M-phase MELK activity is regulated by MPF and MAPK.
ABSTRACT The protein kinase MELK is implicated in the control of cell proliferation, cell cycle and mRNA splicing. We previously showed that MELK activity is correlated with its phosphorylation level, is cell cycle dependent, and maximal during mitosis. Here we report on the identification of T414, T449, T451, T481 and S498 as residues phosphorylated in Xenopus MELK (xMELK) in M-phase egg extract. Phosphorylations of T449, T451, T481 are specifically detected during mitosis. Results obtained in vivo showed that MPF and MAPK pathways are involved in xMELK phosphorylation. In vitro, MPF and MAPK directly phosphorylate xMELK and MPF phosphorylates xMELK on T481. In addition, phosphorylation by MPF and MAPK enhances MELK activity in vitro. Taken together our results indicate that MELK phosphorylation by MPF and MAPK enhance its activity during M-phase.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Maternal embryonic leucine zipper kinase (MELK) functions as a modulator of intracellular signaling and affects various cellular and biological processes, including cell cycle, cell proliferation, apoptosis, spliceosome assembly, gene expression, embryonic development, hematopoiesis, and oncogenesis. In these cellular processes, MELK functions by binding to numerous proteins. In general, the effects of multiple protein interactions with MELK are oncogenic in nature, and the overexpression of MELK in kinds of cancer provides some evidence that it may be involved in tumorigenic process. In this review, our current knowledge of MELK function and recent discoveries in MELK signaling pathway were discussed. The regulation of MELK in cancers and its potential as a therapeutic target were also described.International Journal of Molecular Sciences 01/2013; 14(11):21551-60. · 2.46 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Maternal Embryonic Leucine zipper Kinase (MELK) was recently shown to be involved in cell division of Xenopus embryo epithelial cells. The cytokinetic furrow of these cells ingresses asymmetrically and is developmentally regulated. Two subpopulations of xMELK, the mMELK (for "mitotic" xMELK) and iMELK ("interphase" xMELK), which differ in their spatial and temporal regulation, are detected in Xenopus embryo. How cells regulate these two xMELK populations is unknown. In this study we show that, in epithelial cells, xMELK is present at a higher concentration at the apical junctional complex, in contrast to mesenchyme-like cells, which have uniform distribution of cortical MELK. Interestingly, mMELK and iMELK also differ by their requirements towards cell-cell contacts to establish their proper cortical localization both in epithelial and mesenchyme-like cells. Receptor for Activated protein Kinase C (RACK1), which we identified as an xMELK partner, co-localizes with xMELK at the tight junction. Moreover, a truncated RACK1 construct interferes with iMELK localization at cell-cell contacts. Collectively, our results suggest that iMELK and RACK1 are present in the same complex and that RACK1 is involved in the specific recruitment of iMELK at the apical junctional complex in epithelial cells of Xenopus embryos.Biology open. 01/2013; 2(10):1037-48.
- [Show abstract] [Hide abstract]
ABSTRACT: Elevated MELK expression is featured in multiple tumors and correlated with tumorigenesis and tumor development. This study is aimed to investigate the mechanisms of MELK-mediated development of gastric cancer.Molecular cancer. 05/2014; 13(1):100.
©2006 LANDES BIOSCIENCE. DO NOT DISTRIBUTE.
Catherine Jessus for critical reading of this paper.
This work was supported by the CNRS, the Ligue
Nationale Contre le Cancer (équipe labellisée) and
Cancéropôle Grand Ouest.
MELK is phosphorylated during mitosis indicating that an important regulation of
MELK takes place during M-phase. This view is supported by the recent finding that a
subpopulation of MELK is localised to the cell cortex at the metaphase/anaphase transition.17
Human MELK (hMELK) produced in bacteria is active and auto/trans phosphorylates.3,18,19
[Cell Cycle 5:8, 883-889, 15 April 2006]; ©2006 Landes Bioscience
Erich A. Nigg2
1CNRS UMR 6061 Génétique et Développement; Université de Rennes; Rennes,
2Department of Cell Biology; Max-Planck-Institute of Biochemistry; Martinsried,
3Laboratoire de Biologie du Développement; UMR-CNRS 7622; Université Pierre et
Marie Curie; Paris, France
†Current Address: Centre de Recherche de Biochimie Macromoléculaire; CNRS FRE
2593; 1919 Route de Mende; 34293 Montpellier cedex 5, France
*Correspondence to: Jean-Pierre Tassan; CNRS UMR 6061 Génétique et
Développement; Université de Rennes 1; IFR140 GFAS; Faculté de médecine,
2 avenue du Professeur Léon Bernard; CS 34317; 35043 Rennes Cedex, France;
Tel.: +126.96.36.199.46.89; Fax: +188.8.131.52.44.78; Email : jean-pierre.tassan@
Original manuscript submitted: 02/02/06
Manuscript accepted: 03/06/06
Previously published online as a CellCycle E-publication:
MELK, mitosis, MPF, MAPK, M-phase
hMELK human maternal embryonic leucine
Xenopus maternal embryonic leucine
M-phase promoting factor
mitogen-activated protein kinase
We are very grateful to Catherine Jessus for providing
us with p21cip1-GST and MBP-Mos and for
allowing realisation of part of this work in her
laboratory. In addition, we also would like to thank
M-phase MELK Activity is Regulated by MPF and MAPK
The protein kinase MELK is implicated in the control of cell proliferation, cell cycle and
mRNA splicing. We previously showed that MELK activity is correlated with its phospho-
rylation level, is cell cycle dependent, and maximal during mitosis. Here we report on the
identification of T414, T449, T451, T481 and S498 as residues phosphorylated in
Xenopus MELK (xMELK) in M-phase egg extract. Phosphorylations of T449, T451, T481
are specifically detected during mitosis. Results obtained in vivo showed that MPF and
MAPK pathways are involved in xMELK phosphorylation. In vitro, MPF and MAPK directly
phosphorylate xMELK and MPF phosphorylates xMELK on T481. In addition, phospho-
rylation by MPF and MAPK enhances MELK activity in vitro. Taken together our results
indicate that MELK phosphorylation by MPF and MAPK enhance its activity during
We previously showed that MELK (Maternal Embryonic Leucine zipper Kinase,1that
we previously named pEg3), is a cell cycle dependent protein kinase.2,3MELK catalytic
activity is correlated with its phosphorylation status and is maximal during mitosis in
Xenopus embryos and human cultured cells. Recently, MELK was found to be involved in
proliferation of neural cell progenitors.4MELK knockdown and overexpression, respec-
tively, decreases and increases proliferation of neural cell progenitors. The view that MELK
is involved in cell proliferation is in agreement with higher MELK expression observed in
proliferating cells.1,5-8Transcription of the MELK gene is regulated by Rb family members
and E2F transcription factors.9,10
How could MELK regulate the cell cycle? Human MELK associates with and
phosphorylates phosphatase CDC25B, an activator of the G2/M transition.3,11We
reported that expression of MELK fused with Green Fluorescent Protein (GFP) induced
accumulation of human cultured cells in G2phase of the cell cycle, an effect that was
counteracted by overexpression of CDC25B. This led to the conclusion that MELK is
potentially involved in cell cycle progression as a negative regulator of the G2/M transition.3
It has also been shown that MELK regulates some transcription events: MELK knockdown
affects expression of ZPR9 and B-MYB at the RNA level.4ZPR9 interacts with and is
phosphorylated by MELK.12ZPR9 also binds and regulates B-MYB,13a transcription
factor involved in cell cycle regulation and apoptosis (reviewed in ref. 14). Moreover, in
Zebra fish, a MELK-like kinase is involved in hematopoiesis.15Interestingly, the knock-
down of MELK-like kinase affected the expression of several transcription factors involved
in embryonic hematopoiesis and MELK-like overexpression enhanced the promoter activity
of one of the transcription factors. In addition to its capacity to interact with and phos-
phorylate the cell cycle regulator CDC25B and its role in transcription regulation, MELK
has been implicated in regulating splicing events. Specifically, MELK was shown to interact
in a mitotic phosphorylation dependent manner with the transcription and splicing factor
NIPP1, a protein phosphatase 1 inhibitor.16This interaction inhibits spliceosome assembly
and could contribute to the inhibition of mRNA splicing observed during mitosis. Taken
together these results suggest that MELK is implicated in cell cycle control through different
MELK Activity is Enhanced by MPF and MAPK
Threonine 167 (T167) situated in the activation loop (T-loop) of
the catalytic domain is phosphorylated in bacterially produced hMELK
and is required for kinase activity.18,19This indicated that hMELK
can auto/trans phosphorylate and autoactivate in bacteria. Whether
this is also the case in vertebrate cells is at present unknown.
In the present study, we report the identification of T414, T449,
T451, T481 and S498 phosphorylation sites in Xenopus MELK
(xMELK) in M-phase egg extract. We also identified 14 residues
phosphorylated in recombinant human MELK (hMELK) expressed
and purified from bacteria. Experiments performed in vivo impli-
cated MPF (mitosis promoting factor, the complex of CDK1 and
cyclin B) and MAPK (mitogen-activated protein kinase) pathways in
xMELK phosphorylation. In vitro, MPF and MAPK directly phos-
phorylate xMELK and enhance its kinase activity. In addition, we
show that in vitro MPF phosphorylates T481. M-phase specific
phosphorylations of T449, T451, T481 were detected during
Xenopus oocyte maturation, in embryos, and in Xenopus cultured
cells. Taken together these results indicate that MELK activity is
regulated by phosphorylation by MPF and MAPK during M-phase.
MATERIALS AND METHODS
Preparation of Xenopus oocytes, eggs, embryos and M-phase extracts.
Fully grown oocytes were prepared as described.20Selected oocytes were
microinjected with p21cip1-GST at the intracellular concentration of 1 µM
or incubated in the presence of 50 µM U0126 (Promega). Meiotic maturation
was induced by 1 µM progesterone or by injection of 50 µg of MBP-Mos
(MBP is for maltose binding protein). Oocytes were referred to as GVBD
when a well defined white spot was formed, and were assumed to be in
metaphase II 3 hours after GVBD. CHX was added at 100 µg/ml for
1 hour. For extracts preparation, oocytes were lysed in 5 volumes of EB
(8 mM β-glycerophosphate, 20 mM EGTA, 15 mM MgCl2, 1 mM DTT,
pH 7.3), supplemented with protease cocktail inhibitor (P8340, Sigma),
and then centrifuged at 15000 g for 15 minutes at 4˚C. Xenopus eggs and
embryos were obtained as previously described.2Mitotic egg extracts were
prepared as described by Murray.21
Culture of XL2 cells and mitotic synchronization. Xenopus XL2 cells22
were grown as described previously.2To obtain cells blocked in mitosis, cells
were first arrested in S-phase by addition of 2 µg/ml of aphidicolin (Sigma
Chemicals) to the culture medium for 30 hours. Cells were released from the
aphidicolin block during 7 hours and mitotic cells were collected after 15 hours
of treatment with 40 µg/ml ALLN (N-acetyl-leucyl-leucyl-norleucinal,
Production of recombinant proteins. Plasmids allowing expression of
MBP-xMELK, his-tagged hMELK wild-type (WT) and K/R, p21cip1-GST
and MBP-Mos and purification of hMELK WT and K/R, p21cip1-GST and
MBP-Mos were described previously.2,3,23,24Expression and purification of
MBP-xMELK were performed following manufacturer’s instructions (New
England Biolabs Inc.). Radiolabeled xMELK was produced in TNT T7
Quick rabbit reticulocyte lysate according to the manufacturer’s instructions
(Promega Inc.) from plasmid pT7T- xMELK.2
Incubation of xMELK in mitotic egg extract. Recombinant xMELK
produced in rabbit reticulocyte lysate was added to CSF extract (1/6 vol/vol)
and incubated at 21˚C for one hour. For mass spectrometry, 15 µg of MBP
or MBP-xMELK were incubated with 800 µl of CSF extract or for control
with the same volume of extraction buffer (10 mM Hepes, pH 7.7, 100 mM
KCl, 2 mM MgCl2, 7 mM EGTA, 50 mM sucrose) at 21˚C for two hours.
In case of radiolabeling of xMELK, 10 µCi of [γ-32P] ATP were added to
50 µl of extract. Subsequently, MBP and MBP-xMELK were repurified as
follows: First, extract was diluted 1/10 in column buffer (10 mM phosphate
buffer, 0.5 M NaCl, 10 mM β-mercaptoethanol pH 7.2) supplemented
with pepstatine, leupeptine, chymostatine, PMSF at 10 µM each, a cocktail
of protease inhibitor (Roche), 40 mM NaF, 40 mM β-glycerophosphate,
and 2.5 µM okadaic acid. Diluted extract was incubated with amylose resin
(New England Biolabs Inc.) at 4˚C for one hour. After washing of the beads
with column buffer supplemented with protease and phosphatase
inhibitors, proteins were recovered by addition of protein sample buffer and
denaturated at 95˚C for 5 minutes. Proteins were separated on a 8% poly-
acrylamide gel, stained with 0.02% Coomassie R250 in 10% acetic acid,
and destained in water.
Mass spectrometry. Protein digestion was performed essentially as
reported25and peptides were desalted and concentrated using miniaturized
micro-extraction tips.26Purified peptide mixtures were then analyzed on a
matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF)
mass spectrometer (Reflex III, Bruker Daltonik, Bremen, Germany) in
positive and negative ion reflector mode using 2,5-dihydroxybenzoic acid
(Bruker Daltonik, Bremen, Germany) as matrix. Spectra were searched for
peptide pairs with a mass difference of the phosphate group (80 Dalton)
using an in-house-developed software script. Peptides which showed the
characteristic losses of phosphoric acid (98 Dalton) and phosphate (80
Dalton) upon post-source decay analysis27were considered to be phospho-
rylated. To determine the site of phosphorylation within the peptide sequence,
the phosphopeptides were subjected to collision-induced dissociation (CID)
on a quadrupole time-of-flight mass spectrometer (Q-TOF Ultima, Waters,
Manchester, UK) equipped with a nano electrospray ionization source.28
Spectra were inspected individually and fragments containing the phosphate
group or showing a characteristic strong loss of phosphoric acid were
considered to contain the phosphorylated amino acid.
Production of anti-phosphoxmelk antibodies. Polyclonal antibodies
specific to pT449, pT451, and pT481 were obtained by immunisation of
rabbits with peptides the PAPW(p)TPTPRRKQNEC, PAPWTP(p)
TPRRKQNEC, and QSKE(p)TPTKKPIGTGEC, respectively. Immune
sera were then affinity-purified and antibodies reacting against unphospho-
rylated peptides were discarded by a nonphosphorylated peptide column
MPF, MAPK and hMELK kinase assays. One µg of purified hMELK
was incubated with or without 20 U of MPF or 200 U of MAPK (New England
Biolabs Inc.) in a final volume of 20 µl of kinase buffer (New England
Biolabs Inc.) in the presence of 100 µM ATP. After incubation at 30˚C
during 20 minutes, 30 µl of 5 mM magnesium acetate containing 15 µg of
AMARA peptide29and 5 µCi of [γ-32P] ATP (3000 Ci/mmol) were added
to the reaction followed by an additional incubation at 30˚C for 10 minutes.
Incorporation of 32P into the peptide was measured as described by Browne
Lambda phosphatase treatment. Proteins extracted from five eggs were
incubated at 30˚C for 30 minutes with 500 units of lambda phosphatase in
the appropriate buffer (New England Biolabs, Inc.).
Western blot with L2, anti-phosphoMELK, P-MAPK, cyclin B2 and
β-tubulin. Anti-xMELK antibody (L2) was described previously.2Antibodies
against pT449, pT451 and pT481 were used at 5.5 µg/ml, 7.5 µg/ml, and
2.25 µg/ml, respectively. Anti-phosphoMAPK (monoclonal E-4, Santa-Cruz)
was diluted 1/1000, anti-cyclin B2 (a generous gift from Thierry Lorca) was
used 1/2000 and anti-β-tubulin (TUB 2.1, Sigma) was diluted 1/1000. All
antibodies were incubated in skimmed milk, 5% in TBST,2with the excep-
tion of anti-phospho MAPK and anti-cyclin B2 antibodies which were used
in 3% BSA.
In vivo MELK phosphorylation is dependent on MPF and MAPK
activities. Xenopus oocytes are blocked in prophase (G2-like phase) of meiosis I.
In response to progesterone, MPF and MAPK are activated, leading to
GVBD (germinal vesicle breakdown) and the first meiotic division. MPF
then declines, due to cyclin B degradation, and reincreases due to cyclin B
accumulation, thus inducing entry into metaphase II. In this situation MPF
and MAPK activities are stabilized. Phosphorylation of xMELK during
maturation of Xenopus oocytes results in a shift in electrophoretic mobility
at the time of GVBD (Fig. 1A and see ref. 2). The influence of protein
synthesison xMELK phosphorylation was tested during in vitro oocyte maturation
induced by progesterone. The protein synthesis inhibitor cycloheximidine
2006; Vol. 5 Issue 8
MELK Activity is Enhanced by MPF and MAPK
(CHX) was added to oocytes, either at GVBD, or when oocytes were
blocked in metaphase II (MII). As shown in Figure 1A, cyclin B synthesis
was inhibited when GVBD oocytes were incubated for one hour with CHX.
The MAPK pathway was also inhibited as shown by the large reduction of
MAPK in its active phosphorylated form. In these oocytes, xMELK exhib-
ited the same electrophoretic mobility as in prophase oocytes, whereas in
control GVBD oocytes, the electrophoretic mobility of xMELK was indis-
tinguishable to that observed in matured oocytes. This result indicates that
xMELK phosphorylation is reversible in GVBD oocytes and that xMELK is
dephosphorylated when both MPF and MAPK are inactivated. In contrast,
xMELK electrophoretic mobility remained retarded in CHX treated MII
oocytes, and as expected, cyclin B level was unaffected, and MAPK remained
under its active phosphorylated form. Thus, xMELK phosphorylation
depends on protein synthesis during the metaphase I–metaphase II transition
while it is stabilized during metaphase II arrest. Therefore the phosphorylation
control of xMELK is highly correlated to MPF and MAPK activities.
The respective contributions of the MPF and MAPK pathways on
xMELK phosphorylation were ascertained in vivo. To prevent MAPK
activation, prophase oocytes were treated with MEK (MAPK activating
kinase) inhibitor U0126 and MPF activation was triggered by progesterone.
In control oocytes, MPF was activated as revealed by the electrophoretic
shift of cyclin B2 and MAPK was activated by phosphorylation (Fig. 1B).
As expected, in U0126 treated oocytes, MAPK was not phosphorylated
whereas cyclin B was still shifted indicating that U0126 suppressed MAPK
activation without preventing MPF activation. In U0126 treated oocytes,
xMELK underwent the same electrophoretic mobility as in matured oocytes.
Thus, when MAPK was kept inactive, the phosphorylation dependent shift
in xMELK electrophoretic motility was not inhibited. In order to determine
the contribution of the MAPK pathway on xMELK phosphorylation,
prophase oocytes were first injected with p21cip1to block MPF activation23
followed by a second injection with recombinant MBP-Mos leading to
MAPK activation. Under these conditions, MPF activation was prevented as
revealed by the presence of unshifted cyclin B2 whereas MAPK was active
(Fig. 1C). In this context, xMELK appeared as an intermediate phosphory-
lated form. All together these results suggest that the MPF and MAPK
pathways are involved in xMELK phosphorylation during the G2-M transition
Identification of phosphorylation sites in xMELK and hMELK. To
obtain a more comprehensive knowledge of the phosphorylation of xMELK
during M-phase, a mass spectrometry analysis was undertaken to identify
phosphorylation sites. The amount of endogenous xMELK recovered by
immunoprecipitation from Xenopus eggs was too limited to proceed to
analysis. To solve this problem, recombinant MBP-xMELK produced in
Escherichia coli was incubated with a Xenopus CSF (Cytostatic Factor)
extract naturally blocked in metaphase by CSF activity (reviewed
in ref. 21). As expected, xMELK incubated in CSF extract was
phosphorylated (as demonstrated by 32P incorporation) whereas a
similar amount of MBP protein alone was not (Fig. 2A). Five
phosphopeptides were identified by mass spectrometry in recom-
binant xMELK incubated with CSF extract but not in xMELK
incubated with buffer (Fig. 2B). Fragmentation of the phospho-
peptides allowed the identification of four phosphoamino acids:
threonine 414 (T414) in peptide 1, threonine 449 (T449) and
threonine 451 (T451) in peptide 2 and serine 498 (S498) in
peptide 3. We were unable to distinguish between threonine 481
and 483 phosphorylation in peptide 4 and between serine 505 and
517 in peptide 5. In order to obtain further insight into MELK
phosphorylation, we also identified phosphorylation sites in human
MELK (hMELK) produced in E. coli (Fig. 2C) which is known to
be auto/trans phosphorylated.3,18,19Fourteen phosphoamino
acids were identified (Table 1) of which some were independently
reported recently.18,19These include T167 in the activation loop
of the catalytic domain which was reported to be required for
MELK activity.18,19Remarkably, two tyrosines were found to be
auto/trans phosphorylated even though MELK is classified as a
serine/threonine kinase. Autophosphorylation on a tyrosine
residue was also reported for serine/threonine kinase Aurora A when expressed
in bacteria.31Interestingly, in hMELK the equivalent of Xenopus serine 505
was found auto/trans phosphorylated leading us to suggest that this residue
may be phosphorylated in xMELK. As shown in Figure 3, phosphorylation
sites identified in Xenopus and human MELK were mostly situated in the
median domain suggesting a role of this domain in MELK regulation. Three
out of the four phosphorylation sites identified in xMELK were conserved
in human and one in mouse. Remarkably, all the identified phosphoamino
acids in xMELK are MPF or MAPK phosphorylation consensus sites which
is consistent with the hypothesis that xMELK could be a MPF and MAPK
MPF and MAPK phosphorylate MELK and enhance its activity in vitro.
In order to test if MELK could be a MPF or a MAPK substrate, in vitro
phosphorylation experiments were performed. Radiolabeled xMELK was
produced in reticulocyte lysate and incubated with either Xenopus CSF
extract, MPF or MAPK. Subsequently, the phosphorylation state of xMELK
was monitored by the electrophoretic mobility of xMELK. As described
previously,2xMELK incubated with egg extract showed a reduced elec-
trophoretic mobility which is characteristic of phosphorylated xMELK
(Fig. 4A). After incubation with MPF or MAPK, xMELK electrophoretic
mobility was decreased. xMELK incubated with MPF appeared more shifted
than with MAPK in correlation with the previous in vivo observations (Fig. 1),
but less than with CSF extract. Moreover, incubation with MPF and MAPK
together induced a greater shift in xMELK electrophoretic mobility than
using them separately, but this shift was still smaller than the shift induced
by incubation with CSF extract suggesting that an unidentified third kinase
may be involved. This result indicates that in vitro MPF and MAPK are able
to induce xMELK phosphorylation and that these phosphorylations have
cumulative effects on xMELK electrophoretic motility. However we could
not exclude the possibility that MPF and MAPK did not phosphorylate
directly xMELK but instead activated intermediate kinases present in rabbit
To deepen the study of MELK phosphorylation by MPF and MAPK,
rabbit polyclonal antibodies were raised against phosphorylated sites detected
by mass spectrometry. Phosphopeptide fragmentation did not allow to
distinguish between T481 and T483 phosphorylation. However, since T481
is followed by a proline, it makes this residue a more likely candidate to be
phosphorylated by MPF than T483 which is followed by a lysine. Therefore
we raised antibodies against pT449, pT451, pT481, pS498 and pS505.
Because computer assisted analysis predicted that the sequence surrounding
T414 does not constitute a potential immunogenic environment, we did
not raise antibody against this site. Antibodies were tested and selected for
their ability to specifically recognize endogenous phosphorylated xMELK in
Xenopus egg extract. The antibodies against pS498 and pS505 did not detect
Figure 1. Contribution of MPF and MAPK pathways to xMELK phosphorylation during
oocyte maturation. (A) Oocyte maturation was released by incubation of prophase I
oocytes (PI) with progesterone. Cycloheximidine (CHX) was added at GVBD or at
metaphase II (MII) for 1 hour. (B) Prophase oocytes were incubated with or without
50 µM U0126 (U0) for 1 hour. Then progesterone (Pg) was added or not (-).
(C) Prophase oocytes were microinjected or not (-) with p21cip1 (cip) to an internal
concentration of 1 µM. One hour later they were microinjected with 50 µg of recom-
binant MBP-Mos (Mos) or treated with progesterone (Pg). In each case (A–C), oocytes
were collected after 50% GVBD was reached in progesterone treated control oocytes
and proteins were analysed by Western blot with anti-xMELK, anti-phospho-MAPK
and anti-cyclin B2.
MELK Activity is Enhanced by MPF and MAPK
xMELK. However, as shown in Figure 4B,
antibodies against pT449, pT451 and pT481
recognized endogenous xMELK in Xenopus
egg extract. When Xenopus egg extract was
treated with lambda phosphatase, the signals
were lost, indicating that these antibodies are
specific to phosphorylated xMELK. To con-
tinue the characterization of these antibodies,
recombinant MBP-xMELK was phosphory-
lated by MPF and MAPK, as controlled by 32P incor-
poration (Fig. 4C), further indicating that in vitro,
both kinases directly phosphorylate xMELK. MBP
alone was not phosphorylated (data not shown). In all
cases xMELK was recognized by the L2 polyclonal
antibody. pT481 was detected after xMELK phospho-
rylation by MPF but not by MAPK. This result is in
agreement with sequence surrounding T481 which cor-
responds to a MPF consensus phosphorylation site.
MBP-xMELK was not phosphorylated in vitro by Plx1
or Aurora-A, two mitotic kinases representing good
candidates as xMELK kinases (data not shown). Kinases
which phosphorylate T449 and T451 remain to be
identified as well as the phosphorylation site(s) targeted
We next investigated the effect of phosphorylation
by MPF and MAPK on MELK catalytic activity. We
took advantage of the fact that human recombinant
MELK is already phosphorylated at T167 and active
and the recent finding that the AMARA peptide can be
used as a substrate to measure MELK activity in
vitro.18,19As shown in Figure 4D, the activity of
Figure 2. Identification of sites phosphorylated
in recombinant xMELK after incubation with
mitotic egg extract. (A) Equal amounts of MBP
or MBP-xMELK were incubated with a
Xenopus CSF extract in the presence of [γ-32P]
ATP. Proteins were repurified, resolved by
SDS-PAGE and stained with Coomassie blue.
Phosphorylation was detected by autoradiog-
raphy. (B) MBP-xMELK incubated as in (A),
without addition of radiolabelled ATP, was
submitted to mass spectrometry. Fragmentation
spectrum of the doubly charged phosphopep-
tide GRLEFNSVDSAPApTPVVQR (1), the triply
charged, singly phosphorylated, isobaric
peptides DENVFLHPAPWpTPTPR and DEN-
VFLHPAPWTPpTPR (2) and the triply charged
phosphopeptide KPIGTGEEFANVIpSPER (3).
The most intense N- and C-terminal fragments
(marked b and y, respectively) are marked in
the spectrum, and the phosphorylated fragments
(+ph) and fragments showing a predominant
loss of phosphoric acid (-ph.acid) are shown
in red. A scheme with all observed N- and
C-terminal fragments is shown below the
spectra. (B) phosphopeptides NQSKETPTK (4)
and SVELDLNQAHIDSAQK (5) were detected
singly phosphorylated without unambigouasly
determining which amino acid was phospho-
rylated. For peptide (4), phosphorylation of
the serine residue could be excluded based
on the fragmentation analysis. Potential
phosphoamino acids are indicated in blue.
(C) Coomassie blue staining of wild-type (WT)
and an inactive K/R mutant hMELK produced
in E. coli.
Identification of phosphorylated sites in recombinant human
MELK expressed in bacteria
2006; Vol. 5 Issue 8
MELK Activity is Enhanced by MPF and MAPK
hMELK towards AMARA was considerably increased after phosphorylation
by MPF or MAPK (7 and 10 times respectively). MPF and MAPK alone
phosphorylated AMARA very poorly (Fig. 4D). When hMELK was phos-
phorylated by MPF and MAPK in the same reaction, hMELK activity was
further increased (15 times, Fig. 4D). We conclude that, in vitro, MPF and
MAPK directly phosphorylate MELK and enhance its catalytic activity.
MPF directly phosphorylates T481 residue. Moreover, a third yet unidenti-
fied kinase could be involved in MELK phosphorylation during M-phase.
Phosphorylations of T449, T451 and T481 are cell cycle dependent.
Global changes in the xMELK phosphorylation state during oocyte matu-
ration were evident by the appearance of retarded xMELK bands (Fig. 5).
Antibodies against pT449, pT451 and pT481 detected xMELK in GVBD
oocytes (corresponding to metaphase I) and in maturated oocytes (MII,
metaphase II) but not in prophase oocytes (PI). As shown previously,
xMELK is rapidly dephosphorylated after fertilization, rephosphorylated
during the first embryonic cell cycle, but with a less important shift in its
apparent molecular weight compared to eggs. After mitosis, the level of
xMELK phosphorylation decreases (Fig. 5 and see ref. 2). Anti-phospho-
peptide antibodies did not detect xMELK in embryos 40 minutes after
fertilization, phosphoepitopes reappeared and peaked around 90 minutes
after fertilization corresponding to mitosis as marked by cyclin B2 resynthesis
(Fig. 5). Phosphorylation of xMELK was also detected during the second
embryonic cell cycle (100 minutes) after embryos had passed the two cells
stage. These results indicated that T449, T451 and T481 are phosphory-
lated during meiosis and mitosis in the first embryonic cell cycle. These
antiphosphopeptides were then used to follow xMELK phosphorylation in
Xenopus XL2 cultured cells since xMELK is phosphorylated during mitosis
as shown by the appearance of retarded bands in the Western blot with L2
antibody (Fig. 5 and see ref. 2). Antibodies against pT449, pT451 and pT481
detected xMELK in XL2 cells synchronized in mitosis but not in asynchro-
nously growing cells. This indicates that, in vivo, T449, T451 and T481
residues are phosphorylated specifically during M-phase in maturing oocytes
and somatic cells.
Figure 3. Conservation of phosphorylated sites
in Xenopus and mammalian MELK. Scheme
represents the MELK structure with the N-terminal
kinase catalytic domain and the conserved
C-terminal domain separated by the poorly
conserved median domain (M). The deduced
amino acid sequences of the Xenopus (Xl),
human (Hs) and mouse (Mm) MELK are aligned
into a part of the median domain. Phosphorylated
residues are indicated in red for Xenopus MELK
and underlined in blue for the human protein.
The phosphopeptide NQSKETPTK was identified
by mass spectrometry and phosphorylation of
T481 was subsequently demonstrated (see below).
Circles indicate that the phosphoamino acid
detected in xMELK is conserved in human and
a star indicates that this amino acid is also
conserved in mouse protein. The wave indicates
that the same amino acid is phosphorylated in
Xenopus and human protein.
Figure 4. in vitro MPF and MAPK phosphorylate MELK and enhance its activity.
(A) Radiolabelled xMELK produced in rabbit reticulocyte lysate was incu-
bated with either Xenopus CSF extract (CSF), MPF or MAPK, then resolved
by SDS-PAGE and detected by autoradiography. (B) Proteins extracted from
unfertilized eggs (UFE) were dephosphorylated by treatment with λ-phos-
phatase and then analyzed by Western-blot analysis with anti-xMELK,
anti-pT449, anti-pT451 and anti-pT481 antibodies. (C) Equal amounts of
purified recombinant MBP-xMELK were incubated with either MPF or MAPK
or kinase buffer (-) and [γ-32P] ATP. xMELK phosphorylation was followed by
32P incorporation and Western-blot analysis was performed with anti-xMELK
and anti-pT481 antibodies. (D) hMELK was incubated with MPF, MAPK, MPF
and MAPK and its activity was measured in vitro using the AMARA peptide.
Results are shown as the mean value obtained from three different experiments.
MELK Activity is Enhanced by MPF and MAPK
The present work allowed identification
of T414, T449, T451, T481 and S498 as
residues phosphorylated in xMELK. T449,
T451 and T481 were shown to be specifi-
cally phosphorylated during M-phase in
contexts as different as oocyte maturation,
embryos and cultured cells. A combination
of in vivo and in vitro approaches allowed
the identification of MPF and MAPK as
xMELK-upstream kinases. In vivo, MPF
and MAPK activities are involved in the
apparent molecular weight shift which
accompanies xMELK phosphorylation
during oocyte maturation. In vitro, MPF
and MAPK directly phosphorylate recom-
binant xMELK and induce a shift in the
electrophoretic mobility of xMELK. In
addition, MPF creates the phosphoepitope
recognized by the anti-pT481 antibody,
thus demonstrating that MPF directly
phosphorylates xMELK in vivo. These results
support the view that xMELK is a direct
target of MPF and MAPK. In addition, MELK phosphorylation by
MPF and MAPK enhances recombinant hMELK activity in vitro.
Taken together our results suggest that MPF and MAPK control
xMELK activity during M-phase.
Phosphorylation of human T478 was reported to create an inter-
action site for NIPP1 and prevented spliceosome assembly in mitotic
cell extract.16Recombinant hMELK expressed in asynchronously
growing mammalian cells did not interact with NIPP1 whereas an
interaction was detected in mitotic cells. This result is in agreement
with our finding that phosphorylation of Xenopus T481 (corre-
sponding to human T478) is mitosis specific. Moreover, our results
suggest that MPF may control MELK/NIPP1 interaction by phos-
phorylating T478/T481. Surprisingly, a recombinant hMELK
mutant in which T478 was substituted by alanine (T478A), purified
from asynchronous SF9 insect cells and renaturated, was reported to
be more active than wild-type hMELK thus suggesting that T478
phosphorylation may be inhibitory.16Indeed, in asynchronous cells,
hMELK is expected to be unphosphorylated on T478 and thus as
active as the unphosphorylated T478A mutant. The alanine substi-
tution may induce a conformational change that could increase
T478A mutant activity. We also do not exclude the possibility that
MPF could phosphorylate MELK on additional sites inducing a
global increase in MELK activity.
Results obtained by Vulsteke et al.16also showed that mutation
of T446 (human) in a non phosphorylable residue alters, although
to a lesser extend than T478, MELK/NIPP1 interaction in the yeast
two-hybrid system. Our results show that T449 (corresponding to
human T446) is phosphorylated in vivo during mitosis, thus
reinforcing the view that T446 phosphorylation may be involved in
NIPP1 interaction. Regulation of MELK/NIPP1 interaction is
probably not the only function of MELK phosphorylation by MPF
since MELK activity is stimulated by MPF whereas spliceosome
inhibition was reported to be independent of MELK activity.16
Phosphorylations of T449 and T451 were demonstrated to occur
in vivo during mitosis but the kinase(s) responsible for T449 and
T451 phosphorylations remain(s) to be identified. We could not
identify the phosphorylated site in one additional phosphopeptide
(SVELDLNQAHIDSAQK). This peptide, however, contains S505,
and the equivalent residue was found auto/trans phosphorylated in
recombinant hMELK expressed in E. coli (our results and see ref. 19).
S505 is not followed by a proline, thus excluding MPF and MAPK
as kinases responsible for this phosphorylation. Interestingly, S505 is
preceded by arginines at position -3 and -4 (relative to phosphory-
lated residue). These positions are also occupied by arginines in the
SAMS and AMARA peptides, respectively, two in vitro MELK
substrates.18,19We propose that S505 could be phosphorylated in
xMELK and that it could correspond to an auto/trans phosphoryla-
tion site. T414 and S498 were also found phosphorylated in xMELK
incubated with mitotic egg extract. In recombinant hMELK, we have
detected the phosphorylation of T409, which is equivalent to Xenopus
T414. This result is in contrast to a study of Beullens et al.19in which
S407 was determined as the phosphorylated residue in this peptide.
T414 is not preceded by a arginine at positions -3 or -4 suggesting
that it may not be an in vivo auto/trans phosphorylation site in
xMELK. Instead, T414 and S498 lie in potential MAPK and MPF
phosphorylation sites, respectively, which suggests that they might be
phosphorylated by MAPK or MPF or other proline-directed kinase(s)
that remain(s) to be identified.
Mass spectrometry analysis of active recombinant hMELK
allowed identification of 13 auto/trans phosphorylation sites and in
particular T167 (this study and see refs. 18 and 19). T167 lies in the
activation loop (T-loop) of the catalytic domain of hMELK and
phosphorylation of T167 contributes to MELK activation.18,19
Taking together the results obtained in our study and previous
reports18,19indicates that recombinant hMELK expressed in bacteria
can auto/trans phosphorylate on at least 20 sites including ten
serines, seven threonines and three tyrosines. This broad in vitro
phosphorylation spectrum supports the hypothesis that in vivo, a
mechanism exists to control MELK substrate specificity. This
control could correspond to a post-translational modification or an
interaction with another protein. The identification of this regulatory
mechanism will be important to identify additional relevant in vivo
Figure 5. xMELK phosphorylation in vivo. Oocytes were collected in prophase I (PI), at GVBD and in
metaphase II (MII). Embryos were collected every 10 minutes from 40 to 110 minutes after fertilization.
All embryos were at the two cells stage by 100 minutes. Xenopus XL2 cells were blocked into mitosis
(M) with ALLN or grown asynchronously (As). Western blot analyses were performed with anti-xMELK,
anti-pT449, anti-pT451, anti-pT481, anti-cyclin B2 and anti-β-tubulin antibodies.
2006; Vol. 5 Issue 8
MELK Activity is Enhanced by MPF and MAPK
In xMELK and in recombinant hMELK, most of the phospho-
rylation sites are in the median domain (our results and see ref. 19)
suggesting an important role of this domain in MELK regulation. It
was hypothesised that MELK activity could be controlled by
autoinhibition by its non catalytic domain.19Phosphorylation of the
medium domain may contribute to the release of this MELK autoin-
hibition control. MELK was recently described to localize at the cell
cortex specifically during mitosis.17This localization is dependent
on the C-terminal domain but phosphorylation of the median
domain may regulate the timing of this localization. At present, we
have no insights into the physiological role of xMELK phosphoryla-
tion on T414, T449, T451 and S498. However, it is noteworthy
that four of the five phosphorylated sites identified in xMELK are
conserved in the human protein. Only T451, which was shown to
be phosphorylated in vivo, is not conserved in hMELK. This argues
in favour of an important role of these phosphorylations.
Surprisingly, only one phosphorylated residue identified in this
study is conserved in mouse. Regulation of MELK in mouse might
be different or could require phosphorylation at a more limited
number of sites. It is also possible that, in this species, phosphorylation
takes place at cryptic sites but having identical functional roles.
The mitotic MELK phosphorylation status appears complex.
However, characterization of MELK phosphorylation sites and iden-
tification of upstream kinases are essential steps in the understanding
of MELK regulation and its role in cell proliferation.
1. Heyer BS, Warsowe J, Solter D, Knowles BB, Ackerman SL. New member of the Snf1/
AMPK kinase family, Melk, is expressed in the mouse egg and preimplantation embryo.
Mol Reprod Dev 1997; 47:148-56.
2. Blot J, Chartrain I, Roghi C, Philippe M, Tassan JP. Cell Cycle regulation of pEg3, a new
Xenopus protein kinase of the KIN1/PAR-1/MARK Family. Dev Biol 2002; 241:327-38.
3. Davezac N, Baldin V, Blot J, Ducommun B, Tassan JP. Human pEg3 kinase associates with
and phosphorylates CDC25B phosphatase: A potential role for pEg3 in cell cycle regulation.
Oncogene 2002; 21:7630-41.
4. Nakano I, Paucar AA, Bajpai R, Dougherty JD, Zewail A, Kelly TK, Kim KJ, Ou J, Groszer
M, Imura T, Freije WA, Nelson SF, Sofroniew MV, Wu H, Liu X, Terskikh AV, Geschwind
DH, Kornblum HI. Maternal embryonic leucine zipper kinase (MELK) regulates multi-
potent neural progenitor proliferation. J Cell Biol 2005; 170:413-27.
5. Heyer BS, Kochanowski H, Solter D. Expression of Melk, a new protein kinase, during
early mouse development. Dev Dyn 1999; 215:344-51.
6. Rhodes DR, Yu J, Shanker K, Deshpande N, Varambally R, Ghosh D, Barrette T, Pandey
A, Chinnaiyan AM. Large-scale meta-analysis of cancer microarray data identifies common
transcriptional profiles of neoplastic transformation and progression. Proc Natl Acad Sci
USA 2004; 101:9309-14.
7. Gray D, Jubb AM, Hogue D, Dowd P, Kljavin N, Yi S, Bai W, Frantz G, Zhang Z,
Koeppen H, de Sauvage FJ, Davis DP. Maternal embryonic leucine zipper kinase/murine
protein serine-threonine kinase 38 is a promising therapeutic target for multiple cancers.
Cancer Res 2005; 65:9751-61.
8. Hemmati HD, Nakano I, Lazareff JA, Masterman-Smith M, Geschwind DH,
Bronner-Fraser M, Kornblum HI. Cancerous stem cells can arise from pediatric brain
tumors. Proc Natl Acad Sci USA 2003; 100:15178-83.
9. Black EP, Huang E, Dressman H, Rempel R, Laakso N, Asa SL, Ishida S, West M, Nevins
JR. Distinct gene expression phenotypes of cells lacking Rb and Rb family members. Cancer
Res 2003; 63:3716-23.
10. Verlinden L, Eelen G, Beullens I, Van Camp M, Van Hummelen P, Engelen K, Van
Hellemont R, Marchal K, De Moor B, Foijer F, Te Riele H, Beullens M, Bollen M,
Mathieu C, Bouillon R, Verstuyf A. Characterization of the condensin component Cnap1
and protein kinase Melk as novel E2F target genes downregulated by 1,25-dihydroxyvita-
min D3. J Biol Chem 2005; 280:37319-30.
11. Mirey G, Chartrain I, Froment C, Quaranta M, Bouche JP, Monsarrat B, Tassan JP,
Ducommun B. CDC25B phosphorylated by pEg3 localizes to the centrosome and the
spindle poles at mitosis. Cell Cycle 2005; 4:806-11.
12. Seong HA, Gil M, Kim KT, Kim SJ, Ha H. Phosphorylation of a novel zinc-finger-like
protein, ZPR9, by murine protein serine/threonine kinase 38 (MPK38). Biochem J 2002;
13. Seong HA, Kim KT, Ha H. Enhancement of B-MYB transcriptional activity by ZPR9, a
novel zinc finger protein. J Biol Chem 2003; 278:9655-62.
14. Sala A. B-MYB, a transcription factor implicated in regulating cell cycle, apoptosis and can-
cer. Eur J Cancer 2005; 41:2479-84.
15. Saito R, Tabata Y, Muto A, Arai K, Watanabe S. Melk-like kinase plays a role in
hematopoiesis in the zebra fish. Mol Cell Biol 2005; 25:6682-93.
16. Vulsteke V, Beullens M, Boudrez A, Keppens S, Van Eynde A, Rider MH, Stalmans W,
Bollen M. Inhibition of spliceosome assembly by the cell cycle-regulated protein kinase
MELK and involvement of splicing factor NIPP1. J Biol Chem 2004; 279:8642-7.
17. Chartrain I, Couturier A, Tassan JP. Cell cycle dependent cortical localization of pEg3
protein kinase in Xenopus and human cells. Biol Cell 2006; 98:253-63.
18. Lizcano JM, Goransson O, Toth R, Deak M, Morrice NA, Boudeau J, Hawley SA, Udd L,
Makela TP, Hardie DG, Alessi DR. LKB1 is a master kinase that activates 13 kinases of the
AMPK subfamily, including MARK/PAR-1. Embo J 2004; 23:833-43.
19. Beullens M, Vancauwenbergh S, Morrice N, Derua R, Ceulemans H, Waelkens E, Bollen
M. Substrate specificity and activity regulation of protein kinase MELK. J Biol Chem
20. Jessus C, Thibier C, Ozon R. Levels of microtubules during the meiotic maturation of the
Xenopus oocyte. J Cell Sci 1987; 87:705-12.
21. Murray AW. Cell cycle extracts. Methods Cell Biol 1991; 36:581-605.
22. Anizet MP, Huwe B, Pays A, Picard JJ. Characterization of a new cell line, XL2, obtained
from Xenopus laevis and determination of optimal culture conditions. In Vitro 1981;
23. Frank-Vaillant M, Jessus C, Ozon R, Maller JL, Haccard O. Two distinct mechanisms
control the accumulation of cyclin B1 and Mos in Xenopus oocytes in response to proges-
terone. Mol Biol Cell 1999; 10:3279-88.
24. Roy LM, Haccard O, Izumi T, Lattes BG, Lewellyn AL, Maller JL. Mos proto-oncogene
function during oocyte maturation in Xenopus. Oncogene 1996; 12:2203-11.
25. Shevchenko A, Wilm M, Vorm O, Mann M. Mass spectrometric sequencing of proteins
silver-stained polyacrylamide gels. Anal Chem 1996; 68:850-8.
26. Gobom J, Nordhoff E, Mirgorodskaya E, Ekman R, Roepstorff P. Sample purification and
preparation technique based on nano-scale reversed-phase columns for the sensitive analysis
of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry.
J Mass Spectrom 1999; 34:105-16.
27. Hoffmann R, Metzger S, Spengler B, Otvos Jr L. Sequencing of peptides phosphorylated
on serines and threonines by post-source decay in matrix-assisted laser desorption/ioniza-
tion time-of-flight mass spectrometry. J Mass Spectrom 1999; 34:1195-204.
28. Wilm M, Mann M. Analytical properties of the nanoelectrospray ion source. Anal Chem
29. Dale S, Wilson WA, Edelman AM, Hardie DG. Similar substrate recognition motifs for
mammalian AMP-activated protein kinase, higher plant HMG-CoA reductase kinase-A,
yeast SNF1, and mammalian calmodulin-dependent protein kinase I. FEBS Lett 1995;
30. Browne GJ, Finn SG, Proud CG. Stimulation of the AMP-activated protein kinase leads
to activation of eukaryotic elongation factor 2 kinase and to its phosphorylation at a novel
site, serine 398. J Biol Chem 2004; 279:12220-31.
31. Haydon CE, Eyers PA, Aveline-Wolf LD, Resing KA, Maller JL, Ahn NG. Identification
of novel phosphorylation sites on Xenopus laevis aurora A and analysis of phosphopeptide
enrichment by immobilized metal-affinity chromatography. Mol Cell Proteomics 2003;