C. elegans Sequencing Consortium (1998) Genome sequence of the nematode C. elegans: a platform for investigating biology. Science 282,2012-2018[Abstract/Free Full Text]
-
Hunter, T., Plowman, G. D. (1997) The protein kinases of budding yeast: six score and more. Trends Biochem. Sci. 22,18-22[Medline]
-
Cohen, P. (1997) The search for physiological substrates of MAP and SAP kinases in mammalian cells. Trends Cell Biol 7,353-361
-
Fanger, G. R., Gerwins, P., Widmann, C., Jarpe, M. B., Johnson, G. L. (1997) MEKKs, GLKs, MLKs, PAKs, TAKs, and Tpls: upstream regulators of the c-Jun amino terminal kinases?. Curr. Opin. Genet. Dev. 7,67-74[Medline]
-
Ferrell, J. E., Jr (1996) MAP kinases in mitogenesis and development. Curr. Top. Dev. Biol. 33,1-60[Medline]
-
Marshall, C. J. (1995) Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation. Cell 80,179-185[Medline]
-
Leberer, E., Thomas, D. Y., Whiteway, M. (1997) Pheromone signalling and polarized morphogenesis in yeast. Curr. Opin. Genet. Dev. 7,59-66[Medline]
-
Yang, D. D., Kuan, C.-Y., Whitmarsh, A. J., Rincon, M., Zheng, T. S., Davis, R. J., Rakic, P., Flavell, R. A. (1997) Absence of excitotoxicity-induced apoptosis in the hippocampus of mice lacking the Jnk3 gene. Nature (London) 389,865-870[Medline]
-
Yang, D. D., Conze, D., Whitmarsh, A. J., Barrett, T., Davis, R. J., Rincon, M., Flavell, R. A. (1998) Differentiation of CD4+ T cells to Th1 cells requires MAP kinase JNK2. Immunity 9,575-585[Medline]
-
Dong, C., Yang, D. D., Wysk, M., Whitmarsh, A. J., Davis, R. J., Flavell, R. A. (1998) Defective T cell differentiation in the absence of JNK1. Science 282,2092-2095[Abstract/Free Full Text]
-
Kato, Y., Tapping, R. I., Huang, S., Watson, M. H., Ulevitch, R. J., Lee, J.-D. (1998) Bmk1/Erk5 is required for cell proliferation induced by epidermal growth factor. Nature (London) 395,713-716[Medline]
-
Kornhauser, J. M., Greenberg, M. E. (1997) A kinase to remember: dual roles for MAP kinase in long-term memory. Neuron 18,839-842[Medline]
-
Atkins, C. M., Selcher, J. C., Petraitis, J. J., Trzaskos, J. M., Sweatt, J. D. (1998) The MAPK cascade is required for mammalian associative learning. Nature Neurosci 1,602-609[Medline]
-
Mansour, S. J., Matten, W. T., Hermann, A. S., Candia, J. M., Rong, S., Fukasawa, K., Vande Woude, G. F., Ahn, N. G. (1994) Transformation of mammalian cells by constitutively active MAP kinase kinase. Science 265,966-970[Abstract/Free Full Text]
-
Cowley, S., Paterson, H., Kemp, P., Marshall, C. J. (1994) Activation of MAP kinase kinase is necessary and sufficient for PC12 differentiation and for transformation of NIH 3T3 cells. Cell 77,841-852[Medline]
-
Dudley, D. T., Pang, L., Decker, S. J., Bridges, A. J., Saltiel, A. R. (1995) A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc. Natl. Acad. Sci. USA 92,7686-7689[Abstract/Free Full Text]
-
Reszka, A. A., Bulinski, J. C., Krebs, E. G., Fischer, E. H. (1997) Mitogen-activated protein kinase/extracellular signal-regulated kinase 2 regulates cytoskeletal organization and chemotaxis via catalytic and microtubule-specific interactions. Mol. Biol. Cell. 8,1219-1232[Abstract]
-
Pages, G., Lenormand, P., L’Allemain, G., Chambard, J. C., Meloche, S., Pouyssegur, J. (1993) Mitogen-activated protein kinases p42mapk and p44mapk are required for fibroblast proliferation. Proc. Natl. Acad. Sci. USA 90,8319-8328[Abstract/Free Full Text]
-
Xia, Z., Dickens, M., Raingeaud, J., Davis, R. J., Greenberg, M. E. (1995) Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science 270,1326-1331[Abstract/Free Full Text]
-
Saklatvala, J., Rawlinson, L., Waller, R. J., Sarsfield, S., Lee, J. C., Morton, L. F., Barnes, M. J., Farndale, R. W. (1996) Role for p38 mitogen-activated protein kinase in platelet aggregation caused by collagen or a thromboxane analogue. J. Biol. Chem. 271,6586-6589[Abstract/Free Full Text]
-
Ichijo, H., Nishida, E., Irie, K., Dijke, P. T., Saitoh, M., Moriguchi, T., Takagi, M., Matsumoto, K., Miyazono, K., Gotoh, Y. (1997) Induction of apoptosis by ASK1, a mammalian MAPKKK that activates SAPK/JNK and p38 signaling pathways. Science 275,90-94[Abstract/Free Full Text]
-
Yang, X., Khosravi-Far, R., Chang, H. Y., Baltimore, D. (1997) Daxx, a novel Fas-binding protein that activates JNK and apoptosis. Cell 89,1067-1076[Medline]
-
Verheij, M., Bose, R., Lin, X. H., Yao, B., Jarvis, W. D., Grant, S., Birrer, M. J., Szabo, E., Zon, L. I., Kyriakis, J. M., Haimovitz-Friedman, A., Fuks, Z., Kolesnick, R. N. (1996) Requirement for ceramide-initiated SAPK/JNK signalling in stress-induced apoptosis. Nature (London) 380,75-79[Medline]
-
Zanke, B. W., Boudreau, K., Rubie, E., Winnett, E., Tibbles, L. A., Zon, L., Kyriakis, J., Liu, F.-F., Woodgett, J. R. (1996) The stress-activated protein kinase pathway mediates cell death following injury induced by cis-platinum, UV irradiation or heat. Curr. Biol. 6,606-613[Medline]
-
Alessi, D. R., Gomez, N., Moorhead, G., Lewis, T., Keyse, S. M., Cohen, P. (1995) Inactivation of p42 MAP kinase by protein phosphatase 2A and a protein tyrosine phosphatase, but not CL100, in various cell lines. Curr. Biol. 5,283-295[Medline]
-
Neel, B. G., Tonks, N. K. (1997) Protein tyrosine phosphatases in signal transduction. Curr. Opin. Cell Biol. 9,193-204[Medline]
-
Hooft van Huijsduijnen, R. (1998) Protein tyrosine phosphatases: counting the trees in the forest. Gene 225,1-8[Medline]
-
Denu, J. M., Dixon, J. E. (1998) Protein tyrosine phosphatases: mechanisms of catalysis and regulation. Curr. Opin. Chem. Biol. 2,633-641[Medline]
-
Pulido, R., Zuniga, A., Ullrich, A. (1998) PTP-SL and STEP protein tyrosine phosphatases regulate the activation of the extracellular s
TOP
ABSTRACT
INTRODUCTION
