Hippocampal kindling protects several structures from the neuronal damage resulting from kainic acid-induced status epilepticus

BRAIN RESEARCH ELSEVIER

Brain Research 634 (1994) 245-256

Research Report

Hippocampal kindling protects several structures from the neuronal damage resulting from kainic acid-induced status epilepticus Mary Ellen Kelly, Dan C. McIntyre * Department of Psychology, Carleton University, Life Science Research Building, Ottawa, Ont. K1S 5B6, Canada (Accepted 7 September 1993)

Abstract

In an attempt to study the effects of piriform cortex damage on kindled seizure propagation, we administered kainic acid (12 mg/kg; i.p.) to rats previously kindled from the dorsal hippocampus. Unexpectedly, the ensuing status epilepticus (SE) in the kindled rats did not result in the piriform cortex damage normally observed in naive rats. As a result of this surprising finding, a more comprehensive investigation was undertaken to compare dorsal hippocampal kindled and control rats on their electrographic and behavioral SE development and subsequent brain damage. The SE induction profile and the pattern of brain damage observed in our control rats was similar to previous reports [Neuroscience, 14 (1985) 375-403; Brain Res., 218 (1981) 299-318]. By contrast, although fewer kindled rats than controls responded to the initial dose of kainic acid with electrographic and behavioral seizures, those many kindled rats that did respond, showed a pattern of SE induction that was different from controls. Kindled rats manifested fewer 'wet dog shakes', more generalized convulsions and a faster development of severe limbic status (SLS) than controls. In addition, without pharmacological intervention, the SLS continued longer in kindled rats than in controls. Histological examination revealed brain damage in kindled rats that was markedly different from controls. Unlike controls, kindled rats had no damage in the piriform cortex or substantia nigra reticulata and minimal hippocampal damage, yet showed midline thalamic and anterior olfactory nuclei damage similar to controls. These differences were observed from 1 to 28 days after kindling. Although the mechanism(s) of this kindling-based neuroprotection is not known, its discovery should add importantly to our understanding of epilepsy-induced alterations of subsequent neuronal function.

Key words: Kindling; Neuroprotection; Kainic acid; Status epilepticus; Epilepsy; Hippocampus; Piriform cortex; Substantia nigra

I. Introduction

Status epilepticus (SE), a condition characterized by continuous focal or generalized seizure activity lasting more than 30 min [11], can be observed in both epileptic patients and in those with no prior history of seizures [22]. Although it has been suggested that the course of SE and its neuropathological consequences do not differ between these two populations, this point has not been addressed clinically [22]. Numerous animal models of SE have been developed that allow for detailed study of its many forms. Many of these models include systemic administration of chemoconvulsants such as kainic acid [2,3,20,34] or pilocarpine [14]. Rats administered kainic acid have

* Corresponding author. Fax: (1) (613) 788-3667. 0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0006- 8 9 9 3 ( 9 3 ) E 1251 -W

been shown to manifest severe limbic SE approximately 90 min following injection that ultimately resuits in extensive loss of several limbic structures, in particular the olfactory or piriform cortex [3,12,20]. Because of the ability of kainic acid-induced SE to produce complete bilateral lesions of the piriform cortex, we administered kainic acid to rats previously kindled from the dorsal hippocampus. Our initial goal was to assess whether the kindled dorsal hippocampus could still trigger generalized convulsive seizures in the absence of the piriform cortex [25]. However, in undertaking this experiment, we observed serendipitously that a kindled rat responded differently from controls when challenged systemically with kainic acid. The differences between kindled and controls included (a) the onset and frequency of various seizure-related behaviors, (b) the longevity of electrographic SE and (c) the pattern and extent of acquired brain damage. To

246

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

our knowledge, all studies previously investigating the pathophysiological sequelae of kainic acid-induced SE in adults have concentrated on normal animals. Some of these results have been presented previously in a preliminary form [16,25].

2. Materials and methods 2.1. Subjects A total of 71 male Long-Evans hooded rats, weighing 250-300 g at the time of surgery, were housed individually with food and water available at all times.

2.2. Groups Initially there were 34 non-kindled control rats and 37 kindled rats. The controls were divided into 4 groups: 2 sham-operated groups (electrodes and headplug, n = 10; electrodes only, n = 7) and 2 unoperated groups (handled, n = 7; non-handled, n = 10). Like the kindled rats (see below), the sham-operated controls were placed in the stimulation chamber daily for 6 - 8 weeks, but were never stimulated. The handled-unoperated controls received daily handling for 6 - 8 weeks prior to administration of kainic acid, while the non-handled controls remained in their home-cage until the kainic acid treatment. Eight of the kindled rats were not administered kainic acid and formed a special kindled group that was used only in the analysis of brain damage.

2. 3. Surgical procedures After 1 week of daily handling, the rats to be kindled and the sham-operated controls were anesthetized with sodium pentobarbital (60 m g / k g , i.p.) and implanted with bipolar stimulating/recording electrodes bilaterally in the dorsal hippocampus and olfactory bulbs or, in some cases, unilaterally in the dorsal hippocampus and olfactory bulb and bilaterally in the amygdalae. The stereotaxic coordinates were: dorsal hippocampus, 2.5 m m posterior to bregma, 3.7 mm lateral to midline and 3.0 m m below the skull; olfactory bulb, 8.5 mm anterior to bregma, 1.5 m m lateral to midline and 3.0 m m below the skull; amygdala, 0.2 m m posterior to bregma, 4.5 m m lateral to midline and 8.5 m m below the skull [37]. The electrodes were constructed of two twisted strands of 127/~m diameter Nichrome wire, Diamel-insulated, attached to male Amphenol pins: The electrodes were cut to length, surgically implanted, secured to the skull with 6 jeweUer's screws and a covering of dental acrylic and positioned into a headplug assembly [30].

2.4. Kindling procedure O n e week following surgery, afterdischarge (AD) thresholds were determined at all 4 electrode sites in the group to be kindled. The A D threshold, defined as the m i n i m u m stimulus intensity to provoke an AD, was determined by presenting to each structure a 2 s, 60 Hz sine wave stimulus of progressively increasing intensity (15, 25, 35, 50, 75, 100, 150, 250, 300, and 350 t~A peak-to-peak) until an A D was triggered. The interval between stimulations was 1 rain. Thresholds were determined first for both olfactory bulbs and then both dorsal hippocampi. One h was allotted between assessment of each structure. Twenty-four hours later, dorsal hippocampal kindling was begun. The dorsal hippocampus selected for kindling was based on the successful elicitation of an A D from the ipsilateral olfactory bulb. If an A D had been elicited from both olfactory bulbs, then the dorsal hippocampus with the lower threshold was used as the kindling site. Kindling involved stimulating the dorsal hippocampus at its A D threshold with a 2 s, 60 Hz sine wave twice daily with an interstimulus interval greater than 6 h. The kindling procedure continued until

6 stage 5 [38] generalized convulsions were recorded. One day following kindling, A D thresholds were redetermined for the 4 structures.

2.5. Status epilepticus Either 1 day (n = 10) or 21-28 days (n = 12) following their last kindled convulsion, kindled and associated control rats received a 12 m g / k g (i.p.) injection of kainic acid. Since kainic acid-induced SE was originally intended only as a treatment to create piriform cortex damage, detailed electrographic recordings of SE induction were taken in only 5 kindled rats and 8 controls. This allowed for the administration of kainic acid to several rats simultaneously. However, E E G recordings were taken from all implanted rats intermittently throughout SE for a maximum of 16 hr. Despite the lack of detailed E E G recordings in some animals, all were closely monitered for behavioral seizures for a m i n i m u m of 4 h after kainic acid administration. After this time, if the severity of SE was not lifethreatening, the animals were monitered every 30 min until behavioral signs of SE diminished. If behavioral seizure activity was not evident by 80 min following the kainic acid injection, animals were given a supplemental injection of kainic acid (6 m g / k g ; i.p.). A further supplemental dose was given if clinical signs of seizure again were absent in the next 30 min period. Once SE was initiated, animals were allowed to recover from SE without pharmacological intervention. If a rat appeared as though it would not survive SE, the SE severity was decreased with a 15 m g / k g (i.p.) dose of sodium pentobarbital.

2.6. Histology Seven to 50 days following SE, under deep anesthesia, all rats were perfused intracardially with saline and 10% formalin or 4% paraformaldehyde. Electrodes in the implanted animals were withdrawn 24 h later and the brains stored in fixative for at least 3 days before sectioning. Frozen serial sections of 40 /~m were taken throughout the brain, mounted on gelatinized slides, stained with Cresyl Violet and, in several cases, a modification of the Timm's procedure which was counterstained with Cresyl Violet [41]. The T i m m ' s procedure was used occasionally because, under low magnification, it provided dramatic visual evidence of an innervation loss in the piriform cortex area that was associated with neuronal damage. Determinations of brain damage were made blind and involved both neuron and glia cell counts in a 5 0 x 5 0 /~m grid. The average number of each cell type (neuron or glia) was determined from two randomly chosen locations in each structure or cell field. The average cell counts in each structure of the kindled and control rats with SE was compared to the average cell counts from the special kindled group not administered kainic acid. Cells were considered to be neurons if their cells bodies were larger in diameter than 15 ~zm and to be glia if the cell bodies were smaller in diameter than 10/zm. To simplify the data presentation, the counts of both cell types were combined to form an ordinal scale damage score. On this scale, neuron loss and gliosis was scored as slight (score = 1), moderate (score = 2), severe (score = 3), or no damage (score = 0). The latter was defined as neuron and glia numbers in the m e a s u r e m e n t grid that were statistically similar to the special kindled rats not administered kainic acid. Slight damage was indicated by a 50% increase of glial cells in the grid a n d / o r evidence of a few necrotic neurons in the cell field beyond the grid. Moderate damage involved up to a 200% increase in glial cells a n d / o r a 50% reduction of neurons in the grid, while severe damage was more than a 200% increase of glial cells a n d / o r a 90-100% loss of neurons in the grid.

2. 7. Data analysis The two basic SE groups (kindled and control) were initially compared on their response to a 12 m g / k g dose of kainic acid. To be considered 'sensitive' to kainic acid, after the injection a rat must

M.E. Kelly, D.C. Mclntyre /Brain Research 634 (1994) 245-256 have developed at least one form of behavioral automatism a n d / o r seizure, including 'wet dog shakes', clonic-tonic-clonic seizure or varying stages (1-5) of clonic limbic convulsions [38]. Further comparisons between kindled and control animals involved the latency from kainic acid injection to (a) the first behavioral seizure (regardless of its form), (b) the first stage 5 clonic limbic convulsion and (c) the onset of severe limbic status (SLS). SLS was defined as a continuous stage 2 masticatory seizure [38], which involves head bobbing and drooling, punctuated every 3 - 5 min by stage 3 - 5 clonic convulsions [38]. Only those animals that were sensitive to the initial 12 m g / k g dose of kainic acid were used in the anaylses of SLS induction. All rats eventually developing SLS were used in the analysis of brain pathology. The data were analysed parametrically with A N O V A and, when appropriate, individual group comparisons were m a d e using the N e w m a n - K e u l s test. Proportional data were analysed with the ~2_ test.

3. Results

In 7 kindled rats, the headplug assembly was dislodged before the kainic acid experiment. Histological assessment of the remaining rats indicated the kindling electrode was positioned in the lateral portion of CA1 (n = 20) or CA2-3 (n = 10) in the kindled group. These placements were similar to the sham-operated control group, i.e., CA1 (n = 12) or CA2-3 (n = 5). 3.1. Kindling The AD thresholds, which were determined before and 1 day after dorsal hippocampal kindling (35.1 ___6.0 and 4 8 . 6 _ 11.7/xA, respectively), were not statistically different. The number of stimulations to the first stage 5 convulsion (kindIing rate) was 30.2 _+ 4.1 stimulations. There were no AD threshold or kindling differences related to electrode placements (data not shown). 3.2. Sensitivity to kainic acid Since there were no differences between the 4 nonkindled control groups, they were merged into one group. Ninety-seven percent (33/34) of the collective control group responded to the 12 m g / k g dose of kainic acid with one or more behavioral seizures. In contrast, the kindled rats were significantly ( P < 0.01) less sensitive to kainic acid, since only 63% (14/22) of them showed one or more behavioral seizures after the 12 m g / k g injection. This insensitivity to kainic acid was more strongly manifested in those kindled rats administered kainic acid 1 vs 21-28 days following their last kindled convulsion. The 12 m g / k g dose of kainic acid triggered behavioral seizure activity in only 50% ( 5 / 1 0 ) of the 1 Day group ( P < 0.001) compared to 75% (9/12) of the 21-28 Day group ( P < 0.10). In the rats that did respond with seizure activity to the initial 12 m g / k g dose of kainic acid (kindled, n = 14; controls, n = 33), the first form of that activity was determined. As can be seen in Fig. 1, the first

100 80-

'~

[]

CTC

[]

GLS

247

so4020

O'

i

WDS CTC GLS Control

WDS CTC GLS Kindled

Fig. 1. The percentage of rats showing a generalized limbic seizure (GLS), wet dog shakes (WDS) or a clonic-tonic-clonic (CTC) seizure as their first behavioral seizure form after the kainic acid injection in both the control and kindled groups. * Significantly different from the control group, P < 0.001.

automatisms a n d / o r seizure activity most prevalent in controls was wet-dog shakes at 70%, followed by clonic-tonic-clonic convulsions at 21% and stage 3-5 convulsions in only 9%. In contrast, the first seizure observed in 93% of the kindled rats responsive to the 12 m g / k g dose of kainic acid was a stage 3-5 limbic convulsion. Interestingly, the majority of these initial convulsions in kindled rats appeared indistinguishable in form and duration to those previously recorded from each rat during electrical kindling (data not shown). Despite differences between kindled and control rats in the form of their initial seizure behavior, the latency from kainic acid administration to the onset of the initial behavior did not differ in the two groups (26.7 + 4.1 vs 33.3 _+ 2.8 rain, kindled vs control, respectively).

4. Induction of SLS

4.1. Behavior Not all rats manifesting one or more behavioral seizures to the initial dose of kainic acid went on to develop SLS. Despite the ability of the original dose to trigger discrete seizure activity, 4 of these rats (3 kindled and 1 control) required supplemental injection(s) of kainic acid to trigger SLS. The following comparisons between the kindled and control rats on the behavioral progression into SLS were based only on animals developing SLS to the original 12 m g / k g dose of kainic acid (kindled, n = 11; control, n = 32). 4.1.1. Controls The development of SLS in control rats was similar to that reported by Lothman and Collins [20] following i.v. administration of either a 12 or 16 m g / k g dose of

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

248

Control ---o--

01 n.

/j-

WDS

..... :

100

nodding, called stage 2 masticatory seizure, and increasingly frequent stage 3-5 convulsions. The mean latency to the first stage 5 convulsion in controls was 67.6 + 4.1 min. Animals were judged to be in SLS once the stage 2 seizures became continuous (called masticatory SE by Handforth and Ackerman [10]; and Mclntyre et al. [26]) and were punctuated every 3-5 min with a stage 5 clonic convulsion. In control rats, SLS required 95.3 + 3.5 min to develop (Fig. 2).

Profile

80 60

iI

4)

0

~. ~'~

iI

40

o.

iii II~

20

' o'° ..... .,1°

0

20

100 n-

100

Kindled

Profile

WDS

..... k---

GC

;

SLS

:

t

/

SO" 60"

*

*

• .......

• .......

40

By contrast, in kindled rats, SLS was achieved after 76.4 + 3.9 min. This was significantly faster than controis ( P < 0.005). In addition, the profile of SLS development in kindled rats was unique. Unlike controls, very few kindled rats manifested wet dog shakes. Instead, during the time wet dog shakes were observed in control rats, all kindled rats exhibited two or more discrete stage 5 convulsions. The latency to their first stage 5 convulsion, at 30.0 + 4.6 min, was signficantly faster than controls ( P < 0.001). In a few kindled rats, infrequent wet dog shakes were observed immediately before the onset of SLS but did not increase in frequency like the control rats (Fig. 2).

140

*

•

"°

......

60

---o-•

4.1.2. Kindled rats

"7"-

" ......

c"

/ /

\

ft.

20 0

'

,.o-

• 20 Time

....

--',- o.

4.2. E E G activity "~'"~

60 after

100 KA

Injection

, 140 (rain)

Fig. 2. The percentage of rats showing wet dog shakes (WDS), discrete generalized convulsions (GC) of mixed types and severe limbic status (SLS) at various periods of time after the kainic acid injection in both control and kindled groups. * Significantly different from the other group, P < 0.05.

kainic acid. Immediately following kainic acid injection rats manifested prostrate immobility with their abdomen on the chamber floor, intermittent periods of exaggerated respiration, reduced responsiveness to external stimuli and staring. This behavior was observed from 5-30 min following kainic acid administration and disappeared with the beginnings of wet dog shakes. The latter began as discrete episodes, but rapidly progressed into a near-continuous form. As can be seen in Fig. 2, nearly 80% of control rats were exhibiting repeated wet dog shakes between 40-60 min after the kainic acid injection. During this same period of time, only 12% of controls exhibited discrete generalized convulsive seizures. The latter were usually unorganized, myoclonic jerks of the upper body and forelimbs or brief clonic-tonic-clonic seizures, but were rarely clonic stage 3-5 limbic convulsions. With the decline in wet dog shakes, control rats began to manifest characteristic limbic convulsions that involved excessive salivation, mastication and head

4.2.1. Control rats

Similar to previous descriptions of the genesis of SLS following kainic acid administration in normal rats [21,42], interictal discharges were evident in the hippocampus within 15 min of the kainic acid injection. Except for continuous immobility and staring, these discharges were without behavioral correlates. By 30-40 min post-injection, discrete hippocampal ictal discharges appeared, then lengthened, became more frequent, and recruited the olfactory bulbs a n d / o r amygdala. Even 60 min post-injection, the olfactory bulb and amygdala discharges were without clear behavioral correlates. However, by 90 min, as the electrographic seizures in the different areas became more regular in frequency and merged, the onset of the olfactory b u l b / amygdala discharges became associated with stage 1-2 seizure behavior. Soon after, SLS began (Fig. 3). With the onset of SLS, the discharges in all structures became continuous ictal events and, without pharmacological intervention, remained in this form for more than 1 h. The continuous ictal activity eventually was interrupted by increasing low-voltage 'fiat periods' and, between 4 and 6 h post-injection, changed into periodic epileptiform discharges on a flat background or PEDs [42]. In controls, at the end of SE, the E E G activity in the dorsal hippocampus was considerably reduced in amplitude compared to baseline recordings (Fig. 4).

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256 4. 2.2. Kindled rats

In kindled rats, like controls, hippocampal interictal discharges a p p e a r e d within 15 min of the kainic acid injection. However, unlike controls, within 15-30 min post-injection, many discrete hippocampal ictal discharges a p p e a r e d which were associated with olfactory b u l b / a m y g d a l a discharges and stage 2 - 5 convulsions. Except for the early appearance of olfactory b u l b / amygdala discharges and associated discrete limbic convulsions, the electrographic progression of kindled rats into SLS was similar to controls. Once SLS was established, the E E G severity during the first few h was similar between kindled and control rats. On the other hand, after 4 - 6 h of SLS, the

Baseline RH . . . . . . . . . . .

......

5 min :.

.~ ,

40 min

10.5

mV

RH RO

~"~;'~-": , . T T ' : ; : , ~ ; ; ; , ~ ~

LH

LO

90 min

LH

249

kindled rats maintained strong continuous ictal activity while the controls showed very reduced activity or PEDs. Eventually, SLS in the kindled rats also subsided but, unlike controls, the final E E G baseline in the dorsal hippocampus was similar in amplitude to that seen before kainic acid exposure (Fig. 5).

5. Brain pathology Six control rats and one kindled rat died within 24 h of kainic acid administration and were not included in the histological assessment. The assessment, however, did include those rats given supplemental doses of kainic acid to induce SE. To simplify the data presentation, first the neuron and glia counts of the kindled and control rats after SE were compared to the special kindled rats which did not experience SE. Second, the kindled and control counts were converted into a damage score as indicated in the method section. For example, the average ( ± S . E . M . ) neuron count in layer 2 of the piriform cortex from the special kindled group that did not experience SE, compared to the kindled and control groups that did experience SE (and no pentobarbital), was 12.3 + 0.7, 14.8 + 1.2 and 0.0 + 0.0, respectively, while the complementary glia counts for the same groups were 1.5 + 0.4, 1.6 + 0.6 and 23.8 + 5.9, respectively. Thus, based on the cell counts of the special kindled group, the above counts from the kindled and control SE rats converted into simplified damage scores of 0 and 3, respectively. The range of neuron and glia cell counts in the special kindled comparison group for all reported structures was as follows: anterior olfactory nucleus ( 5 - 9 and 0-2, neurons and glia, respectively), midline thalamic nuclei (3-5 and 1-4), dorsal hippocampus CA1 ( 4 - 8 and 0 - 1 ) and CA3 (3-5 and 0-1), ventral hippocampus CA1 (4-8 and 0 - 1 ) and CA3 (3-5 and 0-2), piriform cortex (8-19 and 0 - 4 ) and the substantia nigra reticulata ( 1 - 4 and 3-9). The locus and severity of brain damage in each rat was highly dependent upon (a) its group membership (kindled vs control), (b) the form of SE induced and (c) the time of pharmacological intervention with pentobarbital (if at all). 5.1. N o pentobarbital

The 16 rats that did not require pentobarbital intervention during SE are presented in Table 1. Fig. 3. Electrographic recordings before injection (basehne) and at indicated times after injection of kainic acid in a control rat (C#8). At 5 min, interictal spiking was apparent in the right hippocampus (RH). By 40 min, discrete seizures in both the RH and left hippocampus (LH) recruited both olfactory bulbs (RO, LO), but with no behavioral correlates. By 90 rain, the discrete EEG seizures merged into near-continuous ictal activity; soon thereafter SLS began.

5.1.1. Control rats

All 8 control bilateral damage substantia nigra both the dorsal

rats developing SLS showed severe to the piriform cortex (Figs. 6 and 7), reticulata (Fig. 7), C A 1 - 3 fields in and ventral hippocampus, anterior

250

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

RH----------

RO

LH

Baseline

a

b

3 Hr

3Hr

6 Hr

16 Hr 0.5 mV 5s

Fig. 4. Electrographic recordings before and at various times after the injection of kainic acid during SLS in a control rat (C#8). At 3 h, the more common ictal pattern (a) was occasionally interrupted by a mild stage 3 convulsion (b). By 6 h, the hippocampal discharge abated and by 16 h was absent. Note the reduced voltage in the hippocampus at 16 h compared to baseline. Twenty days after SE, this rat had CA1 and CA3 hippocampal damage scores of 2 and 3, respectively.

olfactory nuclei, m i d l i n e t h a l a m u s ( i n c l u d i n g the dorsomedial, p a r a t a e n i a l a n d r e u n i e n s nuclei) a n d m a n y o t h e r structures n o t p r e s e n t e d . Interestingly, two of the controls did not develop SLS b u t r a t h e r the less intense c o n t i n u o u s stage 2 masticatory SE. I n b o t h cases, the s u b s t a n t i a nigra reticulata a n d m u c h of the dorsal h i p p o c a m p u s was spared.

rats e x p e r i e n c e d m o t o r a n d electrographic seizures which a p p e a r e d as severe as controls a n d lasted longer t h a n controls, yet the piriform cortex a n d s u b s t a n t i a nigra reticulata a p p e a r e d n o r m a l (Figs. 6 a n d 7). I n addition, while d a m a g e to the dorsal h i p p o c a m p u s was only slight, the v e n t r a l h i p p o c a m p u s was more damaged, b u t significantly less t h a n controls. O n the o t h e r hand, d a m a g e to the a n t e r i o r olfactory n u c l e u s a n d m i d l i n e t h a l a m u s in the kindled rats was similar to controls. A p p a r e n t l y previous dorsal h i p p o c a m p a l kindling provided a d r a m a t i c n e u r o p r o t e c t i o n from kainic

5.1.2. Kindled rats T h e b r a i n d a m a g e in the 6 k i n d l e d rats showing SLS was dramatically different from controls. T h e k i n d l e d

R

H

~

RO ~,~,~,,~.=~*,,~=_:~,.L~,~

LA

. . . . . ..

~--

~

~

a

~

~,~-~. . . . . ~

b IIIrrrlrrr=TrrlrrrrrfTFfrl~rrrrl'P¢l~'~

Baseline

3 Hr

3Hr

6 Hr

16 Hr 0.5 mV 5s

Fig. 5. Electrographic recordings before and at various times after the injection of kainic acid during SLS in a dorsal hippocampus kindled rat (DH#26). Like controls, the more common ictal pattern at 3 h (a) in this kindled rat was occasionally interrupted by a stage 3 convulsion (b). Unlike controls, the kindled rat at 6 h showed strong ictal discharges in all structures, including the right hippocampus (RH) and both amygdalae (RA, LA). Also, unlike controls, the voltage in the RH at 16 h was not decreased compared to baseline. Thirty days after SE, this rat had CA1 and CA3 hippocampal damage scores of 1 and 0, respectively.

M.E. Kelly, D.C. Mcln tyre / Brain Research 634 (1994) 245 - 256

251

Table 1 Comparison of kindled and control groups in the degree of brain damage in selected neural structures following spontaneous recovery from status epilepticus (no pentobarbital intervention) Group

Form

n

of SE

Structure AON

MT

Hippocampus Dorsal

PIR

SNR

Ventral

CA1

CA3

CA1

CA3

Controls

SLS Mast.

8 2

3.0 _+ 0.0 3.0 + 0.0

3.0 _+ 0.0 3.0 __ 0.0

2.5 _+ 0.4 b 0.5 _+ 0.5 c

2.8 _+ 0.3 b 0.0 + 0.0 c

2.9 _+ 0.1 a 2.5 _+ 0.5

3.0 _+ 0.0 b 2.5 _+ 0.5 a

3.0 -~- 0.0 b 3.0 _+ 0.0 b

3.0 _+ 0.0 b 0.0 + 0.0 c

Kindled

SLS

6

2.7 _+ 0.3

3.0 _+ 0.0

0.3 _+ 0.2

0.0 + 0.0

1.7 + 0.3

0.7 + 0.5

0.0 _+ 0.0

0.0 + 0.0

Average + S.E.M. damage score where 0, 1, 2, 3 equals no, minimal, moderate and severe damage, respectively. SLS, severe limbic status; Mast., masticatory status; AON, medial anterior olfactory nuclei; MT, midline thalamic nuclei; PIR, piriform cortex; SNR, substantia nigra reticulata. Significantly different from the kindled group, a p < 0.05, b p < 0.01; c significantly different from the SLS control group, P < 0.05.

acid-induced SE for several neural structures, particularly the piriform cortex and the substantia nigra reticulata. 5.2. Pentobarbital intervention If the severity of SE was determined to be lifethreatening, a low dose of pentobarbital was administered. This occurred at a time when the duration of SE was either less than 1.5 h (Table 2) or between 2 - 4 h (Table 3). Following the injection, SE either stopped (which we called 'none' in Tables 2 and 3) or resumed in the same (SLS) or a milder form (masticatory or ambulatory SE [25]). 5.2.1. Controls In control rats, if the SLS was terminated within 1.5 h, very little evidence of brain damage was noted

(Table 2). On the other hand, if SE resumed after the injection, the resulting brain damage was related to the form of SE which redeveloped. Resumption with the mild ambulatory form [26] was associated with considerable damage to limbic structures, excluding the piriform cortex. The substantia nigra reticulata was also spared. Resumption of the more severe masticatory SE resulted additionally in destruction of the piriform cortex but not the substantia nigra reticulata. Also, inexplicably, the two masticatory SE rats showed little damage to their dorsal hippocampus. In two other rats, SLS resumed after the injection and resulted in brain damage equivalent to controls receiving no pentobarbital intervention. There were 5 rats that endured SE for 2 - 4 h before receiving pentobarbital. Four of these rats exhibited SLS before the injection, and in 3 of the 4 rats SE did not resume. However, similar to the 1 rat that did

Fig. 6. Photomicrographs of the pirilorm cortex of a hippocampal-kindled rat from the special group not administered kainic acid (A) compared to a kindled rat (B) and control rat (C) from the groups which experienced SE without pharmacological intervention (Cresyl Violet stain).

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

252

13

Fig. 7. Photomicrographs of the substantia nigra (A,C: Cresyl Violet stain) and whole brain (B,D: Timm's and Cresyl Violet stain) from a hippocampal-kindled (A,B) and control (C,D) rat following spontaneous termination of SE. Note the extensive gliosis in the substantia nigra reticulata (C) and loss of innervation and destruction in the amygdala-piriform area (D) of the control rat (10 days post-SE) compared to the normal appearance of these areas in the kindled rat (40 days post-SE).

resume with SE (as ambulatory SE), all showed considerable damage to most structures, including the piriform cortex and substantia nigra reticulata. Only one rat showed masticatory SE before the injection and, although SE did not resume, massive loss of limbic

structures were observed, while the substantia nigra reticulata was spared (Table 3). In summary, in control rats, 2 or more h of SLS destroyed the limbic structures and the substantia nigra reticulata. Similar durations of masticatory SE,

Table 2 Comparison of kindled and control groups in the degree of brain damage in selected neural structures after status epilepticus (SE) and pentobarbital intervention within 1.5 h of SE onset Group

Form of SE before (B) & after (A) Pentobarb.

n

Structure AON

MT

Hippocampus

PIR

Dorsal CA1

Controls 1. 2. 3. 4.

B SLS SLS SLS SLS

A none Amb. Mast. SLS

6 3 2 2

0.0 2.3 3.0 3.0

Kindled 1. 2. 3.

B SLS SLS SLS

A none Mast. SLS

4 3 4

0.0 + 0.0 0.0 ± 0.0 2.0 ± 0.7 d

_+ 0.0 a ± 0.3 ± 0.0 ± 0.0

0.2 2.7 3.0 3.0

+ 0.1 a + 0.3 ± 0.0 +_0.0

0.8 ± 0.4 2.7 ± 0.3 2.5 ± 0.3

0.0 3.0 0.5 3.0

± ± ± ±

Ventral CA3

0.0 0.0 b 0.5 0.0 b

0.0 _+ 0.0 0.0 ± 0.0 0.3 ± 0.2

SNR

0.0 1.7 0.5 3.0

+ ± + ±

CA1 0.0 0.7 0.5 0.0 b

0.0 ± 0.0 0.3 + 0.3 0.3 + 0.2

0.0 3.0 3.0 3.0

± ± ± +

CA3 0.0 a 0.0 0.0 0.0

0.0 _+ 0.0 1.0 ± 1.0 2.0 ± 0.4

0.0 3.0 3.0 3.0

± ± ± ±

0.0 a 0.0 0.0 0.0

0.3 ± 0.0 1.0 +_ 1.0 1.3 ± 0.7

0.0 0.6 3.0 3.0

± + ± +

0.0 0.3 0.0 c 0.0 c

0.0 + 0.0 0.0 ± 0.0 0.8 ± 0.8

0.0 0.0 0.0 3.0

± 0.0 ± 0.0 + 0.0 _+ 0.0 a

0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0

Average ± S.E.M. damage score where 0, 1, 2, 3 equals no, minimal, moderate and severe damage, respectively. SLS, severe limbic status; Amb., ambulatory status; Mast., masticatory status; AON, medial anterior olfactory nuclei; MT, midline thalamic nuclei; PIR, piriform cortex; SNR, substantia nigra reticulata. Significantly different from: a all other control groups, b control groups 1 and 3, c control groups 1 and 2, ~ all other kindled groups, P < 0.05.

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

253

between kindled and control rats was not surprising since the kindled rats had a fully developed kindled limbic seizure network before the kainic acid injection while the controls did not. It is believed that kainic acid initially induces seizure activity by activating kainic acid receptors in the hippocampal CA3 region [3]. This region is sensitive to very low concentrations of kainic acid [39] and contains the highest level of kainic acid binding sites in the brain [4,31,36,43]. Although kainic acid likely initiated seizure activity in the same areas in kindled and control rats, it should be recalled that several of the kindled rats were less responsive than controls to the 12 m g / k g dose. This decreased responsivity was most pronounced 1 day (compared to 21-28 days) following kindling, where 50% of the kindled rats required a supplemental dose(s) of kainic acid to develop SLS. This relative insensitivity to kainic acid in the kindled rats might be related to the report that 1 day following either angular bundle or amygdala kindling a significant but transient down-regulation of kainic acid receptors in the hippocampal CA3 area occurred [35,40]. Yet, paradoxically, SLS developed more rapidly in our kindled rats which were responsive to the 12 m g / k g dose than in controls. We suspect, therefore, that recent kindling raised the threshold for kainic acid to recruit hippocampal discharges, but once that threshold was reached, limbic convulsions and SLS developed more readily in the kindled compared to control rats. Regardless of whether or not additional injections of kainic acid were required to develop SLS, once developed, SLS was as severe, or more so, in kindled compared to control rats. In addition, as SLS continued, an equal number of kindled and control rats required pentobarbital intervention to reduce the severity of their seizures so that it was no longer life-threatening. In nearly half of the rats, SE did not

however, destroyed most limbic structures but not the substantia nigra reticulata. 5.2.2. Kindled rats

Like the control rats, the 4 kindled rats experiencing less than 1.5 h of SLS, with no resumption after pentobarbital, showed little or no brain damage. However, in those rats that resumed SE, as either masticatory SE or SLS, severe damage occurred to the anterior olfactory nucleus and midline thalamus and moderate damage to the ventral hippocampus, but minimal or no damage occurred to the dorsal hippocampus, piriform cortex and substantia nigra reticulata (Table 2). In the 4 kindled rats that required pentobarbital 2-4 h after SLS onset, the SE did not resume. Their brain damage was restricted to the anterior olfactory nuceus and midline thalamus with minimal or no involvement of other limbic structures or the substantia nigra reticulata (Table 3).

6. Discussion

When challenged with a 12 m g / k g dose of kainic acid, the response of dorsal hippocampal-kindled rats was different from controls. Response differences were evident in (a) the induction phase of SE, (b) the associated EE G correlates and (c) the neuropathological outcome. During the induction phase of SE, ictal discharges appeared first in the hippocampus and then later in the olfactory bulbs and amygdala. With the early olfactory-amygdala discharges, the kindled rats, unlike the controls, often exhibited stage 5 limbic convulsions. Indeed the stage 5 convulsion was the first form of behavioral seizure evident in 93% of the kindled rats compared to only 9% of the controls. This difference

Table 3 Comparison of kindled and control rats in the degree of brain damage in selected neural structures after status epilepticus (SE) and pentobarbital intervention within 2 - 4 h of SE onset Group

Form of SE before (B) & after (A) Pentobarb.

n

Structure AON

MT

Hippocampus Dorsal CA1

Controls

Kindled

B SLS Mast. SLS

A none none Amb.

3 1 1

2.3 + 0.7 3.0 3.0

3.0 ± 0.0 a 3.0 3.0

2.0 _ 1.0 2.0 1.0

B SLS

A none

4

0.7 _+ 0.4

1.5 _+ 0.6

0.0 +_ 0.0

PIR

SNR

Ventral

a

CA3

CA1

CA3

2.7 _+ 0.3 a 3.0 1.0

2.0 ± 1.0 3.0 2.0

2.7 _+ 0.3 b 3.0 2.0

2.7 _+ 0.3 b 3.0 3.0

3.0 + 0.0 b 0.0 3.0

0.5 ± 0.5

0.8 +_ 0.5

0.3 ± 0.3

0.0 ± 0.0

0.0 ± 0.0

Average _ S.E.M. damage score where 0, 1, 2, 3 equals no, minimal, moderate and severe damage, respectively. SLS, severe limbic status; Amb., ambulatory status; Mast., masticatory status; A O N , medial anterior olfactory nuclei; MT, midline thalamic nuclei; PIR, piriform cortex; SNR, substantia nigra reticulata. Significantly different from the kindled group: a p < 0.05, b p < 0.01.

254

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

resume following this treatment. In the rats that resumed SE (or did not require pentobarbital intervention), the electrographic seizure continued for a much longer period in those kindled than in controls. The earlier offset of SE in the controls speculatively was a product of seizure-induced brain damage that was not seen in the kindled rats. In control rats, the profile of brain damage was consistent with previous reports [2,12,20]. However, the extent of brain damage in various structures was highly dependent on both the duration and the form of SE. In our hands, all control rats with substantia nigra reticulata damage experienced at least 2 or more h of SLS. Rats experiencing less than 2 h of SLS showed no substantia nigra reticulata damage. The damage inflicted by SLS upon these limbic areas and the substantia nigra reticulata in control rats stands in marked contrast to the paucity of damage seen in the kindled rats. SLS in kindled rats was behaviorally and electrographically as severe as in controis. In spite of this, remarkably, there was n o damage to the piriform cortex and substantia nigra reticulata of kindled rats, and only slight damage to the dorsal hippocampus. However, these kindled rats were not without neuronal vulnerability. Damage approximating that of controls was observed in the anterior olfactory nucleus and midline thalamic nuclei of kindled rats. Thus, specific limbic or limbic associated areas were differentially damaged in the kindled rats, while the piriform cortex and substantia nigra reticulata showed complete neuroprotection. Several mechanisms have been proposed whereby kainic acid-induced SE leads to cell death. It is believed that kainic acid activation of kainate receptors leads to increased glutamate release and, eventually, glutamate activation of the N M D A subtype receptors. Prolonged N M D A receptor activation promotes cascading biochemical and ionic changes, including the accumulation of large amounts of intracellular calcium. Evidence that N M D A receptor activation is involved and that the accumulation of intracellular calcium is critical in neurotoxic cell death has been shown by several investigators [6,7,8,10,29,44]. Recently, accumulation of oxygen free radicals has also been implicated in kainic acid neurotoxicity. After systemic treatment with kainic acid, administration of the free radical scavenger, mannitol, protected most neurons [1]. Similarly, after intracerebral administration of kainic acid, allopurinol, an inhibitor of the free radical-generating enzyme, xanthine oxidase, selectively protected the piriform cortex but not the striaturn [7]. Whether alterations to one or more of these mechanisms could account for the selective neuroprotection observed in kindled rats is unknown. The evidence that N M D A receptors might be more important for the

genesis of kindling than its maintenance [28] suggests that previous kindling might bias glutamate transmission in favor of non-NMDA receptors [13]. As a result, the SE induced by kainic acid in kindled rats might be more supported by non-NMDA activation than the SE in controls. Alteration in the mechanisms of scavenging oxygen free radicals during prolonged seizure activity may also be a factor in the kindling-induced protection from kainic acid. Mori et al. [32] found a significant longlasting increase of superoxide dismutase (SOD) in whole brains of amygdala-kindled rats. SOD are important enzymes that remove oxygen radicals from aerobic cells, significantly reducing the potential toxicity of the free radicals. The neuroprotection reported here is perhaps similar but far more extensive than the kindling-based neuroprotection from pilocarpine-sensitized SE reported by Buterbaugh and Hudson [5]. Although numerous methodological differences exist between the 2 studies, kindling seemed to provide varying degrees of neuroprotection for several limbic structures in both studies, while the substantia nigra reticulata was protected only in our study. Also, perhaps phenomenologically similar to our neuroprotection observation is the recent report that a bout of kainic acid SE, terminated after 1 h with pentobarbitat anesthesia, and reinstated 16 h later with additional kainic acid, was associated with significant neuroprotection in the hippocampal CA3 region compared to controls [33]. In addition, the ischemic-tolerance phenomenon, whereby previous exposure to nonlethal ischemia protected hippocampal CA1 neurons against a second, longer ischemic episode [15,18,19] is further demonstration of an 'experience-induced' alteration in the mechanisms leading to cell death. And lastly, the importance of the location of the kindled focus in the neuroprotection phenomenon was highlighted by our recent observation [17] that amygdala kindling provided protection from kainic acid-induced SE only for the ipsilateral amygdala-piriform area and not the hippocampus or contralateral structures. We believe this protection was restricted because amygdala kindling largely modifies the neural circuits only within the ipsilateral hemisphere [27]. On the other hand, by way of the hippocampal commissure [24], kindling in the dorsal hippocampus modifies the associated neural circuits of both hemispheres. In summary, the present results provide dramatic evidence that a kindled and naive brain respond very differently when challenged with kainic acid-induced SE. The neuronal protection observed in the piriform cortex, substantia nigra reticulata and dorsal hippocampus, suggests that previous seizure activity may be beneficial for neuronal survival during subsequent, more prolonged seizure activity. Understanding the

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

factors involved in the kindling-induced protection phenomenon should provide new insights into the mechanisms critical for cell death. The structurally selective nature of the protection may also indicate the pathways or structures that are most important to kindling.

References [1] Baran, H., Lassmann, H., Sperk, G., Seitelberger, F. and Hornkiewicz, O., Effect of mannitol treatment on brain neurotransmitter markers in kainic acid-induced epilepsy, Neuroscience, 21 (1987) 679-684. [2] Ben-Ari, Y., Tremblay, E. and Ottersen, O.P., Injections of kainic acid into the amygdaloid complex of the rat: an electrographic, clinical and histological study in relation to the pathology of epilepsy, Neuroscience, 5 (1980) 515-528. [3] Ben-Ari, Y., Limbic seizure and brain damage produced by kainic acid: mechanisms and relevance to human temporal lobe epilepsy, Neuroscience, 14(2) (1985) 375-403. [4] Berger, M. and Ben-Ari, Y., Autoradiographic visualization of [3H]-kainic acid receptor subtypes in the rat hippocampus, Neurosci. Lett., 39 (1983) 237-242. [5] Buterbaugh, G.G. and Hudson, G.M., The kindling process and vulnerability to status epilepticus. In J.A. Wada (Ed.), Kindling 4, Plenum, New York, 1990, pp. 169-181. [6] Clifford, D.B., Olney, J.W., Benz, A.M., Fuller, T.A. and Zorumski, C.F., Ketamine, phencyclidine, and MK-801 protect against kainic acid-induced seizure-related brain damage, Epilepsia, 31 (1990) 382-390. [7] Facchinetti, F., Virigili, M., Contestabile, A. and Barnabei, O., Antagonists of the NMDA receptor and allopurinol protect the olfactory cortex but not the striatum after intra-cerebral injection of kainic acid, Brain Res., 585 (1992) 330-334. [8] Fariello, R.G., Golden, G.T., Smith, G.G. and Reyes, P.F., Potentiation of kainic acid epileptogenicity and sparing from neuronal damage by an NMDA receptor antagonist, Epilepsy Res., 3 (1989) 206-213. [9] Garthwaite, G. and Garthwaite, J., Neurotoxicity of excitatory amino acid receptor agonists in rat cerebellar slices: dependence on calcium concentration, Neurosci. Lett., 66 (1986) 193-198. [10] Handforth, A. and Ackermann, R.F., Functional [14C]2-deoxyglucose mapping of progressive states of status epilepticus induced by amygdala stimulation in rat, Brain Res., 460 (1988) 94-102. [11] Hauser, W.A., Status epilepticus: frequency, etiology, and neurological sequelae. In A.V. Delgado-Escueta, D.M. Treiman and R.J. Porter (Eds.), Advances in Neurology, Vol 34: Status Epilepticus, Raven, New York, 1983, pp. 3-14. [12] Heggli, D.E., Aamodt, A.A. and Malthe-Sorenssen, D., Kainic acid neurotoxicity: effect of systemic injection on neurotransmitter markers in different brain regions, Brain Res., 230 (1981) 253-262. [13] Hosford, D.A., Crain, B.J., Zhen, C., Bonhaus, D.W., Friedman, A.H., Okainic acidzaki, M.M., Nadler, J.V. and McNamara, J.O., Increased AMPA-sensitive quisqualate receptor binding and reduced NMDA receptor binding in epileptic human hippocampus, J. Neurosci., 11 (1991) 428-434. [14] Jope, R.S., Morrisett, R.A. and Snead, O.C., Characterization of lithium potentiation of pilocarpine-induced status epilepticus in rats, Exp. Neurol., 91 (1986) 471-480. [15] Kato, H., Liu, Y., Araki, T. and Kogure. K., Temporal profile of the effects of pretreatment with brief cerebral ischemia on the neuronal damage following secondary ischemic insult in the

255

gerbil: cumulative damage and protective effects, Brain Res., 553 (1991) 238-242. [16] Kelly, M.E. and Mclnytre, D.C., Previous kindling protects the pyriform cortex from neuronal damage during kainic acid-induced status epilepticus, Epilepsia, 31 (Suppl. 5) (1990) 634. [17] Kelly, M.E. and Mclntyre, D.C., Previous amygdala kindling protects the ipsilateral pyriform cortex but not the hippocampi from neuronal damage during kainic acid-induced status epilepticus, Epilepsia, 32 (Suppl. 3) (1991) 40. [18] Kitagawa, K., Matsumoto, M., Tagaya, M., Hata, R., Ueda, H., Niinobe, M., Handa, N., Fukunaga, R. Kimura, K., Mikoshiba, K. and Kamada, T., 'Ischemic tolerance' phenomenon found in the brain, Brain Res., 528 (1990) 21-24. [19] Kitagawa, K., Matsumoto, M., Kuwabara, K., Tagaya, M., Ohtsuki, T., Hata, R., Ueda, H., Handa, N., Kimura, K. and Kamada, T., 'Ischemic tolerance' phenomenon detected in various brain regions, Brain Res., 561 (1991) 203-211. [20] Lothman, E.W. and Collins, R.C., Kainic acid-induced limbic seizures: metabolic, behavioral, electroencephalographic and neuropathological correlates, Brain Res., 218 (1981) 299-318. [21] Lothman, E.W., Bertram, E.H., Bekenstein, J.W. and Perlin, J.B., Self-sustaining limbic status epilepticus induced by 'continuous' hippocampal stimulation: electrographic and behavioral characteristics, Epilepsy Res., 3 (1989) 107-119. [22] Lothman, E.W., The biochemical basis and pathophysiology of status epilepticus, Neurology, 40(2) (1990) 13-23. [23] Mclntyre, D.C., Nathanson, D. and Edson, N., A new model of partial status epilepticus based on kindling, Brain Res., 250 (1982) 53-63. [24] Mclntyre, D.C. and Edson, N., Facilitation of secondary site kindling in the dorsal hippocampus following forebrain bisection, Exp. Neurol., 96 (1987) 569-579. [25] Mclntyre, D.C. and Kelly, M.E., Is the pyriform cortex important for limbic kindling? In J.A. Wada (Ed.), Kindling 4, Plenum, New York, 1990, pp. 21-32. [26] Mclntyre, D.C., Don, J.C. and Edson. N., Distribution of [laC]2-deoxyglucose after various forms and durations of status epilepticus induced by stimulation of a kindled amygdala focus in rats, Epilepsy Res., 10 (1991) 119-133. [27] Mclntyre, D.C. and Plant, J.R., Long-lasting changes in the origin of spontaneous discharges from amygdala-kindled rats: piriform vs. perirhinal cortex in vitro, Brain Res., 624 (1993) 268-276. [28] McNamara, J.O., Morrisett, R. and Nadler, J.V., Recent advances in understanding mechanisms of the kindling model. In P. Chauvel, A.V. Delgado-Escueta, et al. (Eds.), Advances in Neurology, Vol. 57, Raven, New York, 1992, pp. 555-560. [29] Meldrum, B. and Garthwaite, J., Excitatory amino acid neurotoxicity and neurodegenerative disease, Trends Pharmacol. Sci., 11 (1990) 379-387. [30] Molino, A. and Mclntyre, D.C., Another inexpensive headplug for the electrical recording and/or stimulation of rats, Physiol. Behav., 9 (1972) 273-275. [31] Monaghan, D.T. and Cotman, C.W., The distribution of [3H]kainic acid binding sites in rat CNS as determined by autoradiography, Brain Res., 252 (1982) 91-100. [32] Mori, N., Wada, J.A., Watanabe, M. and Kumashiro, H., Increased activity of superoxide dismutase in kindled brain and suppression of kindled seizure following intra-amygdaloid injection of superoxide dismutase in rats, Brain Res., 557 (1991) 313-315. [33] Najm. I., Schreiber, S.S., Bruce, A., Tocco, G. and Baudry, M., (1992) A short episode of seizure activity protects CA3 neurons from prolonged seizure activity-induced death, Proc. Soc. Neurosci., 18 (1992) 1146. [34] Olney, J.W., Rhee, V. and Ho, O.L., Kainic acid: a powerful neurotoxic analogue of glutamate, Brain Res., 77 (1974) 507-512.

256

M.E. Kelly, D.C. Mclntyre / Brain Research 634 (1994) 245-256

[35] Okazaki, M.M., McNamara, J.O. and Nadler, J.V., Kainate and quisqualate receptor autoradiography in rat brain after angular bundle kindling, Neuroscience, 37 (1990) 135-142. [36] Patel, S., Meldrum, B.S. and Collins, J.F., Distribution of [3H]kainic acid and binding sites in the rat brain: in vivo and in vitro receptor autoradiography, Neurosci. Lett., 70 (1986) 301307. [37] Pellegrino, L.J., Pellegrino, A.S. and Cushman, A.J., A Stereotax& Atlas of the Rat Brain, Plenum, New York, 1979. [38] Racine, R.J., Modification of seizure activity by electrical stimulation: II. Motor seizure, Electroencephalogr. Clin. Neurophysiol., 32 (1972) 281-294. [39] Robinson, J.H. and Deadwyler, S.A., Kainic acid produces depolarization of CA3 pyramidal cells in the in vitro hippocampal slice, Brain Res., 221 (1981) 117-127. [40] Savage, D.D., Nadler, J.V. and McNamara, J.O., Reduced kainic

[41]

[42]

[43]

[44]

acid binding in rat hippocampal formation after limbic kindling, Brain Res., 323 (1984) 128-131. Sloviter, R.S., A simplified Timm stain procedure compatible with formaldehyde fixation and routine paraffin embedding of rat brain, Brain Res. Bull., 8 (1982) 771-774. Treiman, D.M., Walton, N.Y. and Kendrick, C., A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus, Epilepsy Res., 5 (1990) 49-60. Unnerstall, J.R. and Wamsley, J.K., Autoradiographic localization of high affinity [3H]kainic acid binding sites in the rat forebrain, Eur. J. PharmacoL, 323 (1983) 128-131. Virgili, M., Migani, P., Contestabile, A. and Barnabei, O., Protection from kainic acid neuropathological syndrome by NMDA receptor antagonists:effect of MK-801 and CGP 39551 on neurotransmitter and glial markers, Neuropharmacology, 31(5) (1992) 469-474.