Naunyn-Schmiedeberg's Arch Pharmacol (1995) 352:365-373 © Springer-Verlag 1995

Esa R. Korpi - Garry Wong • Hartmut Ltiddens

Subtype specificity of 7-aminobutyric acid type A receptor antagonism by clozapine

Received: 16 February 1995 / Accepted: 1 June 1995

Abstract Clozapine, an atypical neuroleptic, functionally antagonizes the 7-aminobutyric acid-induced chloride up- take via the main central inhibitory receptor, 7-aminobuty- ric acid type A (GABAA) receptor, in brain vesicles. GABAA antagonism by micromolar concentrations of clo- zapine is more efficient in rat cerebrocortical and hippo- campal membranes than in cerebellar membranes, as evidenced by clozapine reversal GABA-inhibition of 35 [ S]t-butylbicyclophosphorothionate @sS]TBPS) binding.

A typical neuroleptic, haloperidol, failed to antagonize GABA in any of these brain regions, while the specific GABAA antagonist 2'-(3'-carboxy-2',3'-propyl)-3-amino-6- p-methoxyphenylpyrazinium bromide (SR 95531) was effi- cient in all three brain regions. Clozapine action on [3sS]TBPS binding was unaffected by the benzodiazepine receptor antagonist flumazenil. Clozapine inhibited the binding of [3H]muscimol and [3H]SR 95531 to the GABA recognition site, but this effect only partially correlated with the regional differences in and the potency of cloza-

• * . . . . . 35 • pine antagonism of GABA-lnhlbltlOn of [ S]TBPS bmd- ing, suggesting that also other than GABA sites may mediate clozapine actions. Autoradiography of [35S]TBPS binding revealed GABA antagonism by clozapine in most brain regions. Main exceptions were cerebellar granule cell and molecular layers, olfactory bulb external plexiform and glomernlar layers and primary olfactory cortex, where clozapine antagonized GABA inhibition less than average, and lateral hypothalamic and preoptic areas where its an- tagonism was greater than average. Recombinant a6f127~2 receptors, the predominant a6 subunit-containing receptor subtype in cerebellar granule cells, failed to show GABA antagonism by clozapine up to 100/~M. In contrast, re- combinant alf1272 receptors, forming the predominant re- ceptor subtype in the brain, were clozapine sensitive.

Esa R. Korpi (~) • G. Wong Biomedical Research Center, Alko Group Ltd, POB 350, FIN-0010I Helsinki, Finland

H. Ltiddens Laboratory of Molecular Neuroendocrinology, Center for Molecular Biology, University of Heidelberg, D-69120 Heidelberg, Germany

Recombinant a6fi272 and a6f1372 receptors resulted in clo- zapine-insensitive receptors, whereas a6f1172 receptors were clozapine sensitive• The efficacy of clozapine to an- tagonize GABA in alflx72 receptors decreased in the or- der of alfll,/2>alfl2~2>alfl3y2. The results indicate that clozapine antagonizes the function of most GABA a recep- tor subtypes, and that the interaction is determined by the interaction of the a and fi subunit variants. GABA antag- onism is a unique property of clozapine, not shared by haloperidol, which might be involved in the pharmacologi- cal mechanism for the increased seizure susceptibility as- sociated with clozapine treatment.

Key words Clozapine - Brain regional heterogeneity - GABA antagonists GABAA receptor subunits Recombinant receptors • [35S]TBPS binding • [35S]TBPS autoradiography

Introduction

The potent antipsychotic clozapine is classified as an aty- pical neuroleptic since it lacks the extrapyramidal side-ef- fects characteristic for typical neuroleptics such as haloper- idol (Angst et al. 1971; for review, see: Baldessarini and Frankenburg 1991). Treatment of negative and positive symptoms with clozapine in patients unresponsive to typi- cal neuroleptics has often been successful (Kane et al. 1988; Kane 1992). Unfortunately, clozapine administration is associated with serious adverse side-effects such as de- lirium, blood dyscrasias, and seizures (Idanp~i-Heikkil~i et al. 1977; Gaertner et al. 1989). Safer treatment of patients with clozapine requires better understanding of the me- chanisms involved in adverse side-effects.

Nanomolar concentrations of clozapine affect several neurotransmitter receptors (Peroutka and Snyder 1980; see: Deutch et al. 1991; Coward 1992; Seeman and Van Tol 1994)• For example, it antagonizes with high affinity the

366

serotonin 5-HT2 and 5-HT1o receptors (Corm and Sanders- Bush 1985; Kuoppam~iki et al. 1993). It blocks dopamine Dj and D2 receptors and potently the recently discovered D 4 receptors (Van Tol et al. 1991; Farde et al. 1992). Clo- zapine antagonizes muscar in ic acetylchol ine receptor ac- tion, which m a y partly explain the low incidence of extra- pyramidal side-effects dur ing clozapine treatment, and its action on his taminergic H~ receptors may be responsible for its sedative effects (Schotte et al. 1993). Some of these actions ma y contr ibute to the un ique therapeutic profile of clozapine.

Clozapine t reatment at high doses or dur ing rapid dose escalat ion is associated with seizures (Devinsky et al. 1991; Pacia and Dev insky 1994). Since an tagonism of the main inhibi tory neurotransmit ter receptors in the brain, )~- aminobutyr ic acid type A (GABAA) receptors, produces increased excitabil i ty and seizures, it is possible that this side-effect of c lozapine is due to inhibi t ion of GABAerg ic activity. Indeed, exper iments by Squires and Saederup (1991; 1993) have shown that c lozapine can partially inhi- bit activity at GA BAA receptors as studied with l igand b ind ing technique on rat brain membranes us ing the con- vulsant [35S]t-butylbicyclophosphorothionate ([35S]TBPS) as a ligand. GA BAA receptors show enormous heterogene- ity derived from different combina t ions of the k n o w n 13 subunits in pentameric receptor protein complexes (Olsen and Tobin 1990; Wisden and Seeburg 1992; Lfiddens and Korpi 1995).

The present study focussed on de termining whether clo- zapine acts as a G ABAA antagonist , in compar i son to ha- loperidol, and whether c lozapine actions differ in brain re- gions k n o w n to conta in different GABAA receptor sub-

types.

Methods

Brain membranes. Adult male Wistar rats (Department of Laboratory Animals, University of Helsinki, Helsinki, Finland) were decapitated and cerebral cortex, hippocampus and cerebellum were dissected and frozen. For each membrane preparation, cerebral cortical, hippocam- pal, and cerebellar tissues were pooled from 2, 4, and 4 rats, respec- tively. Tissues were homogenized with a Polytron in 50 volumes of ice-cold 50 mM Tris-citrate buffer (pH 7.4) supplemented with 1 mM disodium edetate, centrifuged at 20,000g for 20 rain. Pellets were re- suspended in the same buffer and recentrifuged 5 times. The final sus- pension was prepared in 50 mM Tris-citrate buffer, aliquotted, and stored frozen at -80°C.

Recombinant receptors. Human embryonic kidney (HEK) 293 cells were transfected with rat cDNAs encoding al or a6, ill, /?2 or/73, and 72S subunits, subcloned individually into eukaryotic expression vectors (Pritchett et al. 1989b; Ymer et al. 1989; Ltiddens et al. 1990; Korpi et al. 1993) by the calcium phosphate precipitation method (Chen and Okayama 1987). Quantitative ratios of the cDNAs for the a, [3, and 7 subunits were as described by Korpi and Liiddens (1993). Briefly, HEK cells were plated on dishes 15-cm in diameter (Becton Dickinson Labware, Lincoln Park, NJ, USA) and transfected two to three days later. About 20 h after transfection, the medium was chan- ged, and 48 h after transfection, cells were washed and harvested in phosphate-buffered saline. Cell pellets were homogenized with a Polytron in 50 mM Tris-citrate (pH 7.4) buffer, centrifuged, resus- pended, and used immediately or stored frozen at -80°C.

Ligand binding. Frozen membranes were thawed, resuspended and centrifuged once, before final resuspension in 50 mM Tris-citrate buf- fer to give a protein concentration of 100-240 #g/ml (Bio-Rad Pro- tein Assay kit) in a total volume of 0.5 ml per assay tube (Korpi and Lfiddens 1993). After defined incubation times of duplicate samples, bound and free ligands were separated by rapid filtration of the mem- branes onto Schleicher and Schuell #52 or Whatman GF/B glass fiber filters. Samples were rinsed twice with 5 ml of ice-cold 10 mM Tris- HC1 (pH 7.4). The air-dried filters were immersed in 4 ml of scintilla- tion fluid and their radioactivity was determined.

[35S]t-Butylbicyclophosphorothionate ([3sS]TBPS; Du Pont de Ne- mours, NEN, Dreieich, Germany) binding at 2 nM concentration was determined after 90-rain incubation at 22°C in 50 mM Tris-citrate buffer supplemented with 200 mM NaC1. Nonspecific binding was defined in the presence of 20/~M picrotoxinin (Sigma, St. Louis, MO, USA). Clozapine (Sandoz Research Institute, Berne, Switzer- land), haloperidol (Janssen Pharmaceutica, Beerse, Belgium), the spe- cific GABAA antagonist 2'-(3'-carboxy-2',3'-propyl)-3-amino-6-p- methoxyphenylpyrazinium bromide (SR 95531; Research Biochemi- cals, Natick, MA, USA; Heaulme et al. 1986) and the specific benzo- diazepine antagonist flumazenil (Ro 15-1788; Hoffmann-LaRoche, Basle, Switzerland; Hunkeler et al. 1981) were used with or without GABA (Sigma). Clozapine and haloperidol were dissolved in 0.1 M HCI, and then diluted in assay buffer. SR 95531 and GABA were di- rectly dissolved into the buffer.

(3H]Muscimol and [3H]SR 95531 (both from Du Pont de Ne- mours) binding (at 6 nM each) to cerebellar and cerebrocortical mem- branes was determined with and without various concentrations of clozapine in 50 mM Tris-citrate buffer at 0°C after 60 or 30 rain, re- spectively. The nonspecific binding for the radioligands was deter- mined by 100/zM and 1 mM GABA, respectively.

GABA-induced ~6CF flux. Brain vesicles from combined cerebrocorti- cal and cerebellar tissues were prepared as described in Korpi and Uusi-Oukari (1989) with the modification that 100/~M phenylmethyl- suphonyl fluoride (Sigma) was included in the first homogenization step to prevent protein degradation. Aliquots of 250 #1 (about 250/~g protein) were preincubated in assay buffer (NaC1 145 raM, KC1 5 mM, MgC12 1 mM, CaCI2 1 raM, D-glucose 10 raM, Hepes 10 raM, pH 7.5) for 2 (acute actions of clozapine) or 10 (clozapine preincubation) rain at 30°C. Then 250/A of assay buffer containing 0.4 ,uCi 36CF (Du Pont de Nemours) in the presence of 20 ¢zM GABA with and without clozapine (1-100/IM) was rapidly mixed with the vesicles. The flux was terminated after 3 s by quenching the samples with 4 mi of ice-cold assay buffer containing 100/IM picro- toxinin, followed by filtration onto Whatman GF/C glass fiber filters using a Hoefer filtration unit. The filters were washed twice with the same buffer, dried, and the radioactivity measured in Wallac Opti- phase HiSafe scintillation cocktail using a Wallac 1410 liquid scintil- lation counter. Previous experiments have shown that the assay deter- mines GABAA agonist-induced chloride uptake, which is picrotoxin and bicuculline sensitive (Korpi and Uusi-Oukari, 1989). In the pre- sent experiments, the ECs0 for GABA was 22.5+2.0/~M (mean + SEM, n = 4).

Autoradiography. The procedure used was modified from Olsen et al. (1990) and Edgar and Schwartz (1990). Briefly, 14-/~m frontal sec- tions from frozen whole brains of Wistar rats were cut in a Leitz 1720 cryostat at the following levels (in mm) from the bregma according to Paxinos and Watson (1982): 7, 2.7, 1.5, -0.8, -1.8, -3.3, -5.3, -6.3, -8.5, and -10.3. Sections were preineubated in an ice- water bath for 15 min in 50 mM Tris-HC1 (pH 7.4) supplemented with 120 mM NaC1. Incubation with [35S]TBPS (200 dpm//~l, ad- justed to 6 nM with cold TBPS) for 90 rain at room temperature (22°C) was performed in the same buffer, using 600-#1 liquid bubbles over sections on object glasses in a humid chamber in the presence and absence of 5/~M GABA and/or 100 #M clozapine. After incuba- tion, the sections were washed three times for 15 s in ice-cold incuba- tion buffer, dipped into distilled H20, air-dried at room temperature, and exposed to Hyperfilm-flmax (Amersham, UK) for 3-5 days. 20/~M picrotoxinin reduced the signal to background level (not shown). Regional labeling intensities were quantitated from the films

Fig. 1 Effects of clozapine (left), haloperidol (center) and SR 95531 (right) on picrotoxi- nin-sensitive convulsant binding in various brain regions. [35S]TBPS binding without (©) and with 5 #M exogenous GABA (O) in hippocampal (Hi), cerebrocortical (Co), and cerebel- lar (Ce) membranes. Results are expressed as percentages (mean_+SEM, n = 3) of basal binding determined in the ab- sence of added GABA (= 100%). The values on the left of the gaps are in the absence of drags. Significance of the differ- ence from corresponding control values (Student's t-test): * p<0.05, ** p<0.01, *** p<0.001. All SR 95531 concentrations significantly (p<0.001) antagonized GABA-in- hibition of the binding (signifi- cances not shown in the figures)

t- "o t- ..Q

o

v

.c_ "o c.- ..Q O9 13. 133 1-- 6O

150

Hi 100 o- ** **

It"

0 ---~ I -6 -5 -4

150

C x

100 o- . ***

50 * * *

' , 0 ---4 ~ - -5 -4

150

C e

100 o-

50 * * *

log [Clozapine]

15° I Hi 100[ o - - . _ ~

2°[ o ,, .; _; .~

120[

looF

C x

°°I [ , - - ~

*"---4

is° i C e lo0[ 501 -

• ~ ~ . ~ . ~ *

i F i 01 H -6 -5 -4

log [Haloperidol]

t5(:

10C

367

Hi

/

---~I -6 - -4 15o r

0t----~ i I I I -6 -5 -4

150

** C e

100 o ~

50

,/

i i f 0 ~ I -6 -5 -4

log [SR 95531]

by using MCID MI and M4 image analysis devices and programs (Imaging Research, St. Catharines, Canada). Locations of various brain areas on exposed films were identified with the aid of the same brain sections stained with thionin. Binding densities for each brain area were averaged from measurements from two to three sections.

14 Plastic C-standards (Amersham) exposed simultaneously with the brain sections were used as reference with the resulting binding va- lues given as radioactivity levels estimated for gray matter areas (nCi/ g). Sample autoradiographs from horizontal sections were photo- graphed as positive images to illustrate regional effects of clozapine.

Statistics. Statistical significance of the differences between two group means was assessed with Student's or paired t-tests or analysis of variance with Tukey-Kramer post hoc test with Instat program (GraphPad Software, San Diego, CA, USA).

Results

Micromolar concentrations of clozapine inhibited picrotox- inin-sensitive [35S]TBPS binding in the absence of exo- genous GABA and antagonized the inhibition of the bind- ing induced by 5 #M GABA in rat hippocampal and cere- brocortical membranes (Fig. 1). In cerebellar membranes, clozapine failed to affect [35S]TBPS binding in the ab- sence of exogenous GABA and only partially reversed the inhibition of the binding induced by GABA. Already 3 #M clozapine significantly antagonized the inhibition of 5 #M GABA in all tissues studied. Haloperidol signifi- cantly affected [35S]TBPS binding only at 100/zM con- centration, but failed to antagonize GABA inhibition of the binding in the three brain regions (Fig. 1). Instead, ha- loperidol slightly potentiated the GABA inhibition. Our re- sults on the differential actions of clozapine and haloperi-

dol on [35S]TBPS binding confirm previous results by Squires and Saederup (1991) with forebrain membranes. The specific GABAA receptor antagonist SR 95531 signif- icantly (p<0.05) decreased hippocampal [35S]TBPS bind- ing in the absence of GABA, while slightly enhancing the binding in cerebellar membranes (Fig. 1). SR 95531 po- tently antagonized the GABA inhibition of the binding in all brain regions studied, causing a significant (p<0.01) elevation over basal binding (--100%) in hippocampal membranes.

Although modulation of GABA-inhibition of [35S]TBPS binding predicts the functional properties of various ligands on GABAA receptors (Wood et al. 1984; Im and Blakeman 1991; see: Ltiddens and Korpi 1995), we also determined the mode of clozapine action on GA- BAA receptor function using picrotoxinin-sensitive GABA-induced 36C1 flux into brain vesicles. Figure 2 de- picts that 10 and 100/zM clozapine significantly inhibited GABA-induced flux, both when clozapine was added to- gether with GABA and when it was present already dur- ing a 10-rain preincubation. Basal flux was not enhanced by clozapine alone. These results indicate that clozapine is a GABAA antagonist without any partial agonist actions and that antagonism of GABA-inhibition of [35S]TBPS binding can be used to study the GABAA antagonism of clozapine.

Basal [35S]TBPS binding to rat brain sections was widely detected both in cerebellum and forebrain (Fig. 3; Table 1), roughly in agreement with the regional distribu- tion observed by Edgar and Schwartz (1990) and Olsen et al. (1990). The effect of clozapine (100/ tM) without exo- genous GABA on [35S]TBPS binding was similar in all

368

500

400 ..Q

"~ 300

× 200 i

O 100

0

o - - - ~

, , _; _; log [Clozapine]

Fig. 2 Clozapine antagonism of picrotoxinin-sensitive GABA-in- duced 36C1- flux in brain vesicles. Flux in the absence of added GABA ((D); GABA-induced flux with clozapine added together with GABA, (0 ) ; GABA-induced flux with clozapine added during 10- min preincubation, ([]). Results are given as percentages (mean_+SEM, n = 3) of basal flux. Significance of the difference from GABA alone (Student's t-test): * p<0.05 ** p<0.01, *** p<0.001

brain regions analyzed (Table 1). It decreased the binding to about 66% of basal binding. GABA (5/~M) decreased the binding to about 30% of basal binding in all brain re- gions. We detected decreases larger than average in the ol- factory bulb, temporal auditory cortex, medial prefrontal cortex, anterior cingulate cortex, entorhinal cortex and pos- teromedial amygdaloid nucleus (Table 1). Clozapine par- tially antagonized the GABA-induced decrease of [35S]TBPS binding in the forebrain, increasing it from 30% to 58% of basal binding. However, clozapine was in- active or less active in cerebellum and olfactory areas. In the cerebellar granule cell layer clozapine did not reverse the GABA inhibition of [35S]TBPS binding (paired t-test: p>0.8). GABA antagonism of clozapine was more pro- nounced in lateral preoptic and lateral hypothalamic areas. These data suggest that GABA antagonism of clozapine is

GABAA receptor subtype selective, in contrast to the spe- cific GABAA receptor antagonists bicuculline (50/~M) and SR 95531 (50/~M), which reversed the GABA inhibition in all brain regions (Korpi et al. 1992; Korpi and Ltiddens 1993).

Clozapine significantly affected the binding of [3H]muscimol and [3H]SR 95531 in rat cerebellar and cer- ebrocortical membranes (Fig. 4A, B). The binding of GA- BA A agonist [3H]muscimol was less affected by clozapine than the binding of GABAA antagonist [3H]SR 95531. [3H]Muscimol binding was slightly more inhibited by 100 #M clozapine in cerebrocortical than cerebellar mem- banes, whereas no regional difference was observed in clo- zapine-sensitivity of [ H]SR 95531 binding. The benzodia- zepine antagonist flumazenil, at a concentration that blocks all benzodiazepine receptors (10#M), did not block the actions of clozapine on cerebrocortical [35S]TBPS binding in the absence and presence of GABA (Fig. 4C).

Using recombinant GABAA receptors we determined the efficacy of clozapine as a GABA antagonist of alfl2y2 and a6/?272 receptors. Clozapine inhibited [35S]TBPS binding without exogenous GABA similarly in both recep- tor subtypes, but it effectively reversed the GABA-induced inhibition of [35S]TBPS binding only in alf1272 receptors (Fig. 5A). 10 #M GABA inhibited [35S]TBPS binding and 10/~M SR 95531 reversed that inhibition in alfi272 and a6f12y2 receptors (Fig. 5B). However, 100/~M clozapine was effective only in the former receptor subtype.

In a6/?x72 receptors, clozapine did not antagonize GABA inhibition of [35S]TBPS binding in the presence of /?2 or/?3 variants (Fig. 6A, B), but was active on a6/?172 receptor. This indicates that domains necessary for cloza- pine antagonism are determined by both a and /? subunit variants. GABA inhibition of [35S]TBPS binding was sig- nificantly antagonized by clozapine in all al/?xy2 recep-

BASAL GABA GABA

+

CLOZAPINE

Fig. 3 Brain regional antagonism by clozapine of GABA inhibition of [3sS]TBPS binding as revealed by ligand autoradiography. Autora- diographs of picrotoxinin-sensitive [~sS]TBPS binding in rat brain hor- izontal sections depict basal binding, GABA (5 #M), and GABA

(5/~M) + clozapine (100 #M). Ce cerebellum; CPu caudate-putamen; Cx cerebral cortex; Hi hippocampus; IC inferior collicnlus; OB olfac- tory bulb; S septum; Th thalamus

369

Table 1 Effects of clozapine on regional binding of [35S]TBPS in rat brain sections as revealed by autoradiography

Brain region Basal binding Clozapine GABA Clozapine + GABA

Olfactory areas Olfactory bulb, external plexiform layer 799_+126"* 64.0+_22.9 5.2+_0.8*** 17.2+_1.6"**

glomemlar layer 223+_47 67.1+_27.9 8.7.+3.4** 26.9+_3.3** internal granular layer 107+_7'** 59.2+11.5 20.1+4.7 52.2+7.3

Olfactory tubercle 138+.35" 77.2+_19.2 16.9_+4.9 49.4+_20.3 Islands of Calleja 681+140"* 65.1+6.I 7.6+_3.1" 36.9+_7.6 Primary olfactory cortex 193+8"* 61.3+12.4 16.5+2.3" 40.5_+4.7* Cerebral cortical areas Frontoparietal cortex, motor 269.+31 66.3.+5.5 28.9_+7.1 58.0_+12.9

layers MII 232_+13 69.4_+9.8 17.3_+4.3 41.7_+9.7 layers IV-VI 283_+40 65.5_+4.2 33.0_+7.8 62.9+_13.2

Temporal cortex, auditory 258+_3 55.2_+7.5 15.6_+3.9" 38.9+_11.5 layers I-III 196+43 56.2_+14.9 13.7_+4.8" 39.4+_14.1 layers IV-VI 316+61 52.9_+12.8 18.9+_3.7 40.2_+16.1

Limbic areas Medial prefrontal cortex 197_+19" 69.5_+12.8 14.6_+2.1"* 38.0_+6.9 Anterior cingulate cortex 272+16" 68.3_+11.0 14.1_+3.0" 32.9_+10.2 Entorhinal cortex 186+28 54.8_+8.8 14.1+2.6"* 39.0_+7.5 Subiculum 252_+45 64.9_+11.3 32.5_+6.2 60.7_+11.4 Hippocampus, CA1 129_+18"* 65.6_+8.7 24.4_+6.4 51.8_+10.9

CA3 142_+20* * 68.6_+ 12.3 23.4_+7.2 61.9_+ 15.4 dentate gyms 149_+16"* 70.8_+5.1 21.8_+5.3 58.7_+8.5

Bed nucleus stria terminalis 203_+16 66.9_+7.7 23.9_+3.8 61.5_+3.6 N. of horizontal limb of diagonal band 438_+38"* 67.5+ 11.2 31.7_+3.8 59.0_+5.8 Septohippocampal nucleus/teania tecta 95_+22** 82.4+_5.9 21.0+4.6 51.9_+5.1 Lateral septal nuclei 196_+39 69.6_+6.5 38.9_+11.8 76.1_+14.5 Triangular septal nucleus 95_+33** 56.1_+11.8 52.0_+17.6 77.3+_25.6 Bed nucleus of anterior commissura 279_+45 56.4_+13.9 24.7_+6.4 53.2+_6.5 Amygdala 203 _+21 60.6+_ 1.5 * 21.1 -+48 50.7_+9.8 Posteromedial cortical amygdaloid n. 187_+34 61.8_+6.6 13.9_+3.3" 40.1_+7.4 Basal ganglia Nucleus accumbens 185_+31 61.9_+4.9 19.2_+4.1 58.6_+7.7 Caudate/Putamen 146_+25 * * 74.0_+4.8 39.8_+9.4 81.7_+ 15.1 Globus pallidus 329+40* 64.7_+5.7 39.3_+6.6 67.4_+9.2 Clanstrum 311+_29" 68.7_+7.9 24.5_+6.6 48.3_+8.7 Thalamus Paraventricular thalamic nucleus 210_+27 66.6+_1.1 19.8+_11.1 49.4+_15.0 Anterodorsal thalamic nucleus 253_+48 71.2_+9.5 38.0_+12.8 62.5_+16.2 Centrolateral/medial thalamic nucleus 246_+50 64.7_+3.8 35.2+9.3 55.1_+ 15.1 Intermediodorsal thalamic nucleus 196_+27 65.4+_4.4 21.2+5.8 44.9+_9.9 Lateral posterior/laterodorsal thalamic n. 199+26 68.4+.4.5 32.7_+4.9 47.0_+6.5 Ventroposterior thalamic nucleus 206+_36 76.4_+6.0 44.2_+7.3 52.7_+8.4 Zona incerta/Subthalamic nucleus 235_+43 64.1_+1.7 47.2_+10.5 77.2_+16.3 Medial habenular nucleus 110_+36"* 69.3_+10.9 49.2_+21.8 75.5+_36.7 Medial geniculate nucleus 211_+46 69.4_+22.9 45.1 +9.2 72.9_+ 18.3 Hypothalamus Lateral preoptic area 216+_19 60.9_+8.1 39.8_+2.1 * 72.7+4.5* Lateral hypothalamic area 176_+18** 69.2+15.1 45.6_+9.1 88.3+.9.8* Anterior hypothalamic area 191_+18" 68.0-+14.4 31.2_+9.7 76.3_+9.4 Paraventricular hypothalamic nucleus 119+_9'** 67.5_+10.6 20.2_+7.8 53.3_+12.4 Ventromedial hypothalamic nucleus 155+_17"* 61.1+9.4 21.7_+6.5 60.7_+14.1 Midbrain Substantia nigra, pars reticulata 318+.58 63.9_+12.8 31.6_+4.2 59.9_+11.6 Ventral tegmental area 62_+21"** 55.9_+14.4 105.4+39.6 137.4+_34.5 Interpeduncular nucleus 157_+42' 61.7_+11.4 33.8_+7.8 65.0_+12.1 Red nucleus 182+_17" 57.7+_8.1 51.7_+5.7" 83.9_+12.9 Superior colliculus, superior gray layer 263_+50 60.8_+9.9 30.9+.11.5 66.5_+15.6 Superior colliculus, intermediate layer 262_+55 63.9_+10.2 45.5_+10.9 81.2_+16.2 Central gray 222_+58 64.9_+6.0 41.5_+11.7 78.6_+17.4 Inferior colliculus, external nucleus 220_+64 68.8_+13.8 46.8_+11.9 83.4_+19.5 Interior colliculus, central nucleus 319_+84 63.4_+11.3 42.6+_10.2 76.6+_16.9 Cerebellum Granule cell layer 244_+88 130 _+29 23.1_+6.2 22.7_+8.7" Molecular layer 201_+44 68.6_+13.2 12.3_+6.6 19.8+_7.4"

Average 230 66.4 29.7 57.8

The concentrations of clozapine and GABA were 100 and 5/zM, respectively. Optical density values in relation to radioactivity standards are means_STD for 4 brains. Basal binding is in nCi/g, other data is given as percent of basal binding. Significance of the difference from the cor- responding overall average value shown at the bottom of the table for each condition (Student's t-test): * p_<0.05, ** p<0.01, *** p<0.001

370

log [Clozapine] log [Clozapine]

'1~ "O

25 25

~' "5

100

50.

Clozapine 100 ,aM

c

+ + + +

FlumazeNI 10 ,uM + + + +

GABA 5 ~ M . . . . + + + +

Fig. 4 Effects of clozapine on GABA-sensitive [3H]muscimol (A) and [3H]SR 95531 (B) binding in cerebellar (©) and cerebrocortical membranes (I) , and failure of flumazenil to block clozapine actions on [35S]TBPS binding in cerebrocortical membranes (C). Results are expressed as percentages (mean+SEM, n = 3) of basal of each ligand. The highest clozapine concentration significantly (p<0.01; Student's t-

3 3 test) decreased both [ H]muscimol (A) and [ H]SR 95531 (B) binding in both brain regions. Flumazenil neither affected the decrease nor the

35 increase of [ S]TBPS binding by clozapine in the absence and pre- sence of GABA, respectively (ANOVA, p>0.05). Significance of the difference from the corresponding cerebellar value (Student's t-test): ** p < 0 . 0 1

15C

L6 i0C

2 ~ m2

'o 5c rq

.'-%

II I I I

6 -5 -4

log [ C I o z a p i n e ]

GABA GABA GABA + +

SR 95531 CIozapine

Fig. 5 Effects of clozapine on [35S]TBPS binding in a6 and al sub- unit-containing recombinant GABAA receptors. A binding of [35S]TBPS without (©, rq) and with ( 0 , I ) GABA in a6fi272 (©, II) and alf1272 ([2, I ) GABAA receptors. GABA concentration was 1 and 10 ffM with a6f12y2 and alfl2y2 receptors, respectively, because a6f1272 receptors are at least 10 times more sensitive to GABA than alfl2y2 receptors (Korpi and Lfiddens, 1993). B binding in the presence of 10 pM GABA with and without SR 95531 (10/IM) or clozapine (100/IM) in a6fi2y2 (open bars) and alfl2y2 (stippled bars) receptors. Results (mean_+SEM, n = 3) are expressed as percentages of the basal binding for each receptor, Significance of the difference from corresponding control values (Student's t-test): **

p < 0 . 0 1 , *** p < 0 . 0 0 1

tors. The al subunit and the fl variants of cdfix72 recep- tors interacted to modulate the efficacy of clozapine in the order of alfily2>ulf12y2>alf1372 (Fig. 6C, D).

f • 15o

E 100

so

0

o~ 150

25 lOO

CI] I - - 5 0

CO L J

A

I I

rl - 6 --15 -4

I t I II -6 -5 - 4

15o I

I B 1ool.- . . . . . . . . . . . . . . . . . . . . ;,~:;.

O' IF 6 5 -4

250

I50

i00 . . . . . . . . . . . . .

- 6 - 5 - 4

log [Clozapine]

Fig. 6 Effects of fl subunit variants on clozapine modulation of 35, [ S]TBPS binding in a6fll/372 and alfll/3y2 receptors. A modula-

tion of [35S]TBPS binding by clozapine in the absence of GABA in a6flxy2 receptors. B effect of clozapine on [35S]TBPS binding in the presence of 1 #M GABA in a6fixy2 receptors. C modulation of [35S]TBPS binding by clozapine in the absence of GABA in Mflxy2 receptors. D effect of clozapine on TBPS binding in the presence of 10/IM GABA in alflxy2 receptors, al/6flly2 ([2); al/6f12y2 (©); al/ 6f1372 (O)- Results are given as percentages (mean_+SEM, n = 3) of basal binding that is depicted by dotted lines. Significance of the dif- ference from corresponding control values (Student's t-test): * p<0.05, • * p < 0 . 0 1 , *** p < 0 . 0 0 1

Discussion

We found the atypical antipsychotic clozapine to antago- nize central GABAA receptors in most brain regions, which agrees with the [BsS]TBPS binding data of Squires and Saederup (1991; 1993) on rat forebrain membranes. The typical antipsychotic haloperidol did not share this clozapine action. Furthermore, clozapine was inefficient in antagonizing GABAA receptors in the cerebellum, which we delineated to the specific subunit composition of the receptors involved.

Clozapine as a GABAA antagonist

Our results demonstrate that clozapine antagonizes GA- BA a receptor function without any partial agonism (Fig. 2). Findings of Squires et al. (1983) indicate that mi- cromolar concentrations of clozapine have a negligible af- finity to [35S]TBPS recognition site, which can be ex- cluded as a site of clozapine action also by our results as clozapine enhanced [35S]TBPS binding in recombinant re- ceptors over basal binding level in the presence of exogen- ous GABA. Clozapine does not act via the benzodiazepine recognition sites (Fig. 4C), either.

Clozapine is reported to inhibit [3H]GABA binding to brain membranes (Lloyd and Dreksler 1979) suggesting that the mechanism of clozapine action on the GABAA re- ceptor complex involves the GABA recognition site. This

371

interpretation was partially supported by the present ex- periments. First, clozapine inhibited the binding of [3H]muscimol and [3H]SR 95531, an agonist and antago- nist, respectively, that recognize the GABA site of the re- ceptor. Cerebrocortical [3H]muscimol binding was slightly more inhibited than cerebellar binding, which agrees with regional profile of GABA antagonism by clozapine, but no such difference was observed in [3H]SR 95531 bind- ing. Second, in recombinant receptors clozapine elevated [35S]TBPS binding above the basal binding level in the presence of GABA (Figs 5A and 6D) similarly to SR 95531 (Fig. 5B), suggesting that both antagonists de- creased the effective GABA concentration to values that actually enhanced the binding in pre-equilibrium condi- tions of a 90-min incubation (binding equilibrium is ap- proached only at 180 min; Maksay and Simonyi 1986; Korpi and Ltiddens 1993). On the other hand, clozapine was apparently less potent in interfering with [3H]muscimol and [3H]SR 95531 binding than with GABA-inhibition of [35S]TBPS binding in native receptor preparations. This suggests that the GABA recognition sites can not be the only site of clozapine antagonism, the exact site(s) remaining unknown. Molecular pharmacologi- cal studies on mutant and chimeric recombinant GABAA receptors are expected to eventually provide us with the information on critical subunit domains for various ligand binding sites (see below).

To explain decreased [35S]TBPS binding in recombi- nant receptors by clozapine in the absence of GABA (Fig. 5), we may assume that, in cerebellar sections and membranes, the direct effect of clozapine in the absence of exogenous GABA and/or other modulators, such neuro- steroids, was still blocked by residual agonists (Table 1, Fig. 1). Residual GABA has been demonstrated to strongly affect [35S]TBPS binding in brain sections (Edgar and Schwartz 1990), especially in the cerebellar cortical layers (Korpi et al. 1992). The direct effect on recombi- nant receptors may thus reflect intrinsic action of cloza-

• 3 5 pine on [ S]TBPS binding, similar to that of GABAA an- tagonists bicuculline, 3a-hydroxy-ll-oxo-16-imino-17-aza- 5fi-androstane (RU 3135) and SR 95531 (Korpi and Lfid- dens 1993), and not their GABA antagonism. The direc- tion of this action is reminiscent of agonist action, but our functional data indicated that clozapine is devoid of any agonistic activity. Indeed, similar bidirectional modulations of [35S]TBPS binding have also been observed for the benzodiazepine recognition site ligands in the absence and presence of GABA (see: Maksay and Simonyi 1988; Im and Blakeman 1991), indicating that only in the presence of micromolar GABA can the direction of the allosteric li- gand-induced change in [35S]TBPS binding be used to pre- dict pharmacological activity.

Brain region-specific antagonism: dependence on receptor subunits

We excluded some GABAA receptor subunit combinations as targets of clozapine. Each pentameric GABAA receptor

is believed to assemble from members of the a (al-6), fl (ill-3) and y (yl-3) subunit classes (Olsen and Tobin 1990; Wisden and Seeburg 1992). The a6 subunit is un- iquely expressed in cerebellar granule cells (Lfiddens et al. 1990), the region in which GABA inhibition of [35S]TBPS binding was least antagonized by clozapine. Using recom- binant receptors, we found a6fi2/372 receptors to be insen- sitive to clozapine, which agrees with the autoradiographic localization of clozapine action on GABAA receptors in rat brain sections. Since a6flly2 receptors were sensitive to clozapine, their contribution to the granule cell GABAA receptor population must be negligible. We drew the same conclusion in a recent study, in which we established furo- semide as a selective antagonist of cerebellar granule cell GABAa receptors of the form a6fi2/3y2, although it did not affect the a6fily2 receptors (Korpi et al. 1995).

alfixy2 are abundant GABAA receptors in most brain regions (Benke et al. 1994). They are most prominent in cerebellar molecular layer and olfactory bulb external plexiform layer (Laurie et al. 1992), where clozapine sig- nificantly (paired t-test: p<0.01) enhanced the GABA-in-

35 hibited [ S]TBPS binding in contrast to the cerebellar granule cell layer (p>0.8). However, GABA inhibition of [35S]TBPS binding was less affected by clozapine in the former two brain regions than in other regions. Since alfl3y2 receptors were least prone to clozapine action, their proportion in the GABAA receptor population of the olfactory bulb external plexiform layer might be signifi- cant, partially explaining the reduced clozapine efficacy. The/73 variant is abundantly expressed in olfactory bulb, but it is rare in the cerebellar molecular layer (Laurie et al. 1992; Persohn et al. 1992; Wisden et al. 1992). On the other hand, f13 subunit expression is similar in olfactory bulb external plexiform layer (lower than average sensitiv- ity to clozapine) and in lateral hypothalamic areas (higher than average sensitivity to clozapine) (Laurie et al. 1992; Persohn et al. 1992; Wisden et al. t992). Hypothalamic areas are thus enriched in/?3 subunit mRNA, but a2 is the main a subunit rather than al (Wisden et al. 1992), sug- gesting that a2/?372 receptors might show increased cloza- pine efficacy as compared to alfl3y2 receptors.

Our results with recombinantly expressed a6flxy2 and al/?x72 receptors indicate that clozapine antagonism is de- termined by an interaction of a and fl subunits. The a sub- units, together with y variants, play a decisive role in GA- BAA receptor function by determining benzodiazepine pharmacology (Pritchett et al. 1989a, b; Liiddens et al. 1994; see: Lfiddens and Wischen 1991). The present data further indicates the importance of the fi variants in the pharmacological heterogeneity of GABAA receptor sub- types, which was recently suggested by the obligatory pre- sence of the f13 variant in a5flxy213 receptors to form [35S]TBPS bindings sites (Ltiddens et al. 1994). Thus, clo- zapine may be used as a pharmacological probe to define amino acid residues of a and fl subunits critical for their molecular interaction in the function of GABAA receptors. Furthermore, clozapine may be used as a lead structure to develop more potent subtype-selective ligands for GABAA receptors.

372

GABAA antagonism and clozapine-induced seizures

Micromolar concentrations of clozapine are needed to ob- serve GABAA receptor antagonism. If relevant drug con- centrations acting on neurotransmitter receptors in vivo are close to levels estimated to prevail in the cerebrospinal fluid and plasma water (10-20 nM, see: Seeman 1992), it is unlikely that this receptor system is involved in the ther- apeutic effects of clozapine.

Increased risk for seizures in clozapine-treated patients might be associated with interference of forebrain inhibi- tory GABAerg ic activity. It coincides with prolonged high-dose therapy (Devinsky et al. 1991; Pacia and De- vinsky 1994), when clozapine concentrations in blood plasma (and presumably those in the brain; see: Hartvig et al. 1986) are elevated to micromolar levels (Choc et al. 1987; Hating et al. 1994), or with a rapid titration phase at low doses especially in patients with a history of seizures or epilepsy (Pacia and Devinsky 1994). This mechanism cannot account for the proconvulsive properties of neuro- leptic drugs generally (Messing et al. 1984), since haloper- idol failed to antagonize G A B A at micromolar concentra- tions (Fig. 1) that are known to prevail in the brain of ha- loperidol-treated patients (Korpi et al. 1984). Our results suggest further that cerebellar GABAA receptors cannot be involved in seizures associated with clozapine treatment.

Acknowledgements The authors wish to thank Pirkko Johansson and Sabine Grtinewald for expert technical assistance, and Peter H. Seeburg for constructive critique on the project. The study was sup- ported by the Academy of Finland and the Deutsche Forschungsge- meinschaft (SFB 317/B9).

References

Angst J, Jaenicke U, Padrutt A, Scharfetter C (1971) Ergebnisse eines Doppelblindversuches von Clozapin (8-chlo-11-(4-methyl-1- piper- azinyl)-5H-dibenzo-(b,e)-l,4)-diazepin) im Vergleich zu Levome- promazin. Pharmacopsychiatry 4:192-200

Baldessarini RJ, Frankenburg FR (1991) Clozapine - a novel antipsy- chotic agent. N Engl J Med 324:746-754

Benke D, Fritschy J-M, Trzeciak A, Bannwarth W, Mohler H (1994) Distribution, prevalence, and drag binding profile of 7-aminobuty- ric acid type A receptor subtypes differing in the fl-subunit var- iant. J Biol Chem 269:27100-27 107

Chen C, Okayama H (1987) High-efficiency transformation of mam- malian cells by plasmid DNA. Mol Cell Biol 7:2745-2752

Choc MG, Lehr RG, Hsuan F, Honigfeld G, Smith HT, Borison R, Volavka J (1987) Multiple-dose pharmacokinetics of clozapine in patients. Pharm Res 4:402-405

Conn PJ, Sanders-Bush E (1985) Serotonin-stimulated phosphoinosi- tide turnover: mediation by the $2 binding site in rat cerebral cor- tex but not in subcortical regions. J Pharmacol Exp Ther 234:195-203

Coward DM (1992) General pharmacology of clozapine. Br J Psy- chiatry 160 lSuppl 17]:5-11

Deutch AY, Moghaddam B, Innis RB, Krystal JH, Aghajanian GK, Bunney BS, Chamey DS (1991) Mechanisms of action of atypical antipsychotic drugs. Implications for novel therapeutic strategies for schizophrenia. Schizophr Res 4:121-156

Devinsky O, Honigfeld G, Patin J (1991) Clozapine-related seizures. Neurology 41:369-371

Edgar PR Schwartz RD (1990) Localization and characterization of 35S-t-butylbicyclophosphorothionate binding in rat brain: an auto- radiographic study. J Neurosci 10:603-612

Farde L, Nordstr6m AL, Wiesel FA, Pauli S, Halldin C, Sedvall G (1992) Positron emission tomographic analysis of central D, and D2 dopamine receptor occupancy in patients treated with classical neuroleptics and clozapine. Relation to extrapyramidal side ef- fects. Arch Gen Psychiatry 49:538-544

Gaertner HJ, Fischer E, Hoss J (1989) Side effects of clozapine. Psy- chopharmacology 99:$97-$100

Hating C, Neudorfer C, Schwitzer J, Hummer M, Saria A, Hinterhu- ber H, Fleischhacker WW (1994) EEG alterations in patients trea- ted with clozapine in relation to plasma levels. Psychopharmacol- ogy 114:97-100

Hartvig P, Eckern~is S, LindstrOm L, Ekblom B, Bondesson U, Lundqvist H, Halldin C, Nftgren K, L~ngstr6m B (1986) Receptor binding of N-(methyl-11C)clozapine in the brain of rhesus monkey studied by positron emission tomography (PET). Psychopharma- cology 89:248-252

Heaulme M, Chambon JR Leyris R, Molimard JC, Wermuth CG, Bi- ziere K (1986) Biochemical characterization of the interaction of three pyridazinyl-GABA derivatives with the GABAA receptor site. Brain Res 384:224-231

Hunkeler W, M6hler H, Pieri L, Polc R Bonetti ER Cumin R, Schaff- her R, Haefely W (1981) Selective antagonists of benzodiaze- pines. Nature 290:514-516

Idfinp~i~i-Heikkil~i J, Alhava E, Olkinuora M, Palva IP (1977) Agranu- locytosis during treatment with clozapine. Eur J Clin Pharmacol 11:193-198

Im WB, Blakeman DP (1991) Correlation between 7-aminobutyric acidA receptor ligand-induced changes in t-butylbicyclopho-

35 36 sphoro[ S]thionate binding and C1- uptake in rat cerebrocortical membranes. Mol Pharmacol 39:394-398

Kane JM (1992) Clinical efficacy of clozapine in treatment-refractory schizophrenia: an overview. Br J Psychiatry 160 [Suppl 17]:41-45

Kane JM, Honigfeld G, Singer J, Meltzer H (1988) Clozapine for the treatment-resistant schizophrenic: a double-blind comparison with chlorpromazine. Arch Gen Psychiatry 45:789-796

Korpi ER, Ltiddens H (1993) Regional 7-aminobutyric acid sensitiv- 35 ity of t-butylbicyclophosphoro[ S]thionate binding depends on 7-

aminobutyric acidA receptor a subunit. Mol Pharmacol 44:87-92 Korpi ER, Uusi-Oukari M (1989) GABAA receptor-mediated chloride

flux in brain homogenates from rat lines with differing innate al- cohol sensitivities. Neuroscience 32:387-392

Korpi ER, Kleinman JE, Costakos DT, Linnoila M, Wyatt RJ (1984) Reduced haloperidol in the post-mortem brains of haloperidol- treated patients. Psychiatry Res 11:259-269

Korpi ER, Ltiddens H, Seeburg PH, (1992) GABAA antagonists re- veal binding sites for [35S]TBPS in cerebellar granular cell layer. Eur J Pharmacol 211:427-428

Korpi ER, Kleingoor C, Kettenmann H, Seeburg PH (1993) Benzo- diazepine-induced motor impairment linked to point mutation in cerebellar GABAA receptor. Nature 361:356-359

Korpi ER, Kuner T, Seeburg PH, Lfiddens H (1995) Selective antago- nist for the cerebellar granule cell-specific 7-aminobutyric acid type A receptor. Mol Pharmacol 47:283-289

Kuoppam~ki M, Syv~ilahfi E, Hietala J (1993) Clozapine and N-des- methylclozapine are potent 5-HTm receptor antagonists. Eur J Pharmacol - Mol Pharmacol Sect 245:179-182

Laurie DJ, Seeburg PH, Wisden W (1992) The distribution of 13 GA- BAA receptor subunit mRNAs in the rat brain. II. Olfactory bulb and cerebellum. J Neurosci 12:1063-1076

Lloyd KG, Dreksler S (1979) An analysis of [3H]gamma-aminobuty- ric acid (GABA) binding in the human brain. Brain Res 163:77- 87

Ltiddens H, Korpi ER (1995) Biological function of GABAa/benzo- diazepine receptor heterogeneity. J Psychiatry Res 29:77-94

Lfiddens H, Wisden W (1991) Function and pharmacology of multi- ple GABAA receptor subunits. Trends Pharmacol Sci 12:49-51

Lfiddens H, Pritchett DB, Kohler M, Killisch I, Kein~nen K, Monyer H, Sprengel R, Seeburg PH (1990) Cerebellar GABAA receptor selec- tive for a behavioural alcohol antagonist. Nature 346:648-651

373

Ltiddens H, Seeburg PH, Korpi ER (1994) Impact of fl and y variants on ligand-binding properties of y-aminobutyric acid type A recep- tors. Mol Pharmacol 45:810-814

Maksay G, Simonyi M (1986) Kinetic regulation of convulsant (TBPS) binding by GABAergic agents. Mol Pharmacol 30:321- 328

Maksay G, Simonyi M (1988) Nonequilibrium modulation of 3SS- TBPS binding by benzodiazepine agonists and antagonists. Bio- chem Pharmacol 37:2195-2200

Messing RO, Closson RG, Simon RP (1984) Drug-induced seizures: a 10-year experience. Neurology 34:1582-1586

Olsen RW, Tobin AJ (1990) Molecular biology of GABAA receptors. FASEB J 4:1469-1480

Olsen RW, McCabe RT, Wamsley JK (1990) GABAA receptor sub- types: autoradiographic comparison of GABA, benzodiazepine, and convulsant binding sites in the rat central nervous system. J Chem Neuroanat 3:59-76

Pacia SV, Devinsky O (1994) Clozapine-related seizures: experience with 5,629 patients. Neurology 44:2247-2249

Paxinos G, Watson C (1982) The rat brain in stereotaxic coordinates. Academic Press, New York

Peroutka SJ, Snyder SH (1980) Relationship of neuroleptic drug ef- fects at brain dopamine, serotonin, a-adrenergic, and histamine re- ceptors to clinical potency. Am J Psychiatry 137:1518-1522

Persohn E, Malherbe R Richards JG (1992) Comparative molecular neuroanatomy of cloned GABAA receptor subunits in the rat CNS. J Comp Neurol 326:193-216

Pritchett DB, Ltiddens H, Seeburg PH (1989a) Type I and Type II GABAA-benzodiazepine receptors produced in transfected cells. Science 245:1389-1392

Pritchett DB, Sontheimer H, Shivers BD, Ymer S, Kettenmann H, Schofield PR, Seeburg PH (1989b) Importance of a novel GA- BAA receptor subunit for benzodiazepine pharmacology. Nature 338:582-585

Schotte A, Janssen PFM, Megens AAHR Leysen JE (1993) Occu- pancy of central neurotransmitter receptors by risperidone, cloza-

pine and haloperidol, measured ex vivo by quantitative autoradio- graphy. Brain Res 631:191-202

Seeman P (1992) Dopamine receptor sequences. Therapeutic levels of neuroleptics occupy D2 receptors, clozapine occupies D 4. Neurop- sychopharmacology 7:261-284

Seeman R Van Tol HHM (1994) Dopamine receptor pharmacology. Trends Pharmacol Sci 15:264-270

Squires RF, Saederup E (1991) A review of evidence for GABAergic predominance/glutamatergic deficit as a common etiological factor in both schizophrenia and affective psychoses: more support for a continuum hypothesis of "functional" psychosis. Neurochem Res 16:1099-1111

Squires RF, Saederup E (1993) Mono N-aryl ethylenediamine and pi- perazine derivatives are GABAA receptor blockers: implications for psychiatry. Neurochem Res 18:787-793

Squires RF, Casida JE, Richardson M., Saederup E (1983) [35S]t-Bu- tylbicyclophosphorothionate binds with high affinity to brain-spe- cific sites coupled to y-aminobutyric acid-A and ion recognition sites. Mol Pharmacol 23:326-336

Van Tol HHM, Bungow JR, Guan H-C, Sunahara RK, Seeman R Niznik HB, Civelli O (1991) Cloning of a human dopamine D 4

receptor gene with high affinity for the antipsychotic clozapine. Natme 350:614-619

Wisden W, Seeburg PH (1992) GABAA receptor channels: from subu- nits to functional entities. Curr Opin Neurobiol 2:263-269

Wisden W, Laurie DJ, Monyer H, Seeburg PH (1992) The distribu- tion of 13 GABAA receptor subunit mRNAs in the rat brain. I. Telencephalon, diencephalon, mesencephalon. J Neurosci 12:1040-1062

Wood PL, Loo R Braunwalder A, Yokoyama N, Cheney DL (1984) In vitro characterization of benzodiazepine receptor agonists, an- tagonists, inverse agonists and agonist/antagonists. J Pharmacol Exp Ther 231:572-576

Ymer S, Schofield PR, Draguhn A, Werner R K6hler M, Seeburg PH (1989) GABAA receptor fi subunit heterogeneity: functional ex- pression of cloned cDNAs. EMBO J 8:1665-1670