Blockade of spreading depression in chronic epileptic rats: reversion by diazepam

Epilepsy Research 27 (1997) 33 – 40

Blockade of spreading depression in chronic epileptic rats: reversion by diazepam Rubem Carlos Arau´jo Guedes a,*, Esper Abra˜o Cavalheiro b a

Departamento de Nutric¸a˜o, Uni6ersidade Federal de Pernambuco, 50670 -901 Recife PE, Brazil b Neurologia Experimental, UNIFESP-EPM, Sao Paulo SP, Brazil

Received 2 August 1996; received in revised form 18 November 1996; accepted 6 December 1996

Abstract Following pilocarpine-induced status epilepticus, rats become chronically epileptic showing 2 – 3 spontaneous recurrent seizures per week. The aim of this work was to verify the characteristics of spreading depression (SD) in these chronic epileptic rats (n= 16). SD was evoked in one point of the frontal cortex by topical application of KCl solution at 20 min intervals, and recordings were made in two points over the parietal cortex (a ‘near’ point and a ‘remote’ one, about 3 and 8 mm posterior to the stimulating region, respectively). In all control animals (n= 10), KCl stimulation elicited SD which in 100% of the cases propagated regularly to the two recording points. In the chronic epileptic rats only about 50% of the KCl applications were effective in eliciting SD, detected at the ‘near’ recording point. Of these, only 3% propagated to the ‘remote’ recording point. In eight of the above epileptic rats, diazepam (5–10 mg/kg, i.v.) was injected after 1–2 h of recording, when SD incidence in response to KCl stimulation has been established. After diazepam, the incidence of SDs propagating regularly to the ‘near’ and to the ‘remote’ recording points increased significantly, (132 and 53 SDs, respectively, out of 139 KCl stimulations), as compared with the incidence in the pretreatment period (33 and 1, respectively, out of 63 stimulations; P B 0.005). These data indicate an impairment in the susceptibility to cortical SD in chronic epileptic rats suggesting modifications in cortical excitability in the pilocarpine model of epilepsy. They also indicate a facilitatory effect of diazepam on SD, confirming previous observations in non-epileptic rats. © 1997 Elsevier Science B.V. Keywords: Pilocarpine model of epilepsy; Spreading depression; Cortical excitability; Chronic epilepsy; Diazepam; Rats

1. Introduction

* Corresponding author. Tel.: + 55 81 2718470; fax: + 55 81 2718473 or + 55 81 2718500.

Cortical spreading depression (SD) originally was described as a reversible and propagating ‘wave’ across the brain’s surface reflecting a re-

0920-1211/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 9 2 0 - 1 2 1 1 ( 9 6 ) 0 1 0 1 7 - 0

34

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

duction of cortical spontaneous and evoked activity following electrical, mechanical or chemical stimulation at one point on the neural tissue [18]. Additionally, other changes in cortical physiology accompanying SD were found to consist of brain vascular alterations [19], a simultaneous slow potential change of the cortical surface [20], and water and ion translocations [17,23,24]. Despite this good degree of characterisation of the process, the underlying biological events directly responsible for the phenomenon still need to be identified. Several proposals have appeared regarding the involvement of certain ions [7,12,21,25] and synaptic transmitters [9 – 11,28] in SD. It has also been observed that several conditions that influence neural excitability also can interfere with brain susceptibility to SD [8,10,21]. Early on, it was noted by Leao [18] that abnormal electrographic activity, similar to that observed in the EEG of epileptic patients, eventually appeared during the SD phenomenon. The relationship between SD and acutely created brain hyperactivity has been elegantly studied by Bures and co-workers [2,14,15]. These authors have shown that SD does not invade a cortical area in which excitability has been increased either by electrical or by topical chemical stimulation. Also, in non-epileptic adult rats we have previously shown that epileptiform EEG activity induced acutely by the intravenous injection of bicuculline can abolish the effectiveness of KCl in eliciting SD, while i.v. diazepam facilitated SD propagation [11]. While considerable attention has been paid to the influence of acutely-induced hyperexcitability on SD, there is relatively little information on the SD features in chronic epileptic rats. One way to address this subject is to study rats treated with pilocarpine. A single high dose of pilocarpine (PILO; 300– 380 mg/kg, i.p.), a potent muscarinic agonist, acutely induces sequential behavioural and electrographic changes indicative of epileptic activity, resulting in widespread damage to the forebrains of both rats and mice [26,27]. The early phase of the response comprises akinesia, ataxic lurching, facial automatisms and head tremor. After 15–25 min these changes are followed by motor limbic seizures with rearing, forelimb clonus, salivation,

intense masticatory movements and loss of posture. Such episodes recur every 2–8 min and within 50–60 min after PILO administration can lead to status epilepticus that may last for up to 12 h, rendering the animals prostrate or critically ill. The first 24 h during which those events occur, when the lethality rate reaches 30%, has been called the acute period. After a subsequent ‘silent period’ (the seizurefree phase) that varies from 4 to 44 days post-administration (mean 14.893.0 days) all surviving animals begin to exhibit spontaneous recurrent seizures (SRSs) which characterise the chronic period, varying from 2 to 15 seizure episodes per month (mean 2–3 seizures per week). Behaviourally, a spontaneously occurring seizure is characterised by facial automatisms, head nodding, forelimb clonus, rearing and loss of upright posture, and electrographically by paroxysmal hippocampal discharges that rapidly spread to cortical regions [5,22]. No spontaneous remission of SRSs has been observed for survival times even as long as 6 months. Due to their characteristics, SRSs monitored during the chronic period have been considered a good model for human temporal lobe epilepsy, since it reproduces in the rat several aspects of the human condition [3]. The facts mentioned above, as well as the lack of information in the literature on SD in chronic epileptic animals, prompted us to investigate whether rats made chronically epileptic by a single injection of PILO present changes in cortical susceptibility to SD, as compared with normal animals. Our data confirm this hypothesis. Additionally, evidence is presented indicating a facilitatory role played by diazepam on SD in chronic epileptic rats. Some of these data have been communicated to the Annual Meeting of the American Epilepsy Society, in Baltimore, MD (USA) [4].

2. Methods Twenty-three adult Wistar rats (200–250 g body weight) were used. They were housed in groups of 4–5 and maintained on a standard light/dark cycle (light on 07:00–19:00), with free

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

access to rat chow pellets and tap water. Sustained seizures were induced by a single i.p. administration of pilocarpine hydrochloride (320 mg/kg; Sigma, St. Louis, MO). Scopolamine methylnitrate (Sigma, St. Louis, MO) was injected (1 mg/kg, s.c.) 30 min before PILO in order to reduce peripheral cholinergic effects. Approximately 30 min after PILO administration most of the animals (21/23) had entered status epilepticus (SE) which lasted for 8 – 24 h. Five animals died in the first 24 h (acute phase). In the remaining 16 rats, spontaneous seizures have been monitored via a video-EEG system (Stellate systems, Quebec, Canada) as previously reported [5]. Animals were monitored for periods ranging from 60 – 90 days during which spontaneous recurrent seizure frequency was determined. Another group of ten animals (control group) was injected with scopolamine methylnitrate followed by saline (2 ml/kg) and observed for the same period of time as the experimental rats. Sixty to 90 days after PILO-induced status epilepticus, rats showing spontaneous recurrent seizures (n =16) were anaesthetised (Thionembutal, 40 mg/k, i.p.) and three holes (3 – 4 mm in diameter) were drilled on the right side of the skull (two above the parietal cortex, to place two recording electrodes, and one above frontal cortex, to apply KCl to elicit SD). These holes were aligned in the anteroposterior direction and were located parallel to the midline. Two plastic pipettes (4–5 cm long, 1 mm tip diameter) containing chlorided silver wires and filled with agarRinger solution were gently placed in contact with the parietal surface and used as recording electrodes. They were called ‘near’ and ‘remote’ electrodes, according to the distances (about 3 and 8 mm, respectively) separating them from the stimulating place, at the frontal cortex. A third pipette of the same type was placed on the nasal bones and served as a common reference electrode. Spontaneous electrical activity (ECoG) and DC potential changes occurring during SD were recorded continuously for 3 to 5 h on a 4-channel Gould chart paper recorder. SD was elicited by 1 min application of a small cotton pledget (1–2 mm in diameter) soaked in KCl (either 0.2 M or 4.0 M) to the frontal cortex. During the recording

35

session, rectal temperature was monitored and maintained at 3891°C. In eight chronic epileptic rats, a polyethylene cannula was inserted into the femoral vein at the beginning of the surgical procedure. This cannula was used to inject diazepam (5–10 mg/k, i.v.) 1–2 h after the beginning of the recording session. After diazepam, recordings were continued for additional 2–3 h. Mean values obtained from SD propagation rate were compared between groups by the Student’s t-test. In the epileptic rats treated with diazepam, mean SD velocities obtained before and after the treatment were compared by the paired t-test. The number of KCl-elicited SDs which were recorded at the ‘near’ and at the ‘remote’ recording cortical points was determined individually in the diazepam-treated animals and compared by the x 2 test.

3. Results As previously reported [3,5] each spontaneous seizure lasted approximately 50–60 s and was similar to a stage 5 kindled seizure, i.e. seizure began with clonic movements of the vibrissae, mastication, clonic movements of the forepaws and rearing followed by falling with clonus of the four limbs. In the EEG, fast high-voltage spiking activity was initially observed in the hippocampal leads with rapid spread to the cortical recording. All animals which survived the acute status epilepticus phase evolved to the chronic phase of spontaneous seizures after a silent period of approximately 8–27 days. The mean frequency of seizures in the 16 chronic animals used in the experiments reported presently was 2.891.2 seizures/week. Rats treated with saline did not present any acute or chronic manifestation of seizures. In the non-epileptic (control) rats, the topical application of 0.2 M KCl to the frontal cortex for 1 min elicited a single SD ‘wave’, which propagated normally and could be recorded by the two electrodes at the parietal surface. This was found to occur in 100% of the animals of this group. SD elicited in this way propagated normally at a rate of 3.59 0.3 mm/min. After SD, electrographic

36

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

Fig. 1. Electrocorticogram (2 upper traces) and slow potential changes (2 lower traces) accompanying SD in the neocortex of a control rat. The two cortical electrodes (1 and 2) and a reference electrode (R) are shown in the inset. S indicates the KCl stimulation site. In (A) a cotton ball (1–2 mm diameter) soaked in KCl (4 M) was applied at the time indicated by the arrow and left continuously during 160 min. A series of SD was elicited and could be recorded at both cortical points 1 and 2. (B and C), 50 min and 170 min, respectively, after the beginning of stimulation (interrupted 10 min before (C). Note the great reduction in ECoG amplitude after KCl application.

cortical recovery took 5 – 10 min to be completed. An interval of at least 20 min was kept before the next stimulation, as usual in our laboratory [10– 12]. During continuous application of 4 M KCl in non-epileptic rats, successive SDs appeared at intervals of 2 to 5 min, the ECoG activity became strongly depressed and the amplitudes and durations of the DC recording of SD ranged from 10 to 20 mV and from 40 to 80 s, respectively (Fig. 1). In the group made chronically epileptic by PILO, stimulation with 0.2 M KCl failed to elicit SD in 100% of the cases. Examples of recordings in 2 rats of this group are shown in Fig. 2. By using a more concentrated KCl solution (4.0 M), we were able to elicit SD which, however, did not propagate regularly. Fig. 2 shows that, after being recorded at the first electrode (nearest to the stimulation point), SD ceased propagating before reaching the second (‘remote’) recording electrode. Estimations of the SD propagation rate in these animals (based on the time elapsed between KCl application and SD appearance at the ‘near’

recording point) revealed a mean velocity of 2.09 0.5 mm/min, significantly lower than that found in the non-epileptic group (PB 0.005). As compared to control rats, epileptic animals presented a tendency to display longer intervals between successive SD waves (4 to 10 min), lower amplitudes (5–15 mV) and longer durations (60– 100s). The two rats which received PILO but which did not become epileptic (Section 2), presented SD susceptibilities comparable to those of the control rats, i.e. SD could be elicited in them even by 0.2 M KCl and propagated normally, being recorded at the two sites used for that purpose. No correlation was observed between seizure frequency and SD inhibition in individual animals. In eight of the chronic epileptic rats, the resistance to present, and to propagate SD was individually established and confirmed for 1–2 h of recording. Following these tests the animals received diazepam (5–10 mg/kg, i.v.) and the recording sessions were continued for an additional period of 2–3 h. Fig. 3 shows SD record-

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

37

Fig. 2. SD recordings in two rats (A/B and A%/B%, respectively) rendered chronically epileptic by pilocarpine. In this group, stimulation with 0.2 M KCl for 1 min (which in control rats routinely elicits a single SD wave, that propagates normally over the cortical surface), was not effective in eliciting the phenomenon. A more concentrated KCl solution (4 M) elicited SD, which could be recorded at the cortical point 1, but stopped propagating before reaching point 2. Reductions in ECoG amplitude seen in (A) are due to adjusting the sensitivity of the amplifiers to a lower gain.

ings after diazepam administration in the same epileptic rats presented in Fig. 2. Thirty to 100 min following diazepam, regularly propagated SD waves could be observed at the two recording sites, after application of either 0.2 M (C to F) or 4.0 M (C% to F%) KCl solution. In most cases, following diazepam the amplitudes of the DC recordings tended to increase and their duration tended to decrease, approaching those of the control rats. The SD propagation rate after diazepam increased significantly to 3.090.4 (P B 0.005), compared with the predrug rate. Table 1 summarises the changes in the effectiveness of KCl to elicit SD after diazepam, as compared with the pre-treatment values, in these eight epileptic rats.

4. Discussion Our data showed that in PILO-treated, chronic epileptic rats cortical SD is neither produced, nor propagated so easily as it is in control animals. This indicates that probably hyperexcitability produced by PILO makes elicitation and propagation

of SD more difficult, and is in agreement with data obtained from other models of hyperexcitability. For example, Koroleva and Bures [14] showed in rats that SD did not invade a cortical region in which a focus of hyperexcitability had been created by repetitive electrical stimulation of a portion of the neocortex. Also, the same resistance to SD generation and propagation was found in rats in which hyperexcitability was acutely induced by the GABA antagonists, picrotoxin [2] and bicuculline [11], as well as in rats submitted to pentylenetetrazol kindling [16]. Previous papers have characterised the electrographic activity during the interictal period in PILOtreated, chronically seizing rats [5,22]. In these animals isolated spikes can be observed both in hippocampus and in cortex. This interictal abnormal activity could be involved in SD inhibition. Blocking this activity with benzodiazepine could facilitate SD. In agreement to that, we find that treatment with diazepam facilitates SD propagation in the chronic epileptic rats, probably by enhancing GABA neurotransmission. These results are in accordance with our previous observa-

63

Total

33

03 09 02 07 01 00 00 11 52.4

37.5 81.8 22.2 70.0 25.0 0.0 0.0 78.6

B, as % of A

01

00 00 00 00 00 00 00 01

C = Remote point

3.03

0 0 0 0 0 0 0 9.1

C, as % of B

139

21 19 20 46 04 10 06 13

132

19 19 17 45 03 10 06 13

B= Near point

94.5*

90.5 100.0 85.0 97.8 75.0 100.0 100.0 100.0

53

18 10 03 06 03 01 06 06

B, as % of A (C) Remote point

40.2*

94.7 52.6 17.6 13.3 100.0 10.0 100.0 46.2

C, as % of B

For each animal, the following data are presented, before and after a single i.v. injection of diazepam (5–10 mg/kg). A, number of KCl applications; B and C, number of SDs recorded at the nearest recording point (B; about 3 mm from the stimulated region); and at the remote recording point (C; about 8 mm from the stimulated region). Values for B and C are also presented as percentage of A and B, respectively. Values marked with an asterisk are significantly different from the respective values before diazepam (PB0.005, x 2 test).

08 11 09 10 04 03 04 14

B= Near point

No. of SDs recorded at near or remote recording points

A=No. of KCl stimuli

A= No. of KCl stimuli

No. of SDs recorded at near or remote recording points

After diazepam

Before diazepam

01 02 03 04 05 06 07 08

Rat no.

Table 1 Effect of diazepam on the effectiveness of KCl stimulation in eliciting SD on the cortical surface of eight chronically epileptic rats

38 R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

tions in non-epileptic rats showing that SD velocities of propagation are increased in response to intravenous injection of diazepam, the opposite effect being observed when the convulsion-inducing GABA antagonists bicuculline and picrotoxin were administered [11]. The resistance to SD, presented by epileptic rats, can not be attributable to the barbiturate anaesthesia, since the control rats were under the same anaesthetic and did not present such a resistance. In support to that, barbiturate anaesthesia has been similarly used by others, in experiments showing resistance of the rat cortical tissue to

Fig. 3. SD recordings in the same rats of Fig. 2 (C/F and C%/F%, respectively), after a single i.v. injection of 5–10 mg · kg − 1 diazepam (DZP). Times after DZP are given in min. In one rat (C/F), every application of 0.2 M KCl for 1 min elicited a single SD wave, that propagated regularly, as usually seen in normal animals. In the other rat (C%/F%), continuous 4 M KCl provoked a series of SDs, which propagated normally and could be recorded at points 1 and 2.

39

SD, as a consequence of electrically- and chemically-induced hyperexcitability [2,15,16]. Although our data unequivocally show the reversion by diazepam of the SD effect produced by PILO, they are still insufficient to permit definitive conclusions about the underlying mechanisms. The facilitatory effect of diazepam on SD, observed in chronic epileptic rats, could indicate either that the decreased SD susceptibility produced by PILO cannot be attributed to irreversible alterations or that diazepam may simply counterbalance effects of irreversible alterations on SD elicitation and propagation. This remains to be clarified in further investigations. Concerning probable biochemical mechanisms to explain the SD impairment in these animals, one possibility would be a mechanism based on ability of the brain to control extracellular potassium. It has been shown that hyperexcitability provoked by cortical electrical stimulation leads to a kind of metabolic adaptation, in which brain extracellular K + removal is considerably increased [13,15]. Under such condition, SD threshold would be increased, so that a more intense stimulus would be necessary to elicit it. Once elicited, SD could easily stop propagating at a determined cortical region which, at that moment, is in a state of higher excitability. Treatment with DZP could facilitate SD propagation by reducing brain hyperexcitability and consequently decreasing the effectiveness of the mechanisms responsible for extracellular K + removal. In this context, it is also interesting to consider that benzodiazepines can augment depolarising GABA responses in glial cells [1,29], facilitating potassium (and glutamate) release. Benzodiazepines might also block K + conductances in glia [29], thereby facilitating SD induction [6]. In view of the above considerations, it is reasonable to think that some of the behavioural and cognitive alterations observed in epileptic individuals could have hyperexcitability as the underlying mechanism. Those alterations could be subsequent to changes in the biochemical environment in certain brain areas, resulting in a predominance of excitatory mechanisms.

40

R.C.A. Guedes, E.A. Ca6alheiro / Epilepsy Research 27 (1997) 33–40

Acknowledgements This work was supported by the Brazilian Agencies CNPq (No. 52.1706/94-7), FAPESP and CAPES. We would like to thank Dr. T.P. Hicks for critical comments on the manuscript.

[14]

References

[16]

[1] Bureau, M., Laschet, J., Bureau-Heeren, M., Hennuy, B., Minet, A., Wins, P. and Grisar, T., Astroglial cells express large amounts of GABAA receptor proteins in mature brain, J. Neurochem., 65 (1995) 2006–2015. [2] Bures, J., von Schwarzenfeld, I. and Brozek, G., Blockade of cortical spreading depression by picrotoxin foci of paroxysmal activity, Epilepsia, 16 (1975) 111–118. [3] Cavalheiro, E.A., Status epilepticus and secondary epileptogenesis in the pilocarpine model of epilepsy. In: P. Wolf (Ed.), Epileptic Seizures and Syndromes, John Lybbey, New York, 1994, pp. 523–532. [4] Cavalheiro, E.A. and Guedes, R.C.A., Spreading depression in chronically epileptic rats, Epilepsia, 36 (suppl. 4) (1995) 63. [5] Cavalheiro, E.A., Leite, J.P., Bortolotto, Z.A., Turski, W.A., Ikonomidou, C. and Turski, L., Long-term effects of pilocarpine in rats: structural damage of the brain triggers kindling and spontaneous recurrent seizures, Epilepsia, 32 (1991) 778–782. [6] Fujimoto, M. And Yanase, H., The effect of K + -conductance-blocking substances on the occurrence of retinal spreading depression, Exp. Eye Res., 53 (1991) 333– 336. [7] Grafstein, B., Mechanism of spreading cortical depression, J. Neurophysiol., 19 (1956) 154–171. [8] Guedes, R.C.A., On some conditions that influence cortical spreading depression, Anais Acad. Brasil. Cieˆnc., 56 (1984) 445 – 455. [9] Guedes, R.C.A., Andrade, A.F.D. and Cavalheiro, E.A., Excitatory amino acids and cortical spreading depression. In: E.A. Cavalheiro, J. Lehman and L. Turski (Eds.), Frontiers in Excitatory Amino Acid Research, Allan R. Liss, New York, 1988, pp. 667–670. [10] Guedes, R.C.A., Azeredo, F.A.M., Hicks, T.P., Clarke, R.J. and Tashiro, T., Opioid mechanisms involved in the slow potential change and neuronal refractoriness during cortical spreading depression, Exp. Brain Res., 69 (1987) 113 – 119. [11] Guedes, R.C.A., Cabral-Filho, J.E. and Teodo´sio, N.R., GABAergic mechanisms involved in cortical spreading depression in normal and early malnourished rats. In: R.J. Do Carmo, (Ed.), Spreading Depression, Exp. Brain Res. Series, 23, Springer, Berlin, 1992, pp. 17–26. [12] Guedes, R.C.A. and Do Carmo, R.J., Influence of ionic alterations produced by gastric washing on cortical spreading depression, Exp. Brain Res., 39 (1980) 341–349. [13] Heinemann, U. and Lux, H.D., Undershoots following stimulus-induced rises of extracellular potassium concen-

[15]

[17]

[18] [19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

tration in cerebral cortex of cat. Brain Res., 93 (1975) 63 – 76. Koroleva, V.I. and Bures, J., Circulation of cortical spreading depression around electrically stimulated areas and epileptic foci of the neocortex of rats, Brain Res., 173 (1979) 209 – 215. Koroleva, V.I. and Bures, J., Blockade of cortical spreading depression in electrically and chemically stimulated areas of cerebral cortex in rats, Electroenceph. Clin. Neurophysiol., 48 (1980) 1 – 15. Koroleva, V.I., Vinogradova, L.V. and Bures, J., Reduced incidence of cortical spreading depression in the course of pentylenetetrazol kindling in rats, Brain Res., 608 (1993) 107 – 114. Kraig, R.P. and Nicholson, C., Extracellular ionic variation during spreading depression, Neuroscience, 3 (1978) 1045 – 1059. Lea˜o, A.A.P., Spreading depression of activity in the cerebral cortex. J. Neurophysiol., 7 (1944a) 359 – 390. Lea˜o, A.A.P., Pial circulation and spreading depression of activity in the cerebral cortex. J. Neurophysiol., 7 (1944b) 391 – 396. Lea˜o, A.A.P., Further observations on the spreading depression of activity in the cerebral cortex. J. Neurophysiol., 10 (1947) 409 – 414. Lea˜o, A.A.P., On the spread of spreading depression. In: M.A.B. Brazier (Ed.), Brain Function, Vol. I, University of California Press, Berkeley, 1963, pp. 73 – 85. Leite, J.P., Bortolotto, Z.A. and Cavalheiro, E.A., Spontaneous recurrent seizures in rats: an experimental model of partial epilepsy, Neurosci. Biobeha6. Re6., 14 (1990) 511 – 517. Nicholson, C. and Kraig, R.P., Chloride and potassium changes measured during spreading depression in catfish cerebellum, Brain Res., 96 (1975) 384 – 389. Phillips, J.M. and Nicholson, C., Anion diflusion in spreading depression investigated with ion-sensitive microelectrodes, Brain Res., 173 (1979) 567 – 571. Siesjo, B.K. and Bengtsson, F., Calcium fluxes, calcium antagonists and calcium-related pathology in brain ischemia, hypoglycaemia and spreading depression: a unifying hypothesis, J. Cereb. Blood Flow Metab., 9 (1989) 127 – 140. Turski, W.A., Cavalheiro, E.A., Schwarz, M., Czuczwar, S.J., Kleinrok, Z. And Turski, L., Limbic seizures produced by pilocarpine in rats: a behavioural, electroencephalographic and neuropathological study, Beha6. Brain Res., 9 (1983) 315 – 336. Turski, W.A., Cavalheiro, E.A., Bortolotto, Z.A., Moraes Mello, L.E.A., Schwarz, M. and Turski, L., Seizures produced by pilocarpine in mice: a behavioural, electroencephalographic and morphological analysis, Brain Res., 321 (1984) 237 – 253. Van Harreveld, A. and Fifkova´, E., Glutamate release from the retina during spreading depression, J. Neurobiol., 2 (1970) 13 – 29. Walz, W. and MacVicar, B., Electrophysiological properties of glial cells: comparison of brain slices with primary cultures, Brain Res., 443 (1988) 321 – 324.