| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 32, 29901-29912, August 8, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
Department of Molecular Pharmacology,
Albert Einstein College of Medicine, Bronx, New York 10461 and
Biomyx Technology, San Diego, California
92121
Received for publication, April 14, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Although the importance of MAP kinases in cellular signaling is well established, there is limited understanding of the molecular basis for specific MAP kinase recognition by its activators, inactivators, and substrates. Such knowledge is essential for comprehension of the ability of MAP kinases to integrate diverse biological stimuli and to transmit signals to the nucleus, in order to generate appropriate cellular responses. Recent studies (6–8) suggest that MAP kinases are capable of forming complexes with their cognate activating kinases, inactivating phosphatases and substrates. Our previous studies have focused on the interaction between ERK2, the founding member of the MAP kinase family, and MAP kinase phosphatase 3 (MKP3) (9–11). The MKPs are dual specificity phosphatases capable of dephosphorylating both Tyr(P) and Thr(P) in the activation loop of MAP kinases (12, 13). MKP3 is highly specific in dephosphorylating and inactivating ERK2, with a kcat/Km for the double phosphorylated ERK2 that is 106-fold higher than those for the hydrolysis of p-nitrophenyl phosphate (pNPP) or the bisphosphorylated peptide derived from the activation loop of ERK2 (10). The superiority of the double phosphorylated ERK2 to the ERK2-derived phosphopeptide most likely results from specific protein-protein interactions between ERK2 and MKP3 that are not available between the ERK2 peptide and MKP3. Interestingly, the phosphatase activity of the MKP3-catalyzed pNPP reaction can be dramatically increased in the presence of ERK2 (14). Mechanistic and kinetic studies suggest that ERK2 binding to MKP3 elicit activation of MKP3 activity by facilitating the repositioning of active site residues and general acid loop closure in MKP3 (9, 10, 15).
To identify structural features in MKP3 that are important for ERK2 binding and ERK2-induced activation, we carried out a systematic mutational and deletion analysis of MKP3 (11). Because the activation of MKP3 by ERK2 is dose-dependent and saturable, we can determine both the dissociation constant between ERK2 and MKP3 and the extent of MKP3 activation from the concentration dependence of ERK2 on the MKP3-catalyzed pNPP reaction. Furthermore, we have developed a competitive assay to determine the binding affinity of fragments/domains of MKP3 that are important for ERK2 recognition. With these assays, we were able to quantitatively evaluate the contributions that residues/regions within MKP3 make to ERK2 binding and ERK2-induced MKP3 activation. Our results show that recognition and activation of MKP3 by ERK2 involve multiple regions of MKP3 (11).
In the current study, we have identified structural elements in ERK2 that are important for binding and activation of MKP3 using the activation- and competition-based assays with wild-type MKP3 and an ensemble of ERK2 mutants. In addition, we have unveiled structural features in ERK2 that mediate specific Elk1 recognition. Our results show that protein-protein interactions between ERK2 and its cognate regulators and substrates proceed with a bipartite recognition mechanism.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
44–51/S218D/M219D/N221D/S221D) in pRSETa was kindly provided
by Dr. Natalie Ahn. Oligonucleotide primers CATATGACTGAGATCACCCAACC
(NdeI site underlined) and GGATCCTCATGGCTTCTGGGGCCC
(BamHI site underlined) were used to generate the N-terminal
His6-tagged Elk1 (residues 307–428) by PCR using
pGEX-2T/Elk1-(307–428) (a generous gift from Dr. Kun-Liang Guan) as a
template. The PCR product was subcloned into pET14b at
NdeI/BamHI. Mutant ERK2 and MKP3 were generated by PCRs
according to the standard procedure of the QuickChangeTM site-directed
mutagenesis kit (Stratagene) using either pET15b-His6-ERK2 or
pET21a-MKP3-His6 as a template. All mutants were verified by DNA
sequencing. Peptide Synthesis, Protein Expression, and Purification—A synthetic peptide derived from Elk1 (residues 387–399, Ac-Arg-Arg-Pro387-Arg-Ser-Pro-Ala-Lys-Leu-Ser-Phe-Gln-Phe-Pro-Ser399-NH2), which contains an ERK2 phosphorylation site Ser-389 (16) and an ERK2 docking site sequence FQFP (17), was used in the competition-based assay. The Elk1 peptide was synthesized using standard protocol, purified by high pressure liquid chromatography, and characterized by matrix-assisted laser desorption ionization/time of flight mass spectrometry by Alpha Diagnostic International. The purity of the peptide was determined to be close to 100%. Wild-type and mutant N-terminal His6-tagged ERK2s, wild-type and mutant C-terminal His6-tagged MKP3s, N-terminal His6-tagged Elk1-(307–428), and N-terminal His6-tagged MEK1/G7B were expressed in Escherichia coli BL21/DE3 and purified using standard procedures of Ni2+-nitrilotriacetic acid-agarose (Qiagen) affinity purification as described previously (11). Protein concentration was determined using the Bradford dye binding assay (Bio-Rad) diluted according to the manufacturer's recommendations with bovine serum albumin as standard.
Determination of Dissociation Constants—The dissociation
constants of ERK2 and its mutants for MKP3 were determined by the activation
assay at 30 °C and pH 7.0, in 50 mM 3,3-dimethylglutarate
buffer, containing 1 mM EDTA with an ionic strength of 0.15
M adjusted by addition of NaCl
(11). Briefly, the
MKP3-catalyzed hydrolysis of the p-nitrophenyl phosphate
(pNPP) reaction was initiated by the addition 0.2 µM
MKP3 to a reaction mixture (200 µl) containing 40 mM
pNPP (saturating concentration) and various concentrations of ERK2 or
its mutants. The reaction was quenched after 20–40 min by addition 50
µl of 5 N NaOH. After quenching, 200 µl of the reaction
mixture was transferred to a 96-well plate, and the amount of
p-nitrophenol was determined from the absorbance at 405 nm using a
Spectra MAX340 microplate spectrophotometer (Molecular Devices) with a molar
extinction coefficient of 18,000 M–1
cm–1. The dissociation constant
Kd was calculated by fitting the absorbance at
405 nm versus ERK2 concentration data to
Equation 1 using the nonlinear
regression program Kaleidagraph,
 | (Eq. 1) |
To determine the affinity of Elk1 or MEK1 for ERK2, the competitive binding
assay was used (11). In this
assay, the reaction was initiated by the addition of 0.1 µM MKP3
in a mixture (0.2 ml) containing 40 mM pNPP, 1.2
µM ERK2, and various concentrations of the Elk1 peptide,
Elk1-(307–428), or MEK1 at 30 °C and pH 7.0, in 50 mM
3,3-dimethylglutarate buffer, containing 1 mM EDTA with an ionic
strength of 0.15 M adjusted by addition of NaCl. The reaction was
quenched by addition 50 µl of 5 N NaOH after 30 min. The data
were fitted to Equation 2 by
nonlinear regression analysis to obtain the dissociation constants of Elk1 and
MEK1 for ERK2,
 | (Eq. 2) |
is ERK2
concentration which is fixed at 1.2 µM.
is Elk1 or MEK1 concentration during
titration. 
is the dissociation constant for
MKP3 binding to ERK2 with a value of 0.17 µM
(11), and
is the dissociation constant of Elk1 (or
MEK1) for ERK2. GST Pull-down and Western Blot Analysis—The binding of MEK1 and MKP3 to ERK2 was examined by GST pull-down and Western blot analyses. GST-ERK2 (10 µg) or GST-MKP3 (10 µg) in 0.5 ml of phosphate-buffered saline (140 mM, 2.7 mM KCl, 10 mM Na2HPO4, 1.0 mM KH2PO4, 2 mM dithiothreitol, pH 7.4) was immobilized on 20 µl of glutathione-Sepharose 4B beads (Amersham Biosciences), respectively, with gentle agitation at 4 °C for 2 h. In one experiment, different amounts of MKP3 and MEK1 (0–10 µg) were mixed in 200 µl of phosphate-buffered saline containing 0.5% Triton X-100 and incubated with 20 µl of GST-ERK2 bound beads. In another experiment, different amounts of ERK2 and MEK1 (0–20 µg) were mixed in 200 µl of phosphate-buffered saline containing 0.5% Triton X-100 and incubated with 20 µl of GST-MKP3 bound beads. After incubation with gentle agitation at 4 °C for 2 h, the beads were washed with phosphate-buffered saline and then boiled in 25 µl of 2x SDS sample buffer for 5 min to release the proteins from the beads. The sample was microcentrifuged at 10,000 rpm for 2 min, and 10 µl of the supernatant was loaded on 10% SDS-polyacrylamide gel. When the electrophoresis was complete, the proteins on the gel were transferred to nitrocellulose membrane using a Trans-Blot SD semi-dry electrophoretic transfer cell (Bio-Rad) at 150 mA and at room temperature for 1 h. The membrane was blocked in 5% milk in TBS-T (20 mM Tris, 150 mM NaCl, 0.1% Tween 20, pH 7.6) for1hat room temperature and then incubated with mouse anti-His6 monoclonal antibody (sc-8036, Santa Cruz Biotechnology) overnight at 4 °C. After washing with TBS-T, the membrane was incubated with goat anti-mouse antibody conjugated with horseradish peroxidase (sc-2005, Santa Cruz Biotechnology) for 1 h at room temperature. The immunocomplexes were detected by chemiluminescence upon incubation with ECL reagents (Amersham Biosciences). The membrane was immediately expose to film for 1–5 min to visualize the His6-tagged proteins.
ERK2 Kinase Assay—The kinase activity of wild-type and
mutant ERK2s was monitored by a radioisotope assay in which the rate of
incorporation of 32P from [
-32P]ATP into a
substrate was directly measured. Reactions were carried out in 50 µl of the
kinase buffer (20 mM MOPS, 50 mM KCl, 10 mM
MgCl2, 0.1 mM EDTA, and 1 mM dithiothreitol,
pH 7.4), containing ERK2 at a final concentration of 0.5 µM and
varied concentrations of the protein substrates, myelin basic protein (MBP,
Sigma), or Elk1. Reactions were initiated by the addition 1 mM
[-32P]ATP (PerkinElmer Life Sciences; NEG002A) (100
cpm/pmol) and allowed to proceed at 30 °C for 40 min for MBP and 25 min
for Elk1. The reactions were terminated by the addition of 10 µl of 9.0%
(final 1.5%) phosphoric acid. The 32P-labeled product was separated
from [-32P]ATP using P81 phosphocellulose paper (Whatman,
2.1 cm), which binds to the protein or peptide product but not ATP and its
metabolites. Detailed procedures are as follows: 30 µl of the quenched
reaction mixture was spotted onto the 2.1-cm sized P81 paper strips. After
washing the strips with 0.5% phosphoric acid 4 times (2 min each, 10–15
ml of 0.5% phosphoric acid per paper strip) with gentle agitation followed by
1 wash with water and 1 wash with acetone, the P81 papers were dried with a
hair dryer and inserted into a 5-ml scintillation tube. Four ml of
scintillation liquid was added, and the incorporation of 32P into
the product was counted by liquid scintillation spectrometry. Controls were
carried out in which ERK2 and the substrate were replaced by buffer. Each
sample was measured in triplicate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Definition of ERK2 Common Docking Site for MKP3—Many ERK2-interacting proteins such as ERK2 activators (e.g. MEK1/2), inactivators (e.g. MKP3 and HePTP), and substrates (e.g. RSK1 and Elk1) contain a kinase interaction motif (KIM) characterized by a cluster of 2–3 positively charged Arg or Lys residues that are important for ERK2 binding (6, 11, 18–23). Deletion of the KIM sequence (residues 61–75) from MKP3 resulted in a 135-fold reduction in ERK2 binding affinity but did not affect the propensity of MKP3 to be activated by ERK2 (11). Recent data suggest that all ERK2-interacting proteins may bind ERK2 through electrostatic interactions between the positively charged KIM motif and a common docking (CD) domain in ERK2 composed of a stretch of negatively charged amino acids (e.g. Asp-316 and Asp-319) situated opposite to the kinase catalytic cleft (6).
Residue Asp-319 in the CD domain of ERK2 is conserved in all MAP kinases from yeast to man. A dominant gain-of-function mutation of the rolled MAP kinase gene in Drosophila, termed Sevenmaker (rlsevenmaker), contains a single amino acid substitution of the analogous Asp-334 by an Asn (D334N) and activates several developmental pathways (24). The same mutation in mammalian ERK2 (ERK2/D319N) appears to be resistant to inactivation by MKPs in transfected cells (25, 26). Previous studies (6, 14) indicate that ERK2/D319N displays reduced affinity for MKP3 and is unable to activate the MKP3 phosphatase activity. However, using the quantitative activation-based assay, we were able to establish the Kd of ERK2/D319N for MKP3 (14.8 ± 1.2 µM), which is 87-fold larger than that of the wild-type ERK2 (11). More importantly, we found that ERK2/D319N can activate MKP3 to the same level induced by the wild-type ERK2 when a saturating amount of ERK2/D319N is present in the reaction. Thus, like the KIM sequence in MKP3, Asp-319 in ERK2 plays a major role in ERK2 binding to MKP3, but it is not essential for ERK2 induced MKP3 activation.
To further define the contribution of ERK2 CD domain to MKP3 binding and activation, we determined the effects of amino acid substitutions in the CD domain using the activation-based assay outlined under "Experimental Procedures." We chose amino acid residues in the vicinity of Asp-319, based on both primary and tertiary structure. All experiments were performed at pH 7.0, ionic strength of 0.15 M, and 30 °C. As summarized in Table I, the dissociation constant (Kd) for ERK2 and MKP3 is 0.17 ± 0.03 µM. No significant effects were observed for ERK2 mutants Q313E, S218D, E324A, and F329L. However, when Tyr-126 and Tyr-314 were replaced by an Ala, the affinity of ERK2/Y126A and ERK2/Y314A for MKP3 was reduced by 20- and 7.8-fold, respectively. When Asp-160 and Asp-316 was replaced by an Asn, the affinity of ERK2/D160N and ERK2/D316N for MKP3 decreased 18- and 9.6-fold, respectively. Substitution of Glu-79 or Arg-133 with an Ala reduced the affinity of ERK2/E79A and ERK2/R133A for MKP3 by 5.7- and 5.1-fold, respectively. Thus, Asp-319 is the most important residue in ERK2 CD domain as large decreases (60–100-fold) in binding affinity were observed when Asp-319 was changed to Ala, Asn, Glu, or Arg. Interestingly, ERK2/E324Q exhibited a 3.7-fold higher affinity for MKP3 than that of the wild-type ERK2. Taken together, our data suggest that the CD site is likely composed of residues Glu-79, Tyr-126, Arg-133, Asp-160, Tyr-314, Asp-316, and Asp-319, which form a contiguous docking surface on the back side of the ERK2 kinase active site (Fig. 1A). It is important to point out that although mutation of CD site residues decreased MKP3 binding affinity, none of them affected the ability of ERK2 to activate MKP3 (Table I).
|
|
A "Hot Spot" Interaction between the KIM Sequence of MKP3 and the CD Site in ERK2—It has been suggested that the CD domain may make direct contact with the KIM sequence in ERK2-interacting molecules (6, 27). We showed previously that when Arg-64 and Arg-65 in the KIM sequence of MKP3 were replaced by an Ala, the affinity of MKP3/R64A and MKP3/R65A for ERK2 decreased 7.3- and 150-fold, respectively (11). Interestingly, the MKP3/R64K mutant displayed a Kd value for ERK2 identical to that of the wild-type MKP3, indicating that a Lys can effectively replace an Arg at position 64 (Table I). In contrast, the MKP3/R65K exhibited a binding affinity for ERK2 that was nearly 50-fold lower than that of the wild-type MKP3 and only 3-fold better than that of MKP3/R65A (Table I). This result suggests that a guanidinium side chain is required at position 65 of MKP3 for high affinity binding with ERK2.
Because removal of side chains at Arg-65 or Asp-319 results in similar loss in binding affinity (150x for MKP3/R65A and 110x for ERK2/D319A), it is possible that Arg-65 directly interacts with Asp-319. Because Lys fails to substitute for Arg-65 and Asn is unable to replace Asp-319, the guanidinium group of Arg-65 may engage in a bidentate H-bond with the carboxylate group of Asp-319, in which the guanidinium group supplies two H-bond donors and the carboxylate group provides two H-bond acceptors. The importance of the optimal positioning between the side chains of Arg-65 and Asp-319 is highlighted by the large decrease in binding affinity experienced by the conservative mutant ERK2/D319E. The fact that swapping Arg-65 and Asp-319 simultaneously failed to restore the binding affinity between ERK2/D319R and MKP3/R65D indicates that peripheral residues surrounding these two amino acids and the relative orientation of MKP3 and ERK2 are also important for binding (Table I). Further supporting evidence for direct interaction between Arg-65 and Asp-319 came from the observation that switching Arg-65 to Asp in MKP3 or changing Asp-319 to Arg in ERK2 did not result in a further decrease in binding affinity (Table I). Thus, the geometric complementarity between the side chains of Arg-65 and Asp-319 is likely responsible for the formation of a hot spot for high affinity ERK2/MKP3 binding. Indeed, both Arg and Asp residues are enriched in hot spots observed in protein-protein interactions (28). The energetically less important residues in the CD site and KIM sequence likely surround the hot spot interactions and serve to occlude bulk solvent from the hot spot.
Substrate Recognition Region in ERK2 Is Involved in MKP3 Binding and Activation—Although the interaction between CD and KIM is important for high affinity binding between ERK2 and MKP3, it is not essential for the ERK2-induced MKP3 activation. This suggests that, in addition to the CD domain, other regions in ERK2 are also required for specific MKP3 recognition. We discovered previously that the ERK2-induced MKP3 activation requires a putative ERK specific docking sequence 364FTAP367 localized in the C terminus of MKP3 (11). Although mutation of this sequence reduces the affinity of MKP3 for ERK2 by less than 4-fold, this region is essential for ERK2-induced MKP3 activation. Interestingly, the putative ERK docking sequence FXFP can also be found in many ERK2 substrates (17, 29–31). This raises the possibility that the substrate-binding site in ERK2 may also be involved in MKP3 binding and activation.
If the ERK2 substrate-binding region is required for MKP3 activation, then Elk1, a physiological substrate of ERK2, should be able to block the ERK2-induced MKP3 activation. To test this hypothesis, we prepared an Elk1-derived peptide (residues 387–399, Ac-Arg-Arg-Pro387-Arg-Ser-Pro-Ala-Lys-Leu-Ser-Phe-Gln-Phe-Pro-Ser399-NH2), which contain the PRSP phosphoacceptor sequence separated by four residues from the ERK2 docking sequence FQFP. We also prepared the C-terminal fragment of Elk1 (residues 307–428), which contains all of the ERK2 phosphorylation sites, the FQFP docking sequence, and a KIM sequence (residues 312–321, KGRKPRDLEL) (16, 30). Both the Elk1 peptide and Elk1-(307–428) are excellent ERK2 substrates (17, 32, 33). Consistent with the observation that both the FQFP motif and the KIM sequence are important for the ERK2-catalyzed Elk1-(307–428) phosphorylation (30), the Kd value of Elk1-(307–428) for ERK2 (2.1 ± 0.3 µM) is 36-fold lower than that of the synthetic Elk1 peptide (75 ± 9 µM) (Fig. 2A). Importantly, the Elk1 peptide and Elk1-(307–428) can suppress the ERK2-induced MKP3 activation by 80 and 100%, respectively, at saturating peptide/protein concentrations. These results support our hypothesis that regions within ERK2 responsible for substrate binding (as opposed to the CD site) may also play an important role in mediating ERK2-dependent MKP3 catalytic activation.
|
Mapping an Area in ERK2 That Is Important for MKP3 Binding and
Activation—It is tempting to speculate that the C terminus of MKP3,
including the sequence motif that resembles the FXFP sequence, may
bind to ERK2 in a manner analogous to the way by which ERK2 substrates bind to
ERK2. Unfortunately, the peptide/protein substrate-binding region in ERK2 has
not been fully defined. There are no crystal structures available for
ERK2-substrate complexes. However, based on structural comparison with the
cAMP-dependent protein kinase, it has been proposed that the ERK2 substrate
recognition groove may lie between 
D and G
and that the kinase activation loop L12 and the large MAP kinase insertion
(1L14 and 2L14) may be involved in
substrate recognition (34,
35). Subsequent biochemical
studies suggest that D, L11, and L12 may be important for
substrate recognition
(36–38).
In addition, the sequence segment through L13 and G in JNK2
has been shown to be important for c-Jun binding
(39). In the case of p38, two
clusters of amino acids, one in the region defined by
D-L8-E and the other in
6-L10, are also implicated in substrate binding
(40). Finally, recent results
from Nishida and co-workers (7)
indicate that the ED site (defined by Glu-160 and Asp-161 in p38) located in
L11 contributes to the specificity of p38 for MAP kinase-activated protein
kinases. These structural and biochemical data provide an outline for the
putative ERK2 substrate-binding region.
To identify residues in the putative ERK2 substrate binding region that are
important for MKP3 recognition and activation, we selectively removed side
chains from amino acids in D, L8, E, L9,
6, L10, 7, L11, L12, L13, and
G that are surface accessible and/or unique to ERK1/2
(Fig. 3). The functional
significance of these residues in mediating specific MKP3 binding and
activation was quantitatively evaluated using purified wild-type MKP3 and
mutant ERK2s as described
(11). We first examined
residues in D-L8 (Table
II). Moderate effects (2–4-fold decrease in MKP3 binding
affinity) were detected when Lys-112, Leu-113, Lys-115, Thr-116, Gln-117, and
His-118 were individually changed to Ala. With the exception of T116A, none of
the residues are essential for the ERK2-induced MKP3 activation. Y111A and
L119A exhibited a 4.9- and 5.5-fold decrease in MKP3 binding affinity,
respectively, but retained the ability to fully activate MKP3.
|
|
We then focused on the region defined by E, L9,
6, L10, 7, and L11 (residues 121–160).
Although Tyr-126 and Arg-133 reside in E, and Asp-160 is in
L11, they are considered part of the CD site based on their proximity to
Asp-319 in the three-dimensional structure
(Fig. 1A and
Table I). No significant
effects were observed for ERK2 mutants N121A, D122A, N152A, T157A, and T158A.
Moderate decrease (2–5-fold) in binding affinity was observed for S151A
and L154A, although neither failed to activate MKP3 at saturating
concentrations. In contrast, although removal of the side chain from Lys-149
had no effect on MKP3 binding affinity, K149A could only activate MKP3 to 67%
of the phosphatase activity induced by wild-type ERK2.
Next, we studied residues located in F, L13, and
G. As summarized in Table
II, no significant effect was observed when Asn-222 was changed to
an Ala. Only a moderate decrease (2–3.5-fold) in binding affinity was
observed for I225A, P227A, and D233A, although none of them were essential for
the ERK2-induced MKP3 activation. Strikingly, substitutions of Arg-223,
Lys-229, and His-230 with an Ala not only resulted in a large decrease in MKP3
binding affinity (13–49-fold) but also severely impaired the ability of
ERK2 to activate MKP3-catalyzed pNPP reaction
(Table II).
Finally, we made several substitutions in the putative p + 1 site of ERK2
substrate binding region: T179A, E184G, T188A, R189A, and W190A. We also
replaced Ser-264 in the MAP kinase insert with an Ala. No significant effects
were observed for E184G, T188A, and S264A
(Table II). T179A and W190A
exhibited a 4- and 6-fold decrease in MKP3 binding affinity, respectively, but
retained the ability to fully activate MKP3. In contrast, although ERK2/R189A
displayed the same affinity for MKP3 as the wild-type ERK2, it was only able
to activate MKP3 to 70% of the maximum activity even at saturating ERK2/R189A
concentrations. Collectively, the results suggest that in addition to the CD
site, D, L8, L10, L11, L12, F, and L13 are
also involved in MKP3 recognition. Specifically, residues Tyr-111, Leu-119,
and Trp-190 contribute significantly (>5-fold) to high affinity MKP3
binding, whereas residues Thr-116, Lys-149, Arg-189, Glu-218, Arg-223,
Lys-229, and His-230 are important for the ERK2-induced MKP3 activation. These
residues form a novel binding surface area away from the CD site and adjacent
to the active site cleft that binds ATP
(Fig. 1C).
Identification of ERK2 Residues Important for Elk1
Recognition—The results described above suggest that in addition to
the CD site, the putative ERK2 substrate-binding region may also be involved
in MKP3 binding and activation. In order to confirm that the putative
substrate-binding region is indeed important for ERK2 substrate recognition
and to identify amino acid residues that are important for ERK2 substrate
phosphorylation, we determined the kinase activity of the ERK2 mutants using
both MBP and the transcription factor Elk1 as substrates. MBP is a widely used
protein substrate for several protein kinases, including ERK2
(41). The transcription factor
Elk1 is a physiological substrate of ERK2. When phosphorylated by ERK2, Elk1
forms a complex with the serum-response factor and binds the serum-response
promoter element to enhance transcription from the c-fos promoter.
The kinase activity of ERK2 was determined by a radioisotope assay in which
the rate of 32P incorporation from [-32P]ATP into
a substrate was directly measured (see under "Experimental
Procedures"). All steady-state kinetic measurements were performed at pH
7.4 and 30 °C in 1 mM ATP, which is within the range of
physiological ATP concentrations. We determined the kinetic parameters,
kcat and Km, for the wild
type and mutant ERK2s with both MBP and Elk1-(307–428) as a substrate
(Table III). The values for the
overall catalytic efficiency, also known as substrate specificity constant
kcat/Km, are also listed in
Table III.
|
As shown in Table III, substitutions in the CD site and the putative substrate-binding region in ERK2 produce very small changes in the kinetic parameters for the ERK2-catalyzed MBP phosphorylation (<5-fold for most ERK2 mutants). This is consistent with the fact that MBP contains neither a KIM motif nor an FXFP docking sequence. The only exception is D233A, which displays a kcat/Km value 20-fold lower than that of the wild-type ERK2. The modest effects on MBP phosphorylation suggest that mutations in the CD site and the putative substrate-binding region do not introduce gross conformational changes in the ERK2 catalytic site and the overall structure. Because Elk1-(307–428) possesses both KIM and FXFP sequences, much larger changes (i.e. >5-fold) in the kinetic parameters for Elk1 phosphorylation were observed for a number of ERK2 mutants (Table III). Indeed, many residues in the CD site important for MKP3 binding (e.g. Glu-79, Asp-160, Tyr-314, Asp-316, and Asp-319) are also involved in Elk1 recognition. Moreover, a subset of residues (i.e. Tyr-111, Arg-189, Trp-190, Arg-223, Lys-229, and His-230) in the putative substrate binding region shown to be critical for MKP3 binding and activation are also important for Elk1 phosphorylation. Substitution of these residues with Ala led to a 16–130-fold decrease in the kcat/Km values for the ERK2-catalyzed Elk1 phosphorylation. These results indicate that the protein-protein interaction surfaces between ERK2 and MKP3 may overlap with those between ERK2 and Elk1, consistent with the ability of Elk1 to block the ERK2-induced MKP3 activation (Fig. 2A).
Besides residues that are required for both MKP3 and Elk1 binding, there are additional residues that are unique to either MKP3 or Elk1 recognition. For example, Tyr-126 and Arg-133 in the CD site are required for MKP3 but not Elk1 binding (Fig. 1A). In contrast, Thr-157 and Thr-158 are not essential for MKP3 binding but may constitute a part of the CD site for Elk1 (Fig. 1B). Moreover, a large reduction in kcat/Km (10–100-fold) was observed when the side chains of Leu-113, Lys-115, His-118, Asp-233, Ile-225, Pro-227, and Ser-264 were replaced with an Ala (Table III), whereas the same mutations had little effect on the ability of ERK2 to bind and/or activate MKP3 (Table II). On the contrary, substitutions at Thr-116, Leu-119, Lys-149, and Glu-218 resulted in severe impairment of MKP3 binding and activation (Table II), whereas the corresponding mutations had very minor effects on Elk1 phosphorylation (Table III). These differential results strongly suggest that the observed mutational effects are unlikely due to nonspecific structural perturbations. The similarities and differences between the novel MKP3 and Elk1 binding region are highlighted in Fig. 1, C and D.
The large decrease in Elk1 phosphorylation activity observed when Asn-152
and Leu-154 in 7 were replaced with an Ala may result from
perturbations of ATP binding due to their proximity to the ATP-binding pocket
(34). Consequently, Asn-152
and Leu-154 are not considered as part of the binding pocket for protein
substrates. Possible structural changes caused by mutations in these residues
are likely local because removal of side chains from Asn-152 and Leu-154 do
not affect MKP3 binding and activation. Finally, in addition to mutations that
reduce the ERK2 kinase activity, there are several ERK2 mutants (E58Q, D122A,
S151A, and S221A) that display increased kinase activity (3–7-fold)
against both MBP and Elk1 (Table
III).
MKP3 and MEK1 Binding to ERK2 Is Also Mutually Exclusive—In
addition to the complexes formed with its substrates (e.g. Elk1) and
inactivators (e.g. MKP3), ERK2 is also capable of forming a stable
complex with its activator MEK1
(6). Results described above
indicate that the binding sites for MKP3 and Elk1 partially overlap, and the
binding of MKP3 and Elk1 to ERK2 is mutually exclusive. To determine whether
the binding of MEK1 and MKP3 to ERK2 is also mutually exclusive, we measured
the effect of MEK1 on ERK2-induced MKP3 activation using the competitive assay
(see "Experimental Procedures"). For this experiment, we used a
constitutively active MEK1, MEK1/G7B
(MEK1/44–51/S218D/M219D/N221A/S222D)
(42). Kinetic analysis shows
that the combined effect of deleting residues 44–51 and phosphorylation
site substitutions in MEK1 mimic conformational changes normally induced by
phosphorylation (42). As shown
in Fig. 2B, unlike
Elk1-(307–428), MEK1/G7B exerts no significant effect on the
ERK2-induced MKP3 activation even at concentrations up to 32 µM
under the same conditions used for Elk1-(307–428). The result is
consistent with two scenarios. It is possible that MEK1/G7B shares the same
binding sites with MKP3 with a much lower affinity for ERK2. If this is the
case, the binding affinity of MEK1/G7B for ERK2 should be lower than that of
Elk1-(307–428), because the latter at 30 µM could block
the ERK2-induced MKP3 activation by 70%
(Fig. 2). Alternatively, ERK2
can simultaneously associate with both MEK1/G7B and MKP3 to produce a ternary
complex, and the binding of MEK1/G7B to ERK2 does not influence the ability of
ERK2 to activate MKP3.
In order to differentiate these two possibilities, we carried out GST-ERK2 pull-down experiments to measure MKP3 and MEK1/G7B binding to ERK2 directly. In the first experiment GST-ERK2 (10 µg) was used to bind His6-tagged MKP3 and MEK1/G7B. The amount of MKP3 or MEK1/G7B associated with ERK2 was visualized by anti-His6 antibody (for details see "Experimental Procedures"). Consistent with early observations, both MEK1/G7B and MKP3 are capable of binding ERK2 (Fig. 4A, lanes 3–6), although it is clear that more MEK1/G7B is needed to yield the same amount of ERK2 complex. More ERK2·MKP3 complex was formed than ERK2·MEK1/G7B when an equal amount (2 or 10 µg) of MKP3 and MEK1/G7B were mixed with GST-ERK2 in the pull-down experiments (Fig. 4A, lanes 7 and 10). These results indicate that MKP3 binds to ERK2 with a higher affinity than MEK1/G7B does. Moreover, the amount of MKP3 associated with ERK2 is decreased in the presence of MEK1/G7B, whereas the presence of MKP3 also decreases the amount of MEK1/G7B bound to ERK2 (Fig. 4A, lanes 8 and 9). This result suggests that the binding of MKP3 to ERK2 interferes with the association of MEK1/G7B with ERK2, possibly through competition for the same binding sites.
|
To investigate further the binding interactions among MKP3, MEK1/G7B, and ERK2, we also used GST-MKP3 as a pull-down reagent. We determined that MEK1/G7B does not form a complex with MKP3 when 20 µg of MEK1/G7B is mixed with 10 µg of GST-MKP3 (Fig. 4B, lane 3). In contrast, strong complex (ERK2·MKP3) formation is observed when 10 µg of GST-MKP3 is mixed with 2 µg of ERK2 (Fig. 4B, lane 4). If ERK2 were able to bind simultaneously both MKP3 and MEK1/G7B to make a ternary complex, then GST-MKP3 should be able to pull down both ERK2 and MEK1/G7B (through its interaction with ERK2) when the three proteins were incubated together. As shown in Fig. 4B, no MEK1/G7B is detectable from the ERK2·MKP3 complex pulled down by GSTMKP3. Rather, the amount of ERK2 associated with GSTMKP3 decreases with increasing amounts of MEK1/G7B (Fig. 4B, lanes 4–7). Because MEK1/G7B does not bind MKP3, the decreased binding between ERK2 and MKP3 in the presence of MEK1/G7B most likely results from competition by MEK1/G7B and MKP3 to the same binding sites in ERK2. Because ERK2 is still able to form a complex with MKP3 in the presence of a 10-fold excess of MEK1/G7B, the affinity of MKP3 for ERK2 is higher than that of MEK1/G7B. Collectively, the results shown in Fig. 4 suggest that ERK2 displays a higher binding affinity for MKP3 than MEK1/G7B and that the binding of MKP3 and MEK1/G7B to ERK2 is mutually exclusive.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The regulation of ERK2 activity by MKP3 provides a good example for the importance of proper protein-protein interaction in MAP kinase signaling. MKP3 forms a physical complex with ERK2 and is highly specific for ERK2 inactivation (12, 13). Interestingly, purified recombinant ERK2 stimulates the phosphatase activity of MKP3 toward pNPP (14). Moreover, catalytic activation by MAP kinases has also been observed with other MKPs, which mirrors the substrate selectivity of MKPs (12, 13). Biochemical and structural evidence suggest that the active site and general acid loop in MKP3 are deformed in the absence of ERK2 (9, 15, 43, 44). The binding of ERK2 to MKP3 must repair these defects and induce MKP3 to assume an active conformation for ERK2 dephosphorylation (9, 10, 15). These observations suggest a general mechanism by which all members of the MKP family are regulated: the MKPs exist in latent and inactivate states, and upon association with specific MAP kinases, the MKPs are activated leading to selective inactivation of MAP kinases.
To begin to elucidate the molecular basis for specific MKP3 recognition and activation by ERK2, we have carried out a systematic mutational and deletion analysis of MKP3 to evaluate quantitatively the contributions that residues/regions within MKP3 make to ERK2 binding and ERK2-induced MKP3 activation (11). Our results show that recognition and activation of MKP3 by ERK2 involves multiple regions of MKP3. We found that the KIM motif (residues 61–75) in MKP3 plays a major role (135-fold) for high affinity ERK2 binding. In addition, a unique sequence conserved in cytosolic MKPs (residues 161–177 in MKP3) also contributes to ERK2 binding (15-fold). However, these two regions are not essential for ERK2-induced MKP3 activation. A third ERK2-binding site is localized in the C terminus of MKP3 (residues 348–381). Although deletion of this region or mutation of the putative ERK specific docking sequence 364FTAP367 in this region reduces the affinity of MKP3 for ERK2 by less than 10-fold, this region is absolutely required for the ERK2-induced MKP3 activation.
The KIM sequence characterized by a cluster of positively charged residues has been found in many ERK2-binding proteins, including its regulators and substrates (6, 11, 18–23). There is evidence suggesting that the KIM sequences in ERK2-binding proteins compete for binding to the CD domain in ERK2 (residues 311–324), which is decorated with several acidic residues (6, 27). However, it is clear that the KIM-CD interaction alone cannot account for the docking specificity of ERK2 interacting molecules (6). In addition to the KIM sequence (also called D-domain in certain ERK2 substrates), many ERK2 substrates (e.g. transcription factors, KSR, and phosphodiesterase) also have an FXFP docking motif (17, 29–31), whose binding pocket on ERK2 remains to be defined. Interestingly, we noted that similar FXFP sequences also exist in MKPs that are capable of inactivating ERK2 and that the FXFP motif is essential for the ERK2-induced MKP3 activation (11). Thus, our working hypothesis is that the binding of ERK2 to MKP3 and the subsequent ERK2-induced MKP3 activation may be distinct events. The interaction between the KIM motif in MKP3 and the CD site in ERK2 may be important for high affinity binding, but additional interactions with the FXFP motif of MKP3 involving other regions of ERK2 (e.g. the ERK2 substrate binding groove) may be required for MKP3 to adopt a catalytically active conformation.
To test our hypothesis, we have mapped the ERK2-MKP3 interaction surfaces in ERK2 by systematically mutagenizing and characterizing amino acids in and around the CD domain as well as in the putative substrate-binding regions. We have uncovered a contiguous surface area (termed the CD site) including the CD domain responsible for high affinity MKP binding and provided new insight into the interaction between the KIM sequence and the CD site. We have also identified a novel surface area away from the CD site that is important for MKP3 activation. The results support our hypothesis that specific MKP3 recognition and activation by ERK2 involves multiple regions of ERK2. As an added value, we have also identified regions in ERK2 that are essential for Elk1 recognition. Finally, we have studied the binding of ERK2 with MEK1 and MKP3. Collectively, our results indicate that specific binding of MKP3, Elk1, and MEK1 to ERK2 targets both the CD site and the substrate-binding area in ERK2.
The CD Site Responsible for Binding the KIM Sequences in MKP3 and Elk1—The CD domain was originally identified as a linear string of amino acids (residues 311–324 in ERK2) important for docking interactions with ERK2-binding proteins (6). Our results confirmed the requirement of Tyr-314, Asp-316, and Asp-319 in this sequence for MKP3 binding. More importantly, our results revealed additional residues (Glu-79, Tyr-126, Arg-133, and Asp-160), which are topographically close to the CD domain and are involved in MKP3 binding. We propose that together these additional peripheral residues and the CD domain (residues 311–324) form a continuous surface area called the CD site that is responsible for high affinity binding of KIM-containing ERK2 interacting proteins. Although the CD sites for both MKP3 and Elk1 share many of the same amino acids (e.g. Glu-79, Asp-160, Tyr-314, Asp-316, and Asp-319), they also contain residues that are unique to each target (e.g. Tyr-126 and Arg-133 for MKP3 and Thr-157 and Thr-158 for Elk1) (Fig. 1, A and B). Interestingly, Thr-157 and Thr-158 correspond to the ED site required for docking interactions of p38 with several MAP kinase-activated protein kinases (7). It is possible that the CD site for MAP kinase substrate recognition includes both the CD domain and the ED site, which is consistent with the docking groove hypothesis (7).
The most important residue in the CD site for both MKP3 and Elk1
recognition is Asp-319 (Tables
I and
III), whereas the most
important residue in the KIM sequence of MKP3 is Arg-65
(11). Systematic mutagenesis
analyses of both Asp-319 and Arg-65 (Table
I) suggest that these two residues may directly interact with each
other, contributing to an energetic hot spot in the protein-protein binding
interface. It is likely that the guanidinium group of Arg-65 in MKP3 may
engage in a bidentate H-bond with the carboxylate group of Asp-319 in ERK2, in
which the guanidinium group supplies two H-bond donors and the carboxylate
group provides two H-bond acceptors. However, it is of interest to note that
the proposed electrostatic interactions between the basic residues in the KIM
sequence and the acidic residues in the CD domain (Ref.
6; and this study) were not
observed in the structures of p38 in complex with KIM peptides derived from
substrate MEF2 and activating enzyme MKK3b
(45). Instead, the structures
highlight hydrophobic interactions between the KIM peptides with a docking
groove (between helices D and E
and the reverse turn between 7 and 8) near
the ED site. It is possible that different KIM sequences may bind MAP kinases
in slightly different modes. Alternatively, the interactions observed in the
p38 and KIM peptide structures may differ from those in the intact protein
complexes. Clearly, high resolution crystal structures of intact protein
complexes are needed to fully elucidate the precise contacts between MAP
kinase and its interacting proteins.
The Novel MKP3 Binding and Activation Site and the Substrate-binding Site for Elk1—Although the interaction between the KIM sequence and the CD site is important for high affinity ERK2 binding to MKP3, it is not essential for the ERK2-induced MKP3 activation (11) (Table I). Interestingly, substitution of the FXFP sequence in MKP3 by Ala residues reduces ERK2 binding affinity 4-fold, yet the mutant can only be activated to 32% of the wild-type MKP3 (11). Because the FXFP motif is present in many ERK2 substrates and is adjacent and C-terminal to the phosphoacceptor, it raises the possibility that the ERK2 substrate-binding region may also be involved in MKP3 binding and activation. Indeed, both an Elk1-derived peptide (residues 387–399), which contains both the phosphoacceptor sequence and the FQFP motif, and a C-terminal fragment of Elk1 (residues 307–428), which contains all of the ERK2 phosphorylation sites, the FQFP docking motif, and a KIM sequence, can effectively block the ERK2-induced MKP3 activation (Fig. 2A). Furthermore, the involvement of putative ERK2 substrate-binding region in MKP3 activation was also implicated from biochemical analysis of a panel of ERK2/p38 chimeric molecules (23). Thus, it is possible that some structural elements in ERK2 (e.g. the CD site) are important for high affinity MKP3 binding, whereas others (e.g. the substrate-binding region) are required for ERK2-induced activation.
Unfortunately, the ERK2 substrate-binding area has not been fully defined. In order to identify additional factors in ERK2 that contribute to specific MKP3 binding and activation, we have carried out systematic mutagenesis and biochemical analysis of amino acid residues located in regions previously suspected to be involved in MAP kinase substrate recognition. The ERK2 mutants were quantitatively analyzed for their ability to bind MKP3 and for their ability to activate MKP3 (Table II). In addition, the ERK2 mutants were also analyzed for their ability to phosphorylate Elk1 in order to determine whether the putative ERK2 substrate-binding site is indeed important for Elk1 recognition (Table III). Most of the residues selected for mutagenesis have their side chains exposed on the surface, and therefore, their mutations are predicted not to significantly affect the hydrophobic core or the structural integrity of the ERK2 protein. None of the resides have direct contact with catalytically essential residues such as Lys-52, Asp-147, and Asp-165. Indeed, with the exception of Asp-233, substitutions of these residues do not significantly affect the ERK2-catalyzed phosphorylation of MBP, a nonspecific substrate (Table III). Thus, the measured differences in binding and functional activity for the mutants, compared with the wild type, should directly reflect the contributions of particular residues to target recognition of ERK2.
Our results show that Tyr-111 in D, Leu-119 in L8, and
Trp-190 in L12 contribute to high affinity binding to MKP3
(Table II). More importantly,
our results also reveal that Thr-116 in D, Lys-149 in L10,
Arg-189 in L12, and Glu-218 in F are essential for the
ERK2-induced MKP3 activation, whereas Arg-223, Lys-229, and His-230 in L13 are
required for both high affinity MKP3 binding and the ERK2-induced MKP3
activation (Table II).
Together, these residues form a novel MKP3 binding and activation site away
from the CD site (Fig.
1C), which accounts for the ability of ERK2 to activate
MKP3. Our results also suggest that the substrate-binding site for Elk1 may
consist of Tyr-111, Leu-113, and Lys-115 in D, His-118 in
L8, Arg-189 and Trp-190 in L12, Arg-223, Ile-225, Pro-227, Lys-229, and
His-230 in L13, Asp-233 in G, and Ser-264 in the MAP kinase
insert (Fig. 1D).
ERK2 Gain-of-Function Mutations—Genetic screens in several organisms have identified a number of dominant gain-of-function mutations in ERK2. These mutations may affect the intrinsic ERK2 activity and/or interactions with ERK2 activators, inactivators, and substrates. For example, Asp-319 in the CD site is mutated to Asn in the sevenmaker mutant of Drosophila ERK/Rolled, which displays a gain-of-function phenotype (24). We determined that the affinity of ERK2/D319N for MKP3 is 87-fold lower than that of the wild type. It has been proposed that the increased signal sensitivity of the ERK2/D319N mutant is due to a decreased sensitivity to dual specificity phosphatases such as MKP3 (14, 25, 26). ERK2/D160N represents another dominant gain-of-function mutation in the Drosophila MAP kinase, termed rlSu14 (46). Interestingly, Asp-160 is also located in the CD site, and replacement of Asp-160 with Asn reduces the affinity of ERK2 for MKP3 by 18-fold. That Asp-319 plays a more important role than Asp-160 for high affinity MKP3 binding correlates well with the stronger phenotype observed for ERK2/D319N than ERK2/D160N (46).
Substitution of Asp-227 to Asn in Saccharomyces cerevisiae FUS3 MAP kinase produces another dominant gain-of-function mutation (47). We found that replacement of the corresponding His-230 in ERK2 with either an Ala or an Asn causes a large decrease in MKP3 binding affinity and drastically compromises the ability of ERK2 to activate MKP3. These mutations also severely impair the ERK2-catalyzed Elk1 phosphorylation. Because Asp-319, Asp-160, and His-230 are also involved in recognition of both MKP3 and Elk1 and possibly other ERK2-interacting proteins as well, we should emphasize that the observed phenotype may be the result of a compounded effect on all ERK2 interactions in the cell.
In addition to mutations that reduce the ERK2 kinase activity, we also
identified several ERK2 mutants (E58Q, D122A, S151A, and S221A) that display
increased kinase activity (3–7-fold) against both MBP and Elk1
(Table III). In many cases, the
increase in activity results largely from an elevated
kcat. Interestingly, the corresponding E58Q mutation in
S. cerevisiae FUS3 MAP kinase (D48N) exhibits a dominant
gain-of-function mutation phenotype
(47). Substitution of Ser-151
with Asp also yields a kinase with 
10-fold higher basal activity over
wild-type ERK2, possibly through increased autophosphorylation of Tyr-185 in
the activation loop (48). The
mechanisms of activation of E58Q, D122A, S151A, and S221A will require further
investigation.
The Binding of MKP3, Elk1, and MEK1 to EKR2 Is Mutually
Exclusive—A comparison of the CD sites for MKP3 and Elk1 indicates
that they share a number of residues (Glu-79, Asp-160, Tyr-314, Asp-316, and
Asp-319) (Fig. 1, A and
B). Moreover, an inspection of the MKP3 activation site
and the Elk1 substrate-binding site reveals that they share residues Tyr-111,
Arg-189, Trp-190, Arg-223, Lys-229, and His-230
(Fig. 1, C and
D). These results suggest that the binding of MKP3 and
Elk1 to ERK2 is mutually exclusive and are consistent with the ability of Elk1
to block the ERK2-induced MKP3 activation
(Fig. 2A). Our data
also suggest that the binding of MKP3 and MEK1 to ERK2 is also mutually
exclusive (Fig. 4). It is
possible that ERK2 also uses both the CD site and the substrate-binding region
for MEK1 binding. Consistent with this hypothesis, the KIM sequence in MEK1
competes for the same CD site identified in this study
(6,
27). In addition, there is
also evidence suggesting an essential role for residues in G
and the MAP kinase insert in the interaction of ERK2 with MEK1
(49). The region defined by
G and the MAP kinase insert overlap with the Elk1
substrate-binding site uncovered in this study
(Table III and
Fig. 1D).
Although MEK1, MKP3, and Elk1 bind ERK2 in a mutually exclusive manner, they possess a different binding affinity for ERK2. We determined that the Kd values of MKP3 and Elk1 for ERK2 are 0.17 and 2.1 µM, respectively. In addition, our data indicate that the affinity of MKP3 and Elk1 for ERK2 is higher than that of MEK1 (Figs. 2 and 4). This difference in ERK2 binding affinity and the mutual exclusivity in ERK2 association are consistent with the sequence of ERK2 signaling events. The association of ERK2 with MEK1 in the cytosol leads to ERK2 phosphorylation and activation. The activated ERK2 is then sequestered from MEK1 by its substrates for phosphorylation. Because the expression of many MKPs is induced by ERK2 activation, the higher affinity of MKPs for ERK2 ensures efficient termination of ERK2 signaling after MKP induction.
A Bipartite Modular Model for ERK2 Interaction with Its Cognate Regulators and Substrates—We propose that the efficiency and fidelity of ERK2 signaling is achieved by a bipartite recognition process (Fig. 5). In this model, one part of the ERK2-binding proteins (e.g. the KIM sequence) docks to the CD site located on the back side of the ERK2 catalytic pocket for high affinity association. The major function of the KIM/CD interaction may be to increase the "effective concentration" of the interacting molecules. The ERK2 substrate-binding region is close to the ATP-binding pocket and the activation loop (Fig. 1, C and D), and encompasses the p + 1 site for substrate recognition. The interaction of this region with another structural element (e.g. the FXFP motif in MKP3 and Elk1) may not only stabilize binding but also provide contacts crucial for modulating the activity and/or specificity of ERK2 target molecules. Thus, the interaction of the substrate-binding region with ERK2 substrates may be required to organize the ERK2 active site with respect to the phosphoreceptor in the substrate for efficient phosphoryl transfer. Similarly, specific interaction of the substrate-binding region with MEK1 or MKP3 may ensure the precise orientation and positioning of the catalytic residues in MEK1 or MKP3 with respect to the TEY motif in the ERK2 activation loop for efficient phosphorylation or dephosphorylation. Taken together, it appears that the extremely high specificity of ERK2 for its cognate substrates, activators, and inactivators may result from both high affinity binding interactions between the KIM and the CD site and an ERK2 substrate-binding site-induced allosteric mechanism in which specific interactions between ERK2 and its binding proteins enable the attainment of optimal alignment of the catalytic residues with respect to the substrate for efficient catalysis.
|
Finally, it is important to point out that the CD site and the substrate-binding site for MKP3 and Elk1 employ both common and unique amino acid residues (Fig. 1). Definition of these unique protein/protein interactions will enhance our ability to design specific inhibitors to enable precise and highly localized manipulation of MAP kinase signaling. This should ultimately increase our understanding of the detailed mechanisms underlying the extracellular control of cell physiology and provide new insights into diseases associated with malfunctions of MAP kinase signaling activities.
|
FOOTNOTES |
|---|
¶ Irma T. Hirschl Career Scientist. To whom correspondence should be addressed: Dept. of Molecular Pharmacology, Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461. Tel.: 718-430-4288; Fax: 718-430-8922; E-mail: zyzhang{at}aecom.yu.edu.
1 The abbreviations used are: MAP kinase, mitogen-activated protein kinase;
CD site, common docking site; ERK, extracellular signal-regulated protein
kinase; GST, glutathione S-transferase; JNK, c-Jun N-terminal protein
kinase; KIM, kinase interaction motif; MBP, myelin basic protein; MEK, MAP
kinase/ERK kinase; MKP, MAP kinase phosphatase; pNPP,
p-nitrophenyl phosphate; MOPS, 4-morpholinepropanesulfonic acid.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
) and
Elk1-(307–428) (
) to inhibit the ERK2-induced activation of
MKP3-catalyzed pNPP hydrolysis reaction. The data were fit to
