www.elsevier.com / locate / bres
Research report
Seizure-induced neuronal death is associated with induction of c-Jun
N-terminal kinase and is dependent on genetic background
*
Paula Elyse Schauwecker
Department of Cell and Neurobiology, University of Southern California, Keck School of Medicine, BMT 401, 1333 San Pablo Street, Los Angeles, CA90033, USA
Accepted 22 August 2000
Abstract
Previous studies have shown that expression of c-Jun protein, as well as the c-Jun amino-terminal kinase (JNK) group of mitogen-activated protein kinases, may play a critical role in the pathogenesis of glutamate neurotoxicity. In order to define the molecular cascade that leads to c-Jun activation following excitotoxic injury and delineate whether induction of protein synthesis is related to cell death signaling cascades or those changes associated with increased seizure activity, we examined the expression of JNK-1, as well as its substrate, c-Jun and N-terminal phosphorylated c-Jun following kainic acid (KA) administration in two strains of mice. In the present study, we assessed the immunohistochemical expression of these proteins at time points between 2 h and 7 days, in excitotoxic cell death-resistant (C57BL / 6) and -susceptible (FVB / N) mouse strains that were systemically injected with saline or kainic acid. No strain-related differences in the immunohistochemical expression of any of the proteins were observed in intact control mice. However, following KA administration, the magnitude and period of induction of JNK-1 protein was associated with impending cell death, while increased phosphorylation of c-Jun protein was associated with resistance to cell death. In contrast, expression of c-Jun protein does not appear to be a reliable indicator of impending cell death, as it was expressed in resistant and vulnerable subfields in mice susceptible to kainate injury. These results provide the first evidence that JNK-1 expression may be involved in producing the neuronal cell death response following excitotoxin-induced injury. 2000 Elsevier Science B.V. All rights reserved.
Theme: Disorders of the nervous system
Topic: Neurotoxicity
Keywords: Hippocampus; Immunocytochemistry; Excitotoxicity; Mouse strain
1. Introduction neuronal death. Although the mode of neuronal cell death induced by excitotoxins has been widely debated, previous Systemic administration of kainic acid (KA), a model of studies have suggested that this type of cell death exhibits excitotoxicity, induces recurrent seizures and epileptiform some of the characteristics of both necrosis and apoptosis, activity in rodents. The seizures cause selective excitotoxic depending on the severity of the stimulation and the levels
21
cell death of pyramidal neurons in the CA1 and CA3 of intracellular free Ca [14,23,28,42].
subfields and dentate hilar neurons of the hippocampus, While the molecular mechanisms of cell death are not while sparing neurons in the CA2 subfield and dentate well understood, excitotoxic cell death may result from the granule neurons [4,31,34,41]. Excitotoxic cell death is activation of intracellular signaling cascades that may be triggered by a massive release of glutamate, which acti- genetically determined. One of these signaling cascades, vates glutamate receptors leading to dramatic increases in the c-Jun amino-terminal kinase (JNK) / SAPK pathway, is
21 21
intracellular Ca [7]. The increased Ca levels initiate a component of signal transduction cascades that results in signaling cascades within susceptible neurons resulting in cell stress and death, and can be activated by a variety of cellular stresses, such as heat shock, osmotic imbalance, oxidative stress, and excitotoxicity [1,15,21,26,35,48]. The JNK / SAPK kinases are composed of three isoforms, JNK-*Tel.:11-323-442-2116; fax:11-323-442-3466.
E-mail address: [email protected] (P.E. Schauwecker). 1, JNK-2, and JNK-3, all of which have been proposed to
be important mediators of stress-activated signal transduc- served as subjects. All mice were housed individually on a tion pathways [20]. 12-h light / dark schedule. Water and food were available ad
JNKs are also important post-translational modifiers of libitum. c-Jun activity [3,11], in that they can increase the
tran-scriptional activity of c-Jun by phosphorylation on serine 2.2. Drug administration 63 and serine 73, located within the amino terminal
activation domain [8,22,37,43]. Increased c-Jun expression All experimental procedures were performed in accord-has been identified both in neuronal populations with the ance with approved institutional animal research protocols. capacity to survive an excitotoxic insult [44], as well as Kainic acid was dissolved in isotonic saline (pH 7.3) and those populations where selective neuronal degeneration administered subcutaneously. Previous studies have de-occurs [9]. However, the association between expression of fined seizure thresholds and revealed consistent seizures in immediate early genes, such as c-Jun, and their putative both strains with a mortality rate of less than 25% at a dose role as cell death-effectors or -repressors following ex- of 30 mg / kg, s.c. [39]. Following KA administration, mice citotoxic cell death remains unclear. were monitored continuously for 3–4 h for the onset and While the environmental and genetic cues leading to in extent of seizure activity. Seizures were rated according to vivo cell death of neurons are not well understood, our a previously defined scale [38]: Stage 1, immobility; Stage model of strain-related differences in susceptibility to 2, forelimb and / or tail extension, rigid posture; Stage 3, excitotoxic cell death provides a means for differentiating repetitive movements, head bobbing; Stage 4, rearing and between those molecular events that govern cell death and falling; Stage 5, continuous rearing and falling; and Stage protection versus those that occur in response to cell stress. 6, severe tonic-clonic seizures. All mice included in the Previously, we had found that susceptibility to KA-induced study exhibited at least 1 h of continuous Stage 5 seizures. cell death is strain-dependent, yet seizure activity is
comparable between strains [39,40]. Thus, we were inter- 2.3. Histological staining and immunohistochemistry ested in determining how the genetic program in C57BL / 6
mice (a representative resistant strain) is altered to allow Following survival times of 2, 6, 24, 48, 72, or 168 h protection against excitotoxic cell death. The present (n55 or 6 per time point), mice received an overdose of experiments were performed to evaluate the relationship sodium pentobarbital and were perfused transcardially with between seizure induction, expression of the immediate 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) early gene (c-Jun), phosphorylation of c-Jun, and its following saline or KA administration. Brains were imme-upstream regulator (JNK), and cell loss within the hip- diately removed and post-fixed overnight in the same pocampus in mouse strains resistant and susceptible to fixative before being cryoprotected in 30% sucrose in 0.1 kainic acid-induced excitotoxicity. M phosphate buffer (pH 7.3). Horizontal sections (35mm) In this article, we show that increased expression of were cut on a sliding microtome (Leica, Deerfield, IL) and JNK-1 occurred only in mice susceptible to excitotoxic cell incubated in 0.1 M phosphate buffer (pH 7.4) prior to death and was never solely associated with a cell stress immunohistochemical staining. For histological assess-response. In contrast, enhancement of c-Jun phosphoryla- ment, every sixth section was stained with cresyl violet to tion was observed only in those cells resistant to seizure- determine neuronal cell loss and the general histological induced cell death or that survived the excitotoxic insult. features of the tissue. For immunohistochemistry, every Expression of c-Jun protein does not appear to be indica- sixth section was rinsed in 0.1 M phosphate buffer (pH tive of susceptibility or resistance to excitotoxic cell death 7.4) three times for 15 min each, incubated for 2 h on a as protracted expression of c-Jun protein was observed in rotary shaker in a blocking buffer containing 0.5% Triton-both damaged and resistant cell populations in mice X and 5% normal goat serum in 0.1 M phosphate buffer susceptible to excitotoxic cell death. These findings dem- (pH 7.4). Sections were then incubated overnight at 48C onstrate that increased expression of JNK-1 is associated with one of the following polyclonal antibodies: c-Jun with KA-induced cell death and provides further support (1:500; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), for the hypothesis that the JNK / SAPK cascade is involved JNK-1 (1:500; Santa Cruz Biotechnology, Inc., Santa
73
in excitotoxic cell death. Cruz, CA), or phospho-c-Jun (Ser ; 1:1000; Upstate Biotechnology, NY) in the presence of 5% normal goat serum in 0.1 M phosphate buffer (pH 7.4). Sections were
the chromagen with 0.01% H O as peroxidase substrate2 2 areas were below threshold. The percentage of area for 2–5 min in imidazole buffer containing 2.5% nickel occupied by stained cells (proportional area) in hippocam-ammonium sulfate. Immunostained sections were then pal regions (CA1, CA3, and dentate gyrus) was compared mounted onto gelatinized slides, dehydrated in graded between C57BL / 6 and FVB / N mice by performing all alcohols, cleared in xylene and coverslipped. Immuno- measurements with a rectangular shaped box (0.330.05
2
cytochemical controls consisted of the same reaction mm ). Three measurements within each hippocampal procedures described above in the absence of primary region were taken from at least three to five sections from antibody. each animal. Measurements were expressed as ratios of ipsilateral values / intact values in order to control for 2.4. Assessment of neuronal damage variations in immunostaining. Results were expressed as mean6S.E.M. Statistical analyses were performed using a Horizontal sections at the level of the dorsal and ventral one-way ANOVA and Newman–Keuls post-hoc test using hippocampus were examined for cell loss in both strains of the statistical software program, SigmaStat (Jandel Sci-mice. Briefly, the number of degenerating neurons in both entific, San Rafael, CA). Differences were considered the right and left hippocampus from every sixth section significant when P,0.05. Comparisons were always made (240 mm separation distance) was counted and averaged between control C57BL / 6 and FVB / N mice.
into single values for each animal. The number of
de-generating neurons in three brain regions (CA3, CA1, 2.6. Protein extraction, gel electrophoresis and Western dentate hilus) were visually estimated and a histological blotting
damage score was assigned on a 0–3 grading scale: grade
0, none; grade 1.0, mild (10–25%); grade 2.0, moderate For these analyses, fresh brains were rapidly removed (46–54%); and grade 3.0, severe (.75%), according to a from the skulls. Hippocampi from C57BL / 6 and FVB / N previously defined scale [16–18]. All grading was per- mice were micro dissected from intact mice and from mice
¨
formed blindly by an observer who was naıve to strain, 2, 24, and 72 h following KA administration (n56 per time point, and treatment (KA vs. saline injection). Since each time point and per each strain) and stored at2708C histological damage scores were normally distributed, we until use. For gel electrophoresis and Western blotting, were able to use standard parametric methods of data hippocampi were transferred to ice-cold lysis buffer (Tris-analysis. Thus, in order to determine if differences in buffered saline, pH 7.4 containing 1% Triton X-100, 10% histological scores existed among the groups of mice, glycerol, 0.5 mM sodium orthovanadate, and 1 mM results were assessed statistically by one-way ANOVA phenylmethylsulfonyl-fluoride) and the protein content of using the computer program, SigmaStat (Jandel Scientific, all samples was determined using the Bio-Rad DC protein San Rafael, CA), and intergroup differences were analyzed assay kit (Bio-Rad, Hercules, CA). Ten micrograms of by Newman–Keuls post-hoc test. hippocampal tissue was resolved on 10% S.D.S-poly-acrylamide gels (Invitrogen, Carlsbad, CA) and blotted 2.5. Quantitative analysis of immunohistochemistry onto Protran transfer membranes (Schleicher and Schuell, Keene, NH). Membranes were blocked for 1 h at room To assess the distribution and expression of c-Jun, temperature with 5% non-fat dried milk in phosphate-phospho-c-Jun, and JNK-1 immunoreactive cells, sections buffered saline / Tween-20 (0.1% Tween-20) and probed stained for immunoreactivity were examined under bright with a mouse anti-human JNK monoclonal antibody at field illumination with a Nikon microscope and analyzed 1:2000 that recognizes both JNK-1 (JNK-46) and JNK-2 with the Image-Pro image analysis system (Image-Pro (JNK-54; Santa Cruz Biotechnology, Santa Cruz, CA). The Plus, Silverspring, MD) in a blinded fashion. Images were antigen–antibody complexes were visualized with the digitized and the threshold was adjusted manually so that enhanced chemiluminescence ECL system (Amersham, immunostained cells were above threshold and unstained Arlington Heights, IL) after incubating the blot with
HRP-Table 1
a
Classification of seizure parameters in mice after systemic administration of KA
Mouse strain Seizure parameters (% of mice)
Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Duration of
seizures (h)
b c
C57BL / 6 (n556) 10060 10060 10060 10060 87.5065.76 1.1860.11
FVB / N (n547) 10060 10060 10060 10060 97.8264.00 1.3960.09
a
KA induced a similar level of seizure duration and stage irrespective of mouse strain. Data are represented as mean6S.E.M.
b
Statistically significant strain difference: F53.909; P50.051.
c
conjugated (horseradish peroxidase-conjugated) goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA), fol-lowed by apposition of the membranes with autoradio-graphic film (Hyperfilm ECL, Amersham, Arlington Heights, IL).
3. Results
3.1. Seizure effects following systemic administration of
KA
Following systemic administration of KA, all mice (n5 103) underwent a series of behavioral seizure stages as previously described [39,40]. Within 10 min of injection, mice assumed a catatonic phase and were immobile. Within 20–30 min post-injection, mice exhibited forelimb clonus and hindlimb clonus, and within 45 min post-injection, all mice were rearing and exhibited tonic-clonic seizures. Seizures lasted on average 1.3 h and no signifi-cant strain-dependent effects were found with regard to the duration of seizures (Table 1). Table 1 summarizes the percentage of animals from each strain achieving each seizure stage (1 through 5) and the duration of Stage 5 seizures for each of the strains. A nearly significant difference in the percentage of animals achieving Stage 5 seizures was observed between C57BL / 6 and FVB / N mice. However, it is important to note that, regardless of strain, nearly 90% of all animals administered KA ex-hibited Stage 5 seizures.
3.2. Time course of hippocampal cell loss following
systemic administration of KA
In accordance with previous studies [39,40], administra-tion of KA led to the degeneraadministra-tion and loss of CA3 pyramidal neurons and hilar neurons, and occasionally CA1 pyramidal neurons in FVB / N (susceptible) mice, while no cell loss was detectable at any time following KA administration in C57BL / 6 mice (resistant; Figs. 1 and 2). In accordance with previous studies [32,33,45], cells within the dentate granule cell layer and area CA2 of Ammon’s horn were resistant to KA-induced damage in
Fig. 2. Quantitative analysis of KA treatment on hippocampal cell loss in susceptible mice. Within 2 h following administration of C57BL / 6 and FVB / N mice. Strain-dependent differences in KA seizure-KA, a small degree of cell loss was observed within the induced cell loss in areas CA3 (A), CA1 (B), and dentate hilus (C) were hilus and area CA3b of Ammon’s horn (Fig. 2). However, observed between the two strains. Cresyl violet staining and cell counting were performed 2, 6, 24, 48, 72 and 168 h following KA administration by 24 h post-KA administration, significant cell loss was
in both C57BL / 6 and FVB / N mice. Following KA administration, cell observed in the dentate hilus, area CA3, and area CA1. At
loss was significantly decreased in all hippocampal regions assessed at 24, this time point, increased argyrophilic deposits were 48, 72 and 168 h post-administration in FVB / N mice (F512.73; P,.001, present throughout the stratum oriens and stratum Newman–Keuls test). Data represents the mean6S.E.M. of five to six pyramidale of the CA3 and CA1 subfields and within the mice for each strain. * P,.05 as compared to C57BL / 6 mice. dentate hilus. Cell loss was still evident at 7 days post-KA
3.3. Effect of systemic KA administration on expression at the protein level using immunocytochemical and
im-of JNK-1 munoblotting techniques in the brains of saline- and KA-treated resistant (C57BL / 6) and susceptible (FVB / N) Previous studies have suggested that the elevated ex- mice, focusing on the hippocampal region, which is pression of c-Jun and JNK is closely linked to neuronal particularly susceptible to KA-induced cell death. Using an death after glutamate-induced excitotoxicity [21,25,30,49]. anti-JNK-1 antiserum, which specifically reacts with JNK1 In the present study, we examined the expression of JNK-1 p46 and JNK2 p54 of mouse, rat and human origin (Santa
Cruz Biotechnology, Santa Cruz, CA), we found that normal control hippocampus from either strain of mouse showed moderate cytoplasmic and nuclear immuno-reactivity throughout all hippocampal cell areas (dentate granule cells, dentate hilar cells, area CA3, area CA2, and area CA1; Fig. 3). Following administration of KA, increased JNK-1 immunoreactivity began to be detectable as early as 2 h post-KA administration in CA3 and CA1 pyramidal neurons and in dentate hilar neurons of FVB / N mice (susceptible) and became maximal between 72 and 168 h post-administration (Figs. 3 and 4). As shown in Fig. 3, JNK-1 reaction product is present within both the cytoplasm and nucleus of hippocampal neurons at early time points (panels A, B), and it appears that at 72 h following KA administration induction in the remaining CA3 pyramidal neurons may have a more nuclear localiza-tion (panel H), although this was difficult to discern. However, based on our temporal analysis of cell loss following KA administration (Figs. 1 and 2), these data indicate that increased JNK-1 immunoreactivity precedes neuronal loss in areas CA3, CA1, and the dentate hilus. Although increased immunoreactivity was observed in resistant dentate granule cells of FVB / N mice between 48 and 72 h following KA administration (Fig. 4), it was never observed in resistant CA2 pyramidal cells. In contrast, increased expression of JNK-1 protein was never observed in resistant (C57BL / 6) mice at any time point following the administration of KA (Figs. 3 and 4).
Western analysis conducted on samples collected from intact C57BL / 6 and FVB / N mice and at 2, 24 and 72 h following KA administration confirmed the specificity of the JNK antibody that recognized both JNK-1 (46 kDa) and JNK-2 (54 kDa) isoforms (Fig. 5). Immunoblotting of protein extracts revealed that the JNK-1 isoform was significantly induced in FVB / N mice at 2, 24 and 72 h following KA administration, while no significant differ-ence in the mean optical density of the JNK-1 or JNK-2 isoform was observed in C57BL / 6 mice (Fig. 5). These results are consistent with the semi-quantitative analysis of hippocampal immunostaining in C57BL / 6 and FVB / N mice following KA administration (Fig. 4).
Fig. 4. Quantitative temporal analysis of immunoreactivity for JNK-1 in hippocampal cell fields in C57BL / 6 (resistant) and FVB / N (susceptible) mice following KA administration. Comparison of the temporal induction 3.4. Effect of systemic KA administration on expression
of JNK-1 immunoreactivity between mouse strains revealed highly
of c-Jun significant differences at time points between 2 and 168 h in areas CA3
and area CA1 in FVB / N mice as compared to C57BL / 6 mice (F510.37; Basal levels of specific nuclear c-Jun staining were P,0.001, Newman–Keuls test) and at time points between 48 and 72 h in dentate granule cells of FVB / N mice as compared to C57BL / 6 mice found only in neurons of the dentate granule cell layer. No
(F511.34; P,0.001, Newman–Keuls test). Data represent the significant strain differences were observed in any brain
mean6S.E.M. of five to six mice / time point for each strain. * P,0.05 as regions in saline-injected animals. In those mice resistant compared to intact mice of either the C57BL / 6 or FVB / N strain. to KA-induced cell death (C57BL / 6), c-Jun
immuno-reactivity showed a prolonged increase in neurons of the
Fig. 5. Representative Western blot of hippocampal homogenates from intact and KA-treated C57BL / 6 and FVB / N mice at 2, 24 and 72 h post-injection. Western blots to JNK protein show an upper band of about 54 kDa, representing JNK-2, and a lower band of about 46 kDa, representing JNK-1, in both intact and KA-treated mice of both strains. Western blots to JNK show a marked increase in intensity of the 46 kDa band in the hippocampus at 2, 24 and 72 h following KA administration in the FVB / N strain only. No increases in the intensity of either the 46 or 54 kDa band were observed at any time points following KA administration in C57BL / 6 mice.
cells, such as dentate granule cells, at early time points modulators of neuronal cell death resulting from seizure-after KA administration in FVB / N mice, a greater and induced injury [28,39,46,50]. Thus, the aim of this study prolonged increase in c-Jun immunoreactivity was found in was to examine the relationship between excitotoxic cell those neurons destined to die in FVB / N mice at all time death and involvement of the JNK / SAPK signaling path-points examined (Figs. 6 and 7). way following KA administration in mice both resistant and susceptible to excitotoxin-induced injury. We took 3.5. Effect of systemic KA administration on expression advantage of genetic differences in susceptibility to
ex-of phospho-c-Jun citotoxic cell death to provide a means for dissociating those genes induced by increases in KA-induced seizure Using the antibody against human phospho-c-Jun, which activity from those involved in KA-induced hippocampal visualizes N-terminally phosphorylated c-Jun at its serine cell loss. The principal finding of the present study is that 73 residue in adult mouse brain, we found that control immunoreactivity for JNK-1 is only expressed in hip-hippocampus from both mouse strains showed basal levels pocampal neurons from mice susceptible to KA-induced of phospho-c-Jun staining in dentate granule neurons of the injury (FVB / N) following KA administration and is not dentate gyrus. However, following KA administration, induced in mice resistant to KA-induced injury (C57BL / increased expression of phospho-c-Jun was apparent within 6). These results support the concept that JNK-1 may be an all neuronal layers of the hippocampus proper in C57BL / 6 inducible cell death effector in the brain following KA mice (representative resistant strain; Figs. 8 and 9) and administration.
persisted for up to 7 days following KA administration. In Historically, KA-induced seizures have been associated contrast, following KA administration to FVB / N mice with degenerating neurons in the hippocampus, particularly (representative susceptible strain), phospho-c-Jun immuno- within the dentate hilar region and in the CA1, CA3 reactivity was increased dramatically within neurons of the subfields of the hippocampus [32,33,45]. Previous studies dentate granule cell layer as early as 2 h post-administra- have demonstrated that increases in seizure activity result tion and persisted for up to 7 days post-administration in the activation of specific genetic programs that may be (Fig. 9). Increases in phospho-c-Jun immunoreactivity associated with subsequent cell death. Activation of the were also observed in neurons within the CA3 and CA1 JNK / SAPK cascade has been shown to be associated with pyramidal cell layers at early time points, but within 24 h induction of various cellular stresses, such as heat shock, following KA administration, when significant cell loss osmotic imbalance, oxidative stress, and excitotoxicity was observed (Figs. 1 and 2), immunoreactivity was only [1,15,21,26,35,48]. However, the mechanism by which elevated in those remaining neurons in areas CA3 and CA1 stressful stimuli result in induction of JNK remains to be (Figs. 8 and 9). determined. In order to examine the cellular mechanisms that may regulate KA-induced cell death, the induction of the JNK / SAPK cascade was examined in two
Fig. 6. Time course of c-Jun immunoreactivity in the hippocampus of C57BL / 6 and FVB / N mice following saline injection (A, B) and 6 (C, D) and 72 h (E, F), following KA administration, respectively. In control mice of either strain, only dentate granule neurons contain immunoreactive product. Following KA administration, both strains show increased immunoreactivity in dentate granule neurons as well as pyramidal neurons in areas CA3 and CA1 at early time points (a, b, c, d). However, only FVB / N mice show a prolonged induction of c-Jun at later time points within the remaining pyramidal neurons of area CA3 (h) and in dentate granule neurons (g). Scale bar5750mm (A, B, C, D, E, F); Scale bar5150mm (a, b, c, d, e, f, g, h).
unclear whether induction of c-Jun precedes cell death and whether increased c-Jun immunoreactivity occurs only in those cells committed to death. Here, we show that while prolonged induction of c-Jun occurred only in mice susceptible to KA-induced cell death (FVB / N), increased c-Jun protein was present in susceptible and resistant hippocampal neuronal fields. In addition, a transient in-crease in c-Jun protein was also observed throughout the hippocampus in resistant mice. These results suggest that although temporal expression of c-Jun protein can be correlated with either activation of seizure cascades or impending cell death, c-Jun protein is not an accurate predictor of excitotoxic cell death.
Previous studies have suggested a role for c-Jun phos-phorylation in certain forms of cell death [5,11,47], and suggested that activation of the JNK isoforms and c-Jun phosphorylation may play a role in the elicitation of a neuronal death cascade [10,12,48,49]. In the present study, we detected serine-73 phosphorylated c-Jun throughout the hippocampus of C57BL / 6 (resistant) mice following KA administration. In contrast, following KA administration, serine-73 phosphorylated c-Jun was only induced in those cells resistant to excitotoxic cell death (dentate granule neurons) or in those cells that survived the insult (area CA3). Thus, in contrast to our observations with JNK-1, the temporospatial distribution of serine-73 phospho-c-Jun correlated with selective resistance to seizure-induced cell]]] death. Thus, our studies demonstrate incongruence be-tween the spatial expression of JNK-1 and phospho-c-Jun protein. Although the reason for this discrepancy remains to be elucidated, it is possible that a molecule other than JNK-1 may be responsible for phosphorylating c-Jun. Nevertheless, results presented in this paper suggest that expression of serine-73 phosphorylated c-Jun may be related to resistance to seizure-induced cell death and that activation of JNK-1 does not necessarily result in N-terminal serine-73 phosphorylation of c-Jun.
In conclusion, our results demonstrate that expression of JNK-1 is induced by KA-induced cell death and provides further support for the hypothesis that the JNK / SAPK cascade is involved in excitotoxic cell death. Further experiments are necessary to determine the potential role Fig. 7. Quantitative temporal analysis of immunoreactivity for c-Jun in
hippocampal cell fields in C57BL / 6 (resistant) and FVB / N (susceptible) of JNK-1 in KA-induced cell death. In contrast, N-terminal mice following KA administration. Comparison of the temporal induction phosphorylation of c-Jun appears to be involved in resist-of c-Jun immunoreactivity between mouse strains revealed statistically
ance to seizure-induced cell death in the adult mouse brain, significant differences between 48 and 72 h in area CA1 (F511.73;
while c-Jun expression is not an accurate predictor of P,0.001, Newman–Keuls test), and significant strain-related differences
excitotoxic cell death. Although the molecular mechanisms in c-Jun immunoreactivity in areas CA3 and in dentate granule neurons
between 2 and 72 h following KA administration (F58.79; P,0.001, by which cell death is induced and regulated remain Newman–Keuls test). Data represent the mean6S.E.M. of five to six unclear, our findings suggest that differential regulation of mice / time point for each strain. * P,0.05 as compared to intact mice of
genes involved in cell death signaling cascades may either the C57BL / 6 or FVB / N strain.
Fig. 8. Immunodetection of phospho-c-Jun in the hippocampus of C57BL / 6 and FVB / N mice following saline injection (A, B) and 72 (C, D) h following KA administration. High-power photomicrographs of phospho-c-Jun immunoreactivity in the dentate gyrus (E, F) and area CA3 (G, H) 72 h following KA administration. Note the induction of phospho-c-Jun in neurons within the dentate granule cell layer (E, F) of both strains. Although increased immunoreactivity is observed throughout the CA3 pyramidal cell layer of C57BL / 6 mice (G), increased phospho-c-Jun is only present within remaining neurons in area CA3 of FVB / N mice (H). Scale bar5750mm (A, B, C, D); Scale bar5150mm (E, F, G, H).
our model of genetic differences in susceptibility to Acknowledgements
excitotoxic cell death proves a useful tool for dissecting
References
[1] H. Bading, M.E. Greenberg, Stimulation of protein tyrosine phos-phorylation by NMDA receptor activation, Science 253 (1991) 912– 914.
[2] J. Beer, K. Mielke, M. Zipp, M. Zimmermann, T. Herdegen, Expression of c-jun, junB, c-fos, fra-1 and fra-2 mRNA in the rat brain following seizure activity and axotomy, Brain Res. 794 (1988) 255–266.
[3] A. Behrens, M. Sibilia, E. Wagner, Amino-terminal phosphorylation of c-Jun regulates stress-induced apoptosis and cellular proliferation, Nat. Gen. 21 (1999) 326–329.
[4] Y. Ben-Ari, Limbic seizure and brain damage produced by kainic acid: mechanisms and relevance to human temporal lobe epilepsy, Neuroscience 14 (1985) 375–403.
[5] E. Bossy-Wetzel, L. Bakiri, M. Yaniv, Induction of apoptosis by the transcription factor c-Jun, EMBO J. 16 (1997) 1695–1709. [6] N.S. Cheung, F.Y. Carroll, J.A. Larm, P.M. Beart, S.F. Giardina,
Kainate-induced apoptosis correlates with c-Jun activation in cul-tured cerebellar granule cells, J. Neurosci. Res. 52 (1998) 69–82. [7] D.W. Choi, Glutamate receptors and the induction of excitotoxic
neuronal death, Prog. Brain Res. 100 (1994) 47–51. ´
[8] B. Derijard, M. Hibi, I.H. Wu, T. Barrett, B. Su, T. Deng, M. Karin, R.J. Davis, JNK1: a protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun activation domain, Cell 76 (1994) 1025–1037.
[9] M. Dragunow, K. Preston, The role of inducible transcription factors in apoptotic nerve cell death, Brain Res. Rev. 21 (1995) 1–28. [10] A. Eilers, J. Whitfield, C. Babij, L.L. Rubin, J. Ham, Role of the Jun
kinase pathway in the regulation of c-Jun expression and apoptosis in sympathetic neurons, J. Neurosci. 18 (1998) 1713–1724. [11] J.W. Ferkany, R. Zaczek, J.T. Coyle, The mechanism of kainic acid
neurotoxicity, Nature 308 (1984) 561–562.
[12] I. Ferrer, M. Olive, R. Blanco, C. Cinos, A.M. Planas, Selective c-Jun overexpression is associated with ionizing radiation-induced apoptosis in the developing cerebellum of the rat, Mol. Brain Res. 38 (1996) 91–100.
[13] I. Ferrer, A.M. Planas, E. Pozas, Radiation-induced apoptosis in developing rats and kainic acid-induced excitotoxicity in adult rats are associated with distinctive morphological and biochemical c-Jun /AP-1 (N) expression, Neuroscience 80 (1997) 449–458. [14] R.K. Filipkowski, M. Hetman, B. Kaminska, L. Kaczmarek, DNA
fragmentation in rat brain after intraperitoneal administration of kainate, NeuroReport 5 (1994) 1538–1540.
[15] R.S. Fiore, T.H. Murphy, J.S. Sanghera, S.L. Pelech, J.M. Baraban, Activation of p42 mitogen-activated protein kinase by glutamate receptor stimulation in rat primary cortical cultures, J. Neurochem. 61 (1993) 1626–1633.
[16] D.G. Fujikawa, A.H. Daniels, J.S. Kim, The competitive NMDA-receptor antagonist CGP 40116 protects against status epilepticus-induced neuronal damage, Epilepsy Res. 17 (1994) 207–219. Fig. 9. Quantitative temporal analysis of immunoreactivity for
phospho-[17] D.G. Fujikawa, The neuroprotective effect of ketamine administered c-Jun in hippocampal cell fields in C57BL / 6 (resistant) and FVB / N
after status epilepticus onset, Epilepsia 36 (1995) 186–195. (susceptible) mice following KA administration. Comparison of the
[18] D.G. Fujikawa, The temporal evolution of neuronal damage from temporal induction of phospho-c-Jun immunoreactivity between mouse
pilocarpine-induced status epilepticus, Brain Res. 725 (1996) 11– strains revealed statistically significant differences between 6 and 168 h in
22. areas CA1 and CA3 (F515.37; P,0.001, Newman–Keuls test).
Signifi-[19] S. Goodenough, M. Davidson, W. Chen, A. Beckmann, Z. Pujic, M. cant induction of phospho-c-Jun immunoreactivity was observed in
Otsuki, I. Matsumoto, P. Wilce, Immediate early gene expression dentate granule neurons at all time points following KA administration
and delayed cell death in limbic areas of the rat brain after kainic regardless of strain. Data represent the mean6S.E.M. of five to six
acid treatment and recovery in the cold, Exp. Neurol. 145 (1997) mice / time point for each strain. * P,0.05 as compared to intact mice of
451–461. either the C57BL / 6 or FVB / N strain.
[20] S. Gupta, T. Barrett, A.J. Whitmarsh, J. Cavanaugh, H.K. Sluss, B. ´
and activation of c-Jun N-terminal kinases after neuronal injury, J. excitotoxicity is associated with a complex c-Fos and c-Jun response Neurosci. 18 (1998) 5124–5135. which does not preclude either cell death or survival, J. Neurobiol. [22] M. Hibi, A. Lin, T. Smeal, A. Minden, M. Karin, Identification of an 33 (1997) 232–246.
oncoprotein- and UV-responsive protein kinase that binds and [37] B.J. Pulverer, J.M. Kyriakis, J. Avruch, E. Nikolakaki, J.R. potentiates the c-Jun activation domain, Genes Dev. 7 (1993) Woodgett, Phosphorylation of c-Jun mediated by MAP kinases,
2135–2148. Nature 353 (1991) 670–674.
[23] E.M. Johnson Jr., T.L. Deckwerth, Molecular mechanisms of [38] R.J. Racine, Modification of seizure activity by electrical stimula-developmental neuronal death, Annu. Rev. Neurosci. 16 (1993) tion. II. Motor seizure, Electroencephalogr. Clin. Neurophysiol. 32
31–46. (1972) 281–294.
[24] T. Kitayama, K. Ogita, Y. Yoneda, Sustained potentiation of AP1 [39] P.E. Schauwecker, O. Steward, Genetic determinants of suscep-DNA binding is not always associated with neuronal death follow- tibility to excitotoxic cell death: Implications for gene targeting ing systemic administration of kainic acid in murine hippocampus, approaches, Proc. Natl. Acad. Sci. USA 94 (1997) 4103–4108. Neurochem. Int. 35 (1999) 453–462. [40] P.E. Schauwecker, J.J. Ramirez, O. Steward, Genetic dissection of [25] H.W. Ko, K.Y. Park, H. Kim, P.L. Han, Y.U. Kim, B.J. Gwag, E.J. the signals that induce synaptic reorganization, Exp. Neurol. 161
21
Choi, Ca -mediated activation of c-Jun N-terminal kinase and (2000) 139–152.
nuclear factor [kappa]B by NMDA in cortical cell cultures, J. [41] J.E. Schwob, T. Fuller, J.L. Price, J.W. Olney, Widespread patterns Neurochem. 71 (1998) 1390–1395. of neuronal damage following systemic or intracerebral injections of [26] M. Kurino, K. Fukunaga, Y. Ushio, E. Miyamoto, Activation of kainic acid: a histological study, Neuroscience 9 (1980) 991–1014.
mitogen-activated protein kinase in cultured rat hippocampal
neu-[42] N.A. Simonian, R.L. Getz, J.C. Leveque, C. Konradi, J.T. Coyle, rons by stimulation of glutamate receptors, J. Neurochem. 65 (1995)
Kainic acid induces apoptosis in neurons, Neuroscience 75 (1996) 1282–1289.
1047–1055. [27] J.M. Kyriakis, P. Banerjee, E. Nikolakaki, T. Dal, E.A. Rubie, M.F.
[43] T. Smeal, B. Binetruy, D. Mercola, B.A. Grover, G. Heldecker, U.R. Ahmad, J. Avruch, J.R. Woodgett, The stress-activated protein
Rapp, M. Karin, Oncoprotein-mediated signaling cascade stimulates kinase subfamily of c-Jun kinases, Nature 369 (1994) 156–160.
c-Jun activity by phosphorylation of serines 63 and 73, Mol. Cell. [28] H.M. Liu, D.L. Lei, D.L. Yang, Kainate-induced brain lesion: similar
Biol. 12 (1992) 3507–3513. local and remote histopathological and molecular changes as in
[44] C. Sommer, P. Gass, M. Kiessling, Selective c-JUN expression in ischemic brain infarct, J. Neuropath. Exp. Neurol. 55 (1996) 787–
CA1 neurons of the gerbil hippocampus during and after acquisition 794.
of an ischemia-tolerant state, Brain Pathol. 5 (1995) 135–144. [29] K. Mielke, S. Brecht, A. Dorst, T. Herdegen, Activity and
expres-[45] G. Sperk, H. Lassmann, H. Baran, S.J. Kish, F. Seitelberger, O. sion of JNK1, p38 and ERK kinases, c-Jun N-terminal
phosphoryla-Hornykiewicz, Kainic acid induced seizures: Neurochemical and tion, and c-jun promoter binding in the adult rat brain following
histopathological changes, Neuroscience 10 (1983) 1301–1315. kainate-induced seizures, Neuroscience 91 (1999) 471–483.
[46] S. Timsit, S. Rivera, P. Ouaghi, F. Guischard, E. Tremblay, Y. [30] A. Migheli, R. Piva, C. Atzori, D. Troost, D. Schiffer, c-Jun,
Ben-Ari, M. Khrestchatisky, Increased cyclin D1 in vulnerable JNK / SAPK kinases and transcription factor NF-kB are selectively
neurons in the hippocampus after ischaemia and epilepsy: a activated in astrocytes, but not motor neurons, in amylotrophic
modulator of in vivo programmed cell death?, Eur. J. Neurosci. 11 lateral sclerosis, J. Neuropath. Exp. Neurol. 56 (1997) 1314–1322.
(1999) 263–278. [31] J.V. Nadler, Kainic acid as a tool for the study of temporal lobe
[47] A. Watson, A. Eilers, D. Lallemand, J. Kyriakis, L.L. Rubin, J. Ham, epilepsy, Life Sci. 29 (1981) 2031–2042.
Phosphorylation of c-Jun is necessary for apoptosis induced by [32] J.V. Nadler, B.W. Perry, C. Gentry, C.W. Cotman, Loss and
reacquisi-survival signal withdrawal in cerebellar granule neurons, J. Neuro-tion of hippocampal synapses after selective destrucNeuro-tion of CA3–
sci. 18 (1998) 751–762. CA4 afferents with kainic acid, Brain Res. 191 (1980) 387–403.
[48] Z.G. Xia, M. Dickens, J. Raingeaud, R.J. Davis, M.E. Greenberg, [33] J.V. Nadler, G.J. Cuthbertson, Kainic acid neurotoxicity toward
Opposing effects of Erk and JNK-p38 MAP kinases on apoptosis, hippocampal formation: dependence on specific excitatory
path-Science 270 (1995) 1326–1331. ways, Brain Res. 195 (1980) 47–56.
[49] D. Yang, C.Y. Kuan, A.J. Whitmarsh, M. Rincon, T.S. Zheng, R.J. [34] J.V. Nadler, B.W. Perry, C.W. Cotman, Intraventricular kainic acid
Davis, P. Rakic, R.A. Flavell, Absence of excitotoxicity-induced preferentially destroys hippocampal pyramidal cells, Nature 271
apoptosis in the hippocampus of mice lacking the JNK3 gene, (1978) 676–677.
Nature 289 (1997) 865–870. [35] C.M. Pombo, J.V. Bonventre, J. Avruch, J.R. Woodgett, J.M.
