| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
Articles |
Center for Cell Signaling Research, Division of Molecular Life Sciences and College of Pharmacy, Ewha Womans University, Seoul 120-750, Korea [S. Y. K., K-J. L.]; Department of Molecular Pharmacology, Stanford University School of Medicine, Stanford, California 94305-5332 [J. E. F.]; and Division of Life Sciences, Paichai University, Taejon 302-735, Korea [S-K. C.]
Abstract
The Nm23 protein has been implicated in a wide variety of biological processes, including suppression of metastasis, phytochrome responses in plants, and regulation of differentiation. Here we examine whether Nm23 is involved in Xenopus laevis oocyte maturation. We found that Nm23 is present in oocytes, indicating that it has the potential to be a regulator of maturation. Furthermore, modest overexpression of Nm23 inhibited progesterone-induced oocyte maturation. This maturation-inhibitory activity was shared by both the acidic Nm23-H1 isoform and the basic Nm23-H2 isoform and by Nm23 mutants that lack nucleoside diphosphate kinase activity (Nm23-H1 H118F and Nm23-H2 H118F). Expression of Nm23 proteins delayed the accumulation of Mos and the activation of p42 mitogen-activated protein kinase (MAPK) in progesterone-treated oocytes but had no discernible effect on Mos-induced p42 MAPK activation. Therefore, Nm23 appears to act upstream of the Mos/mitogen-activated protein/extracellular signal-regulated kinase kinase/p42 MAPK cascade. These findings suggest a novel biological role for Nm23.
Introduction
Nm23 was originally identified as a possible metastasis suppressor gene through the discovery that its expression is associated with low metastatic potential murine melanoma cell lines (1 , 2) . Nm23 is a member of a gene family that is highly conserved in both prokaryotes and in eukaryotes. Six human Nm23 family members have been identified: (a) Nm23-H1 (3) ; (b) Nm23-H2 (4) ; (c) DR-nm23 (5) ; (d) Nm23-H4 (6) ; (e) Nm23-H5 (7) ; and (f) Nm23-H6 (8) .
Several lines of evidence suggest that Nm23 has a role in regulating differentiation and development. Nm23 is a homologue of the Drosophila awd (abnormal wing disc) gene (9) , and mutations in awd are associated with a variety of developmental defects. During mouse embryonic development, the accumulation of Nm23 protein coincides with differentiation of the embryonic nervous system, heart, and various epithelial tissues during mouse organogenesis (10) , again pointing to a possible role in development. The identification of a differentiation-inhibiting factor (denoted "I factor") in cell lysates and conditioned media from mouse myeloid leukemia cells as a member of the Nm23 family suggests a role for Nm23 in suppressing differentiation (11) . Nm23 expression decreases during the differentiation of hematopoietic cells (12) , again suggesting a role as an antidifferentiation factor. In other contexts, Nm23 may promote differentiation. For example, overexpression of Nm23 promotes neurite outgrowth in PC12 cells (13 , 14) .
The best-characterized Nm23 proteins exist as homo- or hetero-hexamers of two 152-amino acid isoforms, the acidic Nm23-H1 protein and the basic Nm23-H2 protein (15) . These and other Nm23 proteins possess NDP3 kinase activity, catalyzing the transfer of the terminal phosphate of nucleoside triphosphates to NDPs. In some contexts, the NDP kinase activity of Nm23 appears to be critical for its biological effects. For example, whereas wild-type Nm23 promotes neurite outgrowth, a point mutant of Nm23 (H118A) that lacks NDP kinase activity does not promote neurite outgrowth and, in fact, suppresses nerve growth factor-induced neurite outgrowth (13 , 14) .
However, in other contexts, the NDP kinase activity of Nm23 appears to be unimportant. Mutated Nm23 proteins lacking NDP kinase activity still score as the differentiation-inhibiting "I factor" mentioned above (16) . In addition, Nm23-H2 can bind to a nuclease hypersensitive element on the human c-myc promoter and transactivate gene expression in vitro and in vivo (17 , 18) , and this transactivation is independent of NDP kinase activity (19) . Thus, some Nm23 functions appear to be independent of its NDP kinase activity. Nm23 proteins also function as histidine/aspartic acid-specific protein kinases (20) and as serine/threonine-specific protein kinases (21) and have been proposed to link receptors to G proteins (22) . However, it is not yet clear how relevant any of these activities are for the biological effects of Nm23, nor is it clear where Nm23 exerts its effects. It may act in the nucleus (as suggested by its identification as a myc promoter-binding protein), in the cytoplasm, or outside the cell (as suggested by its ability to inhibit differentiation when applied to cells).
We have been studying the biochemical pathways that regulate Xenopus oocyte maturation. This process is interesting from a developmental perspective because maturation is the final, rate-limiting step in the production of a fertilizable egg. The study of oocyte maturation has also yielded important insights into cell cycle regulation, signal transduction cascades, and translational control. Fully grown stage VI Xenopus oocytes are naturally arrested at the G2-M-phase transition of meiosis I. Release from the G2-like arrest is stimulated by progesterone, which initiates a series of events that culminates in GVBD (or nuclear envelope breakdown) and the appearance of a white spot on animal pole of the oocyte (reviewed in Ref. 23 ). Early after progesterone exposure, there is a decrease in oocyte cAMP levels and inhibition of adenylyl cyclase and protein kinase A. After a lag of a few hours, polyadenylation of Mos mRNA occurs, resulting in an increase in the translation of Mos. Synthesis of Mos leads to the activation of MEK and p42/extracellular signal-regulated kinase 2 MAPK, which facilitates the activation of maturation-promoting factor, a protein kinase composed of Cdc2 and cyclin B. Together, p42 MAPK and Cdc2/cyclin B trigger the dramatic cell biological changes of oocyte maturation.
Here we examine whether the ectopic expression of Nm23 proteins can
facilitate or inhibit oocyte maturation. We found that expression of
modest concentrations of Nm23-H1 or Nm23-H2 (to 
3-fold the
endogenous Nm23 level) is sufficient to substantially inhibit
progesterone-induced oocyte maturation. Catalytically inactive Nm23
proteins also inhibited maturation, indicating that this effect does
not depend on the NDP kinase activity of Nm23. Nm23 appears to exert
its effect upstream of the Mos/MEK/MAPK cascade because it inhibited
progesterone-induced Mos accumulation but did not measurably inhibit
Mos-induced maturation. These findings show that Nm23 proteins have the
potential to be maturation-inhibiting factors.
Results
Oocytes Constitutively Express a Nm23-like Protein.
To determine whether Nm23 might be involved in the early
development of Xenopus laevis, we first examined whether
Nm23 proteins are present in Xenopus oocytes and eggs. Three
Xenopus Nm23 cDNAs (X1, X2, and X3) have been identified and
shown to be 82–87% identical to their human counterparts
(24)
. We used antihuman Nm23 antibodies to look for a
cross-reacting Nm23 protein in Xenopus oocytes. A putative
Nm23 band with apparent molecular weight of
Mr 24,000 was seen by
immunoblotting (Fig. 1A)
. Its identification was supported by its binding to
ATP-Sepharose, a characteristic of Nm23 proteins (25)
.
There was no change in the intensity or apparent molecular weight of
the endogenous Nm23 band during progesterone-induced oocyte maturation.
|
,
B—D, both Nm23-H1 and Nm23-H1F caused a
dramatic, dose-dependent delay in progesterone-induced maturation. The
delay was detectable at the lowest doses of RNA used (Fig. 1B)
and increased with increasing RNA doses (Fig. 1, C and D)
. The catalytically inactive Nm23-H1F
protein was more effective, not less effective, than wild-type Nm23-H1
(especially in Fig. 1, C and D
). Therefore, the
maturation-delaying effect of Nm23-H1 does not depend on the NDP kinase
activity of Nm23.
Fig. 1A
shows the final levels of Nm23-H1 and Nm23-H1F after
34 h of expression (24 h in the absence of progesterone followed
by 10 h in the presence of progesterone). The lowest amount of
injected RNA (9 ng/oocyte) resulted in an exogenous Nm23 band about
three times the intensity of the endogenous Nm23 band (Fig. 1A)
. Thus, a relatively modest level of Nm23 overexpression
produced a measurable effect on the kinetics of maturation (Fig. 1B)
, suggesting that the endogenous Nm23 protein may also
function to restrain or oppose maturation. Higher amounts of Nm23 RNA
produced higher levels of Nm23 protein (Fig. 1A)
and had
more dramatic effects on the timing of maturation (Fig. 1, C and D)
. Comparable concentrations (50 ng) of RNA for an
irrelevant protein (catalytically inactive Xmkp-1) had no detectable
effect on
maturation,4
arguing that the inhibition of progesterone-induced maturation
is a Nm23-specific effect.
Next we compared the basic Nm23-H2 and Nm23-H2F proteins with the
acidic Nm23-H1 and Nm23-H1F proteins in terms of their effects on
maturation. Oocytes were microinjected with 50 ng of mRNA for one
of the four Nm23 proteins. About 24 h after injection, when the
accumulation of Nm23 protein reached maximum levels, half of the
microinjected oocytes were placed in 5 µg/ml progesterone. As shown
in Fig. 2
, the H2 and H2F forms of Nm23 were similar to the H1 and H1F forms in
their ability to delay progesterone-induced maturation (compare Fig. 2, A and B
). Once again, the catalytically
inactive form (Nm23-H2F) had a stronger inhibitory effect than the
active NDP kinase (Nm23-H2; Fig. 2, A and B
).
|
, the inhibitory effect of the Nm23 mRNAs was more marked when lower
doses of progesterone were used.
|
Accordingly, we examined the effect of Nm23 overexpression on the time
course of Mos accumulation and p42 MAPK phosphorylation in
progesterone-treated oocytes. Oocytes were microinjected with water or
with Nm23-H1 and Nm23-H1F mRNAs (50 ng) and incubated for 20–24 h. As
shown in Fig. 4
, both Nm23-H1 and Nm23-H1F delayed the phosphorylation of p42 MAPK
(Fig. 4A)
and the accumulation of Mos (Fig. 4B)
.
These findings suggest that Nm23 exerts its effects upstream of the
Mos/MEK/p42 MAPK cascade.
|
, the time
course of p42 MAPK phosphorylation was similar in all three sets of
oocytes. These findings are again consistent with the hypothesis that
Nm23 exerts its effects on oocyte maturation upstream of the
Mos/MEK/p42 MAPK cascade. Discussion
Here we demonstrate that human Nm23 can influence the course of Xenopus oocyte maturation; overexpression of Nm23-H1 or Nm23-H2 delays progesterone-induced GVBD. The effect depended on the amount of Nm23 mRNA microinjected and the concentration of progesterone used to initiate maturation. Substantial delays were seen under conditions in which Nm23 was only modestly overexpressed, suggesting that the endogenous Nm23 protein may also act to oppose or restrain oocyte maturation.
The enzymatically inactive mutants Nm23-H1F and Nm23-H2F were found to inhibit progesterone-induced maturation even more effectively than the wild-type proteins. This finding indicates that NDP kinase activity is not required for Nm23 to negatively regulate progesterone-induced oocyte maturation. Kinase-independent effects of Nm23 have been found in other contexts as well (16 , 19) . However, in some contexts, the NDP kinase activity of Nm23 does appear to be critical for biological function. For example, neurite outgrowth in PC12 cells is promoted by Nm23, but not by kinase-inactive forms of Nm23 (14) . The present results suggest that the mutants may mimic the active rather than inactive conformation of the wild-type proteins in some respects. For example, the mutants might more readily undergo serine phosphorylation than the wild-type proteins. This hypothesis awaits further investigation.
Two lines of evidence indicate that Nm23 acts upstream of Mos to inhibit oocyte maturation: (a) Nm23 expression delays the accumulation of Mos; and (b) Nm23 does not affect Mos-induced p42 MAPK activation. Although the signal transduction pathways upstream of Mos are still poorly understood, previous work suggests two points in the process where Nm23 might exert its antimaturation effect. Nm23 can interact physically with the Gs protein (26 , 27) and has been proposed to participate in G protein activation (22) . Thus, Nm23 might increase basal Gs activity and cAMP levels in the oocyte, accounting for its ability to inhibit progesterone-induced maturation. Nm23 has also been proposed to inhibit polyadenylation of its own message by binding to sequences similar to the c-myc nuclease hypersensitive element in its 3'-untranslated region (24) . Cytoplasmic polyadenylation of Mos and other mRNAs is critical for progesterone-induced oocyte maturation (28, 29, 30, 31) . Thus, Nm23 might inhibit maturation by interfering with polyadenylation. Studies are under way in our laboratory to test these hypotheses. In addition, the identification of Nm23-interacting proteins may provide clues as to how Nm23 impinges on cell signaling in Xenopus oocytes and in other systems as well.
Materials and Methods
Materials.
Plasmids containing wild-type Nm23 were provided by Dr. P. S. Steeg
[National Cancer Institute, Bethesda, MD (21
, 32)
], and
mutants were provided by Dr. H. S. Lee (Kangwon National
University, Chuncheon, Korea). The Xenopus expression
vector pRD67 (33)
, which is derived from pST64T and pcDNA,
was a kind gift from Dr. J. H. Kim (Paichai University, Taejon,
Korea). For expression of Nm23 isotypes in Xenopus oocytes,
the EcoRI-SalI fragments from pT9H1 and pT9H2
(34)
were inserted into the EcoRI and
XhoI sites of pRD67. The Nm23-H1 H118F (Nm23-H1F) and
Nm23-H2 H118F (Nm23-H2F) mutants were inserted into pRD67 in a similar
way and confirmed by sequencing. A fusion protein consisting of
maltose-binding protein and Xenopus Mos fusion (malE-Mos)
was expressed in bacteria from a plasmid provided by Dr. G. Vande Woude
(Frederick Cancer Research and Development Center, Frederick, MD) and
purified as described previously (35)
. Nm23 antibody was
prepared using recombinant Nm23 expressed in Escherichia
coli and purified by ATP affinity chromatography
(25)
.
Oocyte Manipulation.
Female Xenopus laevis were injected with 67 units of
pregnant mare serum gonadotropin (Sigma) 1–2 weeks before surgery to
stimulate development of stage VI oocytes. Pieces of ovary were
surgically removed from frogs anesthetized with tricaine (Sigma). Fully
grown stage VI Xenopus oocytes were sorted manually after
defolliculation of the oocyte for 1–1.5 h with 2 mg/ml collagenase
(Sigma) in calcium-free modified Barth’s solution [88
mM NaCl, 1 mM KCl, 0.82
mM MgSO4, 2.4
mM NaHCO3, and 10
mM HEPES (pH 7.5)]. The oocytes were washed
several times with modified Barth’s solution and then allowed to
recover overnight in OR2 solution [82.5 mM NaCl,
2.5 mM KCl, 1 mM
CaCl2, 1 mM
Na2HPO4, and 5
mM HEPES (pH 7.5)] supplemented with 1 mg/ml BSA
and 50 µg/ml gentamicin.
In Vitro Transcription and Microinjection.
Nm23 mRNA was prepared using an in vitro SP6 transcription
kit (Ambion) and stored in aliquots at -80°C. Oocytes were
microinjected with 50 nl of mRNA (50 ng) around the equator of the
oocyte, transferred, and incubated for 1 day to allow Nm23 protein
accumulation to reach maximal levels. Meiotic maturation was induced by
treating oocytes with 0.1–5 µg/ml progesterone (Sigma), and GVBD was
assessed after various incubation times. In some experiments,
Nm23-expressing oocytes were subjected to a second microinjection with
malE-Mos protein (final concentration, 180 nM) to
induce maturation.
Oocyte Lysis and Immunoblotting.
Oocytes were lysed by pipetting through a 200-µl pipette tip in 60
µl of ice-cold lysis buffer (100 mM NaCl, 50
mM 2-glycerolphosphate, 10 mM EDTA, 2
mM NaF, 1 mM
Na3VO4, 10 mg/ml leupeptin,
10 mg/ml chymostatin, and 10 mg/ml pepstatin) and centrifuged for 2 min
in a Beckman E microcentrifuge with a right angle rotor to remove lipid
and yolk protein. Cytoplasm was added immediately to 0.2 volume of 6x
Laemmli gel sample buffer and boiled for 5 min at 90°C. Samples were
separated by electrophoresis on 10.5% (for MAPK) or 15% (for Nm23)
SDS-polyacrylamide gels (acrylamide:bisacrylamide, 100:1) and
transferred to polyvinylidene difluoride membranes. Nm23 was detected
with polyclonal antiserum (1:1000) and alkaline phosphatase-conjugated
secondary antibody using nitroblue tetrazolium and
5-bromo-4-chloro-3-indolyl phosphate (Promega) as substrate. For p42
MAPK and Mos, the blotting membrane was incubated with polyclonal X15
anti-MAPK serum (1:1000; Ref. 36
) or anti-Mos serum (Santa
Cruz Biotechnology; 1:500) and detected with enhanced chemiluminescence
(Amersham).
Acknowledgments
We thank Dr. H. S. Lee for providing the mutant Nm23 clone.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 Supported by MOE Grant 1998-019-F00070,
by KOSEF through the Center for Cell Signaling Research at Ewha
Womans University, and by NIH Grant GM46383. S. Y. K. is
supported financially by the Brain Korea 21 Project. 
2 To whom requests for reprints should be
addressed, at Center for Cell Signaling Research, Division of Molecular
Life Sciences and College of Pharmacy, Ewha Womans University,
Seoul 120-750, Korea. Phone: 82-2-3277-3038; Fax: 82-2-3277-3760;
E-mail: kjl{at}mm.ewha.ac.kr
3 The abbreviations used are: NDP, nucleoside
diphosphate; GVBD, germinal vesicle breakdown; MAPK, mitogen-activated
protein kinase; MEK, mitogen-activated protein/extracellular
signal-regulated kinase kinase; cAMP, cyclic AMP.
4 M. L. Sohaskey, personal communication.
Received for publication 5/10/00. Revision received 7/ 7/00. Accepted for publication 7/10/00.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
) or mRNAs for Nm23-H1 (•) or Nm23-H1F (
). The amount of RNA
injected ranged from 9–74 ng/oocyte, as indicated. GVBD was scored by
the appearance of a white dot at the oocyte’s animal pole. Ten to 20
oocytes were followed in each experimental group. Data shown are from
one of two similar experiments.
