- Email: [email protected]
Potential locus and mechanism of blockade of conditioned avoidance responding by neuroleptics
Vol. 23,No. 1,pp.73-78,1984 Printed in Great Britain
Neuropharmacology
0028-3908/84 $3.00+O.OO Pergamon Press Ltd
POTENTIAL LOCUS AND MECHANISM OF BLOCKADE OF CONDITIONED AVOIDANCE RESPONDING BY NEUROLEPTICS F. PETTY*t, *Veteran’s TDepartment of Psychiatry,
and A. D. SHERMANt
J. MOTT?
Administration Medical Center, University of Iowa College of
Iowa
City, IA 52240 and
Medicine, 500 Newton Road, Iowa City, IA
52242, U.S.A. (Accepted
1 May
1983)
Summary-In order to assess the possible loci of action of neuroleptics in blocking the acquisition of a one-way conditioned avoidance response, microinjections of three neuroleptics and seven putative neurotransmitters were made into several brain regions and their effects on this behavior were assessed. (0.128 nmol), chlorpromazine When injected into the amygdala, the ED,, values for haloperidol (1.04 nmol) and thioridazine (1.41 nmol) were appropriate in relation to their clinical potency. Injections of neurotransmitters were without effect except in a few cases. Most significantly, the intra-amygdaloid administration of glutamate diethyl ester (an antagonist at quisqualate-type receptors) produced a blockade of avoidance acquisition which, as in the case of the neuroleptics, was not diminished by pretreatment with atropine. Following intraperitoneal injection of chlorpromazine, a statistically-significant blockade of avoidance acquisition and of glutamate, released from slices of amygdala, was obtained at doses of 2 mg/kg or more. With haloperidol, comparable behavioral effects and release of glutamate were found at doses of 0.05 mg/kg or more. The depression of release of glutamate from amygdaloid slices could be attributed to glutamate derived from glutamine. These data suggest a possible role for glutamatergic transmission in the effects of neuroleptics. Key words:
neuroleptics,
amygdala,
glutamate.
Two types of hypotheses relating dopamine to actions of neuroleptics have been proposed. The receptor-blockade hypothesis suggests that neuroleptics increase the firing rate of dopamine-containing neurons (Bunney, Walters, Roth and Aghajanian, 1973) accelerate turnover of dopamine (Carlsson and Lindqvist, 1963) and bind to dopamine receptors (van Rossum, 1966). The coupling-blockade hypothesis is based on data indicating that neuroleptics enhance the spontaneous release of dopamine and modulate stimulus-coupled release (Quastel, Hackett and Okamoto, 1972). The development of the “atypical” neuroleptics has produced substantial modification of both hypotheses. In clinical use, the classical neuroleptics are associated with Parkinsonian-like motor effects, whereas the atypical antipsychotics are essentially devoid of these extrapyramidal side effects (Gerlach, Thorsen and Fog, 1975; Simpson, Lee and Shrivastava, 1975; Burki, 1979). In rats, the atypical neuroleptics fail to produce the catalepsy associated with butytrophenone neuroleptics (Costa11 and Naylor, 1975). Also, while classical antipsychotic drugs typically reduce amphetamineinduced locomotor activity is blocked by atypical neuroleptics. Blockage of conditioned avoidance behavior, however, is characteristic of both classes (Niemegeers, Verbruggen and Janssen, 1969).
On an anatomical basis, research has centered primarily on the dopamine-rich areas of the brain, most notably, the neostriatum (Calne, 1977; Cole, 1978; Feeney and Weir, 1980). A few studies (Costa11 and Naylor, 1974; Oishi, Watanabe, Okmori, Shikata and Ueki, 1979; Rebec, Alloway and Bashore, 1981) have suggested an amygdaloid locus of action. The hypothesized involvement of glutamate in the mechanism of action of neuroleptics (Janssen, 1967) has been difficult to test because of the metabolic complexity of releasable glutamate (Berl, Lajtha and Waelsch, 1961; Tapia and Gonzalez, 1978; van Gelder and Drujan, 1980; Hamberger, Chiang, Nylan, Scheff and Cotman, 1979a; Hamberger, Chiang, Sandoval and Cotman, 1979b). Within the last few years, the possible involvement of glutamate has again received considerable attention on the basis of presynaptic modulation of transmitter release. In striatal slices, glutamate added to medium increased the spontaneous or potassium-induced release of exogenous dopamine (Roberts and Sharif, 1978). This increased release was not present after lesions of the striatum induced by kainic acid (Roberts and Anderson, 1979). Conversely, dopamine or related agonists decreased the release of glutamate from striatal slices (Rowlands and Roberts, 1980; Mitchell and Doggett, 1980). Given the ubiquitous distribution of glutamate, 73
F.
74
PETTY et
and of systemically-administered phenothiazine (de Jaramillo and Guth, 1963), the following study was performed by Sherman, Petty and Sacquitne (1982). Animals were given chlorpromazine at 2.3 mg/kg (i.p.) which is the ED,, for blockade of avoidance acquisition as described. In six of the nine regions studied, the concentration of drug was not correlated with its efficacy. Only in the septum (I = 0.60), caudate (r = 0.90) and amygdala (r = 0.77) were statistically significant correlations observed. Subsequently, 1 lg of chlorpromazine was administered directly into 12 brain areas by stereotaxic injection. Injections of chlorpromazine only into the septum\ caudate, or amygdala blocked conditioned avoidance responding in a manner which was identical to the effects of the drug administered intraperitoneally. The data reveal several important points. First, when rejected into the amygdala, all three neuroleptics produced anti-avoidance effects in proportion to their clinical potency. This was not the case for injections made into the septum (i.e. haldoperidol equipotent with chlorpromazine) or caudate (haloperidol less potent than chlorpromazine). In addition, only in the amygdala did pretreatment with atropine produce no attenuation of the anti-avoidance effects of any of the neuroleptics. For example, the effects of thioridazine observed following injection into the septum or caudate were completely disrupted by pretreatment with atropine. Thirdly, thioridazine injected intraperitoneally, is frequently without significant effects on avoidance acquisition, while administration into the amygdala revealed its potency as approximately equal to chlorpromazine, as is the case in clinical use. These data support the hypothesis of a locus of action of neuroleptics in the amygdala, but the limitations of the intracranial injection procedure need to be considered. For example, diffusion of drug away from the site of injection could have accounted for some of the unpredicted results, and the use of surgical anesthesia prevents an assessment of the action of drug much before one hour has elapsed following stereotaxic surgery. To compliment the information generated by the extension of this type of approach, studies on the release of neurotransmitters were also performed in the following manner. METHODS
Animals
Male Sprague-Dawley rats (20&250 g) were used in all studies. All had free access to food and water during all phases of the experiments except during recovery from stereotaxic surgery. Behavioral
Animals were tested on a grid-floor shuttle-box constructed with one side 10 cm lower than the other. The grid floor sloped between the two levels. After a 1-min period of familiarization, an animal was placed
al.
on the lower grid and the trial begun. If after 10 set the animal was still on the lower grid, a 0.5 mA shock was initiated and continued until the animal escaped to the upper level. After a 20 set intertrial interval, the animal was replaced on the lower level of the grid and another trial begun. Animals moving to the upper grid in less than 10 set were not shocked, having performed an avoidance. Testing was continued until 5 consecutive avoidances were performed as the criterion for having acquired the conditioned avoidance response. Surgical
Stereotaxic injections were performed under ether anesthesia using the coordinates of Pellegrino, Pellegrino and Kushman (1979) with the site of injection of dye. All injections were performed bilaterally. Substances injected (1 pg unless otherwise specified) were dissolved in either physiological saline or, for haloperidol, in propylene glycol in a volume of 0.1 ~1. One hour after stereotaxic injection, the animals were tested as described above. When animals were treated with atropine, 0.8 mg was administered intraperitoneally 40 min before intracranial injection. The analysis of glutamate was carried out by a modification of the method of Lenda and Svennaby. The analysis was performed on an Allteck C-8 column (25cm) using a 13-70% methanol gradient over 20min following derivatization with ophthalaldehyde. Statistical analyses were performed with the Randomization Test for independent samples. Supplies
All materials were purchased from Sigma Chemical Company, St Louis, Missouri except haloperidol, which was donated by McNeil Laboratories, and thioridazine, which was donated by Sandoz. Experiment
I
Groups of six rats each were given intraamygdaloid injections of haloperidol(27, 53, 80, 133, 187 or 267 pmol), chlorpromazine (0.16, 0.31, 0.63, 0.94, 1.25 or 1.57 nmol) or thioridazine (0.27, 0.54, 1.08, 1.62, 2.16 or 2.70 nmol). These stereotaxic injections were made 40 min after the animals had received 3.2 mg of atropine/kg, a dose sufficient to attenuate intra-amygdaloid cholinergic effects. One hour after stereotaxic injections, all inaminals received behavioral testing as described above. Experiment 2
Groups of six rats each were injected with neurotransmitter substances at coordinates corresponding to the caudate, septum, or amygdala. One hour later, they received behavioral testing to determine whether acquisition of avoidance had been affected. In those cases where a significant effect (blockade) was observed, another group of animals was given
Neuroleptics and glutamate
atropine (3.2 mg/kg) and the procedure repeated. The three areas were selected for injection since they were the only three into which direct administration of a neuroleptic was previously observed to have a behavioral effect. The substances injected were the monoamines, dopamine, norepinephrine and serotonin (injected as their hydrochloride salts with doses calculated as the free amine), the amino acids, glutamate, aspartate, and y-aminobutyrate (GABA) (injected as the parent compound), acetylcholine (injected as the hydrobromide salt with the dose calculated as the parent) and glutamic acid diethyl ester (injected as the parent compound). Experiment 3
In the third study, the depolarization-induced release of endogenous glutamate from thick slices of amygdala was the variable. Initially, groups of 6-8 animals were injected with doses of chlorpromazine of between 1 and 4mg/kg (i.p.) or of haloperidol (O.Ol-0.2mg/kg, i.p.). One group served as subjects to assess the behavioral (anti-avoidance) effects of each dose, when tested one hour after injection. Another group of six animals was given the same dose of drug and killed one hour later. The amygdaloid area was dissected and chopped into 300pm slices with a McIlwain tissue chopper. The tissue was washed by incubation in Krebs-Ringer bicarbonate at 38” for 3 min, then centrifuged. The supernatant was decanted, fresh buffer added and the procedure repeated three times. The fifth wash was saved for assay and buffer containing 50 mM KC1 (replacing NaCI) was added. Incubation was continued for 3 mitt, then the supernatant removed following centrifugation. The wash sample and the 50 mM KC1 sample were assayed for glutamate by the hquidchromatographic procedure described. Data were expressed as nmol of substrate per minute per mg protein as determined on the tissue following incubation in 50 mM KC1 buffer. The level in the wash sample (not depolarized) was subtracted from the level in the 50m KC1 (depolarized) sample as a measure of depolarization-induced release. Experiment 4
Groups of six animals each were given chlorpromazine or haloperidol at intraperitoneal doses which had been demonstrated to produce a complete blockade of avoidance acquisition, i.e. 4mg/kg for chlorpromazine and 0.2 mg/kg for haloperidol. One hour later, the amygdalas were dissected free and chopped into 300 pm slices which were then divided into two separate samples. One sample was treated exactly as were samples in a previous study as an attempt to confirm those findings. The other half of the tissue was put through an identical procedure but with all buffers containing glutamine (lo-‘M). As before, glutamate was assayed by high-pressure liquid chromatography with fluorescence detection of the o-phthalaldehyde derivatives.
75
Experiment 5
Three groups of six rats each were used. They received saline (1 ml/kg), haloperidol (0.2 mg/kg) or chlorpromazine (4 mg/kg) (i.p.) one hour before being killed by cervical fracture. The amygdalas were removed and chopped into 300 pm slices. Half of the slices were incubated in Krebs-Ringer bicarbonate as described for experiment 3. The other half was incubated in the same buffer but containing 10-3M glutamine. After incubation for 10 min, the samples were centrifuged and the glutamate and GABA content of the supernatant fluid determined. The amount present in the sample without glutamine was subtracted from the amount in the glutamine-containing supernatant as a measure of hydrolysis of glutamine to glutamate under nondepolarizing conditions.
RESULTS
When estimated by log-probit analysis of the behavioral data obtained one hour after injection into the amygdala, EDSo values of 0.128 + 0.023 nmol, 1.04 f 0.16 nmol and 1.41 f 0.3 1 nmol were obtained for haloperidol, chlorpromazine and thioridazine respectively. These data were calculated as the amount of drug injected into the amygdala at time 0, not as the amount remaining in the amygdala when tested one hour later. Some blockage of avoidance acquisition (Table 1) was observed with administration of GABA or acetylcholine into the amygdala, injection of aspartate into the septum, or acetylcholine into the caudate. These effects were significantly attenuated by pretreatment with atropine. One atropineinsensitive effect was produced in each of the three injected areas. In the caudate, the injection of GABA produced a weak neuroleptic-like effect as did the administration of norepinephine into the septum. In the amygdala, only glutamate diethyl ester disrupted avoidance acquisition either with or without pretreatment with atropine. Doses of chlorpromazine of less than 2mg/kg, given intraperitoneally were without effect on avoidance acquisition or release of glutamate (Table 2). Above this amount, increasing effects on both measures were observed. A similar pattern was observed with doses of haloperidol greater than 0.05 mg/kg. In Table 2, the behavioral measure is given as the number of avoidances per 20 test trials so that the direction of the behavioral effect is the same as that observed for the release of glutamate. The presence of 10 - 5M glutamine in the buffer (Table 3) stimulated the release of glutamate, as reported by others (Hamberger et al., 1979bj. The increased release amounted to about 60% in control, or in drug-treated animais when comparing release in the absence or presence of glutamine. The amount of glutamine-stimulated release in the drug-treated animals amounted to only about half of that seen in
F. Parry Table I. Intracranial
injections
et al.
of neurotransmitters
and avoidance
acquisition
Area Amygdala
Caudate
Septum 12+_4 lOk6 1012 23 + 5* 27 + 9* IO If- 3 14 + 8
11*5 12*7 15+9 11+3
Saline Saline + atropine Dopamine Norepinephrine Norepinepbrine + &opine Serotonin Acetylcholine Acetylcholine + atropine GABA GABA + &opine GDEE GDEE + atropine Aspartate Aspartate + &opine Glutamate
11+5 34 + 9* 13*5 32 f 8* 16k6 28 f 4 26 _t 8* 13+6
-.12+3 14+5 13t5 II k3 11+_4 22 * 3* 14+2 20 + 6* 22 k 6 12+4
1013 1417 29 F IO* 12+3 l6f4
15+5
23 f S 12*2 Ilk5
Data represent mean f SD trials to five consecutive avoidances. N = 6/group tested one hour after intracranial injection. *P i 0.05 versus controls via Randomization Test. GDEE = glutamate diethyl ester.
Table 2. Behavioral
and neurochemical
effects of chlorpromazine
dose 0 1 2 2.5 3 4
b
dose
s
b
1.82kO.19 1.72 + 0.22 1.40 + 0.13’ 1.20 * 0.10’ 1.12 * 0.09’ 1.08+0.11*
0 0.01 0.05 0.1 0.15 0.2
13.8 + 1.1 13.Ok2.1 10.8 f 2.0* 7.5 i 3.2* 4.5 i 1.4* 1.5 i 1.5*
1.82 f 0.19 1.79 kO.10 I .62 f 0.08* 1.53 +_0.17* 1.31 f 0.09: 1.17*0.07*
a 13.5 f 1.0 11.0+3.0 10.8 k 2.3; 7.0 + 4.31 3.2 + 9.5* 0.7 + 0.8.
Data represent mea” + SD of avoidances per minute from slices of amygdala
*Decreased compared to control, b = Release of glutamate.
and haloperidol
Haloperidol
Chlorpromazine
per 20 test trials or the release of glutamate from drug-related animals.
P < 0.05 via Randomization
Table 3. Effect of neuroleptics Control
in nmol/mg
Test. a = Avoidances/20
on glutamine-stimulated Chlorpromazine
trials.
release Haloperidol
Release of glutamate (no glutamine)
233 f 22
126 f 16*
131 * 22*
Release of glutamate (lo-* elutamine) Dikerence
373 + 30
197 * 8*
208 + 29*
140*19
71 f 8*
67 f IO*
Data represent mea” f SD with N = 6/group. Release of glutamate is in pmol/mg minute. *Less than control, P < 0.05via Randomization Test.
controls. The percentage increase in the response to glutamine was not altered by the administration of neuroleptics. As in experiment 3, the minimum dose necessary to abolish avoidance acquisition (within 20 trials) produced a decrease in the release of glutamate of about 40%. Hydrolysis of glutamine to glutamate under nondepolarizing conditions was reduced by 30-50x in drug-treated animals. Since re-uptake of glutamate was not prevented in this procedure, this estimate of the activity of glutaminase is substantially less than values in the literature obtained using different methods. Control levels were 96 + 9 nmol of glutamate and GABA formed per mg of tissue per hour (mean f SD). Treatment with chlorpromazine significantly reduced activity (65 f 18), as did the injection of haloperidol (48 f 20).
protein
tissue per
DISCUSSION
The results of these studies, as they relate to the possible involvement of glutamate in the antiavoidance effects of neuroleptics can be summarized as follows: (1) When injected directly into the amygdala, haloperidol, chlorpromazine and thioridazine all decreased the acquisition of a one-way avoidance response. The relative potency of these agents was appropriate to their relative clinical potency. (2) Within the amygdala, administration of glutamate diethyl ester (GDEE) blocked acquisition of avoidance. A behavioral effect equivalent to about 3 nmol of chlorpromazine was obtained by the administration of about 6 nmol of glutamate diethyl ester. As was true for the administration of a neuroleptic, the behavioral effects of injection with glutamate diethyl
Neuroleptics and glutamate
ester were insensitive
to the administration of atropine. Because of the limitations inherent in the technique of intracranial injection, it cannot be established that both the injected neuroleptic and glutamate diethyl ester were affecting exactly the same site. For example, injections were made at fixed stereotaxic coordinates which, because of normal variation in skull shape, could have lead to a difference of about 1 mm in the actual locus or
injection. Similarly, in the hour between stereotaxic injection and behavioral testing, the diffusion patterns of the more lipid-soluble neuroleptic and the more polar glutamate diethyl ester may have been very different. (3) When injected intraperitoneally, at doses which were behaviorally equivalent, chlorpromazine and haloperidol produced approximately equivalent decreases in the release of glutamate. For example, chlorpromazine at 2 mg/kg (i.p.) produced a 2076 decrease in avoidance acquisition, as did a dose of 0.05 mg/kg of haloperidol. At these doses, chlorpromazine and haloperidol decreased the release of glutamate by 23 and 1lyi respectively. As can also be seen, the response curves relating behavior and the release of glutamate are far from parallel. For example, the largest dose of either neuroleptic reduced avoidance responses by 90-95% and the release of glutamate by 3540%. These differences clearly suggest that some factors other than release of glutamate are also involved in the actions of the drugs. (4) When injected intraperitoneally, chlorpromazine and haloperidol appeared to have effects primarily on glutamate derived from glutamine. In the absence of studies employing labelled precursors, evidence as to the pool affected by the neuroleptics must be considered tentative, but some conclusions can still be drawn. Using the measure of potassium-induced release of glutamate, the administration of a neuroleptic produced a decrease of about 45%. The decrease in glutaminase activity was about 40x, suggesting that the decrease in hydrolysis of glutamine paralleled the decrease in released glutamate. Results such as these could also suggest that neuroleptics decrease the release of glutamate from slices of amygdala by interfering with the uptake of glutamine. This possibility is made more likely by the observation that the inclusion of IO- 5M glutamine in all buffers elevated the release of glutamate by about 60% both in controls and in neuroleptic-treated animals. Thus, although the amount of glutamate released by these two groups was different, the percentage increase (in potassium-stimulated release) remained constant. Clearly, further studies are needed to determine whether absolute or relative changes in the release of glutamate are relevant to the behavioral effects of neuroleptics. In this connection, it should be pointed out that the entire antiavoidance effect of the administration of neuroleptics cannot be attributed to glutamatergic changes in the amygdala. As noted earlier, the dose-response curves relating the release of glutamate to decreased
77
avoidance responding are far from parallel. Thus, glutamatergic effects may be simply one of the behaviorally-relevant changes produced by the administration of neuroleptics. Finally, the fact that the amygdala is a dopaminerich area suggests the possibility that the administration of neuroleptics produces a decrease in the release or glutamate indirectly through a presynaptic inhibition via dopamine (Rowlands and Roberts, 1980). Although the blockade of avoidance acquisition by intracranial administration of glutamate diethyl ester suggests a role for glutamatergic systems, a clear definition of the effects awaits further experimentation. Acknowledgement-The
authors acknowledge the excellent technical help of J. L. Sacquitne. This research was made possible in part by support from the Veterans Administration. REFERENCES Berl S., Lajtha A. and Waelsch H. (1961) Amino acid and protein metabolism. VI. Cerebral compartments of glutamic acid metabolism. J. Neurochem. 7: 18Gl97. Bunney B. S., Walters J. R., Roth R. H. and Aghajanian G. K. (1973) Dopaminergic neurons: effect of antipsychotic drugs and amphetamine on single cell activity. J. Pharmai exp. The;. 185: W-571.
-
Burki H. R. (1979) Extraovramidal Ther. 5: 525-534.
side-effects. Pharmac.
_*
Calne D. B. (1977) Developments in the pharmacology and therapeutics of parkinsonism. Ann. Neural. 1: 11l-l 19. Carlsson A. and Lindqivst M. (1963) Effect of chlorpromazine or haloperidol on formation of 3 methoxytyramine and normetanephrine in mouse brain. Acta pharmac. tox. 20: 140.
Cole S. D. (1978) Brain mechanisms of amphetamineinduced anorexia, locomotion, and stereotypy: A Review. Neurosci. Biobehav. Rev. 2: 89-100.
Costa11 B. and Naylor R. J. (1974) The nucleus amygdaloideus centralis and neuroleptic activity in the rat. Eur. J. Pharmac. 25: 138-146.
Costa11 B. and Naylor R. J. (1975)Neuroleptic antagonism of dyskinetic phenomena. Eur. J. Pharmac. 33: 301-312. de Jaramillo G. A. and Guth P. S. (1963) A study of the localization of phenothiazines in dog brain. Biochem. Pharmac. 12: 525.
Feeney D. M. and Weir C. S. (1980) Sensory neglect after lesions of substantia nigra or lateral hypothalamus: differential severity and recovery of function. Brain Res. 178: 329-346.
Gerlach J., Thorsen K. and Fog R. (1975) Extrapyramidal reactions and amine metabolites in cerebrospinal fluid during haloperidol and clozapine treatment of schizophrenic patients. Psychopharmacology 40: 34 1. Hamberger A. C., Chiang G. H., Nylen E. C., Scheff S. W. and Cotman C. W. (1979a) Glutamate as a CNS transmitter. I. Evaluation of glucose and glutamine as precursors for the synthesis of preferentially released glutamate. Brain Res. 168: 513-530.
Hamberger A., Chiang G. H., Sandoval E. and Cotman C. W. (1979b) Glutamate as a CNS transmitter. II. Regulation of synthesis in the releasable pool. Brain Res. 168: 531-541. Janssen P. A. J. (1967) The pharmacology of haloperidol. 3: 10-18. Mitchell P. R. and Doggett N. S. (1980) Modulation of striatal [3-H]-glutamic acid release by dopaminergic drugs. Lije Sci. 26: 2073-208 1.
78
F. PETTYet al.
Niemegeers C. J. E., Verbruggen F. J. and Janssen P. A. J. (1969) The influence of various neuroleptic drugs on shock avoidance responding in rats. ~~yc~o~~~~a~of~g~ 16: 161. Oishi R., Watanabe S., Ohmori K., Shikata S. and Ueki S. (1979) Effect of stimulation of locus coeruleus on the evoked potential in the amygdala in rats. Jap. J. Pharmac. 29: ios-111. Peliegrino L. J., Pellegrino A. S. and Cushman A. J. (1979) A Stereotaxic At/us of the Rat Brain. Plenum Press, New York. Quastel D. M. J., Hackett J. T. and Okamoto K. (1972) Presynaptic action of central depressant drugs: inhibition of depolarization-secretion coupling. Can. J. Physiol. Pharmac. 50: 219.
Rebec G. V., Alloway K. D. and Bashore T. R. (1981) Differential actions of classical and a typical antipsychotic drugs on spontaneous neuronal activity in the amygdaloid complex. Pha~mac. ~~Qc~ern. Behau. 14: 49-56. Roberts P. J. and Sharif N. A. (1978) Effects of l-glutamate and related aminoacids upon the release of [‘-HI dopamine from rat striatal slices. Brain Res. 157: 391-395, Roberts P. J. and Anderson S. D. (1979) Stimulatory effect
of L-glutamate and relate amino acids on [‘-HI dopamine release from rat striatum: an in vitro model for glutamate actions. J. IVe~r~c~~~. 32: 1539-1545. Rowlands R. J. and Roberts P. J. (19801 Activation of dopamine receptors inhibits calcium-dependent glutamate release from cortico-striatal terminals in vitro. Eur. J. Pharmac. 62: 241-242.
Sherman A. D., Petty F. and Sacquitine J. L. (1982) Regional aspects of the delay in acquisition conditioned avoidance responding by chlorpromazine. L$e Ski. 30: 1811.
Simpson G. M., Lee J. H. and Shrivastava R. K. (1975) Clozapine in tardive dyskinesia. P~ych~ph~~rnuc~~o~y56: 75-80.
Tapia R. and Gonzalez R. M. (1978) Glutamine and glutamate as precursors of the releasable pool of GABA in brain cortex slices. van Gelder N. M. and Drujan B. D. (1980) Alterations in the compartmen~lized metabolism of glutamic acid with changed cerebral conditions. Bruin Res. 200: 443-455. van Rossum J. M. (1966) The significance of dopaminereceptor blockade for the mechanism of action of neuroleptic drugs. Arehs int. Pharmacoc&n. Ther. 160: 492.
Neuropharmacology
0028-3908/84 $3.00+O.OO Pergamon Press Ltd
POTENTIAL LOCUS AND MECHANISM OF BLOCKADE OF CONDITIONED AVOIDANCE RESPONDING BY NEUROLEPTICS F. PETTY*t, *Veteran’s TDepartment of Psychiatry,
and A. D. SHERMANt
J. MOTT?
Administration Medical Center, University of Iowa College of
Iowa
City, IA 52240 and
Medicine, 500 Newton Road, Iowa City, IA
52242, U.S.A. (Accepted
1 May
1983)
Summary-In order to assess the possible loci of action of neuroleptics in blocking the acquisition of a one-way conditioned avoidance response, microinjections of three neuroleptics and seven putative neurotransmitters were made into several brain regions and their effects on this behavior were assessed. (0.128 nmol), chlorpromazine When injected into the amygdala, the ED,, values for haloperidol (1.04 nmol) and thioridazine (1.41 nmol) were appropriate in relation to their clinical potency. Injections of neurotransmitters were without effect except in a few cases. Most significantly, the intra-amygdaloid administration of glutamate diethyl ester (an antagonist at quisqualate-type receptors) produced a blockade of avoidance acquisition which, as in the case of the neuroleptics, was not diminished by pretreatment with atropine. Following intraperitoneal injection of chlorpromazine, a statistically-significant blockade of avoidance acquisition and of glutamate, released from slices of amygdala, was obtained at doses of 2 mg/kg or more. With haloperidol, comparable behavioral effects and release of glutamate were found at doses of 0.05 mg/kg or more. The depression of release of glutamate from amygdaloid slices could be attributed to glutamate derived from glutamine. These data suggest a possible role for glutamatergic transmission in the effects of neuroleptics. Key words:
neuroleptics,
amygdala,
glutamate.
Two types of hypotheses relating dopamine to actions of neuroleptics have been proposed. The receptor-blockade hypothesis suggests that neuroleptics increase the firing rate of dopamine-containing neurons (Bunney, Walters, Roth and Aghajanian, 1973) accelerate turnover of dopamine (Carlsson and Lindqvist, 1963) and bind to dopamine receptors (van Rossum, 1966). The coupling-blockade hypothesis is based on data indicating that neuroleptics enhance the spontaneous release of dopamine and modulate stimulus-coupled release (Quastel, Hackett and Okamoto, 1972). The development of the “atypical” neuroleptics has produced substantial modification of both hypotheses. In clinical use, the classical neuroleptics are associated with Parkinsonian-like motor effects, whereas the atypical antipsychotics are essentially devoid of these extrapyramidal side effects (Gerlach, Thorsen and Fog, 1975; Simpson, Lee and Shrivastava, 1975; Burki, 1979). In rats, the atypical neuroleptics fail to produce the catalepsy associated with butytrophenone neuroleptics (Costa11 and Naylor, 1975). Also, while classical antipsychotic drugs typically reduce amphetamineinduced locomotor activity is blocked by atypical neuroleptics. Blockage of conditioned avoidance behavior, however, is characteristic of both classes (Niemegeers, Verbruggen and Janssen, 1969).
On an anatomical basis, research has centered primarily on the dopamine-rich areas of the brain, most notably, the neostriatum (Calne, 1977; Cole, 1978; Feeney and Weir, 1980). A few studies (Costa11 and Naylor, 1974; Oishi, Watanabe, Okmori, Shikata and Ueki, 1979; Rebec, Alloway and Bashore, 1981) have suggested an amygdaloid locus of action. The hypothesized involvement of glutamate in the mechanism of action of neuroleptics (Janssen, 1967) has been difficult to test because of the metabolic complexity of releasable glutamate (Berl, Lajtha and Waelsch, 1961; Tapia and Gonzalez, 1978; van Gelder and Drujan, 1980; Hamberger, Chiang, Nylan, Scheff and Cotman, 1979a; Hamberger, Chiang, Sandoval and Cotman, 1979b). Within the last few years, the possible involvement of glutamate has again received considerable attention on the basis of presynaptic modulation of transmitter release. In striatal slices, glutamate added to medium increased the spontaneous or potassium-induced release of exogenous dopamine (Roberts and Sharif, 1978). This increased release was not present after lesions of the striatum induced by kainic acid (Roberts and Anderson, 1979). Conversely, dopamine or related agonists decreased the release of glutamate from striatal slices (Rowlands and Roberts, 1980; Mitchell and Doggett, 1980). Given the ubiquitous distribution of glutamate, 73
F.
74
PETTY et
and of systemically-administered phenothiazine (de Jaramillo and Guth, 1963), the following study was performed by Sherman, Petty and Sacquitne (1982). Animals were given chlorpromazine at 2.3 mg/kg (i.p.) which is the ED,, for blockade of avoidance acquisition as described. In six of the nine regions studied, the concentration of drug was not correlated with its efficacy. Only in the septum (I = 0.60), caudate (r = 0.90) and amygdala (r = 0.77) were statistically significant correlations observed. Subsequently, 1 lg of chlorpromazine was administered directly into 12 brain areas by stereotaxic injection. Injections of chlorpromazine only into the septum\ caudate, or amygdala blocked conditioned avoidance responding in a manner which was identical to the effects of the drug administered intraperitoneally. The data reveal several important points. First, when rejected into the amygdala, all three neuroleptics produced anti-avoidance effects in proportion to their clinical potency. This was not the case for injections made into the septum (i.e. haldoperidol equipotent with chlorpromazine) or caudate (haloperidol less potent than chlorpromazine). In addition, only in the amygdala did pretreatment with atropine produce no attenuation of the anti-avoidance effects of any of the neuroleptics. For example, the effects of thioridazine observed following injection into the septum or caudate were completely disrupted by pretreatment with atropine. Thirdly, thioridazine injected intraperitoneally, is frequently without significant effects on avoidance acquisition, while administration into the amygdala revealed its potency as approximately equal to chlorpromazine, as is the case in clinical use. These data support the hypothesis of a locus of action of neuroleptics in the amygdala, but the limitations of the intracranial injection procedure need to be considered. For example, diffusion of drug away from the site of injection could have accounted for some of the unpredicted results, and the use of surgical anesthesia prevents an assessment of the action of drug much before one hour has elapsed following stereotaxic surgery. To compliment the information generated by the extension of this type of approach, studies on the release of neurotransmitters were also performed in the following manner. METHODS
Animals
Male Sprague-Dawley rats (20&250 g) were used in all studies. All had free access to food and water during all phases of the experiments except during recovery from stereotaxic surgery. Behavioral
Animals were tested on a grid-floor shuttle-box constructed with one side 10 cm lower than the other. The grid floor sloped between the two levels. After a 1-min period of familiarization, an animal was placed
al.
on the lower grid and the trial begun. If after 10 set the animal was still on the lower grid, a 0.5 mA shock was initiated and continued until the animal escaped to the upper level. After a 20 set intertrial interval, the animal was replaced on the lower level of the grid and another trial begun. Animals moving to the upper grid in less than 10 set were not shocked, having performed an avoidance. Testing was continued until 5 consecutive avoidances were performed as the criterion for having acquired the conditioned avoidance response. Surgical
Stereotaxic injections were performed under ether anesthesia using the coordinates of Pellegrino, Pellegrino and Kushman (1979) with the site of injection of dye. All injections were performed bilaterally. Substances injected (1 pg unless otherwise specified) were dissolved in either physiological saline or, for haloperidol, in propylene glycol in a volume of 0.1 ~1. One hour after stereotaxic injection, the animals were tested as described above. When animals were treated with atropine, 0.8 mg was administered intraperitoneally 40 min before intracranial injection. The analysis of glutamate was carried out by a modification of the method of Lenda and Svennaby. The analysis was performed on an Allteck C-8 column (25cm) using a 13-70% methanol gradient over 20min following derivatization with ophthalaldehyde. Statistical analyses were performed with the Randomization Test for independent samples. Supplies
All materials were purchased from Sigma Chemical Company, St Louis, Missouri except haloperidol, which was donated by McNeil Laboratories, and thioridazine, which was donated by Sandoz. Experiment
I
Groups of six rats each were given intraamygdaloid injections of haloperidol(27, 53, 80, 133, 187 or 267 pmol), chlorpromazine (0.16, 0.31, 0.63, 0.94, 1.25 or 1.57 nmol) or thioridazine (0.27, 0.54, 1.08, 1.62, 2.16 or 2.70 nmol). These stereotaxic injections were made 40 min after the animals had received 3.2 mg of atropine/kg, a dose sufficient to attenuate intra-amygdaloid cholinergic effects. One hour after stereotaxic injections, all inaminals received behavioral testing as described above. Experiment 2
Groups of six rats each were injected with neurotransmitter substances at coordinates corresponding to the caudate, septum, or amygdala. One hour later, they received behavioral testing to determine whether acquisition of avoidance had been affected. In those cases where a significant effect (blockade) was observed, another group of animals was given
Neuroleptics and glutamate
atropine (3.2 mg/kg) and the procedure repeated. The three areas were selected for injection since they were the only three into which direct administration of a neuroleptic was previously observed to have a behavioral effect. The substances injected were the monoamines, dopamine, norepinephrine and serotonin (injected as their hydrochloride salts with doses calculated as the free amine), the amino acids, glutamate, aspartate, and y-aminobutyrate (GABA) (injected as the parent compound), acetylcholine (injected as the hydrobromide salt with the dose calculated as the parent) and glutamic acid diethyl ester (injected as the parent compound). Experiment 3
In the third study, the depolarization-induced release of endogenous glutamate from thick slices of amygdala was the variable. Initially, groups of 6-8 animals were injected with doses of chlorpromazine of between 1 and 4mg/kg (i.p.) or of haloperidol (O.Ol-0.2mg/kg, i.p.). One group served as subjects to assess the behavioral (anti-avoidance) effects of each dose, when tested one hour after injection. Another group of six animals was given the same dose of drug and killed one hour later. The amygdaloid area was dissected and chopped into 300pm slices with a McIlwain tissue chopper. The tissue was washed by incubation in Krebs-Ringer bicarbonate at 38” for 3 min, then centrifuged. The supernatant was decanted, fresh buffer added and the procedure repeated three times. The fifth wash was saved for assay and buffer containing 50 mM KC1 (replacing NaCI) was added. Incubation was continued for 3 mitt, then the supernatant removed following centrifugation. The wash sample and the 50 mM KC1 sample were assayed for glutamate by the hquidchromatographic procedure described. Data were expressed as nmol of substrate per minute per mg protein as determined on the tissue following incubation in 50 mM KC1 buffer. The level in the wash sample (not depolarized) was subtracted from the level in the 50m KC1 (depolarized) sample as a measure of depolarization-induced release. Experiment 4
Groups of six animals each were given chlorpromazine or haloperidol at intraperitoneal doses which had been demonstrated to produce a complete blockade of avoidance acquisition, i.e. 4mg/kg for chlorpromazine and 0.2 mg/kg for haloperidol. One hour later, the amygdalas were dissected free and chopped into 300 pm slices which were then divided into two separate samples. One sample was treated exactly as were samples in a previous study as an attempt to confirm those findings. The other half of the tissue was put through an identical procedure but with all buffers containing glutamine (lo-‘M). As before, glutamate was assayed by high-pressure liquid chromatography with fluorescence detection of the o-phthalaldehyde derivatives.
75
Experiment 5
Three groups of six rats each were used. They received saline (1 ml/kg), haloperidol (0.2 mg/kg) or chlorpromazine (4 mg/kg) (i.p.) one hour before being killed by cervical fracture. The amygdalas were removed and chopped into 300 pm slices. Half of the slices were incubated in Krebs-Ringer bicarbonate as described for experiment 3. The other half was incubated in the same buffer but containing 10-3M glutamine. After incubation for 10 min, the samples were centrifuged and the glutamate and GABA content of the supernatant fluid determined. The amount present in the sample without glutamine was subtracted from the amount in the glutamine-containing supernatant as a measure of hydrolysis of glutamine to glutamate under nondepolarizing conditions.
RESULTS
When estimated by log-probit analysis of the behavioral data obtained one hour after injection into the amygdala, EDSo values of 0.128 + 0.023 nmol, 1.04 f 0.16 nmol and 1.41 f 0.3 1 nmol were obtained for haloperidol, chlorpromazine and thioridazine respectively. These data were calculated as the amount of drug injected into the amygdala at time 0, not as the amount remaining in the amygdala when tested one hour later. Some blockage of avoidance acquisition (Table 1) was observed with administration of GABA or acetylcholine into the amygdala, injection of aspartate into the septum, or acetylcholine into the caudate. These effects were significantly attenuated by pretreatment with atropine. One atropineinsensitive effect was produced in each of the three injected areas. In the caudate, the injection of GABA produced a weak neuroleptic-like effect as did the administration of norepinephine into the septum. In the amygdala, only glutamate diethyl ester disrupted avoidance acquisition either with or without pretreatment with atropine. Doses of chlorpromazine of less than 2mg/kg, given intraperitoneally were without effect on avoidance acquisition or release of glutamate (Table 2). Above this amount, increasing effects on both measures were observed. A similar pattern was observed with doses of haloperidol greater than 0.05 mg/kg. In Table 2, the behavioral measure is given as the number of avoidances per 20 test trials so that the direction of the behavioral effect is the same as that observed for the release of glutamate. The presence of 10 - 5M glutamine in the buffer (Table 3) stimulated the release of glutamate, as reported by others (Hamberger et al., 1979bj. The increased release amounted to about 60% in control, or in drug-treated animais when comparing release in the absence or presence of glutamine. The amount of glutamine-stimulated release in the drug-treated animals amounted to only about half of that seen in
F. Parry Table I. Intracranial
injections
et al.
of neurotransmitters
and avoidance
acquisition
Area Amygdala
Caudate
Septum 12+_4 lOk6 1012 23 + 5* 27 + 9* IO If- 3 14 + 8
11*5 12*7 15+9 11+3
Saline Saline + atropine Dopamine Norepinephrine Norepinepbrine + &opine Serotonin Acetylcholine Acetylcholine + atropine GABA GABA + &opine GDEE GDEE + atropine Aspartate Aspartate + &opine Glutamate
11+5 34 + 9* 13*5 32 f 8* 16k6 28 f 4 26 _t 8* 13+6
-.12+3 14+5 13t5 II k3 11+_4 22 * 3* 14+2 20 + 6* 22 k 6 12+4
1013 1417 29 F IO* 12+3 l6f4
15+5
23 f S 12*2 Ilk5
Data represent mean f SD trials to five consecutive avoidances. N = 6/group tested one hour after intracranial injection. *P i 0.05 versus controls via Randomization Test. GDEE = glutamate diethyl ester.
Table 2. Behavioral
and neurochemical
effects of chlorpromazine
dose 0 1 2 2.5 3 4
b
dose
s
b
1.82kO.19 1.72 + 0.22 1.40 + 0.13’ 1.20 * 0.10’ 1.12 * 0.09’ 1.08+0.11*
0 0.01 0.05 0.1 0.15 0.2
13.8 + 1.1 13.Ok2.1 10.8 f 2.0* 7.5 i 3.2* 4.5 i 1.4* 1.5 i 1.5*
1.82 f 0.19 1.79 kO.10 I .62 f 0.08* 1.53 +_0.17* 1.31 f 0.09: 1.17*0.07*
a 13.5 f 1.0 11.0+3.0 10.8 k 2.3; 7.0 + 4.31 3.2 + 9.5* 0.7 + 0.8.
Data represent mea” + SD of avoidances per minute from slices of amygdala
*Decreased compared to control, b = Release of glutamate.
and haloperidol
Haloperidol
Chlorpromazine
per 20 test trials or the release of glutamate from drug-related animals.
P < 0.05 via Randomization
Table 3. Effect of neuroleptics Control
in nmol/mg
Test. a = Avoidances/20
on glutamine-stimulated Chlorpromazine
trials.
release Haloperidol
Release of glutamate (no glutamine)
233 f 22
126 f 16*
131 * 22*
Release of glutamate (lo-* elutamine) Dikerence
373 + 30
197 * 8*
208 + 29*
140*19
71 f 8*
67 f IO*
Data represent mea” f SD with N = 6/group. Release of glutamate is in pmol/mg minute. *Less than control, P < 0.05via Randomization Test.
controls. The percentage increase in the response to glutamine was not altered by the administration of neuroleptics. As in experiment 3, the minimum dose necessary to abolish avoidance acquisition (within 20 trials) produced a decrease in the release of glutamate of about 40%. Hydrolysis of glutamine to glutamate under nondepolarizing conditions was reduced by 30-50x in drug-treated animals. Since re-uptake of glutamate was not prevented in this procedure, this estimate of the activity of glutaminase is substantially less than values in the literature obtained using different methods. Control levels were 96 + 9 nmol of glutamate and GABA formed per mg of tissue per hour (mean f SD). Treatment with chlorpromazine significantly reduced activity (65 f 18), as did the injection of haloperidol (48 f 20).
protein
tissue per
DISCUSSION
The results of these studies, as they relate to the possible involvement of glutamate in the antiavoidance effects of neuroleptics can be summarized as follows: (1) When injected directly into the amygdala, haloperidol, chlorpromazine and thioridazine all decreased the acquisition of a one-way avoidance response. The relative potency of these agents was appropriate to their relative clinical potency. (2) Within the amygdala, administration of glutamate diethyl ester (GDEE) blocked acquisition of avoidance. A behavioral effect equivalent to about 3 nmol of chlorpromazine was obtained by the administration of about 6 nmol of glutamate diethyl ester. As was true for the administration of a neuroleptic, the behavioral effects of injection with glutamate diethyl
Neuroleptics and glutamate
ester were insensitive
to the administration of atropine. Because of the limitations inherent in the technique of intracranial injection, it cannot be established that both the injected neuroleptic and glutamate diethyl ester were affecting exactly the same site. For example, injections were made at fixed stereotaxic coordinates which, because of normal variation in skull shape, could have lead to a difference of about 1 mm in the actual locus or
injection. Similarly, in the hour between stereotaxic injection and behavioral testing, the diffusion patterns of the more lipid-soluble neuroleptic and the more polar glutamate diethyl ester may have been very different. (3) When injected intraperitoneally, at doses which were behaviorally equivalent, chlorpromazine and haloperidol produced approximately equivalent decreases in the release of glutamate. For example, chlorpromazine at 2 mg/kg (i.p.) produced a 2076 decrease in avoidance acquisition, as did a dose of 0.05 mg/kg of haloperidol. At these doses, chlorpromazine and haloperidol decreased the release of glutamate by 23 and 1lyi respectively. As can also be seen, the response curves relating behavior and the release of glutamate are far from parallel. For example, the largest dose of either neuroleptic reduced avoidance responses by 90-95% and the release of glutamate by 3540%. These differences clearly suggest that some factors other than release of glutamate are also involved in the actions of the drugs. (4) When injected intraperitoneally, chlorpromazine and haloperidol appeared to have effects primarily on glutamate derived from glutamine. In the absence of studies employing labelled precursors, evidence as to the pool affected by the neuroleptics must be considered tentative, but some conclusions can still be drawn. Using the measure of potassium-induced release of glutamate, the administration of a neuroleptic produced a decrease of about 45%. The decrease in glutaminase activity was about 40x, suggesting that the decrease in hydrolysis of glutamine paralleled the decrease in released glutamate. Results such as these could also suggest that neuroleptics decrease the release of glutamate from slices of amygdala by interfering with the uptake of glutamine. This possibility is made more likely by the observation that the inclusion of IO- 5M glutamine in all buffers elevated the release of glutamate by about 60% both in controls and in neuroleptic-treated animals. Thus, although the amount of glutamate released by these two groups was different, the percentage increase (in potassium-stimulated release) remained constant. Clearly, further studies are needed to determine whether absolute or relative changes in the release of glutamate are relevant to the behavioral effects of neuroleptics. In this connection, it should be pointed out that the entire antiavoidance effect of the administration of neuroleptics cannot be attributed to glutamatergic changes in the amygdala. As noted earlier, the dose-response curves relating the release of glutamate to decreased
77
avoidance responding are far from parallel. Thus, glutamatergic effects may be simply one of the behaviorally-relevant changes produced by the administration of neuroleptics. Finally, the fact that the amygdala is a dopaminerich area suggests the possibility that the administration of neuroleptics produces a decrease in the release or glutamate indirectly through a presynaptic inhibition via dopamine (Rowlands and Roberts, 1980). Although the blockade of avoidance acquisition by intracranial administration of glutamate diethyl ester suggests a role for glutamatergic systems, a clear definition of the effects awaits further experimentation. Acknowledgement-The
authors acknowledge the excellent technical help of J. L. Sacquitne. This research was made possible in part by support from the Veterans Administration. REFERENCES Berl S., Lajtha A. and Waelsch H. (1961) Amino acid and protein metabolism. VI. Cerebral compartments of glutamic acid metabolism. J. Neurochem. 7: 18Gl97. Bunney B. S., Walters J. R., Roth R. H. and Aghajanian G. K. (1973) Dopaminergic neurons: effect of antipsychotic drugs and amphetamine on single cell activity. J. Pharmai exp. The;. 185: W-571.
-
Burki H. R. (1979) Extraovramidal Ther. 5: 525-534.
side-effects. Pharmac.
_*
Calne D. B. (1977) Developments in the pharmacology and therapeutics of parkinsonism. Ann. Neural. 1: 11l-l 19. Carlsson A. and Lindqivst M. (1963) Effect of chlorpromazine or haloperidol on formation of 3 methoxytyramine and normetanephrine in mouse brain. Acta pharmac. tox. 20: 140.
Cole S. D. (1978) Brain mechanisms of amphetamineinduced anorexia, locomotion, and stereotypy: A Review. Neurosci. Biobehav. Rev. 2: 89-100.
Costa11 B. and Naylor R. J. (1974) The nucleus amygdaloideus centralis and neuroleptic activity in the rat. Eur. J. Pharmac. 25: 138-146.
Costa11 B. and Naylor R. J. (1975)Neuroleptic antagonism of dyskinetic phenomena. Eur. J. Pharmac. 33: 301-312. de Jaramillo G. A. and Guth P. S. (1963) A study of the localization of phenothiazines in dog brain. Biochem. Pharmac. 12: 525.
Feeney D. M. and Weir C. S. (1980) Sensory neglect after lesions of substantia nigra or lateral hypothalamus: differential severity and recovery of function. Brain Res. 178: 329-346.
Gerlach J., Thorsen K. and Fog R. (1975) Extrapyramidal reactions and amine metabolites in cerebrospinal fluid during haloperidol and clozapine treatment of schizophrenic patients. Psychopharmacology 40: 34 1. Hamberger A. C., Chiang G. H., Nylen E. C., Scheff S. W. and Cotman C. W. (1979a) Glutamate as a CNS transmitter. I. Evaluation of glucose and glutamine as precursors for the synthesis of preferentially released glutamate. Brain Res. 168: 513-530.
Hamberger A., Chiang G. H., Sandoval E. and Cotman C. W. (1979b) Glutamate as a CNS transmitter. II. Regulation of synthesis in the releasable pool. Brain Res. 168: 531-541. Janssen P. A. J. (1967) The pharmacology of haloperidol. 3: 10-18. Mitchell P. R. and Doggett N. S. (1980) Modulation of striatal [3-H]-glutamic acid release by dopaminergic drugs. Lije Sci. 26: 2073-208 1.
78
F. PETTYet al.
Niemegeers C. J. E., Verbruggen F. J. and Janssen P. A. J. (1969) The influence of various neuroleptic drugs on shock avoidance responding in rats. ~~yc~o~~~~a~of~g~ 16: 161. Oishi R., Watanabe S., Ohmori K., Shikata S. and Ueki S. (1979) Effect of stimulation of locus coeruleus on the evoked potential in the amygdala in rats. Jap. J. Pharmac. 29: ios-111. Peliegrino L. J., Pellegrino A. S. and Cushman A. J. (1979) A Stereotaxic At/us of the Rat Brain. Plenum Press, New York. Quastel D. M. J., Hackett J. T. and Okamoto K. (1972) Presynaptic action of central depressant drugs: inhibition of depolarization-secretion coupling. Can. J. Physiol. Pharmac. 50: 219.
Rebec G. V., Alloway K. D. and Bashore T. R. (1981) Differential actions of classical and a typical antipsychotic drugs on spontaneous neuronal activity in the amygdaloid complex. Pha~mac. ~~Qc~ern. Behau. 14: 49-56. Roberts P. J. and Sharif N. A. (1978) Effects of l-glutamate and related aminoacids upon the release of [‘-HI dopamine from rat striatal slices. Brain Res. 157: 391-395, Roberts P. J. and Anderson S. D. (1979) Stimulatory effect
of L-glutamate and relate amino acids on [‘-HI dopamine release from rat striatum: an in vitro model for glutamate actions. J. IVe~r~c~~~. 32: 1539-1545. Rowlands R. J. and Roberts P. J. (19801 Activation of dopamine receptors inhibits calcium-dependent glutamate release from cortico-striatal terminals in vitro. Eur. J. Pharmac. 62: 241-242.
Sherman A. D., Petty F. and Sacquitine J. L. (1982) Regional aspects of the delay in acquisition conditioned avoidance responding by chlorpromazine. L$e Ski. 30: 1811.
Simpson G. M., Lee J. H. and Shrivastava R. K. (1975) Clozapine in tardive dyskinesia. P~ych~ph~~rnuc~~o~y56: 75-80.
Tapia R. and Gonzalez R. M. (1978) Glutamine and glutamate as precursors of the releasable pool of GABA in brain cortex slices. van Gelder N. M. and Drujan B. D. (1980) Alterations in the compartmen~lized metabolism of glutamic acid with changed cerebral conditions. Bruin Res. 200: 443-455. van Rossum J. M. (1966) The significance of dopaminereceptor blockade for the mechanism of action of neuroleptic drugs. Arehs int. Pharmacoc&n. Ther. 160: 492.