- Email: [email protected]
Characterization of gabaergic seizure regulation in the midline thalamus
Neuropharmacdogy Vol. 29. No. 7. pp. 649-655, 1990 Printed in Great Britain. All nghts reserved
0028.3908/90$3.00+ 0.00 Copyright 0 1990Pergamon Press plc
CHARACTERIZATION OF GABAERGIC SEIZURE REGULATION IN THE MIDLINE THALAMUS J. W. MILLER and J. A. FERRENDELLI Departments of Neurology and Neurological Surgery (Neurology) and Neuropharmacology, Washington University School of Medicine, 660 South Euclid Ave., St Louis, Missouri 63110, U.S.A. (Accepted 29 Noaember 1989)
Summary-This
study characterized the role of GABA in the central medial intralaminar nucleus on seizures induced by pentylenetetrazol given systemically. Injections of the direct selective GABA, agonist. piperidine-4-sulfonic acid or the indirect GABA, agonists, flurazepam and pentobarbital, in this region depressed arousal and facilitated myoclonic and clonic seizures induced by pentylenetetrazol but only caused slight inhibition of tonic seizures. In contrast the GABA, agonist (-)baclofen facilitated all three types of seizures. Recording after injection of piperidine-4-sulfonic acid and (-)baclofen revealed marked suppression and slowing of thalamic and cortical electrical activity. Thalamic injections of the GABA, antagonist, bicuculline methiodide, had opposite behavioral effects, causing hyperactivity and episodes of violent running, not accompanied by EEG discharges. When pentylenetetrazol was infused concommitantly there was marked facilitation of the tonic seizures, which occurred without preceding myoclonic of clonic seizures, or EEG spikes. These results demonstrate that GABA-mediated neurotransmission in the central medial intralaminar nucleus can control the threshold of seizures and that GABA agonists and antagonists have opposite effects. It is suggested that the central medial intralaminar nucleus is not a site of origination or spread of seizures. but controls seizures indirectly by regulating the excitability of other structures and that different synaptic mechanisms and anatomical connections mediate effects on different types of seizures.
Ke,v words-pentylenetetrazol,
central medial nucleus, GABA, seizure, epilepsy, thalamus, arousal.
functions to alert the animal and to inhibit seizures and the injections of muscimol exert their effect by inhibiting these functions. One purpose of the present study was to verify this hypothesis by providing further evidence that such effects result from inhibition mediated by GABA receptors by determining the effects of thalamic injections of a variety of GABAergic agents, including direct agonists and agents which act by modulating the effects of endogenous GABA. Another aim was to differentiate the roles of GABA, and GABA, receptors on the regulation of thalamic seizures by studying the effects of thalamic injections of agonists which are more selective than muscimol. Finally, the effect of injection of the GABA antagonist bicuculline methiodide in the central medial intralaminar nucleus on behavior and seizures induced by PTZ is described.
There is considerable evidence (Ben-Ari, Tremblay, Riche, Ghilini and Naquet, 1981; Miller, McKeon and Ferrendelli, 1987; Mirski and Ferrendelli, 1984, 1986, 1987; Pazdernik, Cross, Giesler, Samson and Nelson, 1985; Rodin, Kitano, Nagao and Rodin, 1977; Velasco, Velasco, Estrada-Villanueva and Machado, 1975) that subcortical structures, including the thalamus, are critical in the production and expression of seizures induced by systematically administered pentylenetetrazol (PTZ). Recently, it has been demonstrated that a restricted medial thalamic region was particularly important for the regulation of seizures (Miller, Hall, Holland and Ferrendelli, 1989). Discrete injections of the inhibitory agent muscimol in this region result in facilitation of myoclonic and clonic seizures induced by PTZ. These injections also affected behavior, depressing the level of arousal and decreasing spontaneous movement. These effects were only seen in the midline dorsal thalamus, in the vicinity of a number of small nuclei. Subsequently it was shown that the central medial intralaminar nucleus was the critical structure in this region which was responsible for such actions on seizures and arousal (Miller and Ferrendelli, 1990). Muscimol is a direct y-aminobutyric acid (GABA) agonist which acts primarily on the GABA, receptor, but also affects the GABA, subtype of receptor (Bowery, Hill and Hudson, 1983). This suggests that the central medial intralaminar nucleus normally
METHODS Under anesthesia with 2.5% halothane in 100% O,, skull holes were drilled in female SpragueDawley rats (15&l 80 g, Sasco) and I /2 inch lengths of 20 gauge stainless steel guide cannulas were stereotaxically placed so that the tips were just above the surface of the brain, 6 mm above the desired target in the anterior midline thalamus, in the vicinity of the central medial intralaminar nucleus (Fig. I), with coordinates provided by the atlas of Paxinos and 649
650
J. W. MILLERand J. A. FERRENDELLI
Fig. 1. Diagram showing the target zone in the central medial nucleus in the thalamus. For inclusion in the study, the entire blue injection site had to be confined to this zone. Abbreviations are as follows: AM-anteromedial nucleus; CeM-central medial nucleus; IAD-interanterodorsal nucleus; IAM-interanteromedial nucleus; PT-paratenial nucleus; PV-paraventricular nucleus; Re-reuniens; Rhrhomboid nucleus.
Watson (1982). The cannulas were fixed in place with acrylic and occluded with wire. The rats were allowed to recover and placed in individual cages with food and water. The next day, catheters were placed in the right jugular vein under halothane anesthesia, sutured under the skin, flushed with heparin and occluded with wire. The next day, the rats were placed in a painless restraining device (Harvard Bioscience) for 2-3 min, while 1 ~1 Hamilton syringes were threaded through the guide cannulas for injection of piperidine-4-sulfonic acid (Sigma), (-) baclofen (CibaGeigy), pentobarbital (Sigma) or flurazepam (Sigma), dissolved in 0.2 p I of 1% alcian blue normal saline at pH 7.4, or control injections of 1% alcian blue saline. The needle of this syringe was cut to the appropriate length so that the tip would reach the desired target. Bicuculline methiodide (Sigma) was dissolved in acetic acid, titrated to pH 5.0 and alcian blue added to produce a concentration of 1%; control injections for these studies also had a pH of 5.0. After an incubation time of 10 min for the injections of piperidine-4-sulfonic acid or baclofen or 2 min for pentobarbital, flurazepam, or bicuculline, the jugular catheter was flushed with heparin and connected to an infusion pump. A 5 mg/ml solution of PTZ in normal saline was infused continuously, at a rate of 1.44 ml/min and the onset of each type of seizure timed. The threshold for myoclonic seizures was defined as the dose of PTZ per kg, calculated from the time at which the first myoclonic seizure occurred, with a correction for the dead space of the catheter. Similarly, the threshold for clonic and tonic seizures was determined from the time to first occurrence. In this report, “clonic seizures” refers to episodes of clonic forelimb movement, with writhing trunk movements, which usually occur shortly after the first myoclonic seizure. It does not refer to the clonic status epilepticus, with movements of all limbs with the animal on its side, which occurs after the tonic seizure.
In one group, brass skull screws were placed at the time of implanting of cannula. A ground screw was placed in the frontal sinus, as well as either two screws over each hemisphere or one screw over the anterior left hemisphere and another over the posterior right. These screws were wired to an amphenol connector which was fastened with acrylic. Electroencephalograph (EEG) recordings were performed on a Grass polygraph in the awake unrestrained animals after bicuculline or control injections. In some animals an intravenous infusion of PTZ was also performed at the time of the EEG recording. In another concentric bipolar recording electrodes group, (Rhodes, SNEX-100) were implanted stereotaxically in the midline thalamus. The next day recordings were performed, before and after injection of baclofen. In all experiments, the animals were sacrificed with 175 mg/kg intravenous sodium pentobarbital. The brains were removed, frozen, and sectioned at 32 pm on a cryostat. Alternate sections were stained with neutral red or a variant of the Koelle and Friedenwald (1949) acetylcholinesterase stain (Hardy, Heimer, Switzer and Watkins, 1976) for precise location of the injections.
RESULTS Comparison of GABA, and GABA, agonists In these experiments, various concentrations of the GABA, agonist piperidine-4-sulfonic acid and the GABA, agonist baclofen, were injected in the anterior midline thalamus. The injections were directed at a zone centered on the central medial intralaminar nucleus (Fig. I), a zone in which injections of musci mol had their greatest effects on seizures and behavior (Miller et al., 1989). Only animals where histological examination revealed confinement of the blue dye to this zone, and no leakage into the ventricular system, were included in the analysis of data.
651
Thalamic seizure regulation (a) 9Or
I
0 Tonic A Clonic
“<
EEG
Th
PIPERIDINE-4-SULFONICACID
(b) 4o r
(nmd.8)
0 TlXllC
A Clonic l Myoclonlc Behavior
SO ?g 70
16
Th
PSA
5 Fig. 3. Examples of recordings of thalamic and EEG activity of a normal alert rat (above) and rat after the injection of 30 nmol of piperidine-4-sulfonic acid (PSA) or (-)baclofen (Bf) in the central medial intralaminar nucleus (behavioral scores = 4). The thalamic activity was recorded by a concentric bipolar electrode, placed just caudally to the injection site. No infusion of PTZ was performed in these animals.
0
0.1
I
I-1 BACLOFEN
3
IO
30
100
(nmdes)
Fig. 2. Effects of injections of piperidine-4-sulfonic acid (a) and (-)baclofen (b) in the thalamic zone defined in Figure 1, on spontaneous behavior and the threshold doses of pentylenetetrazol (PTZ) for different types of seizures. Behavioral depression scores were assessed prior to infusion of PTZ and defined as follows: 0 = no effect; 1 = mild impairment of righting; 2 = righted in more than 3 seconds; 3 = lay on side without righting; 4 = no spontaneous movements other than normal respiration. Points marked * or ** were significantly different from controls by the StudentNewman-Keuls test (P < 0.05 and P < 0.01, respectively). Control values represent the mean of 16 animals, while each experimental point is the mean of 5-9 animals.
Piperidine-4-sulfonic acid had effects on behavior and the threshold of seizures identical to those of muscimol (Fig. 2a). Smaller doses (3-10 nmol) led to a decrease in spontaneous activity and mild impairment of righting, which progressed at larger doses to complete loss of spontaneous movement with the rat lying on its side. For thalamic injections of piperidine-4-sulfonic acid, as well as other agents, such behavioral effects were assessed prior to the administration of PTZ and graded on the scale described in the legend of Fig. 2. The behavioral effects of piperidine-4-sulfonic acid invariably reached a maximum by 10min; therefore, the infusion of PTZ was performed at that time. Injections of piperidine-4-sulfonic acid resulted in a dose-dependent lowering of the threshold dose of PTZ for the first myoclonic and clonic seizure (Fig. 2a). There was a trend for inhibition of tonic seizures. Least’-squares regression of this threshold, as
a linear function of the log of the dose of piperidine4-sulfonic acid, revealed a slope of +7.44, but this was not significantly different from 0. Baclofen had effects on behavior and the threshold for myoclonic and clonic seizures, identical to those of piperidine-4-sulfonic acid, but with approximately 10 times the potency (Fig. 2b). In addition, however, there was a dose-dependent, significant decrease in the threshold for tonic seizures (slope = - 7.11, significantly different from 0 with P > 0.02, threshold as a function of the log of the dose of baclofen). In order to determine what changes in thalamic electrical activity led to these effects on seizures and behavior, regional thalamic electrophysiological activity was recorded before and after the injection of baclofen and piperidine-4-sulfonic acid in the central medial intralaminar nucleus. Before injection, a rhythmic 5-7 Hz cortical EEG rhythm was seen when these animals were alert and exploring the cage (Fig. 3). The thalamic electrode also revealed such activity, although it was not necessarily synchronous with the cortical EEG. After injection of piperidine-C sulfonic acid or baclofen, the cortical activity was much slower, an effect similar to what has been observed previously with muscimol (Miller et al., 1989). Spindles or other rhythmic activity was not seen. Thalamic recording revealed similar slowing, also without rhythmic activity. Efects
of pentobarbital and jlurazepam
These agents, which act on GABA, receptors, to modify the effects of endogenous GABA, were much less potent than the directly acting agents. Injections of either pentobarbital or flurazepam in the anterior
J. W. MILLER and J. A. FERRENDELLI
652 Table
1.Effects
of injection of bicuculline in the anterior midline thalamus (in the zone defined in Fig. I) on seizures and behavioral changes induced by pentylenetetrazol (PTZ) Threshold
Bicuculline dose (nmol)
N
Myoclonic
0
4
* 6
dose (mg PTZ/kg) Clonic k 6
Tonic
Stimulation
score
76-F_
Behavioral stimulation I = hyperactivity demonstrated by sustained walking; 2 = one episode of violent running or jumping; 3 = recurrent running and jumping; 4 = one clonic seizure; 5 = recurrent clonic seizures; 6 = continuous clonic seizure. **P < 0.01, Bonferroni corrected Mann-Whitney Rank Sum Test.
midline thalamus had effects on spontaneous behavior, prior to infusion of the convulsant, identical to those of piperidine-4-sulfonic acid and baclofen, as well as similar effects on the seizure threshold. Infusion of PTZ was carried out when the behavioral effects were maximal, 2 min after injection. Pentobarbital (1 pmol) produced significant behavioral depression, facilitation of myoclonic seizures and inhibition of tonic seizures (behavioral score 1.8 k 0.5, P < 0.01, Bonferroni corrected Mann-Whitney rank Sum Test; myoclonic threshold - 38 k 11%, P < 0.05, Student-Newman-Keuls test; clonic - 11 + 9%, NS; tonic +31 k IO%, P < 0.01, N = 8). Flurazepam (500 nmol) produced trends towards similar effects on the seizure threshold (behavioral score 1.4f 0.5, P < 0.01; myoclonic - 15 + 14%, NS; clonic - 10 + 17%, NS; tonic +6 + 13%, NS; N = 5).
qf bicuculline Thalamic injections of this GABA, antagonist, at small doses (300 pmol) had behavioral effects opposite to those of GABA agonists (Table I), that is, the animals became hyperactive. However, with larger doses episodic fits of violent running and jumping occurred, starting l-2 min after injection. With even larger doses, this activity became nearly continuous
and eventually changed into frank clonic seizures, with rhythmic movements of all extremities with the animals on their sides (Table 1). In one animal, given a dose of 7 nmol of bicuculhne, a generalized tonic seizure occurred with this type of clonic seizure. The effect of thalamic injections of bicuculline on seizures induced by PTZ was tested at a dose (100 pmol) of bicuculhne which did not itself have behavioral effects (Table 1). No effect on the seizure threshold was seen. The control mean threshold dose for myoclonic and clonic seizures were smaller than those observed in other experiments (Fig. 2) possibly because the solutions for bicuculline and the control injections had a pH of 5.0, since bicuculline is insoluble at neutral pH. Further studies of the effects of injections of bicuculline into the thalamus were done with EEG recording. These revealed no epileptiform activity or other EEG change in these animals (N = 3) during the episodes of running and jumping or at other times (Fig. 5). Recordings were also done during infusions of PTZ, which were performed 2 min after the thalamic injections. After control injections of alcian blue saline (N = 3) infusion of PTZ produced a stereotyped sequence of EEG events (Fig. 4). The initial spikes (mean onset 16 mg/kg) were accompanied by slight tremulousness or arrest of movement but eventually (mean onset 31 mg/kg) myoclonic jerks
Fig. 4. Recording of the EEG during PTZ-induced seizures in rats with control injections of saline in the thalamus. The time and dose of PTZ from the onset of the infusion, are given on the lines above the EEG tracing, while the occurrence of seizures is designated by bars and arrows below the tracings. The tracings on the right show the variable appearance of the EEG during tonic seizures.
Thalamic
seizure
1S.C
& Runs. Junw luc
&
5.c &+
12
MG PTZIKG
14 6
lb
18
8
20 10
PTZ Tow
Fig. 5. Recordings of the EEG of animals injected with 1nmol of bicuculline methiodide (Bc) in the thalamus in the zone defined in Figure I. The top two recordings were during running and jumping induced by bicuculhne; some muscle artifact occurred during these events. Pentylenetetrazol was not infused in these 2 animals. The bottom tracing demonstrates a tonic seizure induced by PTZ, infused 2 min after the injection of bicuculline. Some myogenic artifact is superimposed upon the low amplitude rhythmic EEG activity, which occurred during this seizure. occurred
simultaneously
with
the
spikes.
When
the
in more rapid bursts, the myoclonic movements were accompanied by clonic movements of the forelimbs and writhing of the trunk (mean onset 52 mg/kg), which were not locked in time to the spikes. Eventually there occurred a burst of very rapid spiking, with myoclonus changing into explosive running, followed by a generalized tonic seizure with forelimb, but not hindlimb, extension (mean threshold 82 mg/kg). In some animals, the rapid spiking persisted during the tonic seizure (Fig. 4). It was more common, however, for the high amplitude spiking to be replaced by lower amplitude activity sometime during this tonic phase. Following the tonic seizure, a different type of clonic seizure occurred with the animals lying on their sides, with movements of all limbs synchronous with the EEG spikes. In animals infused with PTZ, after thalamic injections of bicuculline (N = 6) there was a profound qualitative change in the effects of PTZ. A generalized tonic seizure occurred with a very small threshold for PTZ (mean 22 mg/kg) with no preceding myoclonic or clonic seizures in 5 animals. In 4 of these animals, the tonic seizure began with an abrupt change in the EEG, to a pattern of low amplitude activity (Fig. 5) without any preceding or concomitant EEG spikes. spikes
occurred
DISCUSSION
This study demonstrated that the central medial intralaminar nucleus regulates various types of seizures induced by PTZ, and arousal and that these different functions can be modified by drugs which affect GABAergic neurotransmission. It was found that the injection of a variety of inhibitory agents,
regulation
653
including direct GABA, and GABA, agonists and indirect GABA, agonists, in this area depressed arousal and facilitated myoclonic and clonic seizures. It was also found that such injections of direct GABA agonists in fact, suppressed local electrophysiological activity. This suggests that this region of the thalamus normally acts to maintain behavioral arousal and to prevent these types of seizures. On the other hand, effects on tonic seizures depended on the subtype of receptor, with thalamic injection of GABA, agonists having no effect or a trend towards inhibition and GABA, agonists markedly facilitating this type of seizure. In contrast to these effects the GABA, antagonist, bicuculline, induced hyperactivity and spontaneous running seizures which were not accompanied by any cortical EEG discharge. When PTZ was infused concomitantly with thalamic injections of bicuculline, the tonic seizure induced by PTZ was markedly facilitated and occurred without preceding myoclonic or clonic seizures or EEG spikes. These results indicate that there is a pharmacological distinction between mechanisms regulating myoclonic and clonic seizures and those regulating tonic seizures. This dichotomy is not surprising, since many feel that tonic seizures have a fundamentally different neuronal substrate than myoclonic and clonic seizures (Browning and Nelson, 1986; Burnham, 1985; Faingold, 1987; Gale, 1988; Miller et al., 1987). Myoclonic and clonic seizures appear to arise from the cerebral cortex and forebrain; these types of seizures are invariably associated with cortical EEG spikes and they are abolished by precollicular transections of the brainstem (Browning and Nelson, 1986). Tonic seizures, on the other hand, apparently arise in the brainstem, since they can be readily induced by electroshock (Bergmann, Costin and Gutman, 1963; Kreindler, Zuckermann, Steriade and Chimion, 1958) or PTZ (Browning and Nelson, 1986) after transection of the brainstem or even by direct electrical stimulation in the reticular formation (Chiu and Burnham, 1982). These concepts can be expanded to generate a hypothesis (modelled in Fig. 6) to explain many of the results of the present study. The first effect of infusion of PTZ is the appearance of EEG spikes, which arise in the cerebral cortex and presumably also involve closely connected structures, such as thalamic relay and association nuclei. As these spikes become better developed, myoclonic jerks begin to accompany them, presumably generated through corticospinal and corticobulbar pathways. Clonic seizures induced by PTZ may include myoclonic movements, which are simultaneous with the cortical spikes, but primarily consist of movements of the forelimb and trunk which are not, suggesting that subcortical structures may help to generate this behavior. Tonic seizures induced by PTZ are normally foreshadowed by rapid bursts of EEG spikes (Fig. 4); presumably when forebrain structures are sufficiently excited by this convulsant, the volleys of descending excitation
J. W. MILLER and J. A. FERRENDELLI
654
agranular neocortex are possible important routes (Jones and Leavitt, 1974). This model can provide directions for future research. It provides insight into the basic processes which produce different types of seizures and demonstrates how neuroanatomical systems, which mediate normal functions, e.g. arousal, can interact with a pathological process such as production of seizures. It may also be useful for understanding the actions of antiepileptic drugs, since complete characterization of the mechanisms of action of any neurotropic drugs requires definition of their preconvulsant and anticonvulsant effects on different neural systems, in addition to the definition of their actions at a molecular and cellular level. MYOCLONIC
CtONiC
TbNlC
Fig. 6. Diagram summarizing the effect of the central medial intralaminar nucleus (CeM) on the seizure process (see text).
then, in turn, activate brainstem structures, which give rise to the tonic seizures. The period of EEG attenuation which is often seen during the later part of the tonic seizure, would then reflect the effect of the ascending output from the reticular formation and other brainstem structures on the cortex (Fig. 6). Since the central medial intralaminar nucleus exerts different effects on different types of seizures and since injections of excitatory agents in this region do not reproduce PTZ-induced seizures, it cannot be a site where these seizures are initiated or propagated. Rather, it is more likely that it acts indirectly to regulate the activity of other structures, which are more directly involved in the production of seizures (Fig. 6). It acts on the forebrain to prevent PTZinduced spiking and myoclonic and clonic seizures under the inhibitory control of both subtypes of GABA receptor. The central medial intralaminar nucleus also exerts influence on tonic seizure mechanisms in the brainstem but GABA, and GABA, receptors have opposite effects. In the present study the injection of most GABA, agonists into the thalamus produced only slight inhibition of tonic seizures but pentobarbital had a stronger effect and it has previously been shown that more marked inhibition of tonic seizures occurred with muscimol (Miller et al., 1989). The effects of the GABA, antagonist, bicuculline, on seizures induced by PTZ were consistent with this model; it facilitated brainstem mechanisms while inhibiting the forebrain, causing tonic seizures, at a low threshold, in the absence of cortical spikes or myoclonic or clonic seizures (Fig. 5). On the other hand, inhibition of GABA, receptors facilitated tonic seizures. The model illustrated in Fig. 6 is at present only a general schematic model. The exact pathways by which the central medial intralaminar nucleus exerts its different effects in seizures are still not defined, but the efferent connections to the striatum and frontal
Acknowledgements-We would like to thank Claire M. Schoen for technical assistance and Patti Nacci for typing the manuscript. We would also like to thank Jack Baty of the Division of Biostatistics, Washington University, for reviewing the analysis of data. This work was supported by NIH grants NS01296 and NS14834 and the Seay Neuropharmacology Fund. REFERENCES Ben-Ari Y., Tremblay E., Riche D., Ghilini G. and Naquet R. (1981) Electrographic, clinical and pathological alterations following systemic administration of kainic acid, bicuculline or pentetrazol: metabolic mapping using the deoxyglucose method with special reference to the pathology of epilepsy. Neuroscience 6: 1361-139 1. Bergmann F., Costin A. and Gutman J. (1963) A low threshold convulsive area in the rabbit’s mesencephalon. Electroenceph. clin. Neurophysiol. 15: 683490. Bowery N. G., Hill D. R. and Hudson A. L. (1983) Characteristics of GABA, receptor binding sites on rat whole brain synaptic membranes. Br. J. Pharmuc. 78: 191-206. Browning R. A. and Nelson D. K. (1986) Modification of electroshock and pentylenetetrazol seizure patterns in rats after precollicular transection. Expl Neural. 93: 546-556. Burnham W. M. (1985) Core mechanisms in generalized convulsions. Fed. Proc. 44: 2442-2445. Chiu P. and Burnham W. M. (1982) The effects of anticonvulsant drugs on convulsions triggered by direct stimulation of the brainstem. Neuropharmacology 21: 355-359. Faingold C. L. (1987) The role of the brain stem in generalized epileptic seizures. Metabolic Brain Dis. 2: 81-112. Gale K. (1988) Progression and generalization of seizure discharge: Anatomical and neurochemical substrates. Epilepsia 29 Suppl. 2: S15-S34. Hardy H., Heimer L., Switzer R. and Watkins D. (1976) Simultaneous demonstration of horseradish peroxidase and acetylcholinesterase. Neurosci. Lett. 3: 1 -5. Jones E. G. and Leavitt R. Y. (1974) Retrograde axonal transport and the demonstration of nonspecific projections to the cerebral cortex and striatum from thalamic intralaminar nuclei in the rat, cat and monkey. J. con~p. Neural. 154: 349-318. Koelle G. B. and Friedenwald J. S. (1949) A histochemical method of locating cholinesterase activity. Proc. Sot. e.xp, Biol. 70: 617-622. Kreindler A., Zuckermann E., Steriade M. and Chimion D. (1958) Electroclinical features of convulsions induced by stimulation of brain stem. J. Neurophysiol. 21: 430436. Miller J. W. and Ferrendelli J. A. (1990) The central medial nucleus: Thalamic site of seizure regulation. Brain Res. 508: 297-300.
Thalamic seizure regulation Miller J. W., Hall C. M., Holland K. D. and Ferrendelli J. A. (1989) Identification of a median thalamic system regulating seizures and arousal. Epilepsiu 30: 493-300. Miller J. W.. McKeon A. C. and Ferrendelli J. A. (1987) Functional anatomy of pentylenetetrazol and electro: shock seizures in the rat brainstem. Ann. Neural. 22: 615421.
Mirski M. A. and Ferrendelli J. A. (1984) Interruption of the mammillothalamic tracts prevents seizures in guinea pigs. Science 226: 72-74. Mirski M. A. and Ferrendelli J. A. (1986) Selective metabolic activation of the mammillary bodies and their connections during ethosuximide-induced suppression of pentylenetetrazol seizures. Epilepsiu 27: 194-203. Mirski M. A. and Ferrendelli J. A. (1987) Interruption of the connections of the mammillary bodies protects against
generalized
655 pentylenetetrazol
seizures in guinea pigs.
J. Neurosci. 7: 662670.
C. (1982) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York.
Paxinos G. and Watson
Pazdernik T. L., Cross R. S., Giesler M., Samson F. E. and Nelson S. R. (1985) Changes in local cerebral glucose utilization induced by convulsants. Neuroscience 14: 823-835.
Rodin E., Kitano H., Nagao B. and Rodin M. (1977) The results of penicillin G administration on chronic unrestrained cats: Electrographic and behavioral observations Electroenceph. cl&~.ieurophysiol. 301 62-72. F. and Velasco F.. Velasco M.. Estrada-Villanueva Machado’J. P. (1975) Specific and nonspecific multiple unit activities during the onset of pentylenetetrazol seizures I. Intact animals. Epilepsia 16: 207-214.
0028.3908/90$3.00+ 0.00 Copyright 0 1990Pergamon Press plc
CHARACTERIZATION OF GABAERGIC SEIZURE REGULATION IN THE MIDLINE THALAMUS J. W. MILLER and J. A. FERRENDELLI Departments of Neurology and Neurological Surgery (Neurology) and Neuropharmacology, Washington University School of Medicine, 660 South Euclid Ave., St Louis, Missouri 63110, U.S.A. (Accepted 29 Noaember 1989)
Summary-This
study characterized the role of GABA in the central medial intralaminar nucleus on seizures induced by pentylenetetrazol given systemically. Injections of the direct selective GABA, agonist. piperidine-4-sulfonic acid or the indirect GABA, agonists, flurazepam and pentobarbital, in this region depressed arousal and facilitated myoclonic and clonic seizures induced by pentylenetetrazol but only caused slight inhibition of tonic seizures. In contrast the GABA, agonist (-)baclofen facilitated all three types of seizures. Recording after injection of piperidine-4-sulfonic acid and (-)baclofen revealed marked suppression and slowing of thalamic and cortical electrical activity. Thalamic injections of the GABA, antagonist, bicuculline methiodide, had opposite behavioral effects, causing hyperactivity and episodes of violent running, not accompanied by EEG discharges. When pentylenetetrazol was infused concommitantly there was marked facilitation of the tonic seizures, which occurred without preceding myoclonic of clonic seizures, or EEG spikes. These results demonstrate that GABA-mediated neurotransmission in the central medial intralaminar nucleus can control the threshold of seizures and that GABA agonists and antagonists have opposite effects. It is suggested that the central medial intralaminar nucleus is not a site of origination or spread of seizures. but controls seizures indirectly by regulating the excitability of other structures and that different synaptic mechanisms and anatomical connections mediate effects on different types of seizures.
Ke,v words-pentylenetetrazol,
central medial nucleus, GABA, seizure, epilepsy, thalamus, arousal.
functions to alert the animal and to inhibit seizures and the injections of muscimol exert their effect by inhibiting these functions. One purpose of the present study was to verify this hypothesis by providing further evidence that such effects result from inhibition mediated by GABA receptors by determining the effects of thalamic injections of a variety of GABAergic agents, including direct agonists and agents which act by modulating the effects of endogenous GABA. Another aim was to differentiate the roles of GABA, and GABA, receptors on the regulation of thalamic seizures by studying the effects of thalamic injections of agonists which are more selective than muscimol. Finally, the effect of injection of the GABA antagonist bicuculline methiodide in the central medial intralaminar nucleus on behavior and seizures induced by PTZ is described.
There is considerable evidence (Ben-Ari, Tremblay, Riche, Ghilini and Naquet, 1981; Miller, McKeon and Ferrendelli, 1987; Mirski and Ferrendelli, 1984, 1986, 1987; Pazdernik, Cross, Giesler, Samson and Nelson, 1985; Rodin, Kitano, Nagao and Rodin, 1977; Velasco, Velasco, Estrada-Villanueva and Machado, 1975) that subcortical structures, including the thalamus, are critical in the production and expression of seizures induced by systematically administered pentylenetetrazol (PTZ). Recently, it has been demonstrated that a restricted medial thalamic region was particularly important for the regulation of seizures (Miller, Hall, Holland and Ferrendelli, 1989). Discrete injections of the inhibitory agent muscimol in this region result in facilitation of myoclonic and clonic seizures induced by PTZ. These injections also affected behavior, depressing the level of arousal and decreasing spontaneous movement. These effects were only seen in the midline dorsal thalamus, in the vicinity of a number of small nuclei. Subsequently it was shown that the central medial intralaminar nucleus was the critical structure in this region which was responsible for such actions on seizures and arousal (Miller and Ferrendelli, 1990). Muscimol is a direct y-aminobutyric acid (GABA) agonist which acts primarily on the GABA, receptor, but also affects the GABA, subtype of receptor (Bowery, Hill and Hudson, 1983). This suggests that the central medial intralaminar nucleus normally
METHODS Under anesthesia with 2.5% halothane in 100% O,, skull holes were drilled in female SpragueDawley rats (15&l 80 g, Sasco) and I /2 inch lengths of 20 gauge stainless steel guide cannulas were stereotaxically placed so that the tips were just above the surface of the brain, 6 mm above the desired target in the anterior midline thalamus, in the vicinity of the central medial intralaminar nucleus (Fig. I), with coordinates provided by the atlas of Paxinos and 649
650
J. W. MILLERand J. A. FERRENDELLI
Fig. 1. Diagram showing the target zone in the central medial nucleus in the thalamus. For inclusion in the study, the entire blue injection site had to be confined to this zone. Abbreviations are as follows: AM-anteromedial nucleus; CeM-central medial nucleus; IAD-interanterodorsal nucleus; IAM-interanteromedial nucleus; PT-paratenial nucleus; PV-paraventricular nucleus; Re-reuniens; Rhrhomboid nucleus.
Watson (1982). The cannulas were fixed in place with acrylic and occluded with wire. The rats were allowed to recover and placed in individual cages with food and water. The next day, catheters were placed in the right jugular vein under halothane anesthesia, sutured under the skin, flushed with heparin and occluded with wire. The next day, the rats were placed in a painless restraining device (Harvard Bioscience) for 2-3 min, while 1 ~1 Hamilton syringes were threaded through the guide cannulas for injection of piperidine-4-sulfonic acid (Sigma), (-) baclofen (CibaGeigy), pentobarbital (Sigma) or flurazepam (Sigma), dissolved in 0.2 p I of 1% alcian blue normal saline at pH 7.4, or control injections of 1% alcian blue saline. The needle of this syringe was cut to the appropriate length so that the tip would reach the desired target. Bicuculline methiodide (Sigma) was dissolved in acetic acid, titrated to pH 5.0 and alcian blue added to produce a concentration of 1%; control injections for these studies also had a pH of 5.0. After an incubation time of 10 min for the injections of piperidine-4-sulfonic acid or baclofen or 2 min for pentobarbital, flurazepam, or bicuculline, the jugular catheter was flushed with heparin and connected to an infusion pump. A 5 mg/ml solution of PTZ in normal saline was infused continuously, at a rate of 1.44 ml/min and the onset of each type of seizure timed. The threshold for myoclonic seizures was defined as the dose of PTZ per kg, calculated from the time at which the first myoclonic seizure occurred, with a correction for the dead space of the catheter. Similarly, the threshold for clonic and tonic seizures was determined from the time to first occurrence. In this report, “clonic seizures” refers to episodes of clonic forelimb movement, with writhing trunk movements, which usually occur shortly after the first myoclonic seizure. It does not refer to the clonic status epilepticus, with movements of all limbs with the animal on its side, which occurs after the tonic seizure.
In one group, brass skull screws were placed at the time of implanting of cannula. A ground screw was placed in the frontal sinus, as well as either two screws over each hemisphere or one screw over the anterior left hemisphere and another over the posterior right. These screws were wired to an amphenol connector which was fastened with acrylic. Electroencephalograph (EEG) recordings were performed on a Grass polygraph in the awake unrestrained animals after bicuculline or control injections. In some animals an intravenous infusion of PTZ was also performed at the time of the EEG recording. In another concentric bipolar recording electrodes group, (Rhodes, SNEX-100) were implanted stereotaxically in the midline thalamus. The next day recordings were performed, before and after injection of baclofen. In all experiments, the animals were sacrificed with 175 mg/kg intravenous sodium pentobarbital. The brains were removed, frozen, and sectioned at 32 pm on a cryostat. Alternate sections were stained with neutral red or a variant of the Koelle and Friedenwald (1949) acetylcholinesterase stain (Hardy, Heimer, Switzer and Watkins, 1976) for precise location of the injections.
RESULTS Comparison of GABA, and GABA, agonists In these experiments, various concentrations of the GABA, agonist piperidine-4-sulfonic acid and the GABA, agonist baclofen, were injected in the anterior midline thalamus. The injections were directed at a zone centered on the central medial intralaminar nucleus (Fig. I), a zone in which injections of musci mol had their greatest effects on seizures and behavior (Miller et al., 1989). Only animals where histological examination revealed confinement of the blue dye to this zone, and no leakage into the ventricular system, were included in the analysis of data.
651
Thalamic seizure regulation (a) 9Or
I
0 Tonic A Clonic
“<
EEG
Th
PIPERIDINE-4-SULFONICACID
(b) 4o r
(nmd.8)
0 TlXllC
A Clonic l Myoclonlc Behavior
SO ?g 70
16
Th
PSA
5 Fig. 3. Examples of recordings of thalamic and EEG activity of a normal alert rat (above) and rat after the injection of 30 nmol of piperidine-4-sulfonic acid (PSA) or (-)baclofen (Bf) in the central medial intralaminar nucleus (behavioral scores = 4). The thalamic activity was recorded by a concentric bipolar electrode, placed just caudally to the injection site. No infusion of PTZ was performed in these animals.
0
0.1
I
I-1 BACLOFEN
3
IO
30
100
(nmdes)
Fig. 2. Effects of injections of piperidine-4-sulfonic acid (a) and (-)baclofen (b) in the thalamic zone defined in Figure 1, on spontaneous behavior and the threshold doses of pentylenetetrazol (PTZ) for different types of seizures. Behavioral depression scores were assessed prior to infusion of PTZ and defined as follows: 0 = no effect; 1 = mild impairment of righting; 2 = righted in more than 3 seconds; 3 = lay on side without righting; 4 = no spontaneous movements other than normal respiration. Points marked * or ** were significantly different from controls by the StudentNewman-Keuls test (P < 0.05 and P < 0.01, respectively). Control values represent the mean of 16 animals, while each experimental point is the mean of 5-9 animals.
Piperidine-4-sulfonic acid had effects on behavior and the threshold of seizures identical to those of muscimol (Fig. 2a). Smaller doses (3-10 nmol) led to a decrease in spontaneous activity and mild impairment of righting, which progressed at larger doses to complete loss of spontaneous movement with the rat lying on its side. For thalamic injections of piperidine-4-sulfonic acid, as well as other agents, such behavioral effects were assessed prior to the administration of PTZ and graded on the scale described in the legend of Fig. 2. The behavioral effects of piperidine-4-sulfonic acid invariably reached a maximum by 10min; therefore, the infusion of PTZ was performed at that time. Injections of piperidine-4-sulfonic acid resulted in a dose-dependent lowering of the threshold dose of PTZ for the first myoclonic and clonic seizure (Fig. 2a). There was a trend for inhibition of tonic seizures. Least’-squares regression of this threshold, as
a linear function of the log of the dose of piperidine4-sulfonic acid, revealed a slope of +7.44, but this was not significantly different from 0. Baclofen had effects on behavior and the threshold for myoclonic and clonic seizures, identical to those of piperidine-4-sulfonic acid, but with approximately 10 times the potency (Fig. 2b). In addition, however, there was a dose-dependent, significant decrease in the threshold for tonic seizures (slope = - 7.11, significantly different from 0 with P > 0.02, threshold as a function of the log of the dose of baclofen). In order to determine what changes in thalamic electrical activity led to these effects on seizures and behavior, regional thalamic electrophysiological activity was recorded before and after the injection of baclofen and piperidine-4-sulfonic acid in the central medial intralaminar nucleus. Before injection, a rhythmic 5-7 Hz cortical EEG rhythm was seen when these animals were alert and exploring the cage (Fig. 3). The thalamic electrode also revealed such activity, although it was not necessarily synchronous with the cortical EEG. After injection of piperidine-C sulfonic acid or baclofen, the cortical activity was much slower, an effect similar to what has been observed previously with muscimol (Miller et al., 1989). Spindles or other rhythmic activity was not seen. Thalamic recording revealed similar slowing, also without rhythmic activity. Efects
of pentobarbital and jlurazepam
These agents, which act on GABA, receptors, to modify the effects of endogenous GABA, were much less potent than the directly acting agents. Injections of either pentobarbital or flurazepam in the anterior
J. W. MILLER and J. A. FERRENDELLI
652 Table
1.Effects
of injection of bicuculline in the anterior midline thalamus (in the zone defined in Fig. I) on seizures and behavioral changes induced by pentylenetetrazol (PTZ) Threshold
Bicuculline dose (nmol)
N
Myoclonic
0
4
* 6
dose (mg PTZ/kg) Clonic k 6
Tonic
Stimulation
score
76-F_
Behavioral stimulation I = hyperactivity demonstrated by sustained walking; 2 = one episode of violent running or jumping; 3 = recurrent running and jumping; 4 = one clonic seizure; 5 = recurrent clonic seizures; 6 = continuous clonic seizure. **P < 0.01, Bonferroni corrected Mann-Whitney Rank Sum Test.
midline thalamus had effects on spontaneous behavior, prior to infusion of the convulsant, identical to those of piperidine-4-sulfonic acid and baclofen, as well as similar effects on the seizure threshold. Infusion of PTZ was carried out when the behavioral effects were maximal, 2 min after injection. Pentobarbital (1 pmol) produced significant behavioral depression, facilitation of myoclonic seizures and inhibition of tonic seizures (behavioral score 1.8 k 0.5, P < 0.01, Bonferroni corrected Mann-Whitney rank Sum Test; myoclonic threshold - 38 k 11%, P < 0.05, Student-Newman-Keuls test; clonic - 11 + 9%, NS; tonic +31 k IO%, P < 0.01, N = 8). Flurazepam (500 nmol) produced trends towards similar effects on the seizure threshold (behavioral score 1.4f 0.5, P < 0.01; myoclonic - 15 + 14%, NS; clonic - 10 + 17%, NS; tonic +6 + 13%, NS; N = 5).
qf bicuculline Thalamic injections of this GABA, antagonist, at small doses (300 pmol) had behavioral effects opposite to those of GABA agonists (Table I), that is, the animals became hyperactive. However, with larger doses episodic fits of violent running and jumping occurred, starting l-2 min after injection. With even larger doses, this activity became nearly continuous
and eventually changed into frank clonic seizures, with rhythmic movements of all extremities with the animals on their sides (Table 1). In one animal, given a dose of 7 nmol of bicuculhne, a generalized tonic seizure occurred with this type of clonic seizure. The effect of thalamic injections of bicuculline on seizures induced by PTZ was tested at a dose (100 pmol) of bicuculhne which did not itself have behavioral effects (Table 1). No effect on the seizure threshold was seen. The control mean threshold dose for myoclonic and clonic seizures were smaller than those observed in other experiments (Fig. 2) possibly because the solutions for bicuculline and the control injections had a pH of 5.0, since bicuculline is insoluble at neutral pH. Further studies of the effects of injections of bicuculline into the thalamus were done with EEG recording. These revealed no epileptiform activity or other EEG change in these animals (N = 3) during the episodes of running and jumping or at other times (Fig. 5). Recordings were also done during infusions of PTZ, which were performed 2 min after the thalamic injections. After control injections of alcian blue saline (N = 3) infusion of PTZ produced a stereotyped sequence of EEG events (Fig. 4). The initial spikes (mean onset 16 mg/kg) were accompanied by slight tremulousness or arrest of movement but eventually (mean onset 31 mg/kg) myoclonic jerks
Fig. 4. Recording of the EEG during PTZ-induced seizures in rats with control injections of saline in the thalamus. The time and dose of PTZ from the onset of the infusion, are given on the lines above the EEG tracing, while the occurrence of seizures is designated by bars and arrows below the tracings. The tracings on the right show the variable appearance of the EEG during tonic seizures.
Thalamic
seizure
1S.C
& Runs. Junw luc
&
5.c &+
12
MG PTZIKG
14 6
lb
18
8
20 10
PTZ Tow
Fig. 5. Recordings of the EEG of animals injected with 1nmol of bicuculline methiodide (Bc) in the thalamus in the zone defined in Figure I. The top two recordings were during running and jumping induced by bicuculhne; some muscle artifact occurred during these events. Pentylenetetrazol was not infused in these 2 animals. The bottom tracing demonstrates a tonic seizure induced by PTZ, infused 2 min after the injection of bicuculline. Some myogenic artifact is superimposed upon the low amplitude rhythmic EEG activity, which occurred during this seizure. occurred
simultaneously
with
the
spikes.
When
the
in more rapid bursts, the myoclonic movements were accompanied by clonic movements of the forelimbs and writhing of the trunk (mean onset 52 mg/kg), which were not locked in time to the spikes. Eventually there occurred a burst of very rapid spiking, with myoclonus changing into explosive running, followed by a generalized tonic seizure with forelimb, but not hindlimb, extension (mean threshold 82 mg/kg). In some animals, the rapid spiking persisted during the tonic seizure (Fig. 4). It was more common, however, for the high amplitude spiking to be replaced by lower amplitude activity sometime during this tonic phase. Following the tonic seizure, a different type of clonic seizure occurred with the animals lying on their sides, with movements of all limbs synchronous with the EEG spikes. In animals infused with PTZ, after thalamic injections of bicuculline (N = 6) there was a profound qualitative change in the effects of PTZ. A generalized tonic seizure occurred with a very small threshold for PTZ (mean 22 mg/kg) with no preceding myoclonic or clonic seizures in 5 animals. In 4 of these animals, the tonic seizure began with an abrupt change in the EEG, to a pattern of low amplitude activity (Fig. 5) without any preceding or concomitant EEG spikes. spikes
occurred
DISCUSSION
This study demonstrated that the central medial intralaminar nucleus regulates various types of seizures induced by PTZ, and arousal and that these different functions can be modified by drugs which affect GABAergic neurotransmission. It was found that the injection of a variety of inhibitory agents,
regulation
653
including direct GABA, and GABA, agonists and indirect GABA, agonists, in this area depressed arousal and facilitated myoclonic and clonic seizures. It was also found that such injections of direct GABA agonists in fact, suppressed local electrophysiological activity. This suggests that this region of the thalamus normally acts to maintain behavioral arousal and to prevent these types of seizures. On the other hand, effects on tonic seizures depended on the subtype of receptor, with thalamic injection of GABA, agonists having no effect or a trend towards inhibition and GABA, agonists markedly facilitating this type of seizure. In contrast to these effects the GABA, antagonist, bicuculline, induced hyperactivity and spontaneous running seizures which were not accompanied by any cortical EEG discharge. When PTZ was infused concomitantly with thalamic injections of bicuculline, the tonic seizure induced by PTZ was markedly facilitated and occurred without preceding myoclonic or clonic seizures or EEG spikes. These results indicate that there is a pharmacological distinction between mechanisms regulating myoclonic and clonic seizures and those regulating tonic seizures. This dichotomy is not surprising, since many feel that tonic seizures have a fundamentally different neuronal substrate than myoclonic and clonic seizures (Browning and Nelson, 1986; Burnham, 1985; Faingold, 1987; Gale, 1988; Miller et al., 1987). Myoclonic and clonic seizures appear to arise from the cerebral cortex and forebrain; these types of seizures are invariably associated with cortical EEG spikes and they are abolished by precollicular transections of the brainstem (Browning and Nelson, 1986). Tonic seizures, on the other hand, apparently arise in the brainstem, since they can be readily induced by electroshock (Bergmann, Costin and Gutman, 1963; Kreindler, Zuckermann, Steriade and Chimion, 1958) or PTZ (Browning and Nelson, 1986) after transection of the brainstem or even by direct electrical stimulation in the reticular formation (Chiu and Burnham, 1982). These concepts can be expanded to generate a hypothesis (modelled in Fig. 6) to explain many of the results of the present study. The first effect of infusion of PTZ is the appearance of EEG spikes, which arise in the cerebral cortex and presumably also involve closely connected structures, such as thalamic relay and association nuclei. As these spikes become better developed, myoclonic jerks begin to accompany them, presumably generated through corticospinal and corticobulbar pathways. Clonic seizures induced by PTZ may include myoclonic movements, which are simultaneous with the cortical spikes, but primarily consist of movements of the forelimb and trunk which are not, suggesting that subcortical structures may help to generate this behavior. Tonic seizures induced by PTZ are normally foreshadowed by rapid bursts of EEG spikes (Fig. 4); presumably when forebrain structures are sufficiently excited by this convulsant, the volleys of descending excitation
J. W. MILLER and J. A. FERRENDELLI
654
agranular neocortex are possible important routes (Jones and Leavitt, 1974). This model can provide directions for future research. It provides insight into the basic processes which produce different types of seizures and demonstrates how neuroanatomical systems, which mediate normal functions, e.g. arousal, can interact with a pathological process such as production of seizures. It may also be useful for understanding the actions of antiepileptic drugs, since complete characterization of the mechanisms of action of any neurotropic drugs requires definition of their preconvulsant and anticonvulsant effects on different neural systems, in addition to the definition of their actions at a molecular and cellular level. MYOCLONIC
CtONiC
TbNlC
Fig. 6. Diagram summarizing the effect of the central medial intralaminar nucleus (CeM) on the seizure process (see text).
then, in turn, activate brainstem structures, which give rise to the tonic seizures. The period of EEG attenuation which is often seen during the later part of the tonic seizure, would then reflect the effect of the ascending output from the reticular formation and other brainstem structures on the cortex (Fig. 6). Since the central medial intralaminar nucleus exerts different effects on different types of seizures and since injections of excitatory agents in this region do not reproduce PTZ-induced seizures, it cannot be a site where these seizures are initiated or propagated. Rather, it is more likely that it acts indirectly to regulate the activity of other structures, which are more directly involved in the production of seizures (Fig. 6). It acts on the forebrain to prevent PTZinduced spiking and myoclonic and clonic seizures under the inhibitory control of both subtypes of GABA receptor. The central medial intralaminar nucleus also exerts influence on tonic seizure mechanisms in the brainstem but GABA, and GABA, receptors have opposite effects. In the present study the injection of most GABA, agonists into the thalamus produced only slight inhibition of tonic seizures but pentobarbital had a stronger effect and it has previously been shown that more marked inhibition of tonic seizures occurred with muscimol (Miller et al., 1989). The effects of the GABA, antagonist, bicuculline, on seizures induced by PTZ were consistent with this model; it facilitated brainstem mechanisms while inhibiting the forebrain, causing tonic seizures, at a low threshold, in the absence of cortical spikes or myoclonic or clonic seizures (Fig. 5). On the other hand, inhibition of GABA, receptors facilitated tonic seizures. The model illustrated in Fig. 6 is at present only a general schematic model. The exact pathways by which the central medial intralaminar nucleus exerts its different effects in seizures are still not defined, but the efferent connections to the striatum and frontal
Acknowledgements-We would like to thank Claire M. Schoen for technical assistance and Patti Nacci for typing the manuscript. We would also like to thank Jack Baty of the Division of Biostatistics, Washington University, for reviewing the analysis of data. This work was supported by NIH grants NS01296 and NS14834 and the Seay Neuropharmacology Fund. REFERENCES Ben-Ari Y., Tremblay E., Riche D., Ghilini G. and Naquet R. (1981) Electrographic, clinical and pathological alterations following systemic administration of kainic acid, bicuculline or pentetrazol: metabolic mapping using the deoxyglucose method with special reference to the pathology of epilepsy. Neuroscience 6: 1361-139 1. Bergmann F., Costin A. and Gutman J. (1963) A low threshold convulsive area in the rabbit’s mesencephalon. Electroenceph. clin. Neurophysiol. 15: 683490. Bowery N. G., Hill D. R. and Hudson A. L. (1983) Characteristics of GABA, receptor binding sites on rat whole brain synaptic membranes. Br. J. Pharmuc. 78: 191-206. Browning R. A. and Nelson D. K. (1986) Modification of electroshock and pentylenetetrazol seizure patterns in rats after precollicular transection. Expl Neural. 93: 546-556. Burnham W. M. (1985) Core mechanisms in generalized convulsions. Fed. Proc. 44: 2442-2445. Chiu P. and Burnham W. M. (1982) The effects of anticonvulsant drugs on convulsions triggered by direct stimulation of the brainstem. Neuropharmacology 21: 355-359. Faingold C. L. (1987) The role of the brain stem in generalized epileptic seizures. Metabolic Brain Dis. 2: 81-112. Gale K. (1988) Progression and generalization of seizure discharge: Anatomical and neurochemical substrates. Epilepsia 29 Suppl. 2: S15-S34. Hardy H., Heimer L., Switzer R. and Watkins D. (1976) Simultaneous demonstration of horseradish peroxidase and acetylcholinesterase. Neurosci. Lett. 3: 1 -5. Jones E. G. and Leavitt R. Y. (1974) Retrograde axonal transport and the demonstration of nonspecific projections to the cerebral cortex and striatum from thalamic intralaminar nuclei in the rat, cat and monkey. J. con~p. Neural. 154: 349-318. Koelle G. B. and Friedenwald J. S. (1949) A histochemical method of locating cholinesterase activity. Proc. Sot. e.xp, Biol. 70: 617-622. Kreindler A., Zuckermann E., Steriade M. and Chimion D. (1958) Electroclinical features of convulsions induced by stimulation of brain stem. J. Neurophysiol. 21: 430436. Miller J. W. and Ferrendelli J. A. (1990) The central medial nucleus: Thalamic site of seizure regulation. Brain Res. 508: 297-300.
Thalamic seizure regulation Miller J. W., Hall C. M., Holland K. D. and Ferrendelli J. A. (1989) Identification of a median thalamic system regulating seizures and arousal. Epilepsiu 30: 493-300. Miller J. W.. McKeon A. C. and Ferrendelli J. A. (1987) Functional anatomy of pentylenetetrazol and electro: shock seizures in the rat brainstem. Ann. Neural. 22: 615421.
Mirski M. A. and Ferrendelli J. A. (1984) Interruption of the mammillothalamic tracts prevents seizures in guinea pigs. Science 226: 72-74. Mirski M. A. and Ferrendelli J. A. (1986) Selective metabolic activation of the mammillary bodies and their connections during ethosuximide-induced suppression of pentylenetetrazol seizures. Epilepsiu 27: 194-203. Mirski M. A. and Ferrendelli J. A. (1987) Interruption of the connections of the mammillary bodies protects against
generalized
655 pentylenetetrazol
seizures in guinea pigs.
J. Neurosci. 7: 662670.
C. (1982) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York.
Paxinos G. and Watson
Pazdernik T. L., Cross R. S., Giesler M., Samson F. E. and Nelson S. R. (1985) Changes in local cerebral glucose utilization induced by convulsants. Neuroscience 14: 823-835.
Rodin E., Kitano H., Nagao B. and Rodin M. (1977) The results of penicillin G administration on chronic unrestrained cats: Electrographic and behavioral observations Electroenceph. cl&~.ieurophysiol. 301 62-72. F. and Velasco F.. Velasco M.. Estrada-Villanueva Machado’J. P. (1975) Specific and nonspecific multiple unit activities during the onset of pentylenetetrazol seizures I. Intact animals. Epilepsia 16: 207-214.