(Received for publication, November 19, 1996, and in revised form, April 14, 1997)
From the In the mammalian central nervous system glutamate
is the major excitatory neurotransmitter and plays a crucial role in
plasticity and toxicity of certain neural cells. We found that
glutamate stimulated activation of p38 and stress-activated protein
kinase (SAPK, also known as c-Jun N-terminal kinase (JNK)), two
subgroup members of the mitogen-activated protein kinase superfamily in matured cerebellar granule cells. The p38 activation was largely mediated by N-methyl-D-aspartate receptors.
Furthermore, we have revealed a novel signaling pathway, that is,
Ca2+-mediated activation of p38 in glutamate-treated
granule cells. The glutamate concentration effective for inducing
apoptosis correlated with that for inducing p38 activation. SB203580, a
specific inhibitor for p38, inhibited glutamate-induced
apoptosis. Thus p38 might be involved in glutamate-induced
apoptosis in cerebellar granule cells.
L-Glutamic acid (glutamate) is the principal
excitatory neurotransmitter in the mammalian central nervous system.
Glutamate not only mediates excitatory neurotransmission but also is
involved in other phenomena such as neuronal plasticity and cell death (neuroexcitotoxicity) (1-5). Cell death induced by glutamate is
believed to be involved in neuronal loss associated with both acute
(e.g. stroke) and chronic (e.g. Alzheimer's
disease) neurodegenerative insults (6-8); thus dissection of glutamate
signal transduction may have clinical significance for neuroprotection.
Glutamate receptors are classified into metabotropic and ionotropic
receptors, and ionotropic glutamate receptors are further categorized
into N-methyl-D-aspartate
(NMDA)1 receptors and non-NMDA receptors.
In many cases, glutamate toxicity (especially later phases of neuronal
degeneration) can be attributed to excessive stimulation of NMDA
subtype of glutamate receptor (9-13).
The integral channel of the NMDA receptor is highly permeable to
Ca2+, and the increase in intracellular Ca2+
concentration is thought to be the key event in evoking NMDA receptor-mediated cell death (7, 14). Several molecules have been shown
to be activated by glutamate through Ca2+ influx including
CaM kinase II, protein kinase C, nitric-oxide synthetase, NF Members of the MAPK superfamily are thought to be important mediators
of signal transduction from the cell surface to the nucleus (16-21).
In addition to classical MAPKs (also referred to as ERKs), the recent
studies revealed two other members of the MAPK superfamily, p38 (also
known as CSBP/RK/MPK2) (22-24) and stress-activated protein kinase
(SAPK, also referred to as c-Jun N-terminal kinase (JNK)) (25, 26). p38
and SAPK are activated by inflammatory cytokines and cellular stresses
such as ultraviolet light and high osmolarity (26-29). p38 and SAPK have recently been shown to be involved in cell death induced by nerve
growth factor deprivation in PC12 cells, by ceramide in U937 and BAE
cells, and by anti-IgM antibody in human B lymphocytes (30-32). In
cultured chick fetal forebrain neurons, p38 activity was down-regulated
by insulin, which can support survival of these cells (33).
These previous reports prompted us to examine whether the
activities of p38 and SAPK are regulated by glutamate in cultured granule cells and whether they are correlated with the death-inducing effects of glutamate. In this study we have shown, for the first time,
that p38 is activated by glutamate and NMDA. In addition, we have shown
evidence suggesting an essential role of p38 in mediating
death-promoting activity of glutamate.
Primary cell
cultures of cerebellar granule cells were prepared from 6-7-day-old
rats as described previously (34). Eight days after plating, matured
granule cells were stimulated. In some experiments, the medium was
replaced with modified Locke's solution (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO3, 1.3 mM CaCl2, 5.6 mM
D-glucose, 5 mM Hepes, pH 7.4) 5 min before
glutamate treatment (35). Modified Locke's solution prepared by
substituting 154 mM Na+ with 154 mM
N-methyl-D-glucamine (NMDG) was also used. The
cell extracts were prepared as described (36).
We produced anti-SAPK antiserum by immunizing
mice with recombinant His-tagged rat SAPK Human c-Jun was
expressed and purified as described (37). ATF2 is a kind gift from Drs.
Suzanne J. Baker and Tom Curran (St. Jude Children's Research
Hospital).
For
immunoprecipitation, anti-p38 antibody (5 µg) and protein A-Sepharose
CL-4B beads (20 µl) were added to an aliquot (200 µl) of the cell
extract and incubated for 2 h at 4 °C. The precipitate was
washed three times with the buffer containing 0.5 M NaCl. The immune complex was suspended in 15 µl of a solution consisting of
20 mM Tris, pH 7.5, 2 mM EGTA, 10 mM MgCl2, 100 µM
[ Coverslips (15 × 15 mm No.
1; Matsunami Glass, Ltd.) were rinsed in ethanol and placed at the
bottom of 35-mm tissue culture plates, and granule cells were plated
directly on the coverslips. Eight days after plating, the granule cells
were exposed to indicated concentrations of glutamate and SB203580 to
examine glutamate-induced cell death. The coverslips were removed from
the plates, and the cells were fixed in fresh 3.7% formaldehyde in
phosphate-buffered saline for 10 min at 37 °C. The cells were
stained with 10 µg/ml DAPI for 30 min at 37 °C. The cells were
washed three times in phosphate-buffered saline, mounted on slides, and
examined by fluorescence microscopy. Cells were scored as apoptotic if
they exhibited margination and condensation of the chromatin and cell shrinkage. For the quantitation of apoptotic cells, more than 500 nuclei from 20 fields were evaluated for each point.
We utilized primary cultures of cerebellar granule cells from
neonatal rats as a model system to investigate the intracellular signal
transductions of glutamate stimulation, since previous studies have
shown that glutamate can induce cell death of these cells (38, 39). We
added triiodothyronine to the medium to reduce apoptotic cells under
unstimulated conditions. In this medium, granule cells were well
differentiated 8 days after plating as described in the previous report
(40). Since recent reports have shown that p38 and SAPK, two members of
the MAPK superfamily, may be involved in apoptosis in some cases (see
introduction), it is an intriguing question if p38 and/or SAPK is
activated downstream of glutamate. Immunoblotting with either anti-p38
antibody or anti-SAPK antiserum revealed that p38 and 54-kDa and 46-kDa
SAPKs were expressed in the granule cells (Fig.
1A). Then we tested whether p38 and SAPK can
be activated by glutamate in the granule cells. The activity of p38 was
detected by an immune complex assay using ATF2 as a substrate, and the
activity of SAPK was detected by an in-gel assay using c-Jun as a
substrate. p38 was markedly activated within 5 min after the addition
of 500 µM glutamate in granule cells, and the activity
remained higher than basal levels for more than 1 h (Fig.
1B). SAPK was activated slightly but reproducibly (about
1.5-fold) by glutamate treatment (Fig. 1C). This relatively
weak activation of SAPK may be partially due to the high basal activity
of SAPK in granule cells in the absence of glutamate (see Fig.
1C, inset). Since the medium may contain a low
level (~50 µM) of glutamate, it is possible that the
endogenous glutamate contributes to the basal activation of SAPK.
Since previous reports showed that Ca2+ influx triggered by
glutamate treatment plays a major role in glutamate toxicity (7, 41),
we examined the role of Ca2+ in glutamate-induced
activation of p38. An influx of Ca2+ from extracellular
medium was required for the glutamate-induced activation of p38,
because addition of 5 mM EGTA to the medium inhibited the
activation (Fig. 2A). Moreover, addition of
Ca2+ ionophore A23187 induced activation of p38 in the
absence of glutamate (Fig. 2B). Thus Ca2+ influx
may be sufficient for activating p38 in granule cells. The slight
activation of SAPK induced by glutamate was also blocked by the
addition of EGTA and thus dependent on Ca2+ influx (Fig.
2A).
Ca2+ influx is mediated by several proteins such as NMDA
receptor and voltage-sensitive Ca2+ channel (VSCC) (42). To
investigate the role of VSCC in glutamate-induced p38 activation, an
impermeant cation NMDG was used to substitute for Na+ to
prevent membrane depolarization. In this experiment, the medium was
replaced by the modified Locke's solution with either NaCl or NMDG
(35). The level of p38 activation by glutamate in modified Locke's
solution with NMDG was almost the same as that in modified Locke's
solution with Na+ (Fig. 3A). In
addition, pretreatment of granule cells with L-type VSCC
inhibitors, nifedipine or verapamil, did not inhibit the glutamate-induced p38 activation (Fig. 3B). These results
suggest that VSCC contributes little, if any, to the glutamate-induced p38 activation in cerebellar granule cells.
The NMDA type of glutamate receptor is thought to play a crucial role
in neurotoxicity (9-12). Especially glutamate-induced apoptosis was
completely blocked by NMDA antagonist in cerebellar granule cells (13).
When granule cells were pretreated with increasing concentrations of an
NMDA antagonist DL-2-amino-5-phosphonopentanoic acid
(DL-AP5), glutamate-induced p38 activation was inhibited in
a dose-dependent fashion (Fig.
4A). Moreover, addition of NMDA to granule
cells induced activation of p38 (Fig. 4B). This NMDA-induced activation of p38 was also inhibited by DL-AP5 dose
dependently (Fig. 4C) and was completely inhibited by
addition of EGTA to the medium (data not shown). Taken together,
Ca2+ influx through NMDA type of glutamate receptor might
be important for glutamate-induced p38 activation.
We then investigated the correlation between the glutamate toxicity and
the activity of p38 in granule cells. Glutamate can induce either early
necrosis or delayed apoptosis in cultures of cerebellar granule cells
(13 and references therein). In this study we counted the number of
neurons displaying cell shrinkage, chromatin condensation, and
formation of typical apoptotic nuclei which are characteristics of
apoptotic cell death. Under unstimulated conditions, the number of
granule cells on the 9th day after plating was 97 ± 4.39% of
that on the 8th day, suggesting that the cells were not dying at the
time the experiments were being done. The number of apoptotic cells
increased by glutamate treatment in a dose-dependent
fashion (Fig. 5A) although the maximal level of population of cells that underwent apoptosis varied among
experiments (from 30% to 95%). In all experiments EC50
(50% effective concentration) of glutamate to induce apoptosis was
about 500 µM, and 100 µM glutamate had a
marginal effect on the viability of the granule cells. Interestingly,
the extent of p38 activation was correlated with the degree of
apoptosis (Fig. 5A), suggesting possible involvement of p38
in the glutamate-induced apoptosis.
To evaluate the significance of p38 activation in the
glutamate-induced apoptosis, we tested the effect of SB203580, a p38 inhibitor (24, 43). SB203580 inhibited p38 with an IC50 of 0.6 µM, and even at 100 µM had no effect on
the activities of 12 other protein kinases tested, including MAPK or
SAPK (43). We also confirmed that 10 µM SB203580 did not
inhibit MAPK activity and SAPK activity (data not shown). Granule cells
were pretreated for 1 h with or without 1 µM or 10 µM SB203580, followed by incubation with 500 µM glutamate for 24 h. The apoptosis induced by
glutamate was partially inhibited by the treatment with SB203580 (Fig.
5B). The extent of the inhibition appeared to depend on the
degree of apoptosis induced by glutamate; SB203580 tended to inhibit apoptosis more effectively when glutamate induced apoptosis
efficiently. On the average of 10 independent experiments SB203580 at
10 µM inhibited the glutamate-induced apoptosis by about
70%. In contrast, PD98059, a MAPKK inhibitor, did not inhibit the
neurotoxic effect of glutamate at all (data not shown). These data
suggest that p38 may mediate the glutamate-induced cell death.
Taken together, we assume that p38 is involved in glutamate
neurotoxicity in cerebellar granule cells based on the following reasons. First, glutamate was a potent activator for p38 in cerebellar granule cells which are sensitive to glutamate toxicity. Second, not
only glutamate but also NMDA could activate p38 efficiently. Third,
glutamate-induced activation of p38 was largely
Ca2+-dependent. Fourth, the level of p38
activation correlated with the degree of apoptosis in matured granule
cells treated with glutamate. While 1-100 µM glutamate
weakly activated p38, it did not induce apoptosis. It is possible that
there is a threshold of p38 activity required for glutamate-induced
apoptosis. Fifth, SB203580 partially inhibited the glutamate toxicity,
although the level of inhibition varied among experiments. Apparently
SB203580 tended to inhibit glutamate toxicity more efficiently when
granule cells were sensitive to glutamate toxicity. The sensitivity
might depend on both the cell density and the population of
contaminated glial cells which may support neuronal survival. The fold
increase of SAPK activation by glutamate or by NMDA was not remarkable. But it is possible that the high basal activity of SAPK is important for glutamate toxicity.
Our results have clearly shown that p38 and SAPK can be activated
through the elevation of intracellular Ca2+, but the
molecular mechanisms of Ca2+-induced activation of SAPK and
p38 are largely unknown. It has previously been shown that several
signaling molecules including PYK2, Src, and RasGRF could transduce
Ca2+ signaling to the Ras-MAPK cascade (44-47). Most
recently, ultraviolet- and high osmolarity-induced activation of SAPK
was reported to depend partially on Ca2+ and PYK2 (48).
These molecules might be involved in glutamate-induced activation of
p38 and/or SAPK.
There are several molecules that have been implicated in mediating
apoptosis-promoting effects of glutamate. For example, nitric-oxide
synthetase and its product NO, and other reactive oxygen species are
implicated as important downstream mediators of NMDA-induced toxicity
(49-52). Here we have provided evidence that two members of the MAPK
superfamily could also be involved in the glutamate-triggered signal
transduction. It is possible that p38 is activated downstream of
reactive oxygen species, since NO and H2O2 are
good activators for p38 (53, 54). Interactions among these molecules
will be elucidated in future studies.
We thank Dr. Yasumasa Bessho for
critical reading of the manuscript.
Department of Genetics and Molecular
Biology,
Physiology, Faculty of Medicine,
Kyoto University, Sakyo-ku, Kyoto 606-01, Japan
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS AND DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENT
REFERENCES
B (15),
and mitogen-activated protein kinase (MAPK). However, the
glutamate-induced signal transduction pathways leading to cell death
are largely unknown.
Cell Cultures and Preparation of Cell Extracts
. The antiserum
specifically recognized 46-kDa and 54-kDa SAPK (36). Anti-p38 antibody
was purchased from Santa Cruz Biotechnology, Inc.
-32P] ATP (3 µCi), and 1.6 µg of recombinant ATF2
and then incubated for 30 min at 30 °C. The reactions were stopped
by addition of Laemmli's sample buffer. After SDS-PAGE,
phosphorylation of ATF2 was quantified by an image analyzer (FUJIX
BAS2000). In-gel kinase assay was performed as described (36).
Fig. 1.
Expression and glutamate-induced activation
of p38 and SAPK. A, cell extracts obtained from rat 3Y1
fibroblastic cells (12 µg of total protein), rat PC12 cells (15 µg
of total protein), and rat cerebellar granule cells (6 µg of total
protein) were immunoblotted with anti-p38 antibody (left
panel) or anti-SAPK antiserum (right panel).
Arrows indicate p38, 46-kDa SAPK, and 54-kDa SAPK.
B, cerebellar granule cells, 8 days after plating, were
treated with 500 µM glutamate for the indicated times.
p38 was immunoprecipitated from the lysates, and its activity was measured by phosphorylation of ATF2 as described under "Materials and
Methods." C, the lysates obtained from 500 µM glutamate-stimulated granule cells for the indicated
times were subjected to an in-gel kinase assay containing c-Jun as a
substrate (inset). Upper bands show the activity
of 54-kDa SAPK, and the lower bands show that of 46-kDa
SAPK. The 54-kDa SAPK activity is shown as fold increase. Values in
B and C are the means ± S.D. of at least
five independent experiments.
[View Larger Version of this Image (26K GIF file)]
Fig. 2.
Role of Ca2+ influx in
glutamate-induced activation of p38 and SAPK. A, effect of
EGTA on glutamate-induced activation of p38 and SAPK in cerebellar
granule cells. Eight days after plating, granule cells were pretreated
with (right panel) or without (left panel) 5 mM EGTA for 5 min, followed by treatment with 500 µM glutamate for 5 min. p38 activity was measured by an
immune complex kinase assay with anti-p38 antibody. SAPK activity was measured by in-gel kinase assay containing c-Jun as a substrate, and
the 54-kDa SAPK activity is shown. B, activation of p38 by Ca2+ ionophore. Granule cells were treated with
Ca2+ ionophore A23187 (5 µM) for 5 min, and
p38 activity was measured.
[View Larger Version of this Image (21K GIF file)]
Fig. 3.
Role of voltage-sensitive Ca2+
channel in glutamate-induced activation of p38. A, the
medium was replaced by modified Locke's solution with NaCl (left
panel) or modified Locke's solution with NMDG (right
panel). In modified Locke's solution with NMDG, NMDG substituted
for Na+ to prevent membrane depolarization by glutamate
treatment. After 5 min, granule cells were treated with 500 µM glutamate for 5 min, and p38 activity was measured by
an immune complex kinase assay. B, granule cells were
pretreated with 30 µM nifedipine or 10 µM
verapamil for 10 min and treated with 500 µM glutamate for 5 min. The p38 activity was measured as in A.
[View Larger Version of this Image (39K GIF file)]
Fig. 4.
Effect of NMDA antagonist on glutamate- or
NMDA-induced activation of p38. A, granule cells were
pretreated with indicated concentrations of NMDA antagonist
DL-AP5 for 10 min and followed by treatment with 500 µM glutamate. p38 activity was measured by an immune
complex kinase assay. B, granule cells were exposed to 100 µM or 500 µM NMDA for 5 min, and p38
activity was measured. C, granule cells were pretreated with
indicated concentrations of DL-AP5 for 10 min and followed
by treatment with 500 µM NMDA. p38 activity was
measured.
[View Larger Version of this Image (47K GIF file)]
Fig. 5.
Inhibition of p38 activation attenuates
glutamate-induced apoptosis. A, correlation between
glutamate-induced apoptosis and activation of p38 in cerebellar granule
cells. Eight days after plating, granule cells were treated with
various concentrations of glutamate. Five minutes after the addition of
glutamate, the cells were extracted and subjected to an immune complex
kinase assay with anti-p38 antibody. Twenty-four hours after the
addition of glutamate, the cells were fixed and stained with DAPI for
quantitation of apoptotic cells as described under "Materials and
Methods." B, effect of SB203580 on glutamate-induced
apoptosis of granule cells. Eight days after plating, granule cells
were pretreated with or without 1 µM or 10 µM SB203580 for 1 h. Then the cells were treated
with 500 µM glutamate for 24 h and stained with DAPI for quantitation of apoptotic cells. At least 500 cells were counted at
each point. We here showed four typical independent experiments in
which glutamate effectively induced apoptosis. Values represent the
average of data from 10 independent experiments. SB203580 had little
effect on untreated granule cells.
[View Larger Version of this Image (22K GIF file)]
*
This work was supported in part by grants-in-aid from the
Ministry of Education, Science and Culture of Japan.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
**
To whom correspondence should be addressed. Tel.: 81-75-751-3992;
Fax: 81-75-751-4019; E-mail: ygotoh{at}virus.kyoto-u.ac.jp.
1
The abbreviations used are: NMDA,
N-methyl-D-aspartate; DL-AP5,
DL-2-amino-5-phosphonopentanoic acid; MAPK,
mitogen-activated protein kinase; SAPK, stress-activated protein
kinase; JNK, c-Jun N-terminal kinase; PAGE, polyacrylamide gel
electrophoresis; NMDG, N-methyl-D-glucamine;
DMEM, Dulbecco's modified Eagle's medium; DAPI,
4,6-diamidino-2-phenylindole; VSCC, voltage-sensitive Ca2+
channel.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.