- Email: [email protected]
Comparison of effects of bilateral injections of bicuculline and muscimol into the caudate-putamen of amygdaloid-kindled rats
ELSEVIER
Neuroscience Letters 169 (1994) 129-132
NEUROSCIENC[ LETT[RS
Comparison of effects of bilateral injections of bicuculline and muscimol into the caudate-putamen of amygdaloid-kindled rats Norio Mori*, Masaru Watanabe Department q/'Neuropsychiatt3~. Fukushima Medical College, 1 Hikariga-oka, [~)tkushima-shi 960-12. JapaH Received 24 May 1993: Revised version received 12 December 1993: Accepted 14 January 1994
Abstract Bicuculline is an antagonist of ig-aminobutyric acid (GABA) receptors, and muscimol is an agonist of GABA receptors. In this study, the effects of bilateral injections of bicuculline and muscimol into the caudate-putamen (CP) were compared in amygdaloidkindled rats. Thirty minutes after the injection of bicuculline (1, 10 and 100 pmol per CP) or muscimol (10, 50 and 100 nmol per CP), the kindled amygdala was stimulated at the previously established generalized seizure triggering threshold (GST). Most doses of bicuculline caused no significant alteration either in the seizure stage or in the afterdischarge duration. Only the 100-pmol dose produced a marked reduction in the afterdischarge duration. With 10 nmol of muscimol, there was no significant change in the kindled seizure stage or in the afterdischarge duration, However, 50 and 100 nmol of muscimol markedly suppressed both parameters. These findings suggest that CP efferent pathways are involved in the mechanism that underlies the development of kindled amygdaloid seizures, and support the concept that GABA acts as an anticonvulsant in the brain.
Key words. y-Aminobutyric acid: Bicuculline; Muscimol: Amygdala; Kindling: Caudate-putamen
y-Aminobutyric acid (GABA) is a major inhibitory amino acid in the mammalian brain [2]. Studies of the role of G A B A in human epilepsy and in animal models of epilepsy have suggested that augmentation of G A B A neurotransmission may be used to control epileptic seizures [2]. Indeed, m a n y anticonvulsant drugs appear to work by increasing the levels of activity in G A B A systems [6,13]. Furthermore, the G A B A receptor agonist muscimol, when administered systemically or intracerebrally, shows a strong anticonvulsant activity in various animal models of epilepsy [2,14]. In recent years, evidence has indicated that the pars reticulata of the substantia nigra (RSN) is a key site in GABA-mediated seizure protection [7,14]. On the other hand, it has also been suggested that G A B A disinhibition could be the mechanism responsible for the development of epileptic seizures. For instance, the G A B A receptor antagonist bicuculline acts as a po-
* Corresponding author. 0304-3940/94/$7.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(94)00051-B
tent convulsant when administered systemically or intracerebrally [2,10]. However, in the caudate-putamen (CP), a subconvulsive dose of bicuculline appears to protect against kindled amygdaloid (AM) seizures in rats [4]. This paradoxical anticonvulsant effect may be explained by enhanced GABA-mediated inhibition within the RSN: intra-CP bicuculline produces a marked increase in the R S N G A B A concentration [8]. This effect arises from the activation of GABAergic efferent projections to the R S N [3,5,8]. However, it is unclear whether the reverse pharmacological manipulation in the CP would result in a convulsant reaction. Information in this regard is necessary for understanding the therapy of epileptic seizures. As the augmentation of G A B A neurotransmission in the brain is regarded as a reliable strategy for seizure suppression [2,15], we believed that it would be of interest to compare the effects of bilateral intra-CP injection of bicuculline and muscimol on kindled AM seizures in rats. Forty-nine male Wistar rats, weighing 200 250 g, were used. The animals were anesthetized with pentobarbital (50 mg/kg, i.p.), and a bipolar electrode, made of twisted
130
~r~". .'lIorL M. lI'atom~bc/Neuro,~'ti'm'c /.citers /6¢' ! I~<)4 , 1~~ 9 I3.3
A 24 h before bicuculline ,,..
,~.
30 mln after blcuculllne L . A M . 7 ~ / ~
GSl"
-,,,-.-*
no atnlle
.,...J 1rllv 2ale¢
B
ZSI~
Fig. 1. EEG samples from bicuculline- (A) and muscimol- (B) group animals, A: in an animal which received 100 pmol of bicuculline into the bilateral caudate-putamen, afterdischarge was totally blocked when the kindled amygdala was stimulated at the previously established generalized seizure triggering threshold (GST) 30 rain after injection. However, re-stimulation at 40 eta above the GST regenerated a kindled Stage S seizure. B: in an animal given 100 nmol of muscimol into the bilateral caudate-putamen, afterdischarge was readily produced at the GST 30 rain after injection. However, a kindled Stage 5 seizure regressed to a Stage 3 seizure. Numbers in the top left corners refer to time (in seconds) after cessation of amygdaloid stimulation, L-AM, left amygdala; L- and R-CX, left and right sensorimotor cortex,
stainless steel wire 200 p m in diameter, was implanted into the left AM. Extradural screw electrodes were bilaterally placed over the sensorimotor cortex. A 23gauge cannula was implanted bilaterally into the CP. One week after surgery, the current intensity needed to produce afterdischarge (AD) was determined at the left AM: the AM was stimulated bipolarly, beginning at 40 pA; the intensity was increased in 40 eta increments at 10-rain intervals, until AD was elicited. The last intensity producing AD was designated the AD threshold, and the AM was stimulated each day at the AD threshold with a 1-s constant current 60-Hz square wave. EEG was monitored every day, and the AD duration was determined by measuring the length of electrographic seizures. The AD was regarded as beginning with the cessation of the A M stimulation. The pattern of clinical seizure development was classified into five stages [18]. All animals were kindled at the left AM until a stable Stage 5 seizure was evoked for five successive days. After the completion of AM kindling, the stimulus intensity was reduced every day in 40-pA decrements, and the last intensity required to induce a Stage 5 seizure was designated the generalized seizure triggering threshold (GST). The kindled animals were divided into 3 groups, which
received vehicle (n = 71, bicuculline (n = 20), and muscitool (n = 22). In their original paper, Cavalheiro et al. [4] showed that bilateral intra-CP injections of 1 pmol bicuculline had an anticonvulsant effect in AM-kindled rats. In a preliminary experiment, we found that bicucuUine doses below 100 pmot injected into the CP did not produce any electroclinical ictal response. Therefore. in this study, we tested 1-, 10-, and 100-pmol doses of bicuculline. Bicuculline methiodide (molecular weight 509.3; Sigma) was dissolved in saline to give 0.5-pl volumes containing 1, 10, and 100 pmol. The bicucullinegroup animals received bilateral injections of 0.5 ul of bicuculline into the CP, at 1 (n = 6), 10 (n = 7), and 100 pmol (n = 7) per side. Muscimol (molecular weight 114.1; Sigma) was dissolved in saline to give 0.5-/.tl volumes containing 10, 50, and 100 nmol. The muscimolgroup animals received bilateral injections of 0.5/~1 of muscimol into the CP, at 10 In = 9), 50 (n = 6), and 100 nmol (n = 7) per side. The vehicle-group animals received bilateral intra-CP injections of 0.5 ¢tl saline. Thirty minutes after the injections, all the animals were snmu[ated at the previously established GST. If AD was not produced, the stimulus intensity was increased in 40-¢tA increments every 10 rnin until AD was elicited. The injec-
N. Mori, M. Watanabe/Neuroseienee Letters 169 (1994) 12~) 132
tion of saline alone did not alter the behavior of the animals or the effect of AM stimulation. Upon completion of the experiment, the animals were deeply anesthetized, and their brains were perfused, serially sectioned, and then stained with Cresyl violet. All electrode tips were confirmed to be in the intended structures. The injection of bicuculline caused no identifiable EEG or behavioral change. When the kindled AM was stimulated at the GST 30 rain after the injection, AD generation was completely suppressed (Fig. 1A). However, stimulation at 40-80/~A above the GST readily generated AD and was accompanied by a Stage 5 seizure. This elevation of the GST occurred in 1 of the 6 animals given 1 pmol, 2 of the 7 animals given 10 pmol, and 2 of the 7 animals given 100 pmol. Generally, however, bicuculline did not affect the mean seizure stage, mean AD duration, or mean GST (or the mean AD threshold) (Table 1). The only exception was a slightly shortened AD duration after the injection of 100 pmol bicuculline. Twenty-four hours after treatment, all animals given bicuculline responded to GST stimulation with kindled Stage 5 seizures. The injection of muscimol caused no EEG or behavioral changes. When the kindled AM was stimulated at the GST, AD was readily produced in all the animals (Fig. I B). However, a regression of kindled seizures to earlier Stages 1 3 was observed in 5 of the 9 animals given 10 nmol, 5 of the 6 animals given 50 nmol, and 5 of the 7 animals given 100 nmol. The mean seizure stages after the injection of 50 and 100 nmol of muscimol were significantly smaller than before treatment (P < 0.05 fi)r 50 nmol; P < 0.01 for 100 nmol) (Table 1). Significant reductions in AD duration were also observed after injections of 50 and 100 nmol of muscimol (P < 0.05 for 50 and 100 nmol) (Table 1). Twenty-four hours after treatment, all the animals given muscimol responded to GST stimulation with kindled seizures.
131
In the present study, the bicuculline dose that produced a reduction in AD duration was 100 pmol. This was much higher than the original dose (1 pmol) used by Cavalheiro et al. [4]. Although the reasons for the differences in effective dose are not known, these could have been related to the differences in experimental procedures between the two studies. In the study of Cavalheiro et al. [4], daily AM stimulation, consisting of a 2-s constant-current 60-Hz square wave, was repeated until kindled seizures were evoked for 3 successive days. In this study, however, a 1-s constant-current 60-Hz square wave was used, and the development of 5 consecutive kindled seizures was regarded as the completion of kindling. Unlike the original study protocol [4], in which the kindled AM was stimulated 15 rain after intra-CP bicuculline, we delivered AM stimulation 30 rain after intra-CP bicuculline. These differences may have contributed to the discrepancy in the doses producing a reduction in AD duration. The CP sends GABAergic projections to the RSN [3,8] and inhibits RSN neuronal activity [5]. Increased firing of CP neurons, induced by the local application of electrical pulses or glutamate, induces a silencing of RSN neurons [5]. In addition, intra-CP bicuculline is known to cause an increase in GABA accumulation within the RSN[8], and this may be involved in GABA-mediated anticonvulsant activity [7,14]. Therefore, the reduction in AD duration after an intra-CP injection of bicuculline may be due, at least in part, to the enhancement of GABA-mediated inhibition in the RSN. Unlike intra-CP bicuculline, however, intra-CP muscimol causes a decrease in GABA concentration in the RSN[8], presumably by diminishing the activity of efferent GABAergic neurons projecting to the RSN [3,8]. Nevertheless, intraCP muscimol showed a strong anticonvulsant effect in AM-kindled rats. As judged by the intact AD generation at the GST, intra-CP muscimol suppressed kindled seizures without affecting AM excitability. This suggests
Table 1 Effects of bilateral injections of bicuculline and muscimol into the c a u d a t e - p u t a m e n in amygdaloid-kindled rats Treatment
24 h before treatment
30 min after treatment
SS
ADD
GST
SS
ADD
GST/ADT
Bieueulline 1 pmol (it = 6) 10 pmol (n = 7) 100 pmol In = 7)
5.0 +_ 0 5.0 -+ 0 5.0 + 0
65.7 _+ 24.7 72.0 _+ 18.5 64.6 _+ 18.0
60.0 -+ 21.9 62.9 +_ 21.4 68.6 -+ 19.5
4.2 + 2.0 3.6 _+ 2.0 3.6 _+ 2.0
54.7 _+ 38.0 52.0 _+ 38.9 40.6 _+ 32.3*
66.7 _+ 20.7 68.6 _+ 30.2 80.0 + 32.7
Muse#no~ 10 nmol In = 9) 50 nmol tn = 6) 100 nmol tn = 7)
5.0 _+ 0 5.0 -+ 0 5.0 _+ 0
80.0 + 19.4 81.3 _+ 32.5 91.4 _+ 32.0
57.8 _+ 21.1 46.7 -+ 7.1) 51.4 + 19.5
3.7 -+ 1.4 3.2 _+ 1.2" 3.0 +_ 2.3**
72.[) + 22.0 52.7 _+ 10.0" 46.3 _+ 16,0"
57.8 +_ 21.1 46.7 +_ 7.0 51.4 +_ 19.5
Values are m e a n _+ S.D. SS, seizure stage: A D D , afterdischarge duration (s): GST, generalized seizure triggering threshold (1/A): ADT, afterdischarge threshold (¢tA). *P < 0.05: **P < 0.001, compared to the corresponding values obtained 24 h before treatment (Student's t-test).
132
N. Mori. M. I~ittanahe! Neuroscience Letters 169 (1994) 129-132
that muscimol in the CP interferes with the mechanism responsible for the expression of the motor manifestations of kindled AM seizures. It has been shown that the CP receives glutamatergic projections from the cerebral cortex, and this may enhance the CP neuronal activity [3,8]. Although no direct CP projections to the cerebral cortex have been identified, it appears that the CP modulates the function of the cerebral cortex through multiple synaptic pathways. Indeed, lesions of the CP cause an alteration in glucose utilization within the cerebral cortex [1,16]. Inhibitory effects of the CP on cortical spiking activity have also been suggested [17]. On the other hand, the thalamus, especially the intralaminar nuclei, sends high density projections to the CP [3,20]. This pathway appears to be mainly excitatory [20]. The CP does not send direct efferent to the thalamus, but functional alteration in the CP causes a change in the activity of the thalamus, presumably via polysynaptic pathways [1,16]. Furthermore, findings of reciprocal pathways between the thalamus and cortex, particularly the demonstration of a monosynaptic thalamocortical-corticothalamic feedback loop at electron microscopic and intracellular levels, provide structural arrangements that favor the reverberation of rhythmic, spindle-related clustered activity in both directions [11,20]. Since both the cerebral cortex and the thalamus are necessary for the expression of the motor manifestations seen in kindled AM seizures [9,21,22], we speculate that disruption of the linkages among the CP, thalamus, and cerebral cortex is responsible for the suppression of kindled seizures after the intra-CP injection of muscimol. A previous study has shown that a strong positive transfer occurs in the CP after the completion of AMkindling in cats [19]. In addition, a biochemical study has revealed a significant reduction in GABA synthesis and a significant increase in the numbers of GABA receptors in the CP of AM-kindled rats [12]. These findings are compatible with the anticonvulsant effect of intra-CP muscimol on kindled AM seizures shown here. This study, therefore, supports the concept that GABA acts as an anticonvulsant in the brain [2,15]. [1] Aiko, Y., Hosokawa, S., Shima, F., Kato, M. and Kitamura, K., Alterations in local cerebral glucose utilization during electrical stimulation of the striatum and globus pallidus in rats, Brain Res., 442 (1988) 43 52. [2] Bradford, H.F. and Dodd, ER., Biochemistry and basic mechanisms in epilepsy. In A.N. Davison (Ed.), Biochemistry and Neurological Disease, Blackwell Press, Oxford, 1976, pp. 114-168. [3] Carpenter, M.B., Interactions between the corpus striatum and brainstem nuclei. In J.S. McKenzie, R.E. Kemm and L.N. Wilcock (Eds.), The Basal Ganglia, Plenum Press, New York, 1984, pp. 1 68.
[4] Cavalheiro, E.A., Bortolotto, Z.A. and Turski, L., Microinjections of the ?'-aminobutyric acid antagonist, bicuculline methiodide, into the caudate-putamen prevent amygdala kindled seizures in rats, Brain Res., 411 (1987) 270272. [5] Chavalier, G. and Deniau, J.M., Disinhibition as a basic process in the expression of striatal function, Trends Neurosci., 13 (1990) 277 280. [6] Costa, A., Guidotti, A. and Toffano, G., Molecular mechanisms mediating the action of diazepam on GABA receptors, Br. J. Psychiatry, 133 (1978) 239-248. [7] Gale, K., Mechanisms of seizure control mediated by y-aminobutyric acid: role of the substantia nigra, Fed. Proc., 44 (1985) 24142424. [8] Gale, K. and Casu, M., Dynamic utilization of GABA in the substantia nigra: regulation by dopamine and GABA in the striaturn, and its clinical and behavioral implications, Mol. Cell. Biochem., 39 (1981) 369J,05. [9] Hiyoshi, T. and Wada, J.A., Midline thalamic lesion placement prior to kindling, Electroencephalogr. Clin. Neurophysiol., 70 (1989) 325-338. [10] Kalichman, M.W., Pharmacological investigation of convulsant y-aminobutyric acid (GABA) antagonists in amygdata-kindled rats, Epilepsia, 23 (1982) 163 171. [11] Kostopoulos, G. and Massino, A., Enhanced response of cortical neurons to the thalamic stimuli precedes the appearance of spike and wave discharges in feline generalized penicillin epilepsy, Brain Res., 278 (1983) 207 217. [12] L6sher, W. and Schwark, W., Further evidence for abnormal GABAergic circuits in amygdala-kindled rats, Brain Res., 420 (1987) 385 390. [13] McDonald, R. and Bergey, G.K., Valproic acid augments GABAmediated postsynaptic inhibition in cultured mammalian neurons, Brain Res., 170 (1979) 558 562. [14] McNamara, J.O., Galloway, M.T., Rigsbee, L.C. and Shin, C., Evidence implicating substantia nigra in regulation of kindled seizure threshold, J. Neurosci., 4 (1984) 2410-2417, [15] Meldrum, B.S., Amino acid transmitters and new approaches to anticonvulsant drug action, Epilepsia, 25 (1984) 379-393. [16] Paul, A.T.K. and McCulloch, J., Cerebral glucose utilization following striatal lesions: the effects of GABA agonist, muscimol, and the dopaminergic agonist, apomorphine, Brain Res., 425 (1987) 290 300. [17] Psatta, D.M., Control of chronic experimental focal epilepsy by feedback caudatumstimulation, Epilepsia, 24 (1983) 444M54. [18] Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroencephalogr. Clin. Neurophysiol., 32 (1972) 281 294. [19] Sato, M., Functional changes in the caudate and accumbens nuclei during amygdaloid and hippocampal seizure development in kindled cats, Folia Psychiat. Neurol. Jap., 31 (1977) 501-512. [20] Steriade, M. and Deschens, M., The thalamus as a neuronal oscillator, Brain Res. Rev., 8 (1984) 1 63. [21] Tanaka, A., Progressive changes of behavioral and electro-encephalographic responses to daily amygdaloid stimulation in rabbits, Fukuoka Acta Med., 63 (1972) 152-164. [22] Wada, J.A. and Komai, S., The effect of the anterior 2/3 callosal bisection upon bisymmetrical and bisynchronous generalized convulsions kindled from the amygdala in epileptic baboon, Papio papio. In G. Reeves (Ed.), Epilepsy and The Corpus Callosum, Plenum Press, New York, 1985, pp. 75 97.
Neuroscience Letters 169 (1994) 129-132
NEUROSCIENC[ LETT[RS
Comparison of effects of bilateral injections of bicuculline and muscimol into the caudate-putamen of amygdaloid-kindled rats Norio Mori*, Masaru Watanabe Department q/'Neuropsychiatt3~. Fukushima Medical College, 1 Hikariga-oka, [~)tkushima-shi 960-12. JapaH Received 24 May 1993: Revised version received 12 December 1993: Accepted 14 January 1994
Abstract Bicuculline is an antagonist of ig-aminobutyric acid (GABA) receptors, and muscimol is an agonist of GABA receptors. In this study, the effects of bilateral injections of bicuculline and muscimol into the caudate-putamen (CP) were compared in amygdaloidkindled rats. Thirty minutes after the injection of bicuculline (1, 10 and 100 pmol per CP) or muscimol (10, 50 and 100 nmol per CP), the kindled amygdala was stimulated at the previously established generalized seizure triggering threshold (GST). Most doses of bicuculline caused no significant alteration either in the seizure stage or in the afterdischarge duration. Only the 100-pmol dose produced a marked reduction in the afterdischarge duration. With 10 nmol of muscimol, there was no significant change in the kindled seizure stage or in the afterdischarge duration, However, 50 and 100 nmol of muscimol markedly suppressed both parameters. These findings suggest that CP efferent pathways are involved in the mechanism that underlies the development of kindled amygdaloid seizures, and support the concept that GABA acts as an anticonvulsant in the brain.
Key words. y-Aminobutyric acid: Bicuculline; Muscimol: Amygdala; Kindling: Caudate-putamen
y-Aminobutyric acid (GABA) is a major inhibitory amino acid in the mammalian brain [2]. Studies of the role of G A B A in human epilepsy and in animal models of epilepsy have suggested that augmentation of G A B A neurotransmission may be used to control epileptic seizures [2]. Indeed, m a n y anticonvulsant drugs appear to work by increasing the levels of activity in G A B A systems [6,13]. Furthermore, the G A B A receptor agonist muscimol, when administered systemically or intracerebrally, shows a strong anticonvulsant activity in various animal models of epilepsy [2,14]. In recent years, evidence has indicated that the pars reticulata of the substantia nigra (RSN) is a key site in GABA-mediated seizure protection [7,14]. On the other hand, it has also been suggested that G A B A disinhibition could be the mechanism responsible for the development of epileptic seizures. For instance, the G A B A receptor antagonist bicuculline acts as a po-
* Corresponding author. 0304-3940/94/$7.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(94)00051-B
tent convulsant when administered systemically or intracerebrally [2,10]. However, in the caudate-putamen (CP), a subconvulsive dose of bicuculline appears to protect against kindled amygdaloid (AM) seizures in rats [4]. This paradoxical anticonvulsant effect may be explained by enhanced GABA-mediated inhibition within the RSN: intra-CP bicuculline produces a marked increase in the R S N G A B A concentration [8]. This effect arises from the activation of GABAergic efferent projections to the R S N [3,5,8]. However, it is unclear whether the reverse pharmacological manipulation in the CP would result in a convulsant reaction. Information in this regard is necessary for understanding the therapy of epileptic seizures. As the augmentation of G A B A neurotransmission in the brain is regarded as a reliable strategy for seizure suppression [2,15], we believed that it would be of interest to compare the effects of bilateral intra-CP injection of bicuculline and muscimol on kindled AM seizures in rats. Forty-nine male Wistar rats, weighing 200 250 g, were used. The animals were anesthetized with pentobarbital (50 mg/kg, i.p.), and a bipolar electrode, made of twisted
130
~r~". .'lIorL M. lI'atom~bc/Neuro,~'ti'm'c /.citers /6¢' ! I~<)4 , 1~~ 9 I3.3
A 24 h before bicuculline ,,..
,~.
30 mln after blcuculllne L . A M . 7 ~ / ~
GSl"
-,,,-.-*
no atnlle
.,...J 1rllv 2ale¢
B
ZSI~
Fig. 1. EEG samples from bicuculline- (A) and muscimol- (B) group animals, A: in an animal which received 100 pmol of bicuculline into the bilateral caudate-putamen, afterdischarge was totally blocked when the kindled amygdala was stimulated at the previously established generalized seizure triggering threshold (GST) 30 rain after injection. However, re-stimulation at 40 eta above the GST regenerated a kindled Stage S seizure. B: in an animal given 100 nmol of muscimol into the bilateral caudate-putamen, afterdischarge was readily produced at the GST 30 rain after injection. However, a kindled Stage 5 seizure regressed to a Stage 3 seizure. Numbers in the top left corners refer to time (in seconds) after cessation of amygdaloid stimulation, L-AM, left amygdala; L- and R-CX, left and right sensorimotor cortex,
stainless steel wire 200 p m in diameter, was implanted into the left AM. Extradural screw electrodes were bilaterally placed over the sensorimotor cortex. A 23gauge cannula was implanted bilaterally into the CP. One week after surgery, the current intensity needed to produce afterdischarge (AD) was determined at the left AM: the AM was stimulated bipolarly, beginning at 40 pA; the intensity was increased in 40 eta increments at 10-rain intervals, until AD was elicited. The last intensity producing AD was designated the AD threshold, and the AM was stimulated each day at the AD threshold with a 1-s constant current 60-Hz square wave. EEG was monitored every day, and the AD duration was determined by measuring the length of electrographic seizures. The AD was regarded as beginning with the cessation of the A M stimulation. The pattern of clinical seizure development was classified into five stages [18]. All animals were kindled at the left AM until a stable Stage 5 seizure was evoked for five successive days. After the completion of AM kindling, the stimulus intensity was reduced every day in 40-pA decrements, and the last intensity required to induce a Stage 5 seizure was designated the generalized seizure triggering threshold (GST). The kindled animals were divided into 3 groups, which
received vehicle (n = 71, bicuculline (n = 20), and muscitool (n = 22). In their original paper, Cavalheiro et al. [4] showed that bilateral intra-CP injections of 1 pmol bicuculline had an anticonvulsant effect in AM-kindled rats. In a preliminary experiment, we found that bicucuUine doses below 100 pmot injected into the CP did not produce any electroclinical ictal response. Therefore. in this study, we tested 1-, 10-, and 100-pmol doses of bicuculline. Bicuculline methiodide (molecular weight 509.3; Sigma) was dissolved in saline to give 0.5-pl volumes containing 1, 10, and 100 pmol. The bicucullinegroup animals received bilateral injections of 0.5 ul of bicuculline into the CP, at 1 (n = 6), 10 (n = 7), and 100 pmol (n = 7) per side. Muscimol (molecular weight 114.1; Sigma) was dissolved in saline to give 0.5-/.tl volumes containing 10, 50, and 100 nmol. The muscimolgroup animals received bilateral injections of 0.5/~1 of muscimol into the CP, at 10 In = 9), 50 (n = 6), and 100 nmol (n = 7) per side. The vehicle-group animals received bilateral intra-CP injections of 0.5 ¢tl saline. Thirty minutes after the injections, all the animals were snmu[ated at the previously established GST. If AD was not produced, the stimulus intensity was increased in 40-¢tA increments every 10 rnin until AD was elicited. The injec-
N. Mori, M. Watanabe/Neuroseienee Letters 169 (1994) 12~) 132
tion of saline alone did not alter the behavior of the animals or the effect of AM stimulation. Upon completion of the experiment, the animals were deeply anesthetized, and their brains were perfused, serially sectioned, and then stained with Cresyl violet. All electrode tips were confirmed to be in the intended structures. The injection of bicuculline caused no identifiable EEG or behavioral change. When the kindled AM was stimulated at the GST 30 rain after the injection, AD generation was completely suppressed (Fig. 1A). However, stimulation at 40-80/~A above the GST readily generated AD and was accompanied by a Stage 5 seizure. This elevation of the GST occurred in 1 of the 6 animals given 1 pmol, 2 of the 7 animals given 10 pmol, and 2 of the 7 animals given 100 pmol. Generally, however, bicuculline did not affect the mean seizure stage, mean AD duration, or mean GST (or the mean AD threshold) (Table 1). The only exception was a slightly shortened AD duration after the injection of 100 pmol bicuculline. Twenty-four hours after treatment, all animals given bicuculline responded to GST stimulation with kindled Stage 5 seizures. The injection of muscimol caused no EEG or behavioral changes. When the kindled AM was stimulated at the GST, AD was readily produced in all the animals (Fig. I B). However, a regression of kindled seizures to earlier Stages 1 3 was observed in 5 of the 9 animals given 10 nmol, 5 of the 6 animals given 50 nmol, and 5 of the 7 animals given 100 nmol. The mean seizure stages after the injection of 50 and 100 nmol of muscimol were significantly smaller than before treatment (P < 0.05 fi)r 50 nmol; P < 0.01 for 100 nmol) (Table 1). Significant reductions in AD duration were also observed after injections of 50 and 100 nmol of muscimol (P < 0.05 for 50 and 100 nmol) (Table 1). Twenty-four hours after treatment, all the animals given muscimol responded to GST stimulation with kindled seizures.
131
In the present study, the bicuculline dose that produced a reduction in AD duration was 100 pmol. This was much higher than the original dose (1 pmol) used by Cavalheiro et al. [4]. Although the reasons for the differences in effective dose are not known, these could have been related to the differences in experimental procedures between the two studies. In the study of Cavalheiro et al. [4], daily AM stimulation, consisting of a 2-s constant-current 60-Hz square wave, was repeated until kindled seizures were evoked for 3 successive days. In this study, however, a 1-s constant-current 60-Hz square wave was used, and the development of 5 consecutive kindled seizures was regarded as the completion of kindling. Unlike the original study protocol [4], in which the kindled AM was stimulated 15 rain after intra-CP bicuculline, we delivered AM stimulation 30 rain after intra-CP bicuculline. These differences may have contributed to the discrepancy in the doses producing a reduction in AD duration. The CP sends GABAergic projections to the RSN [3,8] and inhibits RSN neuronal activity [5]. Increased firing of CP neurons, induced by the local application of electrical pulses or glutamate, induces a silencing of RSN neurons [5]. In addition, intra-CP bicuculline is known to cause an increase in GABA accumulation within the RSN[8], and this may be involved in GABA-mediated anticonvulsant activity [7,14]. Therefore, the reduction in AD duration after an intra-CP injection of bicuculline may be due, at least in part, to the enhancement of GABA-mediated inhibition in the RSN. Unlike intra-CP bicuculline, however, intra-CP muscimol causes a decrease in GABA concentration in the RSN[8], presumably by diminishing the activity of efferent GABAergic neurons projecting to the RSN [3,8]. Nevertheless, intraCP muscimol showed a strong anticonvulsant effect in AM-kindled rats. As judged by the intact AD generation at the GST, intra-CP muscimol suppressed kindled seizures without affecting AM excitability. This suggests
Table 1 Effects of bilateral injections of bicuculline and muscimol into the c a u d a t e - p u t a m e n in amygdaloid-kindled rats Treatment
24 h before treatment
30 min after treatment
SS
ADD
GST
SS
ADD
GST/ADT
Bieueulline 1 pmol (it = 6) 10 pmol (n = 7) 100 pmol In = 7)
5.0 +_ 0 5.0 -+ 0 5.0 + 0
65.7 _+ 24.7 72.0 _+ 18.5 64.6 _+ 18.0
60.0 -+ 21.9 62.9 +_ 21.4 68.6 -+ 19.5
4.2 + 2.0 3.6 _+ 2.0 3.6 _+ 2.0
54.7 _+ 38.0 52.0 _+ 38.9 40.6 _+ 32.3*
66.7 _+ 20.7 68.6 _+ 30.2 80.0 + 32.7
Muse#no~ 10 nmol In = 9) 50 nmol tn = 6) 100 nmol tn = 7)
5.0 _+ 0 5.0 -+ 0 5.0 _+ 0
80.0 + 19.4 81.3 _+ 32.5 91.4 _+ 32.0
57.8 _+ 21.1 46.7 -+ 7.1) 51.4 + 19.5
3.7 -+ 1.4 3.2 _+ 1.2" 3.0 +_ 2.3**
72.[) + 22.0 52.7 _+ 10.0" 46.3 _+ 16,0"
57.8 +_ 21.1 46.7 +_ 7.0 51.4 +_ 19.5
Values are m e a n _+ S.D. SS, seizure stage: A D D , afterdischarge duration (s): GST, generalized seizure triggering threshold (1/A): ADT, afterdischarge threshold (¢tA). *P < 0.05: **P < 0.001, compared to the corresponding values obtained 24 h before treatment (Student's t-test).
132
N. Mori. M. I~ittanahe! Neuroscience Letters 169 (1994) 129-132
that muscimol in the CP interferes with the mechanism responsible for the expression of the motor manifestations of kindled AM seizures. It has been shown that the CP receives glutamatergic projections from the cerebral cortex, and this may enhance the CP neuronal activity [3,8]. Although no direct CP projections to the cerebral cortex have been identified, it appears that the CP modulates the function of the cerebral cortex through multiple synaptic pathways. Indeed, lesions of the CP cause an alteration in glucose utilization within the cerebral cortex [1,16]. Inhibitory effects of the CP on cortical spiking activity have also been suggested [17]. On the other hand, the thalamus, especially the intralaminar nuclei, sends high density projections to the CP [3,20]. This pathway appears to be mainly excitatory [20]. The CP does not send direct efferent to the thalamus, but functional alteration in the CP causes a change in the activity of the thalamus, presumably via polysynaptic pathways [1,16]. Furthermore, findings of reciprocal pathways between the thalamus and cortex, particularly the demonstration of a monosynaptic thalamocortical-corticothalamic feedback loop at electron microscopic and intracellular levels, provide structural arrangements that favor the reverberation of rhythmic, spindle-related clustered activity in both directions [11,20]. Since both the cerebral cortex and the thalamus are necessary for the expression of the motor manifestations seen in kindled AM seizures [9,21,22], we speculate that disruption of the linkages among the CP, thalamus, and cerebral cortex is responsible for the suppression of kindled seizures after the intra-CP injection of muscimol. A previous study has shown that a strong positive transfer occurs in the CP after the completion of AMkindling in cats [19]. In addition, a biochemical study has revealed a significant reduction in GABA synthesis and a significant increase in the numbers of GABA receptors in the CP of AM-kindled rats [12]. These findings are compatible with the anticonvulsant effect of intra-CP muscimol on kindled AM seizures shown here. This study, therefore, supports the concept that GABA acts as an anticonvulsant in the brain [2,15]. [1] Aiko, Y., Hosokawa, S., Shima, F., Kato, M. and Kitamura, K., Alterations in local cerebral glucose utilization during electrical stimulation of the striatum and globus pallidus in rats, Brain Res., 442 (1988) 43 52. [2] Bradford, H.F. and Dodd, ER., Biochemistry and basic mechanisms in epilepsy. In A.N. Davison (Ed.), Biochemistry and Neurological Disease, Blackwell Press, Oxford, 1976, pp. 114-168. [3] Carpenter, M.B., Interactions between the corpus striatum and brainstem nuclei. In J.S. McKenzie, R.E. Kemm and L.N. Wilcock (Eds.), The Basal Ganglia, Plenum Press, New York, 1984, pp. 1 68.
[4] Cavalheiro, E.A., Bortolotto, Z.A. and Turski, L., Microinjections of the ?'-aminobutyric acid antagonist, bicuculline methiodide, into the caudate-putamen prevent amygdala kindled seizures in rats, Brain Res., 411 (1987) 270272. [5] Chavalier, G. and Deniau, J.M., Disinhibition as a basic process in the expression of striatal function, Trends Neurosci., 13 (1990) 277 280. [6] Costa, A., Guidotti, A. and Toffano, G., Molecular mechanisms mediating the action of diazepam on GABA receptors, Br. J. Psychiatry, 133 (1978) 239-248. [7] Gale, K., Mechanisms of seizure control mediated by y-aminobutyric acid: role of the substantia nigra, Fed. Proc., 44 (1985) 24142424. [8] Gale, K. and Casu, M., Dynamic utilization of GABA in the substantia nigra: regulation by dopamine and GABA in the striaturn, and its clinical and behavioral implications, Mol. Cell. Biochem., 39 (1981) 369J,05. [9] Hiyoshi, T. and Wada, J.A., Midline thalamic lesion placement prior to kindling, Electroencephalogr. Clin. Neurophysiol., 70 (1989) 325-338. [10] Kalichman, M.W., Pharmacological investigation of convulsant y-aminobutyric acid (GABA) antagonists in amygdata-kindled rats, Epilepsia, 23 (1982) 163 171. [11] Kostopoulos, G. and Massino, A., Enhanced response of cortical neurons to the thalamic stimuli precedes the appearance of spike and wave discharges in feline generalized penicillin epilepsy, Brain Res., 278 (1983) 207 217. [12] L6sher, W. and Schwark, W., Further evidence for abnormal GABAergic circuits in amygdala-kindled rats, Brain Res., 420 (1987) 385 390. [13] McDonald, R. and Bergey, G.K., Valproic acid augments GABAmediated postsynaptic inhibition in cultured mammalian neurons, Brain Res., 170 (1979) 558 562. [14] McNamara, J.O., Galloway, M.T., Rigsbee, L.C. and Shin, C., Evidence implicating substantia nigra in regulation of kindled seizure threshold, J. Neurosci., 4 (1984) 2410-2417, [15] Meldrum, B.S., Amino acid transmitters and new approaches to anticonvulsant drug action, Epilepsia, 25 (1984) 379-393. [16] Paul, A.T.K. and McCulloch, J., Cerebral glucose utilization following striatal lesions: the effects of GABA agonist, muscimol, and the dopaminergic agonist, apomorphine, Brain Res., 425 (1987) 290 300. [17] Psatta, D.M., Control of chronic experimental focal epilepsy by feedback caudatumstimulation, Epilepsia, 24 (1983) 444M54. [18] Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroencephalogr. Clin. Neurophysiol., 32 (1972) 281 294. [19] Sato, M., Functional changes in the caudate and accumbens nuclei during amygdaloid and hippocampal seizure development in kindled cats, Folia Psychiat. Neurol. Jap., 31 (1977) 501-512. [20] Steriade, M. and Deschens, M., The thalamus as a neuronal oscillator, Brain Res. Rev., 8 (1984) 1 63. [21] Tanaka, A., Progressive changes of behavioral and electro-encephalographic responses to daily amygdaloid stimulation in rabbits, Fukuoka Acta Med., 63 (1972) 152-164. [22] Wada, J.A. and Komai, S., The effect of the anterior 2/3 callosal bisection upon bisymmetrical and bisynchronous generalized convulsions kindled from the amygdala in epileptic baboon, Papio papio. In G. Reeves (Ed.), Epilepsy and The Corpus Callosum, Plenum Press, New York, 1985, pp. 75 97.