Neurophysiol Clin (1990) 20, 189- 201 © Elsevier, Paris
18 9
Original article
Nigral influence on focal epilepsy M Sabatino, G Ferraro, N Vella, V La Grutta* llstituto di Fisiologia umana dell'Universitd di Palermo, Corso Tukory, 129, 90134 Palermo, Italy
(Received 18 April 1989; accepted 21 April 1990)
Summary - The substantia nigra (SN) has been proposed as a structure involved in epileptiform phenomena.
Previous investigations demonstrated that SN is able to elicit hippocampal rhythmic slow activity (RSA) as well as to inhibit hippocampal interictal spikes induced by parenteral administration of penicillin. The present series of experiments was carried out in order to characterize the action of SN on a focal model of hippocampal epilepsy. Experiments were performed on enc~phale isol~ cats in which steady epileptiform activity was induced by locally applied penicillin. Electrical stimulation o f SN pars reticulata (pr) caused a statistically significant decrease of hippocampal spike frequency and amplitude in 30070 of the total number of stimulation sessions. Stimulation of SN pars compacta (pc) was even more effective. It induced inhibitory effects on hippocampal spikes in 91 °/0 of the cases. In 30°7o of the cats, RSA was noted on hippocampal recordings in correspondence to nigral activation. Experimental data support the hypothesis that the SNpc influences hippocampal excitability: a differential role may be played by SNpc and SNpr in the control of seizure processes. substantia nigra / hippocampus / epilepsy / inhibition / cat
R~sum~ - Le r61e de la substantia nigra dans I'~pilepsie hippocampale. La substantia nigra (SN) est une
structure probablement active dans les ph6nom~nes 6pilepfiformes. Des recherches ant6rieures ont d6montr6 que la SN peut d6clencher une activit6 rythmique lente (ARL) de l'hippocampe ainsi qu'inhiber les spikes interictaux de l'hippocampe provoqu6s par l'administration parent6rale de p6nicilline. Le protocole suivant a 6f6 6labor6 afin de caract6riser l'action exerc6e par la SN sur un mod61e d'6pilepsie hippocampale. Les exp6riences ont 6t6 conduites sur des chats h enc6phale isol6 chez lesquels on provoquait l'activit6 6pileptiforme par des applications locales de p6niciUine. La stimulation 61ectrique de la pars reticulata (pr) de la SN causait une r6duction statistiquement signifiante de la fr6quence et de l'amplitude des spikes hippocampaux dans 30°70 du nombre total des sessions de stimulation. La stimulation de la pars compacta (pc) de la SN 6tait encore plus efficace et provoquait des effets d'inhibition sur les spikes hippocam-
Abbreviations. - HS: hippocampal spikes; RSA rhythmic slow activity;SNpc: substantia nigra pars compacta; SNpr: substantia nigra pars reticulata. * Correspondence and reprints.
190
M Sabatino et al
paux dans 91% des cas. Chez 30% des chats, on observait une ARL en enregistrant l'activit6 hippoc~ pale correspondant h I'activation de la SN. Les donn6es exp6rimentales confirment l'hypoth6se qu pc de la SN exerce une influence sur l'excitabilit6 de I'hippocampe. La pc et la pr de la SN jouent prc blement un r61e diff6rentiel darts le contr61e des attaques 6pileptiques.
substantia nigra / hippocampe / ~pilepsie I inhibition I chat
Introduction
Many authors have pointed out the involvement of substantia nigra (SN) in the o trol of abnormal activity. In particular, pharmacological studies suggested that rood cation of nigral GABA activity can facilitate generalization of seizure in the (Iadarola and Gale, 1982; De Sarro et al, 1984). Other authors (Le Gal La Salh al, 1983) proposed that the SN has a role in a negative feedback system suppress paroxysmal activity initiated from the amygdala. Furthermore, both [C deoxyglucose autoradiographic investigations and electroencephalographic recc ings showed a prominent activation pattern of SN during chemically-induced lira seizures together with a direct paroxysmal coupling between SN and hippocam (Collins et al, 1983 ; Lothman et al, 1985). Recently it has been demonstrated t in the rat there is nigral modulation of pilocarpine-induced generalized seizures, ing 2-amino-7-phosphonoheptanoic acid (Turski et al, 1986), and a protection aga: electroshock and pilocarpine-induced seizures by nigral injection of -D-glutan aminomethylsulphonic acid (De Sarro et al, 1986). Some other electrophysiological investigations, studying inter-relationships bctw basal ganglia and hippocampus (La Grutta et al, 1971, 1985, 1986a, 1986b, 19 Le Gal La Salle et al, 1983), indicated that SN strongly inhibits abnormal hippoc~ pal activity in feline generalized penicillin epilepsy (FGPE) (Sabatino et al, 19~ 1986b, 1988). Due to parenteral administration of the convulsant drug, mar~y'bl sites may be activated and co-participate in th~ observed effect. For this reason, undertook a series of experiments in which nigral action was evaluated in a f( model of hippocampal epileptiform activity and its effectiveness was analyzed topographic mapping of the stimulation sites. The 2 functional compartments of are in fact the source of different projections, namely GABAergic and dopami~ gic. Since both GABA and dopamine are recognized as having anticonvulsant fects (Kilian and Frey, 1973 ; L6scher, 1985 ; Gale, 1986; L6scher and Czuczv 1986) we investigated whether a prevalence exists in the control of a focus in hippocampus. To this end, modifications in frequency and amplitude of hippoc. pal penicillin spikes during electrical stimulation of SN were studied and are rep ed here.
Epilepsy and substantia nigra
191
Methods Experiments were conducted on 16 adult cats, weighing 2.5-4.1 kg. TheY were initially anaesthetized with halothan~ (1-3% in Na0-0z mixture) and a vein and the trachea were cannulated. Then the animal was mounted in a DKI 1404 stereotaxic frame and the spinal cord was transected at C a (encdphale isol~). After that, volatile anaesthesia was discontinued and artificial ventilation was performed. Wound areas and pressure points were infiltrated ~with- 1% xylocaine and.,, where possible, xylocaine-soaked sponges were enclosed in the wounds. Rectal temperature was maintained between 37.5-38.5°C by a heating pad. The expired CO 2 was monitored and kept constant at 3.5-4.0%. Throughout the experiment the general state of the cat was monitored by'~ontrolling the heart rate, pupil size and the~:electrocorticogram (ECoG). Bioelectrical activity'of the dorsal hippocampus was bilaterally recorded by means of bipolar stainless-steel,coaxial electrodes (having an external diameter of 0.5 mm and tip 25-50 ~tm) inserted in or near the CA 1 pyramidal layer according to the atlas of Snider and Niemer (A4.0-5.5 ; L3:5"-5.0; H18-19) (1961). Amplification and record -ing of electrical activity was obtained through an 8-channel polygraph (GRASS, mod. 7B). Other simultaneous recordings from cortical (post~sygmoidal gyri and anterior supra-sylvian gyri) and subcortical regions (globus pallidus, caudate nucleus) allowed monitoring of epileptic focalization and eventual diffusion phenomena or seizure generalization. Coaxial bipolar stimulating electrodes were bilaterally lowered into'either SN pars. compacta (pc) or SN pars reticulata (pr) according to pre-determined antero-posterior and mediolateral plannihg. In each animal a variable number of stimulation points was chosen as a function of the effect obtained and of the duration of PCN focus. In any case, the points were equally distributed in SNpc:and SNpr. The parameters of stimulation employed were: 0.5-1.0 ms; 5-30 Hz frequency and 0.1-0.5 mA intensity. Train duration ranged between 5 and 30 ~ s. In each animal the various parameters of stimulation were slowly increased until they reached a significant effect. Epileptiform activity was induced by unilateral intrahippocampal application of penicillin by means of a stereotaxically, placed Hamilton microsyringe (250-1000 ~IU,of peaicillin.in 1-5 tzl solution at neutralrpH) (fig' 1). In all the animals stereotaxic microinjecti0ns were constantly performed at 30, 45, and 60 min after the beginning of the recording session in order to elicit steady focal epileptiform activity. Hippocampal bioelectrical activity was distinguished as: i) rhythmic slow activity (RSA, theta) consisting of trains of roughly sinusoidal waves with a frequency r a n g - i n g J r o m 4 to 7 c/s ; ii) large amplitude irregular activity (IA), lacking the rhythmic character of RSA; iii) penicillin-induced epilepfiform complexes, that is to say, sharp waves, spikes and spike and wave complexes (collectivelly called hippocampal spikes, HS). The time course of HS frequency was analyzed in 6 cats, in which no stimulation was delivered. In the other cats (n = 10) HS frequency per, minlandlHS~amptitudein/~V (negative spike amp~ifude)were c,o~sidered in'~the 1 min epoch preceding stimulation, during the period of stimulation and in the 1 min f011owing stimulation. A, peri0d'ofat least 2 min divided the stimulation sessions. Student's t test for paired'data was employed for comparison of the mean values of HS frequency and amplitude before, during and after SN stimulation. At the end of the experiments fixed brains were removed and frozen ; the electrode placements were identified by Prussian bleu deposits in serial 30/~m coronal section obtained with Nissl thionine and/or Nissl cresyl violet methods (fig 1).
+5 +4
0
P~M
-2
SNpr//
,-f--~--
-6 I
I
I
10
I
l,
8
I
I
6
,I,
a
q
'
I
2
.
I
0
,
2
I
,
I
tl
.
=
6
.
8
I
10
B +6 +4 +2_20
'
~ C
-4 -6
SNDr
~.Z.'~'
" PP
I~CB
~'~,
-10 -,12
]
I
z
I
u
I
J
I
i
I
.
I
,
|
•
I
~
|
•
I
J
n
a
I
A16 Alq A12 AIO A8 A6 Aq A2 0 P2 pll P6 Fig l. Graphic reconstruction of Hamilton microsyringe position within hippocampus and recording el trodes in the contralateral mirror focus. A: Graphic reconstruction of histological sections obtained fr A5.1 frontal plane. B : Diagram obtained from L5.5 sagittal plane. Hatched areas show the regions wh traces of microsyringe and recording electrodes were histologically identified in the different anim~ Numerals reported in abscissae and ordinates indicate distances in millimeters a m o n g different plar CB : cerebellar cortex; F : fornix ; H I P : hippocampus ; ICC : inferior colliculus central nucleus ; ICP : ferior colliculus pericentral nucleus ; LV : lateral ventricle; M G M : medial geniculate magnocellular nucle M G P : medial geniculate principal nucleus ; PP : pes pedunculi ; P U : pulvinar ; SNpc : substantia ni pars compacta; SNpr: substantia nigra pars reticulata.
Epilepsy and substantia nigra
193
Results ECoG features in the encdphale isol~ preparation have been previously described. Briefly, hippocampal bioelectrical activity consisted of large amplitude IA with intermittent trains of RSA. About 3--4 min after 100-300 IU/kg (average 181.12 ___ 98.96 IU/kg) applications of penicillin, hippocampal sharp waves, high amplitude monophasic negative or diphasic spikes appeared bilaterally in the recording, showing a gradual increase in frequency (fig 2A). Spikes attained their peak frequency at about 30 min, stabilized for about 90 min and then disappeared, the evoked penicillin focus lasting approximately 150 min (fig 2B). In a small proportion of cases (10o70), maximal discharges occurred with diffusion to other structures and data collection was consequently stopped. Occasionally (15070 of the cases) typical spike and wave complexes (fulfilling all the criteria as defined by Weir (1965)) and poly-spikes were recorded. When fully developed, HS presented a periodical distribution ranging between 16 and 22 every 1 min (average 19.33 + 2.42) and their amplitude varied from 200 to 550 ~zV (average 371.85 + 115.99/~V; n -- 6 cats). In the cats submitted to electrical stimulation of SN (n = 10 cats), either ipsilaterally or contralaterally to the penicillin focus, a marked decrease of both HS frequency and amplitude was noted. The effect was an all-or-none phenomenon with a specific array of stimulation parameters for each animal. Stimulation of SNpc was more effective than stimulation of SNpr. In particular, SNpc induced inhibition in 39 out of 43 stimulated points (90.69°7o) and had no effect on 4 points of the 43 stimulated (9.3107o). SNpr was responsible for inhibitory effects in 13 of the 44 stimulated points (29.55°70) and had no effect on 31 of the 44 stimulated (70.45o70) (fig 3). The stimulated points giving inhibitory effects were equally distributed in planes A2, A3, A4 and A5. Due to the few inhibitory points within SNpr, data of SNpr stimulation resulted to be statistically not significant. On the contrary, in the cats having a low spike frequency (n = 3), SNpc activation caused total disappearance of spikes and a clear hippocampal RSA (fig 4). In the other cats (n = 7), spikes became smaller and less frequent (fig 5A). In the 10 animals considered, we calculated the individual variability coefficient (a/M X 100) for the average values of HS frequency per min and amplitude in the 1 min epoch preceding SN stimulation and during SN stimulation. The analysis of population of values whose variability was within acceptable limits (n = from 14 to 33 in each animal) showed that during nigral stimulation (20-30 Hz), HS frequency decreased from 26.28 spike per min to 10.63 spike per min (-59.5507o) (fig 5B). The significance was tested by the " F " o f Snedecor (higher variability2/lower variability 2) before applying Student's t test. The difference is highly significant at level P < 0.001 for all cats (n = 10) explored. During the 1 min epoch following SNpc stimulation, frequency almost returned to the base value ( - 1.21 o70, with a significance of P < 0 . 0 1 in 2 cats and not significant in the other 8). Furthermore, SNpc stimulation induced a reduction in HS amplitude from 372.43 ~zV to 302.37 ~zV (-18.81°70); this variation being statistically significant at the level of P < 0 . 0 5 in 1 cat and highly significant ( P < 0.001) in the other 9 (fig 5C). In the
194
M Sabatino et al
1 m i n epoch following s t i m u l a t i o n , H S a m p l i t u d e showed control-like values, th is - 3 . 5 3 ° / 0 , a difference which was statistically highly significant ( P < 0.001) in cats a n d n o t significant in the other 8. It should be emphasized that either SN-induc, theta (fig 4) or the rare R S A which occurred s p o n t a n e o u s l y caused i n h i b i t i o n o f hi p o c a m p a l e p i l e p t i f o r m activity.
PCN
I
l
I
,
l
l
l
~
I I
~
"
11
25 Z
>r.5 z w
2O PCN
0 15 w o~ LL W
!
2
1
lO
j,z
(3_ CO
0
500
2 sec
m I
0
e
• i
I
60
l
I
120
i
I
180
i
I
240
MIN
Fig 2. Penicillin effect on hippocampal spike frequency. A : Enc~phale isol~ cat with local anaesth of painful points. Topical application of penicillin solution on one side caused the appearance of int* tal activity in the contralateral mirror focus. B: Time course of hippocampal spike frequency, n cats. Abscissa : duration of experimental observation. Ordinate : hippocampal spike frequency per (average _+ SD). PCN 1, 2 and 3: intrahippocampal penicillin injections at constant time.
7-
A5
6 5
6 576 5 7
A2
6 5
1
4I
I
I
I
I
I
2
3
4
5
6
7
Fig 3. Reconstruction from histological sections of the stimulated points within substantia nigra in the coronal A2, A3, A4 and A5 planes. Black circles : stimulation points causing inhibitory effects on hippocampal interictal activity. Hollow circles : points in which stimulation failed to induce effects on hippocampar i~a~.erictal activity. It may be noted that inhibitory points are mostly distributed in SNpc; on the contrary ineffective points are predominantly distributed in SNpr. A hypothetical division is traced by a dotted line.
M Sabatiiao et al
196 Discussion
Involvement o f nigral-induced R S A in the antiepileptic effect
Topical penicillin administration allows high concentration of the convulsant dru to reach the hippocampus. Consequently antagonism to inhibitory mechanisms w thin the hippocampal formation is enhanced. Although the penicillin-GABA al tagonism is not the sole mechanism of penicillin-induced hyperactivity (Andersc et al, 1977 ; Davenport et al, 1978), the convulsant drug is capable of altering GAB? mediated inhibition (Hochner et al, 1976; Mac Donald and Barker, 1978; Ribac et al, 1978; Dingledine and Gjerstadt, 1979; Wong and Prince, 1979; Avoli e t a 1982). Thus, since hippocampal RSA depends, at least in part, upon inhibitor GABAergic interneurons (theta cells) (Andersen et al, 1964; Mori et al, 1987), ot present observation that nigral stimulation induced a hippocampal RSA in only 30 ~ of cases whereas in FGPE condition RSA was linked to nigral stimulation in almo 100% of the cases (La Grutta et al, 1986c) confirms theoretical expectations. A neg'. tive interaction between the rare spontaneous RSA and HS was also noted, in agre ment with recent findings of hippocampal sharp wave suppression by RSA-associate behaviours (Buzs~iki, 1986). Inhibitory phenomena were of a lesser degree with respe
PCN
QOQOOQOQOO0 ~ , SNpc 20 Hz : . . . . . . .
] 500 p 5 sec
i
Fig 4. Effect of substantia nigra pars compacta (SNpc) stimulation on hippocampal epileptiform activi: Encdphale isold cat with local anaesthesia of painful points. Moderate interictal activity of the hippoca: pus recorded about 60 min after intrahippocampal injection of penicillin ; in the record 2 follows 1. ] silateral SNpc stimulation (A3, L5.5; H6; l ms, 20 Hz 0.4 rnA, 10 s train duration) induced a regu theta rhythm and completely suppressed spike activity. CGS : central grey substance ; IN : interpedun~ lar nucleus; RN: red nucleus; SC: superior colliculus; SN: substantia nigra.
Epilepsy and substantia nigra
197
A
~'e''''S~eC';Ol~z'''''" •
500 pV 2 sec
o o o e o
>C5 zlO0 uJ (%)
B m
CI LtJ
n U
Lt.I C3
I
100
B
I---
I
(%) 13_ ES
18 9
Original article
Nigral influence on focal epilepsy M Sabatino, G Ferraro, N Vella, V La Grutta* llstituto di Fisiologia umana dell'Universitd di Palermo, Corso Tukory, 129, 90134 Palermo, Italy
(Received 18 April 1989; accepted 21 April 1990)
Summary - The substantia nigra (SN) has been proposed as a structure involved in epileptiform phenomena.
Previous investigations demonstrated that SN is able to elicit hippocampal rhythmic slow activity (RSA) as well as to inhibit hippocampal interictal spikes induced by parenteral administration of penicillin. The present series of experiments was carried out in order to characterize the action of SN on a focal model of hippocampal epilepsy. Experiments were performed on enc~phale isol~ cats in which steady epileptiform activity was induced by locally applied penicillin. Electrical stimulation o f SN pars reticulata (pr) caused a statistically significant decrease of hippocampal spike frequency and amplitude in 30070 of the total number of stimulation sessions. Stimulation of SN pars compacta (pc) was even more effective. It induced inhibitory effects on hippocampal spikes in 91 °/0 of the cases. In 30°7o of the cats, RSA was noted on hippocampal recordings in correspondence to nigral activation. Experimental data support the hypothesis that the SNpc influences hippocampal excitability: a differential role may be played by SNpc and SNpr in the control of seizure processes. substantia nigra / hippocampus / epilepsy / inhibition / cat
R~sum~ - Le r61e de la substantia nigra dans I'~pilepsie hippocampale. La substantia nigra (SN) est une
structure probablement active dans les ph6nom~nes 6pilepfiformes. Des recherches ant6rieures ont d6montr6 que la SN peut d6clencher une activit6 rythmique lente (ARL) de l'hippocampe ainsi qu'inhiber les spikes interictaux de l'hippocampe provoqu6s par l'administration parent6rale de p6nicilline. Le protocole suivant a 6f6 6labor6 afin de caract6riser l'action exerc6e par la SN sur un mod61e d'6pilepsie hippocampale. Les exp6riences ont 6t6 conduites sur des chats h enc6phale isol6 chez lesquels on provoquait l'activit6 6pileptiforme par des applications locales de p6niciUine. La stimulation 61ectrique de la pars reticulata (pr) de la SN causait une r6duction statistiquement signifiante de la fr6quence et de l'amplitude des spikes hippocampaux dans 30°70 du nombre total des sessions de stimulation. La stimulation de la pars compacta (pc) de la SN 6tait encore plus efficace et provoquait des effets d'inhibition sur les spikes hippocam-
Abbreviations. - HS: hippocampal spikes; RSA rhythmic slow activity;SNpc: substantia nigra pars compacta; SNpr: substantia nigra pars reticulata. * Correspondence and reprints.
190
M Sabatino et al
paux dans 91% des cas. Chez 30% des chats, on observait une ARL en enregistrant l'activit6 hippoc~ pale correspondant h I'activation de la SN. Les donn6es exp6rimentales confirment l'hypoth6se qu pc de la SN exerce une influence sur l'excitabilit6 de I'hippocampe. La pc et la pr de la SN jouent prc blement un r61e diff6rentiel darts le contr61e des attaques 6pileptiques.
substantia nigra / hippocampe / ~pilepsie I inhibition I chat
Introduction
Many authors have pointed out the involvement of substantia nigra (SN) in the o trol of abnormal activity. In particular, pharmacological studies suggested that rood cation of nigral GABA activity can facilitate generalization of seizure in the (Iadarola and Gale, 1982; De Sarro et al, 1984). Other authors (Le Gal La Salh al, 1983) proposed that the SN has a role in a negative feedback system suppress paroxysmal activity initiated from the amygdala. Furthermore, both [C deoxyglucose autoradiographic investigations and electroencephalographic recc ings showed a prominent activation pattern of SN during chemically-induced lira seizures together with a direct paroxysmal coupling between SN and hippocam (Collins et al, 1983 ; Lothman et al, 1985). Recently it has been demonstrated t in the rat there is nigral modulation of pilocarpine-induced generalized seizures, ing 2-amino-7-phosphonoheptanoic acid (Turski et al, 1986), and a protection aga: electroshock and pilocarpine-induced seizures by nigral injection of -D-glutan aminomethylsulphonic acid (De Sarro et al, 1986). Some other electrophysiological investigations, studying inter-relationships bctw basal ganglia and hippocampus (La Grutta et al, 1971, 1985, 1986a, 1986b, 19 Le Gal La Salle et al, 1983), indicated that SN strongly inhibits abnormal hippoc~ pal activity in feline generalized penicillin epilepsy (FGPE) (Sabatino et al, 19~ 1986b, 1988). Due to parenteral administration of the convulsant drug, mar~y'bl sites may be activated and co-participate in th~ observed effect. For this reason, undertook a series of experiments in which nigral action was evaluated in a f( model of hippocampal epileptiform activity and its effectiveness was analyzed topographic mapping of the stimulation sites. The 2 functional compartments of are in fact the source of different projections, namely GABAergic and dopami~ gic. Since both GABA and dopamine are recognized as having anticonvulsant fects (Kilian and Frey, 1973 ; L6scher, 1985 ; Gale, 1986; L6scher and Czuczv 1986) we investigated whether a prevalence exists in the control of a focus in hippocampus. To this end, modifications in frequency and amplitude of hippoc. pal penicillin spikes during electrical stimulation of SN were studied and are rep ed here.
Epilepsy and substantia nigra
191
Methods Experiments were conducted on 16 adult cats, weighing 2.5-4.1 kg. TheY were initially anaesthetized with halothan~ (1-3% in Na0-0z mixture) and a vein and the trachea were cannulated. Then the animal was mounted in a DKI 1404 stereotaxic frame and the spinal cord was transected at C a (encdphale isol~). After that, volatile anaesthesia was discontinued and artificial ventilation was performed. Wound areas and pressure points were infiltrated ~with- 1% xylocaine and.,, where possible, xylocaine-soaked sponges were enclosed in the wounds. Rectal temperature was maintained between 37.5-38.5°C by a heating pad. The expired CO 2 was monitored and kept constant at 3.5-4.0%. Throughout the experiment the general state of the cat was monitored by'~ontrolling the heart rate, pupil size and the~:electrocorticogram (ECoG). Bioelectrical activity'of the dorsal hippocampus was bilaterally recorded by means of bipolar stainless-steel,coaxial electrodes (having an external diameter of 0.5 mm and tip 25-50 ~tm) inserted in or near the CA 1 pyramidal layer according to the atlas of Snider and Niemer (A4.0-5.5 ; L3:5"-5.0; H18-19) (1961). Amplification and record -ing of electrical activity was obtained through an 8-channel polygraph (GRASS, mod. 7B). Other simultaneous recordings from cortical (post~sygmoidal gyri and anterior supra-sylvian gyri) and subcortical regions (globus pallidus, caudate nucleus) allowed monitoring of epileptic focalization and eventual diffusion phenomena or seizure generalization. Coaxial bipolar stimulating electrodes were bilaterally lowered into'either SN pars. compacta (pc) or SN pars reticulata (pr) according to pre-determined antero-posterior and mediolateral plannihg. In each animal a variable number of stimulation points was chosen as a function of the effect obtained and of the duration of PCN focus. In any case, the points were equally distributed in SNpc:and SNpr. The parameters of stimulation employed were: 0.5-1.0 ms; 5-30 Hz frequency and 0.1-0.5 mA intensity. Train duration ranged between 5 and 30 ~ s. In each animal the various parameters of stimulation were slowly increased until they reached a significant effect. Epileptiform activity was induced by unilateral intrahippocampal application of penicillin by means of a stereotaxically, placed Hamilton microsyringe (250-1000 ~IU,of peaicillin.in 1-5 tzl solution at neutralrpH) (fig' 1). In all the animals stereotaxic microinjecti0ns were constantly performed at 30, 45, and 60 min after the beginning of the recording session in order to elicit steady focal epileptiform activity. Hippocampal bioelectrical activity was distinguished as: i) rhythmic slow activity (RSA, theta) consisting of trains of roughly sinusoidal waves with a frequency r a n g - i n g J r o m 4 to 7 c/s ; ii) large amplitude irregular activity (IA), lacking the rhythmic character of RSA; iii) penicillin-induced epilepfiform complexes, that is to say, sharp waves, spikes and spike and wave complexes (collectivelly called hippocampal spikes, HS). The time course of HS frequency was analyzed in 6 cats, in which no stimulation was delivered. In the other cats (n = 10) HS frequency per, minlandlHS~amptitudein/~V (negative spike amp~ifude)were c,o~sidered in'~the 1 min epoch preceding stimulation, during the period of stimulation and in the 1 min f011owing stimulation. A, peri0d'ofat least 2 min divided the stimulation sessions. Student's t test for paired'data was employed for comparison of the mean values of HS frequency and amplitude before, during and after SN stimulation. At the end of the experiments fixed brains were removed and frozen ; the electrode placements were identified by Prussian bleu deposits in serial 30/~m coronal section obtained with Nissl thionine and/or Nissl cresyl violet methods (fig 1).
+5 +4
0
P~M
-2
SNpr//
,-f--~--
-6 I
I
I
10
I
l,
8
I
I
6
,I,
a
q
'
I
2
.
I
0
,
2
I
,
I
tl
.
=
6
.
8
I
10
B +6 +4 +2_20
'
~ C
-4 -6
SNDr
~.Z.'~'
" PP
I~CB
~'~,
-10 -,12
]
I
z
I
u
I
J
I
i
I
.
I
,
|
•
I
~
|
•
I
J
n
a
I
A16 Alq A12 AIO A8 A6 Aq A2 0 P2 pll P6 Fig l. Graphic reconstruction of Hamilton microsyringe position within hippocampus and recording el trodes in the contralateral mirror focus. A: Graphic reconstruction of histological sections obtained fr A5.1 frontal plane. B : Diagram obtained from L5.5 sagittal plane. Hatched areas show the regions wh traces of microsyringe and recording electrodes were histologically identified in the different anim~ Numerals reported in abscissae and ordinates indicate distances in millimeters a m o n g different plar CB : cerebellar cortex; F : fornix ; H I P : hippocampus ; ICC : inferior colliculus central nucleus ; ICP : ferior colliculus pericentral nucleus ; LV : lateral ventricle; M G M : medial geniculate magnocellular nucle M G P : medial geniculate principal nucleus ; PP : pes pedunculi ; P U : pulvinar ; SNpc : substantia ni pars compacta; SNpr: substantia nigra pars reticulata.
Epilepsy and substantia nigra
193
Results ECoG features in the encdphale isol~ preparation have been previously described. Briefly, hippocampal bioelectrical activity consisted of large amplitude IA with intermittent trains of RSA. About 3--4 min after 100-300 IU/kg (average 181.12 ___ 98.96 IU/kg) applications of penicillin, hippocampal sharp waves, high amplitude monophasic negative or diphasic spikes appeared bilaterally in the recording, showing a gradual increase in frequency (fig 2A). Spikes attained their peak frequency at about 30 min, stabilized for about 90 min and then disappeared, the evoked penicillin focus lasting approximately 150 min (fig 2B). In a small proportion of cases (10o70), maximal discharges occurred with diffusion to other structures and data collection was consequently stopped. Occasionally (15070 of the cases) typical spike and wave complexes (fulfilling all the criteria as defined by Weir (1965)) and poly-spikes were recorded. When fully developed, HS presented a periodical distribution ranging between 16 and 22 every 1 min (average 19.33 + 2.42) and their amplitude varied from 200 to 550 ~zV (average 371.85 + 115.99/~V; n -- 6 cats). In the cats submitted to electrical stimulation of SN (n = 10 cats), either ipsilaterally or contralaterally to the penicillin focus, a marked decrease of both HS frequency and amplitude was noted. The effect was an all-or-none phenomenon with a specific array of stimulation parameters for each animal. Stimulation of SNpc was more effective than stimulation of SNpr. In particular, SNpc induced inhibition in 39 out of 43 stimulated points (90.69°7o) and had no effect on 4 points of the 43 stimulated (9.3107o). SNpr was responsible for inhibitory effects in 13 of the 44 stimulated points (29.55°70) and had no effect on 31 of the 44 stimulated (70.45o70) (fig 3). The stimulated points giving inhibitory effects were equally distributed in planes A2, A3, A4 and A5. Due to the few inhibitory points within SNpr, data of SNpr stimulation resulted to be statistically not significant. On the contrary, in the cats having a low spike frequency (n = 3), SNpc activation caused total disappearance of spikes and a clear hippocampal RSA (fig 4). In the other cats (n = 7), spikes became smaller and less frequent (fig 5A). In the 10 animals considered, we calculated the individual variability coefficient (a/M X 100) for the average values of HS frequency per min and amplitude in the 1 min epoch preceding SN stimulation and during SN stimulation. The analysis of population of values whose variability was within acceptable limits (n = from 14 to 33 in each animal) showed that during nigral stimulation (20-30 Hz), HS frequency decreased from 26.28 spike per min to 10.63 spike per min (-59.5507o) (fig 5B). The significance was tested by the " F " o f Snedecor (higher variability2/lower variability 2) before applying Student's t test. The difference is highly significant at level P < 0.001 for all cats (n = 10) explored. During the 1 min epoch following SNpc stimulation, frequency almost returned to the base value ( - 1.21 o70, with a significance of P < 0 . 0 1 in 2 cats and not significant in the other 8). Furthermore, SNpc stimulation induced a reduction in HS amplitude from 372.43 ~zV to 302.37 ~zV (-18.81°70); this variation being statistically significant at the level of P < 0 . 0 5 in 1 cat and highly significant ( P < 0.001) in the other 9 (fig 5C). In the
194
M Sabatino et al
1 m i n epoch following s t i m u l a t i o n , H S a m p l i t u d e showed control-like values, th is - 3 . 5 3 ° / 0 , a difference which was statistically highly significant ( P < 0.001) in cats a n d n o t significant in the other 8. It should be emphasized that either SN-induc, theta (fig 4) or the rare R S A which occurred s p o n t a n e o u s l y caused i n h i b i t i o n o f hi p o c a m p a l e p i l e p t i f o r m activity.
PCN
I
l
I
,
l
l
l
~
I I
~
"
11
25 Z
>r.5 z w
2O PCN
0 15 w o~ LL W
!
2
1
lO
j,z
(3_ CO
0
500
2 sec
m I
0
e
• i
I
60
l
I
120
i
I
180
i
I
240
MIN
Fig 2. Penicillin effect on hippocampal spike frequency. A : Enc~phale isol~ cat with local anaesth of painful points. Topical application of penicillin solution on one side caused the appearance of int* tal activity in the contralateral mirror focus. B: Time course of hippocampal spike frequency, n cats. Abscissa : duration of experimental observation. Ordinate : hippocampal spike frequency per (average _+ SD). PCN 1, 2 and 3: intrahippocampal penicillin injections at constant time.
7-
A5
6 5
6 576 5 7
A2
6 5
1
4I
I
I
I
I
I
2
3
4
5
6
7
Fig 3. Reconstruction from histological sections of the stimulated points within substantia nigra in the coronal A2, A3, A4 and A5 planes. Black circles : stimulation points causing inhibitory effects on hippocampal interictal activity. Hollow circles : points in which stimulation failed to induce effects on hippocampar i~a~.erictal activity. It may be noted that inhibitory points are mostly distributed in SNpc; on the contrary ineffective points are predominantly distributed in SNpr. A hypothetical division is traced by a dotted line.
M Sabatiiao et al
196 Discussion
Involvement o f nigral-induced R S A in the antiepileptic effect
Topical penicillin administration allows high concentration of the convulsant dru to reach the hippocampus. Consequently antagonism to inhibitory mechanisms w thin the hippocampal formation is enhanced. Although the penicillin-GABA al tagonism is not the sole mechanism of penicillin-induced hyperactivity (Andersc et al, 1977 ; Davenport et al, 1978), the convulsant drug is capable of altering GAB? mediated inhibition (Hochner et al, 1976; Mac Donald and Barker, 1978; Ribac et al, 1978; Dingledine and Gjerstadt, 1979; Wong and Prince, 1979; Avoli e t a 1982). Thus, since hippocampal RSA depends, at least in part, upon inhibitor GABAergic interneurons (theta cells) (Andersen et al, 1964; Mori et al, 1987), ot present observation that nigral stimulation induced a hippocampal RSA in only 30 ~ of cases whereas in FGPE condition RSA was linked to nigral stimulation in almo 100% of the cases (La Grutta et al, 1986c) confirms theoretical expectations. A neg'. tive interaction between the rare spontaneous RSA and HS was also noted, in agre ment with recent findings of hippocampal sharp wave suppression by RSA-associate behaviours (Buzs~iki, 1986). Inhibitory phenomena were of a lesser degree with respe
PCN
QOQOOQOQOO0 ~ , SNpc 20 Hz : . . . . . . .
] 500 p 5 sec
i
Fig 4. Effect of substantia nigra pars compacta (SNpc) stimulation on hippocampal epileptiform activi: Encdphale isold cat with local anaesthesia of painful points. Moderate interictal activity of the hippoca: pus recorded about 60 min after intrahippocampal injection of penicillin ; in the record 2 follows 1. ] silateral SNpc stimulation (A3, L5.5; H6; l ms, 20 Hz 0.4 rnA, 10 s train duration) induced a regu theta rhythm and completely suppressed spike activity. CGS : central grey substance ; IN : interpedun~ lar nucleus; RN: red nucleus; SC: superior colliculus; SN: substantia nigra.
Epilepsy and substantia nigra
197
A
~'e''''S~eC';Ol~z'''''" •
500 pV 2 sec
o o o e o
>C5 zlO0 uJ (%)
B m
CI LtJ
n U
Lt.I C3
I
100
B
I---
I
(%) 13_ ES
