Epilepsy Res., 9 (1991) 97-104 Elsevier
97
EPIRES 00415
Are rats with genetic absence epilepsy behaviorally impaired?
Marguerite Vergnes, Christian Marescauxa, Any Boehrer and Antoine Depaulis Centre de Neurochimie du CNRS, 67000 Strasbourg (France) and a CHU, Clinique Neurologique Strasbourg (France)
(Received 15 February 1991; revision received 1 May 1991; accepted 6 May 1991) Key words: Rat; Absence epilepsy; Behavior; Learning; EEG
Absence seizures in humans are characterized by unresponsiveness to external stimuli and inactivity. However, in typical generalized non-convulsiveepilepsyin children, intellectual capacities are considered to be normal. Wistar rats from an inbred strain with spontaneous absence-like seizures were compared with rats from the outbred control strain in various behavioral tasks in order to detect possible impairments related either to the absence epilepsyor to occurrence of spike and wave discharges (SWD). Spontaneous circadian locomotion, exploratory activity in an open field, social interactions with an unfamiliar conspecific and mouse killing behavior were similar in both strains. Avoidance learning in a shuttle box or food reinforced learning in a Skinner test were unimpaired or even improved in epileptic rats. During performance of a learned task either in the Skinner box or in a conditioned sound-bar pressing task, SWD were suppressed in epileptic rats as long as they were working for reinforcement. SWD reappeared when the motivation to perform the task had declined: unresponsiveness to a conditioned stimulus was then observed during SWD. These data are in agreement with observations commonly described in children with typical genetic absence epilepsy.
INTRODUCTION Absence epilepsy (or petit mal) is a generalized n o n convulsive epilepsy characterized by the spontaneous occurrence of bilateral, synchronous spike and wave discharges (SWD) on the cortical E E G with concomitant unawareness, unresponsiveness a n d cessation of activities. Classically, in children with pure typical absence epilepsy, despite behavioral impairments which occur during SWD, intelligence is considered to be n o r m a l or even superior in some cases 1,8. A strain of Wistar rats with spontaneous seizures typical of absence epilepsy has been inbred in our laboratory (GAERS = genetic absence epilepsy rats from Strasbourg). I n these animals seizures are characterized by SWD which occur o n a b a c k g r o u n d of quiet wakefulness with a mean frequency of 1 / m i n
Correspondence to: Marguerite Vergnes, LNBC, Centre de Neurochimie du CNRS, 5 rue Blaise Pascal, 67084 Strasbourg Cedex, France.
and durations varying from 1 to 60 s with a mean a r o u n d 20 s. The frequency of the bilateral and synchronous spike and waves varies from 7 to 9 Hz 23. The SWD involve all the neocortex and the m a i n relay nuclei of the thalamus, whereas no SWD are recorded in the limbic structures 24. During the SWD the animals are immobile with reduction of neck muscle tone and occasional twitching of the vibrissae and the face. The pharmacological reactivity of absence seizures has been shown to be similar in this rat strain and h u m a n s 1°,11. A control strain devoid of seizures was outbred from the same Wistar colony25. The 2 strains do not apparently differ in size, growth, reproductive capacity or any morphological characteristics. The behavior of an adult is a dynamic construct resulting from the interaction of i n b o r n capacities and experience. The genetic abnormality which produces the development of a considerable n u m b e r of absence seizures during all the adult life in rats 25 could also directly or indirectly cause some behavioral impair-
0920-1211/91/$03.50 © 1991 Elsevier Science Publishers B.V. All rights reserved
98 ment due to dysfunction of some brain areas. Alternatively, the repeated occurrence of SWD concomitant with loss of attention and responsiveness may provoke reduced efficiency and success in various behaviors, especially in social activities or learning tasks. An analysis of behavioral parameters was undertaken to reveal possible differences in the abilities of epileptic GAERS and non-epileptic controis in various spontaneous and learned behaviors, namely in spontaneous and evoked locomotor activity, intra- and interspecific social interactions, and ability to learn and perform various tasks. Moreover, the EEG was recorded during performance of learned tasks in order to detect impairments related to the incidence of SWD. METHODS
Animals Adult male Wistar rats ( 3 0 0 - 400 g) from 2 strains selected over 15 generations either for spontaneous occurrence of absence-like seizures (GAERS) or for lack of such seizures (controls) were used. All adult GAERS have prolonged SWD, whereas SWD are never recorded in any rat from the control strain. The animals were age-matched in each experiment. Ad lib fed animals were placed in individual cages and maintained on a 12-h dark-light regimen. In the experiments using EEG control the animals were equipped under pentobarbital anesthesia (40 mg/kg i.p.) with 4 stainless steel electrodes screwed into the skull over the fronto-parietal cortex and connected to a microconnector embedded in dental acrylic cement. The EEG was recorded between 2 ipsilateral electrodes by an electroencephalograph (Alvar).
Behavioral tests Circadian locomotion was measured in the home cage (30 × 24 × 35 cm) placed in front of 2 photocells recording the interruption of an infra-red light beam. Food was available. Measures started during the usual light period, and locomotor activity was recorded for 24 h. A circular open field (diameter 1 m) lit with a 60-W bulb above the field was used to measure activity in an unfamiliar environment. The floor of the open field was divided into 12 parts and the number of parts crossed during a 6-min period was quantified as
locomotor activity. In addition the number of rearings on the hind feet and the number of defecations were counted during this period. The open field was carefully cleaned after each measure. Interspecific aggression was evaluated by introducing a mouse into the rat's familiar home cage. Rats were naive with respect to mice. Most killer rats attack and kill rapidly, but some rats kill only overnight. Therefore the mouse was left in the rat's cage for 24 h. Social interactions with a non-familiar weightmatched male Conspecific were recorded in a single 8min session on video tape. The experimental rat was either in a situation of resident, the partner being introduced into its home cage, or in a situation of intruder in the partner's home cage. The different behavioral items were encoded from the video recordings using a microcomputer-based technique 2. The data were processed for each animal and each item in terms of cumulated duration per session (for details see Vergnes et al.22). The following behavioral items were coded: (1) non-social activities: cage exploration, selfgrooming, immobile posture; (2) social activities: partner investigation, allogrooming; (3) offense: attack, offensive sideways, offensive upright; (4) defense: defensive sideways, defensive upright. Avoidance learning was performed in a 2-way shuttle box of 35 × 16 × 20 cm divided into 2 equal compartments by a partition with an opening allowing the rat to cross to the opposite side. Electric foot shocks (0.5 mA) were delivered through the floor grid of the shuttle box. A 9-s sound announced the scrambled shocks associated with the sound for a further 11 s, followed by a 10-s rest. Ten shocks per session were alternately delivered in the 2 compartments when crossing between the compartments had occurred. Four groups of 3 sessions were performed in 2 days. For each session of 10 sound-shock associations, the total number of shocks received, the latency to cross to the opposite compartment, and the number of such crossings were recorded. Animals learned either to avoid the shock by stepping to the safe compartment during the 9-s sound exposure, or to escape the shock during the 11 s of delivery of shocks. Some animals never learned to move to the opposite side in order to escape the shock. Avoidance was considered to be learned when the animals received fewer than 2 shocks per session. Escape was learned when the animals received the
99 shock but crossed within the duration of the shock in more than 8 trials. Instrumental learning with a fixed ratio reinforcement was performed in a Skinner box (28 x 24 x 30 cm). The animals maintained at 85% of their body weight were trained to press a bar to obtain a 45-mg food pellet. A program of reinforcement was scheduled: successively, fixed reinforcement 1 (FR1, each bar press delivers 1 food pellet) until 25 pellets are delivered; FR2 (2 bar presses are necessary to deliver 1 pellet) for 25 pellets, and FR5 to the criterion of 50 pellets. For each reinforcement ratio the time necessary to reach the criterion was measured. Once trained in the Skinner box to obtain food on a FR5 schedule the EEG was recorded in epileptic rats during 3 consecutive 20-min FR5 sessions signaled by a light and interrupted by 2 10-min periods without reinforcement with the light off. The cumulative duration of SWD per session was measured. Instrumental conditioning using a sound-bar pressing association was performed in epileptic rats under continuous EEG control. The rats maintained at 85°7o of their body weight were trained to respond within 1.0 s to a mild sound (0.8-s duration) with a bar press delivering a food pellet. The intensity of the sound was low enough not to interrupt ongoing SWD when EEG was recorded prior to conditioning. The sound occurred randomly at a mean frequency of 2/min. As a control, the EEG of these rats was recorded after the same food deprivation for 20-min periods in a cage without the conditioning equipment, either with no access to food or with free access to food.
0--0 t--I
4030-
.
0
Controls GAERS
L/
.
.
.
.
.
.
.
.
.
.
5
.
.
10
.
.
.
.
.
.
15
.
.
.
20
24 (h)
Dork period
Fig. 1. Mean circadian locomotion in 14 GAERS and 14 controls. Ordinate: number of interruptions of the infra-red light beams per hour.
Cumulative SWD duration per 20 min was measured. The animals were kept awake by gentle stimulation when necessary.
Statistics Comparison of behavioral data obtained in GAERS and in controls were made using the non-parametric Mann-Whitney test. RESULTS
Circadian locomotion The locomotor activity over 24 h was measured in 14 GAERS and 14 non-epileptic controls. All animals displayed the usual circadian rhythmicity with an increased locomotion during the dark period. No difference between the 2 strains was noticed (Fig. 1).
Open field behavior Locomotion, rearing and defecation did not differ between 10 GAERS and 10 controls during the 6-min observation in a open field: locomotion 41.3 + 20.9 and33.8 _+ 13.4; rearings 17.2 _+ 9.6and 14.3 + 8.8; defecation 2.4 +_ 2.8 and 4.5 + 2.9 respectively in controls and GAERS.
Interspecific aggression (mouse killing behavior) When a mouse was introduced into the rat's home cage, 5/ 10 controls and 5/10 GAERS killed the mouse within a few minutes. One GAERS killed the mouse overnight. The non-killers investigated the mouse, but never attacked it.
Social interactions with a conspecific No difference in social interactions was apparent between GAERS and controls, whether the animals were in a situation of residents (n -- 10) in their home cage or introduced into the cage of the unfamiliar partner as an intruder (n = 10). Aggression or defense occurred only occasionally in a few animals. Social investigations as well as non-social activities appeared similar in both groups (Table I).
Two-way active avoidance: shuttle box GAERS (n = 10) and controls (n = 10) underwent training in a 2-way shuttle box. During sessions 1 0 - 12, 4/10 controls had learned to escape the shock by stepping to the safe compartment once the foot
100 TABLE I Mean cumulative duration (s) o f behavioral items during 8 min of social interaction with a non-familiar conspecific (n = lO/group) Residents
Cage exploration Self-grooming Immobility Partner investigation Allogrooming Attack Offensive sideways Offensive upright Defensive sideways Defensive upright
Intruders
GAERS
Controls
GAERS
Controls
202 31 108 92 15 0 11 5 10 1
251 _ 79 18 + 14 100 + 87 88 _+ 32 10 + 18 0 4+7 1+_2 1+_2 0
189 + 47 75 ± 55 113 ___ 61 73 _+ 28 20 +_ 13 0 1+_4 0 6+11 0
257 + 58 44 _+ 27 81 _ 56 69 _ 26 17 + 16 0 8+_16 0 0 0
+ 67 + 21 +- 51 ± 28 +- 8 +_ 19 ± 13 ± 27 ± 4
shock was initiated. N o n e o f t h e m learned to a v o i d the
rain
E::::] Controls I ~ GAERS
shock by stepping to the o p p o s i t e c o m p a r t m e n t d u r i n g the 9-s sound a n n o u n c i n g the shock. A m o n g the 10 G A E R S 3 learned to a v o i d (fewer t h a n 2 shocks received per session) and 5 escaped the shock in the last 3 sessions. G A E R S received fewer shocks t h a n the controis, the d i f f e r e n c e b e c o m i n g significant after the sixth session (Fig. 2).
100
806040200
I n s t r u m e n t a l learning: S k i n n e r b o x
'1 FR
L e a r n i n g to o b t a i n a f o o d pellet in a Skinner b o x by pressing a bar was similar in 10 G A E R S a n d in 10 controis w h e n FR1 a n d F R 2 schedules were applied. W h e n
FR 2
FR 5
Fig. 3. Instrumental learning in a Skinner box: time spent to reach the criterion of 25 pellets in FR 1 and FR2, 50 pellets in FR5 schedule, in 10 GAERS and 10 controls.
5 bar presses were r e q u i r e d to o b t a i n the r e i n f o r c e m e n t (FR5) the G A E R S were significantly faster in r e a c h i n g
0--0
Controls
0--0
GAERS
the criterion o f 50 pellets t h a n the controls (Fig. 3). A f t e r c o m p l e t i o n o f this training E E G was r e c o r d e d in FR5 sessions i n t e r r u p t e d by n o n - r e i n f o r c e d sessions. D u r i n g the r e i n f o r c e d sessions S W D were sup-
S 10 ~ 8
o-o-o-Q-i~_A_o_O~l/o-o-o 1
•
1
O_ *
~O,~ •
6
, •
•
, 0.~,
pressed as l o n g as rats w o r k e d for f o o d . D u r i n g the last 20-rain session, bar pressing was occasionally discontinued, p r e s u m a b l y as a result o f satiation or fatigue. T h e n s o m e S W D a p p e a r e d . N o bar press ever o c c u r r e d
4
d u r i n g a S W D . D u r i n g the n o n - r e i n f o r c e d intersessions bar pressing ceased rapidly a n d m a n y S W D ap-
2
p e a r e d (Fig. 4).
0 2
4
6
8
10
12
Sessions Fig. 2. Mean + SEM number of shocks (S) received in a 2-way avoidance shuttle box (10 trials per session). In controls (n = 10) no avoidance was learned, whereas in GAERS (n = 10) 3/10 animals learned to avoid the shock.
I n s t r u m e n t a l conditioning: s o u n d - b a r pressing The p u r p o s e o f this e x p e r i m e n t , using 5 epileptic rats u n d e r c o n t i n u o u s E E G c o n t r o l , was to investigate a possible i m p a i r m e n t in the p e r f o r m a n c e o f a l e a r n e d task associated with a sensory stimulus due to in-
101
E Q_
0 ED
20
-12
15.
9
10.
6
E O ~'~
5-
3
o
D
~:
C) (1)
these conditions m a n y S W D were recorded with a mean o f 155 s cumulative d u r a t i o n per 20 min. W h e n f o o d pellets were introduced into their cage the animals ate immediately and the total d u r a t i o n o f S W D then d r o p p e d to 3.5 s/20 min. DISCUSSION
if')
FR5
Intersessions
B
Fig. 4. Mean duration of SWD per min and mean number of bar presses per min in GAERS during FR5 sessions and nonreinforced intersessions in a Skinner box.
terference with SWD. The f o o d deprived animals learned to press a b a r to o b t a i n a f o o d pellet less than 1 s after the delivery o f a mild sound. The trained animals remained in close proximity to the bar. As long as the animals worked for food, no S W D was recorded, during sessions lasting up to 2 h. A t the end o f prolonged sessions bar pressing in response to the sound occurred irregularly and finally was suppressed. Simultaneously, S W D appeared. W h e n a sound was produced during a S W D , no b a r pressing occurred, even though a later sound was followed by a b a r press in the absence o f S W D (Fig. 5). In order to control for the effect o f hunger on the occurrence o f S W D , the E E G was recorded while the animals were separately placed in empty cages. Under
The G A E R S present spontaneously a high level o f S W D . The largest number o f seizures occurs when the animals are awake and inactive: 2 0 0 - 6 0 0 s o f S W D per 20 min are usually recorded• However, in these rats, as in humans or in other animal models 3,5,1s,19,26 the occurrence o f S W D is closely related to the state of arousal: they d i s a p p e a r during active behaviors as well as during deep slow wave and paradoxical sleep 7 and they are interrupted by sudden and unexpected sensory stimuli. The behavioral observations obtained in various situations show that animals with absence seizures do not differ from controls. Spontaneous l o c o m o t o r activity over a 24-h light-dark cycle in an a c t o g r a p h or over 6 min in an unfamiliar open field does not differ between the 2 strains, demonstrating that the occurrence o f seizures does not reduce the general level o f activity or the capacity to explore a novel environment• E m o t i o n a l i t y as expressed by defecation in the open field also appears similar in both strains. Social activities including aggressive behaviors can
ECoG .
.
.
.
.
.
.
h .......
J--
SWD Bar press j Sound
n
n
E
E
I R
I r]
Fig. 5. Electrocorticogram (ECoG) recorded in a GAERS at the end of a reinforced session (2 lines continued). The animal had learned to press a bar in response to a 0.8-s sound to obtain a food pellet (E: artifacts due to eating). During the SWD no bar press occurred in response to the sound. Calibration: 200 #V, 2 s.
102 to some extent be modified by experiences which reinforce or suppress further activities according to their past success or failure. All our animals were raised in groups of 50 male rats until the age of 5 - 6 months. In rat colonies each animal occupies a certain social position due to continuous interactions with its conspecifics. When these animals were used in our experiments they had already been exposed to social experience. The ethological analysis of our observations showed that in social interactions adult GAERS behave similarly to controls, whether they are in their own cage or intruding in the partner's home cage. Similarly mouse killing behavior, which is influenced by the current state of reactivity and by past experience with mice21, appears unchanged in GAERS versus controls. These results suggest that the possibility of SWD occurring does not affect social experience or further social behavior in rats. Learning involves attention and arousal which are both impaired during absence seizures. Therefore it was important to know whether learning as well as performance of already learned tasks would be impaired in epileptic rats. Learning of a negatively reinforced task as in a 2-way avoidance shuttle box, or of a positively reinforced task as in a Skinner box, was not impaired in epileptic rats. Moreover our data show that GAERS compared to controls performed better in avoiding a foot shock in the shuttle box as well as in obtaining food with a FR5 schedule in the Skinner box. This latter result cannot be explained on the basis of our current knowledge concerning the 2 strains and it may be independent of the epileptic characteristics of these animals. The selection of the 2 strains may also have produced unknown and uncontrolled selection of other traits. However, possible interactions of cortical SWD with processes of memory consolidation are still unexplored. These data are in agreement with clinical data showing that in children with typical benign absence epilepsy intellectual capacities appear unimpaired or even well developed 8. Although some clinical studies mentioned mental defects in patients with absence seizures, such differences are probably explained by the inclusion of cases with atypical absence epilepsies by various investigators 1. In order to examine possible impairments during the occurrence of SWD, EEG was recorded in epileptic rats during the performance of a learned task, either in the Skinner box or in the conditioned sound-bar press-
ing task. The most striking observation was that of usual fast activity desynchronized EEG without any SWD as long as the animal was waiting or working for the reward. However, many SWD appeared as soon as the animals had no task to perform, as in the nonreinforced intersessions in the Skinner box, or when they were placed, while hungry, in an empty cage. We had previously observed that food deprivation per se does not modify the number of recorded SWD. These results account for the lack of interference of absence epilepsy with behavioral performance: as long as the animal is motivated to perform the task, SWD are suppressed. When motivation decreased, SWD occur, probably as a result of decreased attention. The state of quiet wakefulness was shown to be the most favorable for SWD to occur7. In other animal models of absence seizures, SWD are also suppressed during attentive behavior4,17. In children also absences are less frequent during the performance of tasks requiring active cooperation, such as projective tests. By contrast, when vigilance is reduced or during transitional states to sleep, seizures become more frequent. When a learning task is not really motivating it becomes likely that SWD then appear 5,6,9 In humans, unresponsiveness to environmental stimuli is classically described during absence seizures 13,16. This may be due to loss of attention, impaired processing of information or altered sensory input, possibly associated with a direct effect on motor performance 12,14,15. In rats, SWD reappeared only during signaled intersessions without reinforcement in the Skinner box or at the end of prolonged sessions of conditioned bar pressing when motivation for bar pressing was reduced by satiation, and when working was already reduced. In such situations it happened that SWD prevented responding to a conditioned sound or retarded performance of a FR responding. In fact it appears difficult to demonstrate experimentally the behavioral concomitants of SWD in rats with spontaneous absence seizures, as such seizures appear mainly during inactivity. In cats with penicillininduced absence seizures unresponsiveness to external stimuli is a common concomitant of SWD 20. This discrepancy with our own results may be due to the fact that chemically induced seizures are less sensitive to environmental situations than endogenous ones. In conclusion, our data show that in rats with genetic absence epilepsy the ability to perform spon-
103 taneous and learned behavioral activities is unimpaired. During performance of active and motivated behaviors SWD are suppressed and consequently cannot interfere with such behaviors. Only in borderline states, when motivation to act is low, the occurrence of SWD may suppress perception of information and behavioral responsiveness. These data are in agreement with observations commonly described in children with genetic idiopathic absence epilepsy.
ACKNOWLEDGEMENTS We are grateful to Professor B. Will for his help and advice in the realization of the conditioning experiments. This work was supported by a grant, Contrat de Recherche externe 866017, from INSERM and by 'La Fondation pour la Recherche M6dicale'.
REFERENCES 1 Aicardi, J. (Ed.), Epilepsy in Children, Raven Press, New York, 1986, 413 pp. 2 Depaulis, A., A microcomputer method for behavioural data acquisition and subsequent analysis, PharmacoL Biochem. Behav., 19 (1983) 7 2 9 - 732. 3 Gastaut, H., Batini, G., Broughton, R., Fressy, J., Roger, J. and Tassinari, C.A., Etude 61ectrographique du sommeil nocturne chez les ~pileptiques, Rev. Neurol., 110 (1964) 311-313. 4 Guerrero-Figueroa, R., Barros, A. and De Balbian Verster, F., Some inhibitory effects of attentive factors on experimental epilepsy, Epilepsia, 4 (1963) 2 2 5 - 240. 5 Guey, J., Bureau, M., Dravet, C. and Roger, J., A study of the rhythm of petit real absences in children in relation to prevailing situations. The use of EEG telemetry during psychological examinations, school exercises and periods of inactivity, Epilepsia, 10 (1969) 441 -451. 6 Jung, R., Blocking of petit-mal attacks by sensory arousal and inhibition in attacks by an active change in attention during the epileptic aura, Epilepsia, 3 (1962) 4 3 5 - 437. 7 Lannes, B., Micheletti, G., Vergnes, M., Marescaux, C., Depaulis, A. and Wafter, J.M., Relationship between spikewave discharges and vigilance levels in rats with spontaneous petit mal-like epilepsy, Neurosci. Lett., 94 (1988) 187 - 191. 8 Lennox, W.G. and Lennox, M.A., Epilepsy and Related Disorders, 1, Little, Brown and Co, Boston, MA, 1960, 574 pp. 9 Lennox, W.G., Gibbs, F.A. and Gibbs, E., Effect on the electroencephalogram of drugs and conditions which influence seizures, Arch. Neurol. Psychiatry, 36 (1936) 1236- 1245. 10 Marescaux, C., Micheletti, G., Vergnes, M., Depaulis, A., Rumbach, L. and Warter, J.M., A model of chronic spontaneous petit mal-like seizures in the rat: comparison with pentylenetetrazol-induced seizures, Epilepsia, 25 (1984) 3 2 6 - 331. 11 Marescaux, C., Vergnes, M., Depaulis, A., Micheletti, G. and Warter, J.M., Neurotransmission in rats' spontaneous generalized nonconvulsive epilepsy. In: G. Avanzini et al. (Eds.), Neurotransmitters in Epilepsy, Demos, New York, 1991, pp. 4 5 3 - 465. 12 Mirsky, A.F., Behavioral and psychological effects of petit real epilepsy in the light of a neuropsychologically-based theory of attention. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 311-340.
13 Mirsky, A.F., Duncan, C.C. and Myslobodsky, M.S., Petit mal epilepsy: a review and integration of recent information, J. Clin. Neurophysiol., 3 (1986) 179- 208. 14 Myslobodsky, M.S., Petit real status as a paradigm of the functional anatomy of awareness. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 71 - 104. 15 Myslobodsky, M.S. and Mirsky, A.F., Update on petit mal: the case of heterogeneity. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 365-392. 16 Panayiotopoulos, C.P., Obeid, T. and Waheed, G., Differentiation of typical absence seizures in epileptic syndromes, Brain, 112 (1989) 1039- 1056. 17 Pollen, D.A., Perot, P. and Reid, K.H., Experimental bilateral wave and spike from thalamic stimulation in relation to level of arousal, Electroenceph. Clin. Neurophysiol., 15 (1963) 1017- 1028. 18 Sato, S., Dreifuss, F.E. and Penry, J.K., The effect of sleep on spike-wave discharges in absence seizures, Neurology, 23 (1973) 1335- 1345. 19 Shouse, M.N., Differences between two feline models in sleep and waking state disorders, state dependency of seizures and seizure susceptibility: amygdala kindling interferes with systemic penicillin epilepsy, Epilepsia, 28 (1987) 3 9 9 - 408. 20 Taylor-Courval, D. and Gloor, P., Behavioral alterations associated with generalized spike and wave discharges in the EEG of the cat, Exp. Neurol., 83 (1984) 167- 186. 21 Vergnes, M., Amygdaloid control over interspecies aggression in the rat. In: M. Girgis and L.G. Kiloh (Eds.), Limbic Epilepsy and the Dyscontrol Syndrome, Elsevier, Amsterdam, 1980, pp 8 5 - 9 2 . 22 Vergnes, M., Depaulis, A., Boehrer, A and Kempf, E., Selective increase of offensive behavior in the rat following intrahypothalamic 5,7-DHT-induced serotonin depletion, Behav. Brain Res., 29 (1988) 8 5 - 9 1 . 23 Vergnes, M., Marescaux, C., Micheletti, G., Reis, J., Depaulis, A., Rumbach, L. and Warter, J.M., Spontaneous paroxysmal electroclinical patterns in rat: a model of generalized non-convulsive epilepsy, Neurosci. Lett., 33 (1982) 9 7 - 101. 24 Vergnes, M., Marescaux, C. and Depaulis, A., Mapping of spontaneous spike and wave discharges in Wistar rats with genetic generalized non-convulsive epilepsy, Brain Res., 523 (1990) 87 - 91. 25 Vergnes, M., Marescaux, C., Depaulis, A., Micheletti, G.
104 and Warter, J.M., Spontaneous spike-and-wave discharges in Wistar rats: a model of genetic generalized non-convulsive epilepsy. In: M. Avoli, P. Gloor, G. Kostopoulos and R. Naquet (Eds.), Generalized Epilepsy: Neurobiological Approaches, Birkh~user, Boston, Basel, Berlin, 1990, pp. 238 - 253.
26 Zembilci, N., Dervent, A. and Sarloglu, A.C., Influence of sleep and wakefulness on experimental petit mal activity. In: W.P. Koella and P. Levin (Eds.), Sleep 1976, 3rd Eur. Congr. Sleep Res., Montpellier, Karger, Basel, 1977, pp. 246 - 249.
97
EPIRES 00415
Are rats with genetic absence epilepsy behaviorally impaired?
Marguerite Vergnes, Christian Marescauxa, Any Boehrer and Antoine Depaulis Centre de Neurochimie du CNRS, 67000 Strasbourg (France) and a CHU, Clinique Neurologique Strasbourg (France)
(Received 15 February 1991; revision received 1 May 1991; accepted 6 May 1991) Key words: Rat; Absence epilepsy; Behavior; Learning; EEG
Absence seizures in humans are characterized by unresponsiveness to external stimuli and inactivity. However, in typical generalized non-convulsiveepilepsyin children, intellectual capacities are considered to be normal. Wistar rats from an inbred strain with spontaneous absence-like seizures were compared with rats from the outbred control strain in various behavioral tasks in order to detect possible impairments related either to the absence epilepsyor to occurrence of spike and wave discharges (SWD). Spontaneous circadian locomotion, exploratory activity in an open field, social interactions with an unfamiliar conspecific and mouse killing behavior were similar in both strains. Avoidance learning in a shuttle box or food reinforced learning in a Skinner test were unimpaired or even improved in epileptic rats. During performance of a learned task either in the Skinner box or in a conditioned sound-bar pressing task, SWD were suppressed in epileptic rats as long as they were working for reinforcement. SWD reappeared when the motivation to perform the task had declined: unresponsiveness to a conditioned stimulus was then observed during SWD. These data are in agreement with observations commonly described in children with typical genetic absence epilepsy.
INTRODUCTION Absence epilepsy (or petit mal) is a generalized n o n convulsive epilepsy characterized by the spontaneous occurrence of bilateral, synchronous spike and wave discharges (SWD) on the cortical E E G with concomitant unawareness, unresponsiveness a n d cessation of activities. Classically, in children with pure typical absence epilepsy, despite behavioral impairments which occur during SWD, intelligence is considered to be n o r m a l or even superior in some cases 1,8. A strain of Wistar rats with spontaneous seizures typical of absence epilepsy has been inbred in our laboratory (GAERS = genetic absence epilepsy rats from Strasbourg). I n these animals seizures are characterized by SWD which occur o n a b a c k g r o u n d of quiet wakefulness with a mean frequency of 1 / m i n
Correspondence to: Marguerite Vergnes, LNBC, Centre de Neurochimie du CNRS, 5 rue Blaise Pascal, 67084 Strasbourg Cedex, France.
and durations varying from 1 to 60 s with a mean a r o u n d 20 s. The frequency of the bilateral and synchronous spike and waves varies from 7 to 9 Hz 23. The SWD involve all the neocortex and the m a i n relay nuclei of the thalamus, whereas no SWD are recorded in the limbic structures 24. During the SWD the animals are immobile with reduction of neck muscle tone and occasional twitching of the vibrissae and the face. The pharmacological reactivity of absence seizures has been shown to be similar in this rat strain and h u m a n s 1°,11. A control strain devoid of seizures was outbred from the same Wistar colony25. The 2 strains do not apparently differ in size, growth, reproductive capacity or any morphological characteristics. The behavior of an adult is a dynamic construct resulting from the interaction of i n b o r n capacities and experience. The genetic abnormality which produces the development of a considerable n u m b e r of absence seizures during all the adult life in rats 25 could also directly or indirectly cause some behavioral impair-
0920-1211/91/$03.50 © 1991 Elsevier Science Publishers B.V. All rights reserved
98 ment due to dysfunction of some brain areas. Alternatively, the repeated occurrence of SWD concomitant with loss of attention and responsiveness may provoke reduced efficiency and success in various behaviors, especially in social activities or learning tasks. An analysis of behavioral parameters was undertaken to reveal possible differences in the abilities of epileptic GAERS and non-epileptic controis in various spontaneous and learned behaviors, namely in spontaneous and evoked locomotor activity, intra- and interspecific social interactions, and ability to learn and perform various tasks. Moreover, the EEG was recorded during performance of learned tasks in order to detect impairments related to the incidence of SWD. METHODS
Animals Adult male Wistar rats ( 3 0 0 - 400 g) from 2 strains selected over 15 generations either for spontaneous occurrence of absence-like seizures (GAERS) or for lack of such seizures (controls) were used. All adult GAERS have prolonged SWD, whereas SWD are never recorded in any rat from the control strain. The animals were age-matched in each experiment. Ad lib fed animals were placed in individual cages and maintained on a 12-h dark-light regimen. In the experiments using EEG control the animals were equipped under pentobarbital anesthesia (40 mg/kg i.p.) with 4 stainless steel electrodes screwed into the skull over the fronto-parietal cortex and connected to a microconnector embedded in dental acrylic cement. The EEG was recorded between 2 ipsilateral electrodes by an electroencephalograph (Alvar).
Behavioral tests Circadian locomotion was measured in the home cage (30 × 24 × 35 cm) placed in front of 2 photocells recording the interruption of an infra-red light beam. Food was available. Measures started during the usual light period, and locomotor activity was recorded for 24 h. A circular open field (diameter 1 m) lit with a 60-W bulb above the field was used to measure activity in an unfamiliar environment. The floor of the open field was divided into 12 parts and the number of parts crossed during a 6-min period was quantified as
locomotor activity. In addition the number of rearings on the hind feet and the number of defecations were counted during this period. The open field was carefully cleaned after each measure. Interspecific aggression was evaluated by introducing a mouse into the rat's familiar home cage. Rats were naive with respect to mice. Most killer rats attack and kill rapidly, but some rats kill only overnight. Therefore the mouse was left in the rat's cage for 24 h. Social interactions with a non-familiar weightmatched male Conspecific were recorded in a single 8min session on video tape. The experimental rat was either in a situation of resident, the partner being introduced into its home cage, or in a situation of intruder in the partner's home cage. The different behavioral items were encoded from the video recordings using a microcomputer-based technique 2. The data were processed for each animal and each item in terms of cumulated duration per session (for details see Vergnes et al.22). The following behavioral items were coded: (1) non-social activities: cage exploration, selfgrooming, immobile posture; (2) social activities: partner investigation, allogrooming; (3) offense: attack, offensive sideways, offensive upright; (4) defense: defensive sideways, defensive upright. Avoidance learning was performed in a 2-way shuttle box of 35 × 16 × 20 cm divided into 2 equal compartments by a partition with an opening allowing the rat to cross to the opposite side. Electric foot shocks (0.5 mA) were delivered through the floor grid of the shuttle box. A 9-s sound announced the scrambled shocks associated with the sound for a further 11 s, followed by a 10-s rest. Ten shocks per session were alternately delivered in the 2 compartments when crossing between the compartments had occurred. Four groups of 3 sessions were performed in 2 days. For each session of 10 sound-shock associations, the total number of shocks received, the latency to cross to the opposite compartment, and the number of such crossings were recorded. Animals learned either to avoid the shock by stepping to the safe compartment during the 9-s sound exposure, or to escape the shock during the 11 s of delivery of shocks. Some animals never learned to move to the opposite side in order to escape the shock. Avoidance was considered to be learned when the animals received fewer than 2 shocks per session. Escape was learned when the animals received the
99 shock but crossed within the duration of the shock in more than 8 trials. Instrumental learning with a fixed ratio reinforcement was performed in a Skinner box (28 x 24 x 30 cm). The animals maintained at 85% of their body weight were trained to press a bar to obtain a 45-mg food pellet. A program of reinforcement was scheduled: successively, fixed reinforcement 1 (FR1, each bar press delivers 1 food pellet) until 25 pellets are delivered; FR2 (2 bar presses are necessary to deliver 1 pellet) for 25 pellets, and FR5 to the criterion of 50 pellets. For each reinforcement ratio the time necessary to reach the criterion was measured. Once trained in the Skinner box to obtain food on a FR5 schedule the EEG was recorded in epileptic rats during 3 consecutive 20-min FR5 sessions signaled by a light and interrupted by 2 10-min periods without reinforcement with the light off. The cumulative duration of SWD per session was measured. Instrumental conditioning using a sound-bar pressing association was performed in epileptic rats under continuous EEG control. The rats maintained at 85°7o of their body weight were trained to respond within 1.0 s to a mild sound (0.8-s duration) with a bar press delivering a food pellet. The intensity of the sound was low enough not to interrupt ongoing SWD when EEG was recorded prior to conditioning. The sound occurred randomly at a mean frequency of 2/min. As a control, the EEG of these rats was recorded after the same food deprivation for 20-min periods in a cage without the conditioning equipment, either with no access to food or with free access to food.
0--0 t--I
4030-
.
0
Controls GAERS
L/
.
.
.
.
.
.
.
.
.
.
5
.
.
10
.
.
.
.
.
.
15
.
.
.
20
24 (h)
Dork period
Fig. 1. Mean circadian locomotion in 14 GAERS and 14 controls. Ordinate: number of interruptions of the infra-red light beams per hour.
Cumulative SWD duration per 20 min was measured. The animals were kept awake by gentle stimulation when necessary.
Statistics Comparison of behavioral data obtained in GAERS and in controls were made using the non-parametric Mann-Whitney test. RESULTS
Circadian locomotion The locomotor activity over 24 h was measured in 14 GAERS and 14 non-epileptic controls. All animals displayed the usual circadian rhythmicity with an increased locomotion during the dark period. No difference between the 2 strains was noticed (Fig. 1).
Open field behavior Locomotion, rearing and defecation did not differ between 10 GAERS and 10 controls during the 6-min observation in a open field: locomotion 41.3 + 20.9 and33.8 _+ 13.4; rearings 17.2 _+ 9.6and 14.3 + 8.8; defecation 2.4 +_ 2.8 and 4.5 + 2.9 respectively in controls and GAERS.
Interspecific aggression (mouse killing behavior) When a mouse was introduced into the rat's home cage, 5/ 10 controls and 5/10 GAERS killed the mouse within a few minutes. One GAERS killed the mouse overnight. The non-killers investigated the mouse, but never attacked it.
Social interactions with a conspecific No difference in social interactions was apparent between GAERS and controls, whether the animals were in a situation of residents (n -- 10) in their home cage or introduced into the cage of the unfamiliar partner as an intruder (n = 10). Aggression or defense occurred only occasionally in a few animals. Social investigations as well as non-social activities appeared similar in both groups (Table I).
Two-way active avoidance: shuttle box GAERS (n = 10) and controls (n = 10) underwent training in a 2-way shuttle box. During sessions 1 0 - 12, 4/10 controls had learned to escape the shock by stepping to the safe compartment once the foot
100 TABLE I Mean cumulative duration (s) o f behavioral items during 8 min of social interaction with a non-familiar conspecific (n = lO/group) Residents
Cage exploration Self-grooming Immobility Partner investigation Allogrooming Attack Offensive sideways Offensive upright Defensive sideways Defensive upright
Intruders
GAERS
Controls
GAERS
Controls
202 31 108 92 15 0 11 5 10 1
251 _ 79 18 + 14 100 + 87 88 _+ 32 10 + 18 0 4+7 1+_2 1+_2 0
189 + 47 75 ± 55 113 ___ 61 73 _+ 28 20 +_ 13 0 1+_4 0 6+11 0
257 + 58 44 _+ 27 81 _ 56 69 _ 26 17 + 16 0 8+_16 0 0 0
+ 67 + 21 +- 51 ± 28 +- 8 +_ 19 ± 13 ± 27 ± 4
shock was initiated. N o n e o f t h e m learned to a v o i d the
rain
E::::] Controls I ~ GAERS
shock by stepping to the o p p o s i t e c o m p a r t m e n t d u r i n g the 9-s sound a n n o u n c i n g the shock. A m o n g the 10 G A E R S 3 learned to a v o i d (fewer t h a n 2 shocks received per session) and 5 escaped the shock in the last 3 sessions. G A E R S received fewer shocks t h a n the controis, the d i f f e r e n c e b e c o m i n g significant after the sixth session (Fig. 2).
100
806040200
I n s t r u m e n t a l learning: S k i n n e r b o x
'1 FR
L e a r n i n g to o b t a i n a f o o d pellet in a Skinner b o x by pressing a bar was similar in 10 G A E R S a n d in 10 controis w h e n FR1 a n d F R 2 schedules were applied. W h e n
FR 2
FR 5
Fig. 3. Instrumental learning in a Skinner box: time spent to reach the criterion of 25 pellets in FR 1 and FR2, 50 pellets in FR5 schedule, in 10 GAERS and 10 controls.
5 bar presses were r e q u i r e d to o b t a i n the r e i n f o r c e m e n t (FR5) the G A E R S were significantly faster in r e a c h i n g
0--0
Controls
0--0
GAERS
the criterion o f 50 pellets t h a n the controls (Fig. 3). A f t e r c o m p l e t i o n o f this training E E G was r e c o r d e d in FR5 sessions i n t e r r u p t e d by n o n - r e i n f o r c e d sessions. D u r i n g the r e i n f o r c e d sessions S W D were sup-
S 10 ~ 8
o-o-o-Q-i~_A_o_O~l/o-o-o 1
•
1
O_ *
~O,~ •
6
, •
•
, 0.~,
pressed as l o n g as rats w o r k e d for f o o d . D u r i n g the last 20-rain session, bar pressing was occasionally discontinued, p r e s u m a b l y as a result o f satiation or fatigue. T h e n s o m e S W D a p p e a r e d . N o bar press ever o c c u r r e d
4
d u r i n g a S W D . D u r i n g the n o n - r e i n f o r c e d intersessions bar pressing ceased rapidly a n d m a n y S W D ap-
2
p e a r e d (Fig. 4).
0 2
4
6
8
10
12
Sessions Fig. 2. Mean + SEM number of shocks (S) received in a 2-way avoidance shuttle box (10 trials per session). In controls (n = 10) no avoidance was learned, whereas in GAERS (n = 10) 3/10 animals learned to avoid the shock.
I n s t r u m e n t a l conditioning: s o u n d - b a r pressing The p u r p o s e o f this e x p e r i m e n t , using 5 epileptic rats u n d e r c o n t i n u o u s E E G c o n t r o l , was to investigate a possible i m p a i r m e n t in the p e r f o r m a n c e o f a l e a r n e d task associated with a sensory stimulus due to in-
101
E Q_
0 ED
20
-12
15.
9
10.
6
E O ~'~
5-
3
o
D
~:
C) (1)
these conditions m a n y S W D were recorded with a mean o f 155 s cumulative d u r a t i o n per 20 min. W h e n f o o d pellets were introduced into their cage the animals ate immediately and the total d u r a t i o n o f S W D then d r o p p e d to 3.5 s/20 min. DISCUSSION
if')
FR5
Intersessions
B
Fig. 4. Mean duration of SWD per min and mean number of bar presses per min in GAERS during FR5 sessions and nonreinforced intersessions in a Skinner box.
terference with SWD. The f o o d deprived animals learned to press a b a r to o b t a i n a f o o d pellet less than 1 s after the delivery o f a mild sound. The trained animals remained in close proximity to the bar. As long as the animals worked for food, no S W D was recorded, during sessions lasting up to 2 h. A t the end o f prolonged sessions bar pressing in response to the sound occurred irregularly and finally was suppressed. Simultaneously, S W D appeared. W h e n a sound was produced during a S W D , no b a r pressing occurred, even though a later sound was followed by a b a r press in the absence o f S W D (Fig. 5). In order to control for the effect o f hunger on the occurrence o f S W D , the E E G was recorded while the animals were separately placed in empty cages. Under
The G A E R S present spontaneously a high level o f S W D . The largest number o f seizures occurs when the animals are awake and inactive: 2 0 0 - 6 0 0 s o f S W D per 20 min are usually recorded• However, in these rats, as in humans or in other animal models 3,5,1s,19,26 the occurrence o f S W D is closely related to the state of arousal: they d i s a p p e a r during active behaviors as well as during deep slow wave and paradoxical sleep 7 and they are interrupted by sudden and unexpected sensory stimuli. The behavioral observations obtained in various situations show that animals with absence seizures do not differ from controls. Spontaneous l o c o m o t o r activity over a 24-h light-dark cycle in an a c t o g r a p h or over 6 min in an unfamiliar open field does not differ between the 2 strains, demonstrating that the occurrence o f seizures does not reduce the general level o f activity or the capacity to explore a novel environment• E m o t i o n a l i t y as expressed by defecation in the open field also appears similar in both strains. Social activities including aggressive behaviors can
ECoG .
.
.
.
.
.
.
h .......
J--
SWD Bar press j Sound
n
n
E
E
I R
I r]
Fig. 5. Electrocorticogram (ECoG) recorded in a GAERS at the end of a reinforced session (2 lines continued). The animal had learned to press a bar in response to a 0.8-s sound to obtain a food pellet (E: artifacts due to eating). During the SWD no bar press occurred in response to the sound. Calibration: 200 #V, 2 s.
102 to some extent be modified by experiences which reinforce or suppress further activities according to their past success or failure. All our animals were raised in groups of 50 male rats until the age of 5 - 6 months. In rat colonies each animal occupies a certain social position due to continuous interactions with its conspecifics. When these animals were used in our experiments they had already been exposed to social experience. The ethological analysis of our observations showed that in social interactions adult GAERS behave similarly to controls, whether they are in their own cage or intruding in the partner's home cage. Similarly mouse killing behavior, which is influenced by the current state of reactivity and by past experience with mice21, appears unchanged in GAERS versus controls. These results suggest that the possibility of SWD occurring does not affect social experience or further social behavior in rats. Learning involves attention and arousal which are both impaired during absence seizures. Therefore it was important to know whether learning as well as performance of already learned tasks would be impaired in epileptic rats. Learning of a negatively reinforced task as in a 2-way avoidance shuttle box, or of a positively reinforced task as in a Skinner box, was not impaired in epileptic rats. Moreover our data show that GAERS compared to controls performed better in avoiding a foot shock in the shuttle box as well as in obtaining food with a FR5 schedule in the Skinner box. This latter result cannot be explained on the basis of our current knowledge concerning the 2 strains and it may be independent of the epileptic characteristics of these animals. The selection of the 2 strains may also have produced unknown and uncontrolled selection of other traits. However, possible interactions of cortical SWD with processes of memory consolidation are still unexplored. These data are in agreement with clinical data showing that in children with typical benign absence epilepsy intellectual capacities appear unimpaired or even well developed 8. Although some clinical studies mentioned mental defects in patients with absence seizures, such differences are probably explained by the inclusion of cases with atypical absence epilepsies by various investigators 1. In order to examine possible impairments during the occurrence of SWD, EEG was recorded in epileptic rats during the performance of a learned task, either in the Skinner box or in the conditioned sound-bar press-
ing task. The most striking observation was that of usual fast activity desynchronized EEG without any SWD as long as the animal was waiting or working for the reward. However, many SWD appeared as soon as the animals had no task to perform, as in the nonreinforced intersessions in the Skinner box, or when they were placed, while hungry, in an empty cage. We had previously observed that food deprivation per se does not modify the number of recorded SWD. These results account for the lack of interference of absence epilepsy with behavioral performance: as long as the animal is motivated to perform the task, SWD are suppressed. When motivation decreased, SWD occur, probably as a result of decreased attention. The state of quiet wakefulness was shown to be the most favorable for SWD to occur7. In other animal models of absence seizures, SWD are also suppressed during attentive behavior4,17. In children also absences are less frequent during the performance of tasks requiring active cooperation, such as projective tests. By contrast, when vigilance is reduced or during transitional states to sleep, seizures become more frequent. When a learning task is not really motivating it becomes likely that SWD then appear 5,6,9 In humans, unresponsiveness to environmental stimuli is classically described during absence seizures 13,16. This may be due to loss of attention, impaired processing of information or altered sensory input, possibly associated with a direct effect on motor performance 12,14,15. In rats, SWD reappeared only during signaled intersessions without reinforcement in the Skinner box or at the end of prolonged sessions of conditioned bar pressing when motivation for bar pressing was reduced by satiation, and when working was already reduced. In such situations it happened that SWD prevented responding to a conditioned sound or retarded performance of a FR responding. In fact it appears difficult to demonstrate experimentally the behavioral concomitants of SWD in rats with spontaneous absence seizures, as such seizures appear mainly during inactivity. In cats with penicillininduced absence seizures unresponsiveness to external stimuli is a common concomitant of SWD 20. This discrepancy with our own results may be due to the fact that chemically induced seizures are less sensitive to environmental situations than endogenous ones. In conclusion, our data show that in rats with genetic absence epilepsy the ability to perform spon-
103 taneous and learned behavioral activities is unimpaired. During performance of active and motivated behaviors SWD are suppressed and consequently cannot interfere with such behaviors. Only in borderline states, when motivation to act is low, the occurrence of SWD may suppress perception of information and behavioral responsiveness. These data are in agreement with observations commonly described in children with genetic idiopathic absence epilepsy.
ACKNOWLEDGEMENTS We are grateful to Professor B. Will for his help and advice in the realization of the conditioning experiments. This work was supported by a grant, Contrat de Recherche externe 866017, from INSERM and by 'La Fondation pour la Recherche M6dicale'.
REFERENCES 1 Aicardi, J. (Ed.), Epilepsy in Children, Raven Press, New York, 1986, 413 pp. 2 Depaulis, A., A microcomputer method for behavioural data acquisition and subsequent analysis, PharmacoL Biochem. Behav., 19 (1983) 7 2 9 - 732. 3 Gastaut, H., Batini, G., Broughton, R., Fressy, J., Roger, J. and Tassinari, C.A., Etude 61ectrographique du sommeil nocturne chez les ~pileptiques, Rev. Neurol., 110 (1964) 311-313. 4 Guerrero-Figueroa, R., Barros, A. and De Balbian Verster, F., Some inhibitory effects of attentive factors on experimental epilepsy, Epilepsia, 4 (1963) 2 2 5 - 240. 5 Guey, J., Bureau, M., Dravet, C. and Roger, J., A study of the rhythm of petit real absences in children in relation to prevailing situations. The use of EEG telemetry during psychological examinations, school exercises and periods of inactivity, Epilepsia, 10 (1969) 441 -451. 6 Jung, R., Blocking of petit-mal attacks by sensory arousal and inhibition in attacks by an active change in attention during the epileptic aura, Epilepsia, 3 (1962) 4 3 5 - 437. 7 Lannes, B., Micheletti, G., Vergnes, M., Marescaux, C., Depaulis, A. and Wafter, J.M., Relationship between spikewave discharges and vigilance levels in rats with spontaneous petit mal-like epilepsy, Neurosci. Lett., 94 (1988) 187 - 191. 8 Lennox, W.G. and Lennox, M.A., Epilepsy and Related Disorders, 1, Little, Brown and Co, Boston, MA, 1960, 574 pp. 9 Lennox, W.G., Gibbs, F.A. and Gibbs, E., Effect on the electroencephalogram of drugs and conditions which influence seizures, Arch. Neurol. Psychiatry, 36 (1936) 1236- 1245. 10 Marescaux, C., Micheletti, G., Vergnes, M., Depaulis, A., Rumbach, L. and Warter, J.M., A model of chronic spontaneous petit mal-like seizures in the rat: comparison with pentylenetetrazol-induced seizures, Epilepsia, 25 (1984) 3 2 6 - 331. 11 Marescaux, C., Vergnes, M., Depaulis, A., Micheletti, G. and Warter, J.M., Neurotransmission in rats' spontaneous generalized nonconvulsive epilepsy. In: G. Avanzini et al. (Eds.), Neurotransmitters in Epilepsy, Demos, New York, 1991, pp. 4 5 3 - 465. 12 Mirsky, A.F., Behavioral and psychological effects of petit real epilepsy in the light of a neuropsychologically-based theory of attention. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 311-340.
13 Mirsky, A.F., Duncan, C.C. and Myslobodsky, M.S., Petit mal epilepsy: a review and integration of recent information, J. Clin. Neurophysiol., 3 (1986) 179- 208. 14 Myslobodsky, M.S., Petit real status as a paradigm of the functional anatomy of awareness. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 71 - 104. 15 Myslobodsky, M.S. and Mirsky, A.F., Update on petit mal: the case of heterogeneity. In: M.S. Myslobodsky and A.F. Mirsky (Eds.), Elements of Petit Mal Epilepsy, Peter Lang, New York, 1988, pp. 365-392. 16 Panayiotopoulos, C.P., Obeid, T. and Waheed, G., Differentiation of typical absence seizures in epileptic syndromes, Brain, 112 (1989) 1039- 1056. 17 Pollen, D.A., Perot, P. and Reid, K.H., Experimental bilateral wave and spike from thalamic stimulation in relation to level of arousal, Electroenceph. Clin. Neurophysiol., 15 (1963) 1017- 1028. 18 Sato, S., Dreifuss, F.E. and Penry, J.K., The effect of sleep on spike-wave discharges in absence seizures, Neurology, 23 (1973) 1335- 1345. 19 Shouse, M.N., Differences between two feline models in sleep and waking state disorders, state dependency of seizures and seizure susceptibility: amygdala kindling interferes with systemic penicillin epilepsy, Epilepsia, 28 (1987) 3 9 9 - 408. 20 Taylor-Courval, D. and Gloor, P., Behavioral alterations associated with generalized spike and wave discharges in the EEG of the cat, Exp. Neurol., 83 (1984) 167- 186. 21 Vergnes, M., Amygdaloid control over interspecies aggression in the rat. In: M. Girgis and L.G. Kiloh (Eds.), Limbic Epilepsy and the Dyscontrol Syndrome, Elsevier, Amsterdam, 1980, pp 8 5 - 9 2 . 22 Vergnes, M., Depaulis, A., Boehrer, A and Kempf, E., Selective increase of offensive behavior in the rat following intrahypothalamic 5,7-DHT-induced serotonin depletion, Behav. Brain Res., 29 (1988) 8 5 - 9 1 . 23 Vergnes, M., Marescaux, C., Micheletti, G., Reis, J., Depaulis, A., Rumbach, L. and Warter, J.M., Spontaneous paroxysmal electroclinical patterns in rat: a model of generalized non-convulsive epilepsy, Neurosci. Lett., 33 (1982) 9 7 - 101. 24 Vergnes, M., Marescaux, C. and Depaulis, A., Mapping of spontaneous spike and wave discharges in Wistar rats with genetic generalized non-convulsive epilepsy, Brain Res., 523 (1990) 87 - 91. 25 Vergnes, M., Marescaux, C., Depaulis, A., Micheletti, G.
104 and Warter, J.M., Spontaneous spike-and-wave discharges in Wistar rats: a model of genetic generalized non-convulsive epilepsy. In: M. Avoli, P. Gloor, G. Kostopoulos and R. Naquet (Eds.), Generalized Epilepsy: Neurobiological Approaches, Birkh~user, Boston, Basel, Berlin, 1990, pp. 238 - 253.
26 Zembilci, N., Dervent, A. and Sarloglu, A.C., Influence of sleep and wakefulness on experimental petit mal activity. In: W.P. Koella and P. Levin (Eds.), Sleep 1976, 3rd Eur. Congr. Sleep Res., Montpellier, Karger, Basel, 1977, pp. 246 - 249.
