C-Fos, Jun D and HSP72 immunoreactivity, and neuronal injury following lithium-pilocarpine induced status epilepticus in immature and adult rats

Ž .Molecular Brain Research 63 1998 139–154

Research report

C-Fos, Jun D and HSP72 immunoreactivity, and neuronal injuryfollowing lithium-pilocarpine induced status epilepticus in immature and

adult rats

Celine Dube a, Veronique Andre a, Luciene Covolan b, Arielle Ferrandon a,´ ´ ´ ´Christian Marescaux a, Astrid Nehlig a,)

a INSERM U398, Faculte de Medecine, UniÕersite Louis Pasteur, 11 rue Humann, 67085 Strasbourg Cedex, France´ ´ ´b Departamento de Fisiologia, Escola Paulista de Medicina-UNIFESP, Sao Paulo, Brazil

Accepted 13 October 1998

Abstract

In order to follow the maturation-related evolution of neuronal damage, cellular activation and stress response subsequent to Li-PiloŽ . Ž .seizures in the 10- P10 , 21-day-old P21 and adult rat, we analyzed the expression of the c-Fos protein as a marker of cellular

activation, HSP72 immunoreactivity as the stress response and silver staining for the assessment of neuronal damage in 20 selected brainregions. The early wave of c-Fos measured at 2 h after the onset of seizures was present in most structures of the animals at the three agesstudied and particularly strong in the cerebral cortex, hippocampus and amygdala. The late wave of c-Fos measured at 24 h after the onsetof seizures and that was shown to correlate to neuronal damage was absent from the P10 rat brain, and present mainly in the cerebralcortex and hippocampus of P21 and adult rats. The expression of the stress response, assessed by the immunoreactivity of HSP72 at 24 hafter the seizures was absent from the P10 rat brain and present in the entorhinal cortex, amygdala, hippocampus and thalamus of P21 andadult rats. The expression of Jun D at 24 h after the seizures was discrete and present in most brain regions at all ages. Neuronal injuryassessed by silver staining at 6 h after the onset of seizures was very discrete in the brain of the P10 rat and limited to a few neurons inthe piriform and entorhinal cortices. In older animals, marked neuronal degeneration occurred in the cerebral cortex, amygdala,hippocampus, lateral septum and thalamus. Thus the immediate cell activation induced by lithium-pilocarpine seizures which is present atall ages translates only into a late wave of c-Fos and the expression of HSP72 in P21 and adult animals in which there will be extensivecell damage. q 1998 Elsevier Science B.V. All rights reserved.

Keywords: Seizure; Transcription factor; Neuronal damage; Development

1. Introduction

Clinical and experimental studies strongly suggest thatŽseizure susceptibility threshold, latency, propagation and

. Žduration as well as consequences severity and topogra-.phy of brain damage related to seizures dramatically

w xchange with potsnatal development 24,45,54,56 . How-ever, the pathophysiological mechanisms underlying theseage-dependent differences remain poorly understood. Inthis respect, the model induced in rats by high doses ofpilocarpine or by low doses of pilocarpine associated to

Ž .lithium Li-Pilo is an attractive tool which seems to

) Corresponding author. Fax: q33-388-24-33-60; E-mail:[email protected]

reproduce some of the developmental, clinical and neuro-physiological features of human temporal lobe epilepsyw x10,57 .

The consequences of pilocarpine-induced status epilep-Ž . w xticus SE are age-dependent 11,45,57 . In 7–11 day-old

rats, SE lasts for 2–4 h. The lethality is lower than 5% andneither neuronal damage nor spontaneous recurrent seizures

w xdevelop 45 . When SE is induced between 18 and 24days, SE lasts for 6–8 h. About 15–25% of the pre-pubescent rats which develop SE die in the course of theseizures. In the surviving animals, the presence of neuronaldamage is not an invariant consequence of SE. A moderateto severe pattern of neuronal damage may be observed inthe hippocampus, amygdala, olfactory, piriform and en-torhinal cortices, and some thalamic nuclei. One fifth ofthe surviving rats display spontaneous recurrent seizures

0169-328Xr98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved.Ž .PII: S0169-328X 98 00282-4

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´140

which appear after a ‘silent’ seizure-free phase of a meanw xduration of 37 days 11,45,57 . In adult rats, the adminis-

tration of pilocarpine leads to SE which lasts for up to8–12 h and is followed by recurrent seizure episodes forup to 24–48 h. The lethality rate reaches 50–80% duringthe first days. In the surviving adult rats, morphologicalalterations in the brain are more intense, more extendedand more reproducible than in prepubescent rats. Severeneuronal damage predominates within the hippocampalformation, the piriform and entorhinal cortices. Less in-tense damage is observed in the septum, olfactory tubercle,thalamus, amygdaloid complex, neocortex and substantia

w xnigra 10,25,57 . After the acute seizure period, the surviv-ing adult rats experience a ‘silent’ seizure-free phase last-ing for a mean duration of 14 days after which all animals

w xexhibit spontaneous recurrent seizures 10,57 .We showed recently that in 21-day-old and adult rats,

Li-Pilo induced SE leads to very large acute metabolicŽ .increases 400–850% in the cerebral regions prone to

neuronal damage while in damage-resistant areas, rates ofŽ .metabolism increase rather moderately 20–250% . How-

ever, the same distribution of metabolic increases isrecorded in 10-day-old rats not developing any neuronaldamage after Li-Pilo SE, demonstrating an age-dependentchange in the sensitivity to the neuropathological conse-quences of large metabolic increases induced by severe

w xseizures 16 . In fact, the pathophysiological mechanismsunderlying the age-related differences to SE-induced neu-ronal damage and epileptogenesis remain to be clarified.There is no clear understanding of the age-related neuronalcircuits involved by the initial SE and the role of theduration of the initial SE on the consequences of SE.Therefore, in order to investigate the age-specific effects ofLi-Pilo induced SE on neuronal death and chronic epilep-togenesis, we performed a comparative study of the conse-quences of Li-Pilo in three age-groups of rats using differ-ent markers.

The c-fos proto-oncogene is a reliable marker of cellu-lar activation that undergoes a rapid and transient increasefollowing a variety of external stimuli. The expression ofits encoded protein product, Fos, appears in specific brainregions after the induction of various types of seizuresw x15,28,38,39,41,47 . This early phase of gene expression

w xappears to reflect a sudden burst in spiking activity 29 .The expression of c-Fos and of members of the fos familyincreases transiently and goes back to low basal levels

w xwithin 6–8 h after the seizures 27–29,53 . However,seizure models that induce neuronal damage such as kainateand electrically-induced SE lead to a second wave of thefos family transcription factors that occurs between 24 and

w x72–96 h after the initial seizure episode 27–29,32,52 .The early phase related to neuronal activation seems tolead to the expression of c-Fos, Fra-2, Fos B and Jun Bproteins. There is some controversy about the componentsof the late phase correlating with cell death after severeseizures. Some authors relate neuronal damage with the

late expression of c-Fos and others with the delayedw xexpression of Fos B and Jun D 15,27–29,32,52 .

Brain cells are also able to express the 72 kDa heat-Ž .shock protein HSP72 as a secondary mode of activation

in response to various types of insults, such as hyperther-w x w xmia 33 , hypoxia–ischemia 17,21,55 and seizures

w x34,41,42,50 . The expression of HSP72 represents a re-sponse to excitation-induced stress, potential brain cellinjury and neuronal plasticity. The HSP72 has been charac-terized as both a primary and a partially secondary acti-vated gene in response to the activation of immediate early

w xgenes 32 . The expression of HSP72 seems to be neces-w xsary but not sufficient to ensure cell survival 43 and its

expression after seizures has been related either to cellw x w x w xdeath 3,42 or survival 59,62 or both 35,41,42 .

Therefore, the purpose of the present study was to mapthe expression of the c-Fos protein, both the early and thelate wave, the late expression of Jun D, together with the

Ž .expression of HSP72 in rats at postnatal day 10 P10 , 21Ž .P21 and at the adult stage after Li-Pilo seizures in orderto correlate the expression of these factors with seizure-in-duced cell death which is highly age-dependent in thismodel. Because of the short half-life of the mRNAs, westudied the immunohistochemical expression of the c-Fos,Jun D and the HSP72 proteins that allow more easytemporal assessment of the neuronal response to activation

w xthan in situ hybridization techniques 30 . Neuronal dam-age was assessed at the three ages by the silver stainingmethod allowing the identification of degenerating neuronsw x19,20 .

2. Materials and methods

2.1. Animals

For breeding purposes, adult Sprague–Dawley ratsŽ .Janvier Breeding Center, Le Genest-St-Isle, France , onemale and two females by cage were housed together inmating groups for 5 days. After delivery, litters were

Žreduced to 10 pups for homogeneity day of birth was.considered as day 0 . Adult male Sprague–Dawley rats

Ž .60–70 days, 320–350 g were provided by the samebreeder. Rats were maintained under standard laboratory

Žconditions on a 12r12 h lightrdark cycle lights on at.0600 h . The immunohistochemical experiments allowing

the detection of c-Fos, Jun D and HSP72 were performedon a total number of 60 rats, 18 at P10, 21 at P21 and 21 atthe adult stage. The silver staining method for the assess-ment of neurons undergoing degeneration was performedon a total number of 29 rats, 9 at P10, 10 at P21 and 10 atthe adult stage. All animal experiments were performed inaccordance with the rules of the European Committee

Ž .Council Direction of November 24, 1986 86r69rEECŽand the French Department of Agriculture Licence Num-

.ber 00733 .

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 141

2.2. Induction of SE

Ž .Lithium chloride 3 meqrkg was administered intra-peritoneally to all rats 18–24 h before the subcutaneous

Žinjection of pilocarpine 60 mgrkg in P10 rats and 30.mgrkg in P21 and adult rats . P21 and adult rats received

1 mgrkg methylscopolamine 30 min before the convulsantto reduce the peripheral consequences of pilocarpine ad-ministration. This injection was not necessary in P10 rats.Control P21 and adult animals received also methylscopo-lamine and control animals of all ages received an equiva-lent volume of saline instead of pilocarpine. To improvethe survival rates, the adult rats received deep intramuscu-lar injections of 1–2 mgrkg diazepam starting at 2 h afterthe onset of SE and administered 2–3 times every 3 h.These injections induced muscle relaxation but did notchange the EEG expression of the seizures.

2.3. Fos, Jun and HSP72 immunocytochemistry

The immunohistochemical detection of Fos protein wasperformed at three different times, 2, 24 and 48 h after theonset of Li-Pilo-induced SE. In each time group, therewere two control rats plus four SE-exposed rats at P10Ž .ns18 , two control animals plus five SE-exposed rats at

Ž .P21 ns21 and two control animals plus five SE-ex-Ž .posed rats at adult stage ns21 . At each age, one control

animal was used at 2 h and the other one at 24 h after thesaline injection. The study of Jun D and HSP72 expressionwas performed at 24 h after the onset of SE in sectionsadjacent to those used for the immunohistochemical detec-tion of c-Fos at 24 h. The time of 2 h was representative ofthe full expression of the c-Fos protein in the Li-Pilo

w xmodel of SE in adult rats 41 while the times of 24 and 48h were representative of the late wave of expression of the

w ximmediate early genes 27–29,32,52 and of the maximalw xexpression of HSP72 21 . Within the groups of adult rats

exposed to 2 or 24 h of SE, sections from two rats wereused for the immunohistochemical control of c-Fos at 2 h,and Jun D and HSP72 at 24 h in the absence of primaryantibody. After 2, 24 or 48 h of SE, brains were removed,immediately frozen in isopentane and stored at y808Cuntil being cut into 20 mm serial coronal sections in acryostat.

Brain sections were fixed for 7 min in 4% paraformal-Ž .dehyde dissolved in phosphate buffer saline PBS at pH

7.4. Sections were then sequentially incubated twice inPBS, once in 0.6% hydrogen peroxide in PBS and twice in

ŽPBS containing 0.4% normal goat serum Vector Labora-.tories, Burlingame, CA, USA for Fos or horse serum

Ž .Vector for HSP, 0.25% Triton-X100 and 1.5% serumŽ .albumin BSA . Sections were then incubated overnight at

208C with the primary antibody, a rabbit affinity-purifiedŽpolyclonal antibody Santa Cruz Biotechnology, Santa

Cruz, CA, USA, dilution 1:500 in PBS containing goat.serum for c-Fos, a rabbit affinity-purified polyclonal anti-

Žbody Santa Cruz Biotechnology, dilution 1:400 in PBS.containing goat serum for Jun D or a mouse affinity-puri-

Žfied monoclonal anti-72 Kd HSP Amersham, Les Ulis,.France, dilution 1:200 in PBS containing horse serum for

the detection of HSP72. The sections were rinsed twice inPBS containing the appropriate serum and incubated for 1h at 208C with the secondary antibody, biotinylated goatanti-rabbit antibody, dilution 1:400 for c-Fos and Jun DŽ .Vector and biotinylated anti-mouse antibody, dilution

Ž .1:50 for HSP72 Vector in the corresponding serumrTri-ton-X100rBSArPBS mixture. Sections were rinsed twicein the latter medium and covered with the ABC reagent

Fig. 1. Effects of lithium-pilocarpine-induced SE on c-Fos and HSP72immunoreactivity, and silver staining in the cerebral cortex of immatureand adult rats. Values are expressed as arbitrary labeled cell density unitsas defined in the methods section and shown in Fig. 7. Abbreviations:PIR: piriform cortex, ENT: entorhinal cortex, ACING: anterior cingulatecortex, PRH: perirhinal cortex.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´142

Fig. 2. Effects of lithium-pilocarpine-induced SE on c-Fos and HSP72immunoreactivity, and silver staining in the forebrain of immature andadult rats. Values are expressed as arbitrary labeled cell density units asdefined in Section 2 and shown in Fig. 7. Abbreviations: AMY: amyg-dala, MS: medial septum, LS: lateral septum, PO: primary olfactorycortex.

Ž .Vectastain Kit, Vector for 1 h at 208C. Sections wererinsed twice in PBS and incubated for 5–8 min in amixture of 0.02% diaminobenzidine, 0.5% nickel chlorideand 0.05% hydrogen peroxide in PBS. Thereafter, sectionswere dehydrated in ethanol and coverslipped. In additionto control animals exposed to lithium and saline, immuno-histochemical control sections from two animals subjected

Ž .to SE two for c-Fos, two for Jun D and two for HSP72underwent the procedures described above except for theexposure to the primary antibody.

The distribution of positive neurons was recorded fromthe forebrain to the cerebellum by light microscopy. Direct

visual counting of the density of Fos, Jun D or HSP72expressing neurons was performed in 20 brain regions byusing a grading scale of 0–3, with 0 corresponding to theabsence of reactive cells and 1, 2 and 3 to a low, moderateand high density of labeled cells, respectively, according tothe examples shown in Fig. 8 in the anterior cingulatecortex of adult rats subjected to Li-Pilo SE.

2.4. SilÕer staining

Silver staining for neurons undergoing degenerationwere performed at 6 h after the injection of saline in three

Fig. 3. Effects of lithium-pilocarpine-induced SE on c-Fos and HSP72immunoreactivity, and silver staining in the hippocampus of immatureand adult rats. Values are expressed as arbitrary labeled cell density unitsas defined in Section 2 and shown in Fig. 7. Abbreviations: CA1 andCA3: pyramidal layer of the hippocampal CA1 and CA3 areas, DGGL:dentate gyrus granular layer, DGPL: dentate gyrus polymorphic layer.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 143

Fig. 4. Effects of lithium-pilocarpine-induced SE on c-Fos and HSP72immunoreactivity, and silver staining in the thalamus of immature andadult rats. Values are expressed as arbitrary labeled cell density units asdefined in Section 2 and shown in Fig. 7. Abbreviations: AV: anteroven-tral nucleus, VM: ventromedian nucleus, LAT: lateral nucleus, POST:posterior nucleus.

P10, P21 and adult control animals receiving lithium andsaline and at 6 h after the onset of SE in rats exposed to

ŽLi-Pilo-induced SE ns6 at P10, ns7 at P21 and ns7.at the adult stage . At 6 h after the onset of SE, the animals

were deeply anesthetized and transcardiacally perfusedw x19,20,58 . The perfusion procedures took into account the

w xtechnical approaches designed by Cammermeyer 8,9 toavoid ‘dark’ neurons artifacts. Briefly, the chests of theanimals were opened under deep anesthesia and the perfu-sion of physiological saline was started rapidly at all ageswhen the heart was still actively beating. The perfusionbuffer was perfused at high speed and followed by 250

mlr100 g of a perfusion fixative solution, according to thew xprocedure described 37 . The whole heads were then kept

for 24 h in the perfusion fixative in order to avoid ‘dark’w xneurons artifacts 8,9 after which time the brains were

removed from the skull and allowed to stay in the perfu-sion fixative for at least 1 week.

On the day prior to sectioning, the brains were im-mersed in 30% sucrose dissolved in the perfusion fixativesolution. Coronal sections of 60 mm were cut in a cryostat,mounted on gelatin coated slides and allowed to dry for2–3 h. The sections were then sequentially rinsed twice in1% acetic acid and dehydrated in 1-propanol solutions of

Fig. 5. Effects of lithium-pilocarpine-induced SE on c-Fos and HSP72immunoreactivity, and silver staining in the midbrain and brainstem ofimmature and adult rats. Values are expressed as arbitrary labeled celldensity units as defined in Section 2 and shown in Fig. 7. Abbreviations:SN: substantia nigra, CG: central grey, DR: dorsal raphe, NST: nucleus ofthe solitary tract.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´144

Ž .increasing concentration 50–100% for 5 min each. Thesections were then esterified at 568C for 16 h in a solutioncontaining 0.8 ml of water and 1.2 ml of H SO in 98 ml2 4

of 1-propanol, and sequentially rehydrated in 1-propanoland distilled water. The sections were finally washed in3% acetic acid for 10 min to reduce background stainingand placed in the physical developer until the color of the

Ž .sections became lightly brown about 10 min . Finally, thesections were washed in 1% acetic acid for 30 min to stop

the silver reaction, dehydrated with 1-propanol solutions ofincreasing concentration and xylene, and embedded inCanada Balsam. The physical developer was prepared just

w xbefore use as previously described 37 .The sections were examined by light microscopy and

the density of silver stained cells was determined in 20brains regions by using a grading scale of 0–3, accordingto a scale similar to the one used for the immunohisto-chemical procedures.

Ž .Fig. 6. Immunoreactivity of c-Fos in sections of P10, P21 and adult rats taken at the level of the entorhinal cortex. a P10 rat sacrificed at 2 h after theŽ . Ž . Ž .onset of SE, b P21 rat sacrificed at 2 h after the onset of SE, c adult rat sacrificed at 2 h after the onset of SE, d P10 rat sacrificed at 24 h after theŽ . Ž .onset of SE, e P21 rat sacrificed at 24 h after the onset of SE and f adult rat sacrificed at 24 h after the onset of SE. At all ages, control rats were devoid

Ž .of any c-Fos immunoreactivity data not shown . Note the strong immunoreactivity in all layers of the entorhinal cortex at the three ages studied at 2 hŽ . Ž .after the onset of SE a–c . At 24 h, there is no expression of c-Fos at P10 d . At P21, the c-Fos immunoreactivity is rather located in the basal part of the

Ž .cortex which is the most damaged e and in the adult rat, the late immunoreactivity of c-Fos is mostly expressed outside of the entorhinal cortex in whichŽ .most cells are already irreversibly injured at 24 h f . Scale bar: 100 mm.

()

C.D

ubeet

al.rM

olecularB

rainR

esearch63

1998139

–154

´145

Ž . Ž .Fig. 7. Immunoreactivity of c-Fos in sections of P10, P21 and adult rats taken at the level of the hippocampus. a P10 rat sacrificed at 2 h after the onset of SE, b P21 sacrificed at 2 h after the onset of SE,Ž . Ž . Ž . Ž .c adult rat sacrificed at 2 h after the onset of SE, d P10 rat sacrificed at 24 h after the onset of SE, e P21 rat sacrificed at 24 h after the onset of SE and f adult rat sacrificed at 24 h after the onset of SE.

Ž .At all ages, control rats were devoid of any c-Fos immunoreactivity data not shown . Note the strong immunoreactivity at the level of the granular layer of the dentate gyrus at 2 h after the onset of SE and atŽ . Ž . Ž . Ž .the three ages a–c . The expression of the protein is also strong in the pyramidal cell layers of the hippocampus of P10 a and P21 rats b and in the polymorphic area of the dentate gyrus of P10 rats a

Ž . Ž .compared to adult ones c . At 24 h after the onset of SE, there is no expression of c-Fos in the P10 rat d . In P21 and adult animals, the dentate gyrus is devoid of c-Fos expression which is mainly presentŽ . Ž .in the CA3 pyramidal cell layer of the P21 rat e and the CA1–CA4 pyramidal cell layers of adult rats f . In the latter animals a few cells were also labelled in the polymorphic area of the dentate gyrus. In

the present study, the density of labelled cells in the CA4 area of the P21 rat represents grade 1, the density in the CA3 pyramidal cell layer of the P21 rat represents grade 2 while the density of the granuleŽ .cell layer of the dentate gyrus of the P21 rat represents grade 3 b . Scale bar: 100 mm.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´146

3. Results

3.1. BehaÕioral expression of seizures

Behavioral alterations induced by the administration oflithium and pilocarpine were time- and age-dependent.After the administration of Li-Pilo, P10 rats developedimmediately hyper- or hypoactivity, masticatory automa-tisms, more or less continuous scratching, intense bodytremor without falling or rearing and tonic extension of thetail. SE occurred within 60–80 min after pilocarpine and

lasted for 3–4 h. The rats were then hypoactive andrecovered within a few hours after the end of the seizures.Recurrent episodes of seizures were rare at that age. Thelethality was low, only one rat out of the 13 originallyprepared for the study at 24 and 48 h died. These charac-teristics were in accordance with those previously de-scribed after the injection of a high dose of pilocarpine in

w xP8–P11 rats 11,24,45 .Within 5 min after pilocarpine injection, P21 and adult

rats developed diarrhea, piloerection and other signs ofcholinergic stimulation. During the following 15–20 min,

Ž . Ž . Ž . Ž .Fig. 8. Immunoreactivity of HSP72 in sections taken at the level of the entorhinal cortex from an adult control rat a and P10 b , P21 c and adult rats dŽ .sacrificed at 24 h after the onset of SE. In P10 rats b , there is no HSP72 immunoreactivity which is mainly apparent in layer IV of the entorhinal cortex

Ž . Ž .of P21 rats c . The adult rat is devoid of HSP72 immunoreactivity probably since the cells in that area appear to be dead d . Scale bar: 100 mm.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 147

rats exhibited head bobbing, scratching, masticatory au-tomatisms, wet-dog shakes and exploratory behavior. Re-current seizures started around 20–25 min after pilocarpineadministration. These seizures which associated episodesof head and bilateral forelimb myoclonus with rearing andfalling progressed to SE at about 50 min after pilocarpine,

w xas previously described 10,11,24,45,57 . After the initialphase of sustained seizure activity, recurrent seizures oc-curred less and less frequently but could still be recorded

up to 24–48 h after the induction of SE by pilocarpine. InP21 and adult rats, 14 animals survived the 24 or 48 h

Ž .following the induction of SE while 5 P21 26% and 17Ž .adult rats 55% died within the same period.

3.2. Expression of c-Fos

The control animals of all ages were devoid of anyŽc-Fos expression, either the early or the late wave data not

.shown .

Fig. 9. Immunoreactivity of HSP72 in sections taken at the level of the hippocampus from P10, P21 and adult rats sacrificed at 24 h after the onset of SE.Ž . Ž . Ž . Ž . Ž .a control adult rat; b P10 rat, c P21 rat and d adult rat subjected to lithium-pilocarpine SE, e enlargement at the level of the CA1 pyramidal cell

Ž .layer in an P21 rat and f enlargement at the level of the polymorphic layer of the dentate gyrus in a P21 rat. The P10 rat is devoid of any HSP72Ž .immunoreactivity b which is quite strong at P21 especially in CA1 and CA4. Some neurons are also HSP72 positive in the upper blade of the granular

cell layer of the dentate gyrus. In the adult rat, there is a strong HSP72 expression in the granular and polymorphic layers of the dentate gyrus and a moreŽ . Ž .discrete immunoreactivity in the pyramidal cell layers of the hippocampus. Scale bar: a–d 100 mm, e and f 400 mm.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´148

3.2.1. Early waÕe of c-FosAt 2 h after the induction of SE, the expression of c-Fos

was present in all brain structures, except in the anteroven-tral thalamic nucleus of P10 rats and in the posterior

Ž .thalamus of adult rats Figs. 1–5 . In P10 rats, the earlyŽwave of c-Fos was strong in the cerebral cortex Figs. 1

. Ž .and 6 , the amygdala Fig. 2 , the CA1 area of thehippocampus and the polymorphic layer of the dentate

Ž .gyrus Figs. 3 and 7 , the ventromedian and lateral thala-Ž .mic nuclei Fig. 4 . In P10 rats, the early expression of

Ž .c-Fos was weak in the medial septum Fig. 2 and mostŽ .midbrain and brainstem areas Fig. 5 .

At P21, the early wave of c-Fos was strong in mostbrain areas, reaching maximal values in the cerebral cortexŽ . Ž .Figs. 1 and 6 , the amygdala Fig. 2 , the CA1 hippocam-

Žpal area and the granular layer of the dentate gyrus Figs. 3

Ž .Fig. 10. Silver staining of degenerating neurons in sections from P10, P21 and adult rats taken at the level of the entorhinal cortex. a adult control animal;Ž . Ž . Ž . Ž .b P10 rat, c P21 rat and d adult rat subjected to lithium-pilocarpine SE and sacrificed at 6 h after the onset of seizures. The control adult rat a is

Ž . Ž . Ž .devoid of any dark cells. At P10 b , a few scattered dark cells are visible in the entorhinal cortex while in P21 c and adult rat d , most cells are labelledin the entorhinal cortex. Scale bar: 100 mm.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 149

. Ž .and 7 , as well as in the thalamus Fig. 4 and theŽ .substantia nigra Fig. 5 .

In adult rats, the early expression of c-Fos was moder-ate to strong in all brain areas but reached lower levels

Ž .than those recorded at P21 in most regions Figs. 1–5 .The expression of c-Fos in the adult brain was high in the

Ž . Ž .cerebral cortex Figs. 1 and 6 , the amygdala Fig. 2 , theCA1 hippocampal area and the granular layer of the

Ž .dentate gyrus Figs. 3 and 7 .

3.2.2. Late waÕe of c-FosThe late wave of c-Fos measured at 24 and 48 h after

the onset of Li-Pilo SE was totally absent from the brain ofP10 rats. In P21 and adult animals, the expression of theprotein was more moderate and less widespread than theearly wave.

In the P21 rat, the expression of c-Fos measured at 24 hafter the onset of SE was present in eight of the 20 regionsstudied. These were the piriform and entorhinal corticesŽ .Figs. 1 and 6 , the amygdala and the primary olfactory

Ž .cortex Fig. 2 , the CA1 and CA3 hippocampal areas asŽwell as the polymorphic layer of the dentate gyrus Figs. 3

. Ž .and 7 , and the central grey Fig. 5 . The thalamus wasŽ .devoid of the late wave of c-Fos at that age Fig. 4 . At 48

h after the onset of SE, c-Fos was still expressed in sixregions, the cerebral cortex, the amygdala, the CA1 andthe central grey but the immunoreactivity of the protein

Ž .was weaker than at 24 h data not shown .In adult rats, the late wave of c-Fos was expressed in 13

of the 20 brain regions studied. These were all corticalŽ .regions Figs. 1 and 6 , the amygdala and the primary

Ž .olfactory cortex Fig. 2 , all hippocampal subfields exceptŽ .the polymorphic layer of the dentate gyrus Figs. 3 and 7 ,

Ž . Ž .three thalamic nuclei Fig. 4 and the dorsal raphe Fig. 5 .At 48 h after the onset of SE, c-Fos was still expressed infive regions of adult rats, namely the entorhinal cortex, theCA1 and CA3 hippocampal areas, and two thalamic nucleiŽ .data not shown .

3.3. Expression of Jun D

Conversely to the lack of expression of c-Fos, Jun Dimmunoreactivity was present in saline-treated rats at 24 h

Ž . Ž .Fig. 11. Silver staining of degenerating neurons in sections from P10, P21 and adult rats taken at the level of the hippocampus. a adult control animal; bŽ . Ž . Ž .P10 rat, c P21 rat and d adult rat subjected to lithium-pilocarpine SE and sacrificed at 6 h after the onset of seizures. The control adult rat a is devoid

Ž .of any dark cells. In the P10 rat b , a few labelled neurons can be seen, especially over the pyramidal cell layer and on the inner border of the dentateŽ . Ž .gyrus granule cell layer. In P21 c and adult rats d , dark cells are found mostly in the polymorphic layer of the dentate gyrus while the dendrites of the

Ž .pyramidal cells appear also to be labelled, especially in the adult rat d . Scale bar: 100 mm.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´150

after the injection. At all ages, the basal expression of JunD was weak in the cerebral cortex, the granular layer ofthe dentate gyrus and the thalamus and virtually absent

Ž .from the other brain regions data not shown . The distri-bution of Jun D was scattered, present only in some cellswithin the different brain areas while that of c-Fos wasrather clustered labeling most neurons of a given brainregion.

In P10 rats, Jun D immunoreactivity was slightly in-duced at 24 h after the onset of Li-Pilo SE in the entorhi-nal cortex, the amygdala, and the midbrain–brainstemregions. In P21 rats, Jun D expression was induced by SEin the entorhinal and perirhinal cortices, amygdala, lateralseptum, and substantia nigra. In adult rats, Jun D immuno-reactivity was moderately induced by Li-Pilo SE in all

Žbrain regions with c-Fos immunoreactivity data not.shown .

3.4. Expression of HSP72

The control animals of all ages were devoid of anyŽ .HSP72 expression Figs. 8 and 9 . The HSP72 immuno-

reactivity measured at 24 h after the onset of SE wastotally absent from the brain of P10 rats. Within the 20brain regions studied, HSP72 expression was low to mod-erate and present in seven and six regions of P21 and adultrats, respectively. At P21 the regions expressing HSP72

Ž .were the entorhinal cortex of some animals Fig. 8 , theŽ . Žamygdala Fig. 2 , all hippocampal subfields Figs. 3 and

.9 , and the anteroventral and ventromedian nuclei of theŽ .thalamus Fig. 4 . In P21 rats, the cerebral cortex and

midbrain–brainstem were devoid of any HSP72 expres-sion. In adult rats, the expression of HSP72 was low to

Ž .moderate in the piriform and perirhinal cortices Fig. 1 ,Ž .the amygdala Fig. 2 and the hippocampus, except the

Ž .polymorphic layer of the dentate gyrus Figs. 3 and 9 . Thethalamus and midbrain–brainstem were devoid of anyHSP72 immunoreactivity in adult rats.

3.5. Distribution of injured neurons

The control animals of all ages were devoid of anysilver staining, confirming the lack of effect of the lithium

Ž .treatment on cell integrity Figs. 10 and 11 . In the P10 ratexposed to Li-Pilo SE, degenerating neurons could befound only in two regions, the piriform and entorhinalcortices but their density was very low and not present in

Ž .all animals Figs. 1, 10 and 11 .In P21 rats subjected to SE, degenerating neurons could

be observed in 16 regions of the 20 studied. Their densityŽ .was high in the cerebral cortex Figs. 1 and 10 and the

Ž .amygdala Fig. 2 and low to moderate in all other regions,Ž .the lateral septum, the primary olfactory cortex Fig. 2 ,

Ž . Žthe hippocampus Figs. 3 and 11 and the thalamus Fig..4 . The midbrain–brainstem was devoid of injured neurons

except the substantia nigra where their density was weakŽ .Fig. 5 .

In adult rats subjected to Li-Pilo, degenerating neuronscould be found in 10 out of the 20 regions studied. Their

Ž .density was high in the cerebral cortex Figs. 1 and 10Ž .and the amygdala Fig. 2 and low to moderate in all other

Ž .regions, i.e., the lateral septum Fig. 2 , the two blades ofŽ .the dentate gyrus Figs. 3 and 11 , the ventromedian

Ž . Ž .thalamic nucleus Fig. 4 and the substantia nigra Fig. 5 .In the caudate-putamen, there was a quite variable degreeof silver staining from animals with almost no degenerat-ing cells to animals with heavily silver staining in the

Žcaudate-putamen, especially in the ventral part data not.shown .

4. Discussion

The results of the present study show that the responseto Li-Pilo induced SE is age-dependent. The early wave ofc-Fos and the delayed expression of Jun D are expressed atall ages while the late wave of c-Fos and the expression ofHSP72 occur only in P21 and adult animals. This sec-ondary expression of transcription factors in older animalscorrelates with the early neuronal lesions assessed bysilver staining that are quite discrete at P10 and prominentin P21 and in adult animals.

4.1. Effects of lithium-pilocarpine SE in P10 rats

The response of the brain of the P10 rat to Li-Piloinduced SE is characterized by a strong and generalizedinduction of the c-Fos protein at 2 h after the onset ofseizures which is not followed 24 h later by a second waveof the transcription factor or the expression of HSP72.Conversely the expression of Jun D is slightly induced atthat age in a few regions. The damage recorded after theepisode of SE is very limited in that age group, not presentin all animals and is mainly recorded at the level of thepiriform and entorhinal cortices.

The widespread, early wave of the c-Fos protein inresponse to sustained seizures has also been reported inother models of seizures at that age, mainly after

w xpentylenetetrazol- 26,40,47 or kainate-induced seizuresw x47 . Although the basal expression of the c-fos mRNAstays quite low until P10 in the cerebral cortex and the

w xhippocampus and until P13 in the cerebellum 23,48 , itappears from the present study and the studies cited abovethat the transcription and translation apparatus that facili-tates the stimulus transcription coupling of the c-fos geneand protein can be quite strongly activated by seizures in

w w x xthe P10 rat brain Refs. 26,40,47 and the present study .Moreover, the translation of the c-fos gene into the proteinat early times in the P10 rat subjected to Li-Pilo SE occursin as many regions as those activated at later ages, P21 and

w w x xthe adult rat Ref. 41 and the present study . This was not

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 151

the case in our previous study on pentylenetetrazol-in-duced SE where the pattern of distribution of c-Fos im-munoreactivity was more limited in P10 than in P21 and

w xadult rats, mainly in the hippocampus and the midbrain 2 .In the present study, Li-Pilo SE induces an early c-Fos

expression in the same regions at all ages with only aslightly weaker expression of the protein, mainly in theseptum and the midbrain–brainstem regions and its ab-sence from the anteroventral thalamus in P10 compared toP21 and adult rats. This data shows that the circuitry ofactivation of Li-Pilo seizures seems to involve almost the

w14 xsame areas at all ages. Using C 2-deoxyglucose auto-radiography to measure rates of glucose utilization duringthe acute episode of Li-Pilo SE, we noticed, as in thepresent study, a metabolic activation in the same brainareas of the P10, P21 and adult rat, except in the thalamus

w xwhich was not activated by Li-Pilo seizures at P10 16 .Thus our c-Fos and 2-deoxyglucose data show that Li-PiloSE activates almost the same brain circuits at all agesstudied.

In the P10 rat, the early phase of c-fos immunoreactiv-ity induced by Li-Pilo SE is quite strong but does not leadto subsequent cell death. Most studies have shown that inthe adult rodent brain, c-Fos can be expressed in neurons

w xdestined to survive or to die 4,41,44 . Thus, the immaturerat brain is able to undergo widespread and strong neu-ronal activation, as shown by the marked c-Fos immuno-

Ž .reactivity the present study and the large metabolic in-w xcrease 16 without neuropathological consequences, con-

versely to what occurs in the mature rodent brain.In the P10 rat, there is no late wave of c-Fos expression

either at 24 or 48 h and no expression of HSP72 at 24 hafter Li-Pilo SE. Since the late wave of c-Fos has beenrelated to neuronal damage occurring after severe seizuresw x27–29,32 , the lack of expression of c-Fos at 24 or 48 hafter the episode of Li-Pilo SE is likely related to theabsence of or to the very limited neuronal damage recordedin the P10 rat brain. An alternative explanation may alsorelate to the relatively short duration of the seizures andthe absence of recurrent seizures at that age compared toP21 and adult rats. Moreover, in P10 rats, Jun D which isone of the major components of the late phase of transcrip-tion factors related to neuronal damage in the kainate

w xmodel of SE 29 , is expressed at 24 h after the onset ofLi-Pilo SE although there is no seizure-induced neuronaldamage at that age. Conversely, Li-Pilo seizures do nottrigger the expression of HSP72 at P10. The lack ofHSP72 expression is not related to the immaturity of thestimulus-transcription coupling since members of theHSP70 family, both genes and proteins, are inducible asearly as P7 in the brain of the rat subjected to hypoxia-

w xischemia 7,17,31 . As in the present study, we showedpreviously that pentylenetetrazol-induced SE does not lead

w xto the expression of HSP72 in P10 rats 40 , and to ourknowledge, there is no report in the literature on theinduction of HSP72 by seizures in rats less than 3-week-old.

The expression of HSP72 has been shown to influence theability of cells to survive to various forms of toxicity,

w xincluding seizures 34 . Therefore, either the seizures oc-curring in very immature rats do not provide the criticalstimulus for the expression of HSP72 andror the absenceof expression of HSP72 could reflect the very limitedneuronal damage in Li-Pilo-induced SE in P10 rats recordedin the present study. Likewise, in P10 rats subjected topentylenetetrazol-induced SE there is no HSP72 expres-

w xsion and no neuronal damage 40 . In turn, the inability toinduce a critical stimulus to either trigger cell damageandror the expression of members of the HSP70 family inP10 rats subjected to Li-Pilo seizures may reflect thedifference in the duration, severity and type of seizuresrecorded in P10 rats compared to P21 and adult rats.Moreover, the development of the cholinergic neurons andreceptors is still incomplete at P10 and is only functionally

w xmature around P18–P20 14,46,51 . This immaturity maylimit the severity of the seizures since the number ofbinding sites for the muscarinic agonist is still reduced,therefore limiting the cellular hyperactivation responsiblefor calcium loads, heavy glutamate release and excitotoxicdamage.

4.2. Effects of lithium-pilocarpine SE in P21 and adult rats

In P21 and adult rats, the early wave of c-Fos immuno-reactivity recorded at 2 h after the onset of Li-Pilo SE ispresent in the same brain regions as in P10 rats. Thedistribution of seizure-activated regions in P21 and adultrats is also in accordance with previous reports on thedistribution of c-fos mRNA or c-Fos immunoreactivity inadult rats subjected to high dose of Pilo or Li-Pilo SEw x4,5,41,60,61 . However, the intensity of expression of thec-Fos protein is higher in most regions of the P21 ratcompared to the adult one. This data correlates with thehigher degree of metabolic activation recorded at the sameage in the same seizure model. Indeed, in the P21 rat, theincreases in cerebral metabolic rates for glucose measuredduring the second hour of Li-Pilo-induced SE are about1.5–2 fold higher than those recorded in adult animalsw x16 . Likewise, the induction of c-Fos expression is alsomore marked in P21 than in adult rats subjected to

w xpentylenetetrazol seizures of increasing severity 2 . Thisvery large brain activation in response to seizure activitymay reflect the increased sensitivity of rodents in the thirdweek of life to seizures such as those induced by pilo-

w xcarpine or kainate 1,11,45,56 .As in our study on the temporal evolution of c-Fos

immunoreactivity after Li-Pilo induced seizures in theadult brain, in the present study there is no clear correla-tion between the early c-Fos immunoreactivity and subse-quent cell death. Indeed, c-Fos is expressed in areas that

Žwill undergo neuronal damage hippocampal pyramidal

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´152

.layers, dentate gyrus hilus, cerebral cortex and thalamusŽas well as in regions with no neuronal damage dentate

.gyrus granular cell layer, posterior regions . This data is inagreement with several other studies showing that in theadult rodent brain, the early expression of c-Fos occurs in

w xneurons destined to survive or to die 4,41,44 . Thus, itseems that in the more mature brain, the immediate expres-sion of c-Fos reflects cellular hyperactivation but is not aconsistent marker of neuronal damage.

The second wave of c-Fos immunoreactivity recorded at24 h after the onset of SE occurs mainly in the cerebralcortex, the basolateral amygdala, and hippocampal regionsin P21 and adult rats and in the thalamus of adult animals.The second wave of c-Fos expression has been shown tocorrelate with the occurrence of neuronal damage inseizures induced by kainate and electrically-induced SEw x6,27,28,32 while the intrahippocampal c-fos antisenseoligonucleotide injection prevents neuronal degeneration

w xinduced by NMDA 36 . Likewise, cycloheximide, an in-hibitor of protein synthesis, is able to block both kainate-induced neuronal degeneration in the hippocampus and theprolonged expression of the c-fos gene, suggesting a possi-ble involvement of protein synthesis in excitotoxicity-

w xmediated brain damage 49 . In the present study, althoughthe second wave of c-Fos is present only in P21 and adultanimals that experience neuronal damage after Li-Pilo SE,the anatomical correlation between the late expression ofthe transcription factor and neuronal damage is not soclear. In some regions, there is both cellular degenerationand late c-Fos expression. This is the case of piriform andentorhinal cortices and the polymorphic layer of the den-tate gyrus at both P21 and the adult stage. In other regionssuch as the anterior cingulate and perirhinal cortices aswell as the lateral septum, all thalamic nuclei and thesubstantia nigra in P21 rats, neuronal degeneration is foundin the absence of the late c-Fos expression. The reversesituation occurs the CA1 and CA3 areas of the hippocam-pus in adult rats. However, the lack of degeneration in thepyramidal cell layers of the hippocampus may relate to thefact that we chose to apply silver staining at 6 h after theonset of SE, at which time the maximal damage can berecorded, especially in the thalamus and the cerebral cortexw x13 . At the latter time, almost no degeneration can berecorded in the hippocampal pyramidal cell layers of adultrats since most of it occurs earlier in these very vulnerablecell layers and neuronal degeneration is already veryprominent, rendering these cell layers completely loadedwith degenerating cells at 2 h after the onset of Li-Pilo SEŽ .data not shown . Thus, it appears from the present studythat although the late expression of c-Fos occurs only inthe brains of P21 and adult rats in which damage can berecorded and is totally absent from the brain of P10 ratswhere almost no damage occurs, the anatomical distribu-tion of the late wave of c-Fos immunoreactivity andneuronal damage is not overlapping in all brain regions.The late expression of c-Fos in the brain of P21 and adult

rats could also relate to the recurrence of seizure activityup to 24-48 h at those ages. It must be noticed that thebrain regions expressing c-Fos at 24 or 48 h after the onsetof SE are in very limited number and interest mostlyamygdala, hippocampus, entorhinal and piriform corticesthat belong to the circuitry of limbic seizures.

Moreover, there is no consensus on the relation betweenthe expression of the various transcription factors and theoccurrence of neuronal death. Some authors show that theprolonged expression of c-Fos correlates with seizure-in-duced neuronal damage and programmed cell deathw x12,15,22 . Conversely, it has been shown recently that theprotein composition of the AP-1 complex with includesseveral members of the Fos, Fra and Jun families changesover time. During the early period related to neuronalactivation by kainate-induced seizures, c-Fos, Fra-2, Fos Band Jun B proteins are mostly detected while the late phaseis mainly characterized by the presence of Fos B and JunD. Conversely, in pentylenetetrazol-induced seizures thatdo not lead to neuronal death, the composition of the AP-1

w xcomplex does not change over time 29 . However, in thepresent study, there is no clear correlation between thedistribution of either the c-Fos or the Jun D protein at 24 hand neuronal injury in P21 and adult rats. Moreover, in thepresent study, while c-fos was not expressed at 24 h, theJun D protein was expressed in some regions of the P10rat in the absence of neuronal damage.

Likewise, the anatomical distribution of HSP72 doesnot correlate clearly with the absence or presence of silverstaining in the various brain areas of P21 and adult rats.Indeed, HSP72 is expressed in regions undergoing neu-ronal damage such as the basolateral amygdala and thevarious hipppocampal layers of the P21 and adult rats. It isessentially not expressed in most cortical and thalamicregions that experience neuronal damage. This data is ingood accordance with our previous study on the temporalevolution of the expression of transcription factors aftervarious durations of Li-Pilo-induced SE in the adult rat.Likewise, we were not able to find a clear correlationbetween the expression of HSP72 in various brain regions

w xand the presence or absence of neuronal damage 41 .Other authors have shown that the members of the HSP70family are transiently expressed in both injured neurons as

w xwell as in dying neurons 3,43,63,64 . In fact, the stressresponse is a dynamic response switched on in neural cellssensitive to an insult. This response is needed by the cellbut only transiently since its persistence will not allow the

w xcell to survive 42 . The prolonged co-expression of c-Fosand HSP72 has also been shown to relate to cell death

w xafter kainate-induced SE 49 . In the present study, thiscorrelation also exists in some areas such as the amygdala,the perirhinal cortex, and the hippocampus, but not inother cortical regions or the thalamus. However, the lackof expression of HSP72 and also of the late wave of c-Fosmay partly relate to the very rapid cell death occurring insome brain regions after Li-Pilo SE, i.e., the entorhinal

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´ 153

w w x xcortex Ref. 18 and Figs. 6 and 8 rendering the cellsunable to express any of the transcription factors.

5. Conclusions

The cellular hyperactivation induced by Li-Pilo SEtranslates into a strong, generalized early wave of c-Fospresent at all ages studied at 2 h after the onset of seizures.This early wave of the transcription factor is followed 24 hlater by a second wave of c-Fos and by the expression ofHSP72 in P21 and adult animals in which neuronal degen-eration can be recorded in many brain regions. The lack oflate c-Fos and HSP72 immunoreactivity in P10 animalscorrelates with the lack of lithium-pilocarpine SE-inducedcell damage in this age group. However, in P21 and adultrats, the late expression of c-Fos and HSP72 does notpreclude either cell death or survival.

References

w x1 B.J. Albala, S.L. Moshe, R. Okada, Kainic-acid-induced seizures: a´Ž .developmental study, Dev. Brain Res. 13 1984 139–148.

w x2 V. Andre, N. Pineau, J. Motte, C. Marescaux, A. Nehlig, Mapping´of neuronal networks underlying generalized seizures induced byincreasing doses of pentylenetetrazol in the immature and adult rat,

Ž .Eur. J. Neurosci. 10 1998 2094–2106.w x3 J.N. Armstrong, J.C.L. Plumier, H.A. Robertson, R.W. Currie, The

inducible 70,000 molecularrweight heat shock protein is expressedin the degenerating dentate hilus and piriform cortex after systemic

Ž .administration of kainic acid in the rat, Neuroscience 74 1996685–693.

w x4 V. Baille-Le Crom, J.M. Collombet, M.F. Burckhart, A. Foquin, I.Pernot-Marino, G. Rondouin, G. Lallement, Time course and re-gional expression of c-Fos and HSP70 in hippocampus and piriform

Ž .cortex following soman-induced seizures, J. Neurosci. Res. 45 1996513–524.

w x5 P. Barone, M. Morelli, G. Cicarelli, A. Cozzolino, G. DeJoanna, G.Campanella, G. DiChiara, Expression of c-fos protein in the experi-

Ž .mental epilepsy induced by pilocarpine, Synapse 14 1993 1–9.w x6 G. Bing, M. McMillian, H. Kim, K. Pennypacker, Z. Feng, Q. Qi,

L.Y. Kong, M. Iadarola, J.S. Hong, Long-term expression of the35,000 mol. wt Fos-related antigen in rat brain after kainic acid

Ž .treatment, Neuroscience 73 1996 1159–1174.w x7 K.S. Blumenfeld, F.A. Welsh, V.A. Harris, M.A. Pesenson, Re-

gional expression of c-fos and heat shock protein-70 mRNA follow-ing hypoxia–ischemia in the immature rat brain, J. Cereb. Blood

Ž .Flow Metab. 12 1992 987–995.w x8 J. Cammermeyer, The importance of avoiding ‘dark’ neurons in

Ž .experimental neuropathology, Acta Neuropathol. 1 1961 245–270.w x9 J. Cammermeyer, Is the solitary neuron a manifestation of post-

mortem trauma to the brain inadequately fixed, Histochemistry 56Ž .1978 97–115.

w x10 E.A. Cavalheiro, The pilocarpine model of epilepsy, Ital. J. Neurol.Ž .Sci. 16 1995 33–37.

w x11 E.A. Cavalheiro, D.F. Silva, W.A. Turski, L.S. Calderazzo-Filho, A.Bortolotto, L. Turski, The susceptibility of rats to pilocarpine-in-

Ž .duced seizures is age-dependent, Dev. Brain Res. 37 1987 43–58.w x12 F. Colotta, N. Polentarutti, M. Sironi, A. Mantovani, Expression and

involvement of c-fos and c-jun protooncogenes in programmed celldeath induced by growth factor deprivation in lymphoid cell lines, J.

Ž .Biol. Chem. 267 1992 18278–18283.w x13 L. Covolan, L.E.A.M. Mello, Neuronal degeneration after pilo-

carpine-induced status epilepticus: time-related vulnerability, Soc.Ž .Neurosci. Abstr. 23 1997 2423.

w x14 J.T. Coyle, H.I. Yamamura, Neurochemical aspects of the ontogene-Ž .sis of cholinergic neurons in the rat brain, Brain Res. 118 1976

429–440.w x15 M. Dragunow, R. Faull, The use of c-fos as a metabolic marker in

Ž .neuronal pathway tracing, J. Neurosci. Meth. 29 1989 261–265.w x16 M.J.S. Fernandes, S. Boyet, C. Marescaux, A. Nehlig, Correlation

between hypermetabolism and neuronal damage during status epilep-ticus induced by lithium-pilocarpine in immature and adult rats, J.

Ž .Cereb. Blood Flow Metab. 19 1999 in press.w x17 D.M. Ferriero, H.Q. Soberano, R.P. Simon, F.R. Sharp, Hypoxia–

Ž .ischemia induces heat shock protein-like HSP72 immunoreactivityŽ .in neonatal rat brain, Dev. Brain Res. 53 1990 45–150.

w x18 D.G. Fujikawa, The temporal evolution of neuronal damage fromŽ .pilocarpine-induced status epilepticus, Brain Res. 725 1996 11–22.

w x19 F. Gallyas, F.H. Guldner, G. Zoltay, J.R. Wolff, Golgi-like demon-¨stration of ‘dark’ neurons with an argyrophil III method for experi-

Ž .mental neuropathology, Acta Neuropathol. 79 1990 620–628.w x20 F. Gallyas, M. Hsu, G. Buszaki, Four modified silver methods for

thick sections of formaldehyde-fixed mammalian central nervoustissue: ‘dark’ neurons, perikarya of all neurons, microgial cells and

Ž .capillaries, J. Neurosci. Meth. 50 1993 159–164.w x21 M.F. Gonzalez, K. Shiraishi, K. Hisanaga, S.M. Sagar, M. Mand-

abach, F.R. Sharp, Heat shock proteins as markers of neuronalŽ .injury, Mol. Brain Res. 93 1989 3–100.

w x22 G. Gonzalez-Martin, I. de Diego, D. Crespo, A. Fairen, Transient´c-fos expression accompanies naturally occurring cell death in thedeveloping interhemispheric cortex of the rat, Dev. Brain Res. 68Ž .1992 83–95.

w x23 R.M. Gubits, J.L. Hazelton, R. Simantov, Variations in c-fos geneŽ .expression during rat brain development, Mol. Brain Res. 3 1988

197–202.w x24 E. Hirsch, T.Z. Baram, O.C. Snead III, Ontogenetic study of

lithium-pilocarpine-induced status epilepticus in rats, Brain Res. 583Ž .1992 120–126.

w x25 M.P. Honchar, J.W. Olney, W.R. Sherman, Systemic cholinergicagents induce seizures and brain damage in lithium-treated rats,

Ž .Science 220 1983 323–325.w x26 F.E. Jensen, I.R. Firkusny, G.D. Mower, Differences in c-fos im-

munoreactivity due to age and mode of seizure induction, Mol.Ž .Brain Res. 17 1993 185–193.

w x27 B. Kaminska, R.K. Filipowski, G. Zurkowska, W. Lason, R.Przewlocki, L. Kaczmarek, Dynamic changes in the composition ofthe AP-1 transcription factor DNA-binding activity in rat brainfollowing kainate-induced seizures and cell death, Eur. J. Neurosci.Ž .6 1994 1558–1564.

w x28 B. Kaminska, K. Lukasiuk, L. Kaczmarek, Seizures-evoked activa-Ž .tion of transcription factors, Acta Neurobiol. Exp. 54 1994 65–72.

w x29 B. Kaminska, R.K. Filipkowski, I.W. Biedermann, D. Konopka, D.Nowicka, M. Hetman, M. Dabrowski, D.C. Gorecki, K. Lukasiuk,A.W. Szklarczyck, L. Kaczmarek, Kainate-evoked modulation of

Ž .gene expression in rat brain, Acta Bioch. Pol. 44 1997 781–790.w x30 M. Kiessling, P. Gass, Immediate early gene expression in experi-

Ž .mental epilepsy, Brain Pathol. 3 1993 381–393.w x31 S. Kobayashi, F.A. Welsh, Regional alterations of ATP and heat-

shock protein-72 mRNA following hypoxia-ischemia in neonatal ratŽ .brain, J. Cereb. Blood Flow Metab. 15 1995 1047–1056.

w x32 D. Konopka, D. Nowicka, R.K. Filipkowski, L. Kaczmarek,Kainate-evoked secondary gene expression in the rat hippocampus,

Ž .Neurosci. Lett. 185 1995 167–170.w x33 Y. Li, M. Chopp, Y. Yoshida, S.R. Levine, Distribution of 72-kDa

heat-shock protein in rat brain after hyperthermia, Acta Neuropathol.Ž .84 1992 94–99.

w x34 D.H. Lowenstein, The stress protein response and its potentialrelationship to prolonged seizure activity, Clin. Neuropharmacol. 18Ž .1995 148–158.

( )C. Dube et al.rMolecular Brain Research 63 1998 139–154´154

w x35 D.H. Lowenstein, R.P. Simon, F.R. Sharp, the pattern of 72-kDaheat shock protein-like immunoreactivity in the rat brain following

Ž .flurothyl-induced status epilepticus, Brain Res. 531 1990 173–182.w x36 X.C.M. Lu, F.C. Tortella, H.S. Ved, G.E. Garcia, J.R. Dave, Neuro-

protective role of c-fos antisense oligonucleotide: in vitro and inŽ .vivo studies, NeuroReport 8 1997 2925–2929.

w x37 L.E.A.M. Mello, L. Covolan, Spontaneous seizures preferentiallyinjure interneurons in the pilocarpine model of chronic spontaneous

Ž .seizures, Epilepsy Res. 26 1996 123–129.w x38 J.I. Morgan, D.R. Cohen, J.L. Hempstead, T. Curran, Mapping

patterns of c-fos expression in the central nervous system afterŽ .seizure, Science 237 1987 192–197.

w x39 J.I. Morgan, T. Curran, Proto-oncogene transcription factors andŽ .epilepsy, Trends Neurosci. 12 1991 459–462.

w x40 J. Motte, M.J.S. Fernandes, C. Marescaux, A. Nehlig, Effects ofpentylenetetrazol-induced status epilepticus on c-Fos and HSP72immunoreactivity in the immature rat brain, Mol. Brain Res. 50Ž .1997 79–84.

w x41 J. Motte, M.J.S. Fernandes, T.Z. Baram, A. Nehlig, Spatial andtemporal evolution of neuronal activation, stress and injury in

Ž .lithium-pilocarpine seizures in adult rats, Brain Res. 793 199861–72.

w x42 A.M. Planas, M.A. Soriano, I. Ferrer, E.R. Farre, Regional expres-´sion of inducible heat shock protein-70 mRNA in the rat brainfollowing administration of convulsant drugs, Mol. Brain Res. 27Ž .1994 127–137.

w x43 A.M. Planas, M.A. Soriano, A. Estrada, O. Sanz, F. Martin, I.Ferrer, The heat shock stress response after brain lesions: induction

Žof 72 kDa heat shock protein cell types involved, axonal transport,.transcriptional regulation and protein synthesis inhibition, Progr.

Ž .Neurobiol. 51 1997 607–636.w x44 E. Pozas, J. Ballabriga, A.M. Planas, I. Ferrer, Kainic acid-induced

excitotoxicity is associated with a complex c-fos and c-jun responsewhich does not preclude either cell death or survival, J. Neurobiol.

Ž .33 1997 232–246.w x45 M.R. Priel, N. Ferreira dos Santos, E.A. Cavalheiro, Developmental

aspects of the pilocarpine model of epilepsy, Epilepsy Res. 26Ž .1996 115–121.

w x46 A. Rotter, P.M. Field, G. Raisman, Muscarinic receptors in thecentral nervous sytem of the rat: III. Postnatal development of the

w3 xbinding of H propylbenzilylcholine mustard, Brain Res. Rev. 1Ž .1979 185–205.

w x47 Y. Sakurai-Yamashita, P. Sassone-Corsi, G. Gombos, Immunohisto-chemistry of c-fos in mouse brain during postnatal development:basal levels and changing response to metrazol and kainate injection,

Ž .Eur. J. Neurosci. 3 1991 764–770.w x48 S.S. Schreiber, G. Tocco, I. Najm, C.E. Finch, S.A. Johnson, M.

Baudry, Absence of c-fos induction in neonatal rat brain afterŽ .seizures, Neurosci. Lett. 136 1992 31–35.

w x49 S.S. Schreiber, G. Tocco, I. Najm, R.F. Thompson, M. Baudry,

Cycloheximide prevents kainate-induced neuronal cell death andŽ .c-fos expression in the adult brain, J. Mol. Neurosci. 4 1993

149–159.w x50 S. Shimosaka, T.S. Yuen, R.P. Simon, Distribution of HSP72 induc-

tion and neuronal death following limbic seizures, Neurosci. Lett.Ž .138 1992 202–206.

w x51 V.K. Singh, E.G. McGeer, Choline acetyltransferase in developingŽ .rat brain and spinal cord, Brain Res. 127 1977 159–163.

w x52 R.J. Smeyne, M. Vendrell, M. Hayward, S.J. Baker, G.G. Miao, K.Schilling, L.M. Robertson, J.I. Morgan, Continuous c-fos expression

Ž .precedes programmed cell death in vivo, Nature 363 1993 166–169.w x53 J.L. Sonnenberg, P.F. Macgregor-Leon, J.I. Morgan, T. Curran,

Dynamic alterations occur in the levels and composition of transcrip-Ž .tion factor AP-1 complexes after seizure, Neuron 3 1989 359–365.

w x54 E.F. Sperber, K.Z. Haas, P.K. Stanton, S.L. Moshe, Resistance of´the immature hippocampus to seizure-induced synaptic reorganiza-

Ž .tion, Dev. Brain Res. 60 1991 88–93.w x55 K. Tanaka, R.P. Simon, The pattern of neuronal injury following

seizures induced by intranigral kainic acid, Neurosci. Lett. 176Ž .1994 205–208.

w x56 C. Tomioka, K. Nishioka, K. Kogure, A comparison of induced heatshock protein in neurons destined to survive and those destined to

Ž .die after transient ischemia in rats, Brain Res. 612 1993 216–220.w x57 E. Tremblay, L. Nitecka, M.L. Berger, Y. Ben Ari, Maturation of

kainic acid seizure-brain damage in the rat: I. Clinical, electro-Ž .graphic and metabolic observations, Neuroscience 13 1984 1051–

1072.w x58 L. Turski, C. Ikonomidou, W.A. Turski, Z.A. Bortolotto, E.A.

Cavalheiro, Review: cholinergic mechanisms and epileptogenesis.The seizures induced by pilocarpine: a novel experimental model of

Ž .intractable epilepsy, Synapse 3 1989 154–171.w x59 A. Van den Pol, F. Gallyas, Trauma-induced Golgi-like staining of

neurons: a new approach to neuronal organization and response toŽ .injury, J. Comp. Neurol. 296 1990 654–673.

w x60 K. Vass, M.L. Berger, T.S. Nowak Jr., W.J. Welsh, H. Lassmann,Induction of stress protein HSP 70 in nerve cells after status

Ž .epilepticus in the rat, Neurosci. Lett. 100 1989 259–264.w x61 E.D. Weiner, V.D. Kalasapudi, D.F. Papolos, H.M. Lachman,

Lithium augments pilocarpine-induced fos gene expression in ratŽ .brain, Brain Res. 553 1991 117–122.

w x62 M.B. Williams, R.S. Jope, Distinctive rat brain immediate early generesponses to seizures induced by lithium plus pilocarpine, Mol.

Ž .Brain Res. 25 1994 80–89.w x63 M.L. Wong, S.R.B. Weiss, P.W. Gold, S.Q. Doi, S. Banerjee, J.

Licinio, R. Lad, R.M. Post, M.A. Smith, Induction of constitutiveheat shock protein 73 mRNA in the dentate gyrus by seizures, Mol.

Ž .Brain Res. 13 1992 19–25.w x64 X. Zhang, A.A. Boulton, P.H. Yu, Expression of heat shock protein-

70 and limbic seizures-induced neuronal death in the rat brain, Eur.Ž .J. Neurosci. 8 1996 1432–1440.