Neurochem. Int. Vol. 10, No. 4, pp. 405-411, 1987 0197-0186/87 $3.00 + 0.00 Printed in Great Britain. All rights reserved © 1987 Pergamon Journals Ltd

THE GABAA RECEPTOR COMPLEX IN THE DEVELOPING CHICK OPTIC TECTUM: CHARACTERIZATION OF

[3H]MUSCIMOL BINDING SITES

ALICIA BATUECAS, ANA CUBERO, ANA BARAT and GALO RAMiREZ* Centro de Biologia Molecular (CSIC-UAM), Canto Blanco, 28049 Madrid, Spain

(Received I0 July 1986; accepted 18 November 1986)

Abstract--As part of an ongoing study on the GABAA receptor complex in the developing chick optic tectum we describe some properties of the agonist site, as labeled by [3H]muscimol, including meth- odological, kinetic and pharmacological aspects. 16-day embryos and 10-day chicks have been selected as representative age points for the initial characterization of the receptor, prior to more detailed developmental studies. Our data indicate the existence, in both embryos and young birds, of a single class of statistically equivalent, high-affinity, saturable binding sites, with a dissociation constant (Kd) of 80-90 nM in freeze-thawed/washed membranes, and about 8 nM in membranes additionally extracted with low concentrations of Triton X-100. Maximal densities of binding sites are nearly identical in both membrane preparations, ranging from ~2 to ~3 pmol/mg for the two age points considered.

The pharmacological profiles suggest that avian receptors for [3H]muscimol are generally similar to the corresponding mammalian sites, behaving as typical bicuculline-sensitive, baclofen-insensitive type A GABA receptor sites. However, bicuculline and its derivatives are less efficient displacers of [3H]muscimol in detergent-extracted membrane preparations, being in all cases, as usually, much less effective displacers than GABA agonists.

The effect of Triton X-100 on the muscimol site in the GABA A receptor, increasing the affinity for the radioligand by a factor of 10, and diminishing the efficiency of antagonists, is considered here in terms of structural changes in the receptor, induced by the action of the detergent on the membrane microenvironment.

The retinotectal visual system of the chick has been widely used as a developmental model to analyze the orderly expression of different neurotransmission sys- tems, and the regulatory influences of cellular inter- actions and onset of function thereupon (see Ramirez et al., 1980, 1981, 1983; Villafruela et al., 1981) for examples and further discussion). For instance, our own previous work on transmitter amino acid recep- tors in the chick optic tectum suggests that local excitatory and inhibitory processes might be develop- mentally coordinated so to favor the stability of the chick central nervous system upon eye opening (Ramirez et al., 1983).

The choice of [3H]muscimol as a ligand in the above studies was in agreement with the widespread involvement of chloride channel-linked, GABA A re- ceptors in inhibitory mechanisms (Enna, 1983; Sim- monds, 1983). It is generally accepted that GABA A receptor complexes contain modulatory sites for ben-

*To whom correspondence should be addressed.

zodiazepines and barbiturates, in addition to the agonist site (Olsen, 1981; Simmonds, 1983), although a simple three-site model may not be adequate to accommodate some recent findings concerning the pharmacology of the GABA A receptor complex (Leeb-Lundberg and Olsen, 1983; Trifiletti et al., 1985). It is, furthermore, well established that oc- cupancy of one of the sites in the complex modifies the affinity of the other sites for their respective ligands in an aliosteric fashion, this effect being enhanced in the presence of anions that permeate chloride channels (Olsen, 1981; Simmonds, 1983; Olsen et al., 1984; Harris and Allan, 1985).

As a further step in our studies on inhibitory mechanisms in the developing chick optic tectum, we have consequently undertaken a more detailed char- acterization of the different sites in the GABAA receptor complex. The present paper deals with the agonist site, specifically labeled by [3H]muscimol (Beaumont et aL, 1978; Snodgrass, 1978; DeFeudis, 1980), a potent GABAmimetic drug that selectively acts on the A receptor (Simmonds, 1983).

405

406 ALICIA BATUECAS et al.

EXPERIMENTAL PROCEDURES

Tecta from embryos or young chicks (White-Leghorn, raised in our own facilities on a 12h light/12h darkness schedule) were homogenized, by means of a Dounce tissue grinder (Kontes; 10 times pestle A, 6 times B), in 20 volumes of I mM potassium phosphate, pH 7, containing 0,32 M sucrose and 1 mM MgCI 2. The homogenates were centri- fuged at 72,000g. for 30 min. The resulting pellets were lysed (by osmotic shock) in 40 volumes of l mM potassium phosphate, pH 7, for 30min, and centrifuged again at 72,000 g. These pellets were frozen at -80°C, for at least 24 h and, after thawing, they were washed twice in the lysis buffer, or extracted with 0.05% (v/v) Triton X-100, at 37°C, for 30 min, and then washed as above. The final membrane preparations were, in all cases, resuspended in 10raM potassium phosphate buffer, pH 7, and used as such for the binding studies. All preparative procedures were carried out at 4°C. Departures from this standard procedure are de- tailed in the text. No attempt was made at further sub- fractionation of tectal membranes, so to analyze all [3H]muscimol receptors, in the same preparation, irre- spective of their subcellular localization.

Binding assays were performed at 4°C, in small poly- carbonate tubes containing, in a total volume of 1 ml, 0.5 mg of membrane protein and the radioactive ligand ([methylamine-3H]muscimol, 19 Ci/mmol, from the Radio- chemical Centre, Amersham, England), with our without the displacer. The concentration of radioligand varied be- tween 1 nM and 400 nM in the saturation experiments, and was kept constant at 4 nM (Triton X-100-extracted/washed membranes), or 40 nM (washed membranes), in all other cases. After 30 rain of incubation, the binding assay was terminated by centrifugation of the membranes at 72,000 g. The supernatant was discarded, and the walls of the tubes and the top of the pellets were quickly rinsed with cold buffer. The pellets were finally processed for bound radio- activity and protein measurements, as previously described (Grmez-Barriocanal et aL, 1982).

Non-specific binding was routinely measured in the pres- ence of non-radioactive muscimol, at a concentration l0 s times that of the radioactive ligand, and substracted from total [3H]muscimol binding, Whenever possible, the different displacers used were added as neutral potassium salts, and the final pH of the incubation mixture was checked in all c a s e s .

Tectal lobes from a minimum of 6 birds (or 12 embryos) were pooled for any single measurement or binding assay. In this way, the dispersion of experimental data from different preparations was kept below 10% of the mean values given in figures and tables.

RESULTS

Effect of membrane washing and other treatments on the measured binding of [3H]muscimol

A tho rough washing of tectal membranes is neces- sary to obta in reproducible results. Binding is fur- thermore enhanced by freeze-thawing, and by de- tergent extract ion of the membranes , all these procedures being potential ly helpful in removing endogenous G A B A , or o ther b inding inhibi tors , f rom

Table 1. Effect of different membrane washing procedures and other treatments on the specific binding of [3H]muscimol to chick tectal

membranes

Binding Treatment (%)

One additional high-speed washing 60 Two additional washings 58 Freeze-thawing 23 Freeze-thawing/1 washing 82 Freeze-thawing/2 washings 100" Freeze-thawing/3 washings 100 Freeze-thawing/overnight dialysis 70 Freeze-thawing/Triton X-100 extraction/2 washings 223~: Tectal lobes from 10-day chicks were homogenized as described in

Experimental Procedures. After high speed centrifugation of these homogenates, the pellets were rehomogenized and lysed by osmotic shock, then centrifuged again. These pellets were further processed as described in the Table. Dialysis was against lysis buffer. Binding was carried out in the presence of 40 nM [3H]muscimol, for 30 rain, at 4'~C. (In later experiments, 4 nM radioligand was used when incubating Triton X-100-extracted membranes). The results are based on 3 independent experi- ments, with pooled material from 6 chicks, and duplicate tubes. Results using embryonic (16-day) material were qualitatively similar. *Adopted procedure for regular membrane prepara- tions, tAdopted procedure for Triton X-100-extracted mem- brane preparations.

the membrane preparat ion. Table 1 shows the effect of some of these t reatments , alone or in combinat ion , on the measured b inding of [3H]muscimol. Two sets of condi t ions were finally chosen for s tandard experi- ments with or wi thout extract ion of the membrane prepara t ion with Tr i ton X-100. No binding could be detected in chick liver membranes processed as shown in Table 1.

Influence o f incubation conditions on [3H]muscimol binding to chick tectal membranes

Four different incubat ion parameters were ex- plored for their influence on [3H]muscimol binding, bo th in regular and detergent-extracted membranes . The results of these experiments are shown in Fig. 1. The condi t ions adopted for rout ine experiments, namely 30 min /4°C/pH 7/0.5 + 0.1 mg protein (ar- rows), have been taken as 100% in the graphs. W h e n assessing the effect of protein concent ra t ion (Fig. I D), an addi t ional exper iment was carried out in which increasing amount s of liver membranes were added to a fixed concent ra t ion (0.5 mg/ml in terms of protein) of tectal membranes . In this case, only tectal prote in was taken into account, when calculat ing the radiol igand bound per mg o f protein, since liver membranes do not bind [JH]muscimol. The results plot ted in Fig. 1D show tha t higher membrane concentra t ions , either active tectal membranes or mixed tectum/l iver material , give lower [~H]muscimol binding values.

Characterization of [3H]muscimol binding sites 407

t , / ) LIJ

- J

X UJ

Z W n,,"

I,L 0

120 l

IOOF___o._• ..-8 . . . . 8 . . . . 8

60 I I , ,

0 15 30 5 60

140 -(~

120 100

80; 60 i

40

TIME (rain)

1 i i

I I I

6.6 ZO 74 78

pH

0 10 20 30 40

TEMPERATURE (*C)

. . . .

f %'.0"°°.° .. o o

I • I I : I I. 0 0.5 1.0 15 2.0 25

PROTEIN CONCENTRATION (mglml)

120 lO0

80 60

140 120 100 80 60 40

Fig. 1. Optimization of incubation conditions for binding of [3H]muscimol to chick tectal membranes. 10-day chick tectum membranes were prepared, as described in Experimental Procedures, with (O) or without (0) Triton X-100 extraction, and incubated with [aH]muscimol (40 nM for regular membranes and 4 nM for detergent-extracted membranes) under different conditions. The effect of changes in incubation time (A), temperature (B), pH of medium (C) and membrane protein concentration (D) were systematically explored. Three parameters were maintained at reference values (arrows), derived from preliminary experiments, and the fourth parameter was varied within the limits shown. The results are expressed as % binding efficiency relative to adopted reference values (100%). In the case of protein concentration (D) both the effect of increasing tectal protein (0 , O) and of the addition of (detergent- extracted) liver membranes to a fixed amount of tectal membranes (t-I) were analyzed. % values have been calculated as means of two independent experiments, with duplicate samples in each one, pooling tecta from 6 different birds in the initial homogenates. Similar results were obtained when using membranes

from embryonic tecta (16-day).

Other requirements for optimal measurement of binding, such as the use of centrifugation to separate unbound radioligand, have been established in pre- vious papers (Ramirez et al., 1983; G6mez- Barriocanal et al., 1982).

Using the reference conditions established in Table 1 and Fig. 1, close to 0.9 pmol of [3H]muscimol per mg of protein is bound to regular membranes in the presence of 44t nM radioligand, or to Triton X-100-extracted membranes when using 4 nM radio- ligand.

Saturation kinetics of [3H]muscimol binding to chick tectal membranes

The kinetic behavior of the agonist site in the tectal GABAA receptor complex was analysed, in satur- ation experiments, using a radioligand concentration range of 1 nM to 400 nM. Computer-fitted Scatchard plots of the results of these saturation experiments are given in Fig. 2. Embryonic (16-day; El6) and post- natal (10-day; P10) tectal lobes have been indepen- dently studied, using both regular (Fig. 2, left) and

Triton X-100-extracted (Fig. 2, right) membrane preparations. The kinetic parameters derived from graphs in Fig. 2, and also from Hill plots of the same data (not shown), are given in Table 2. Taken together, these results suggest that we are dealing with a single population of statistically equivalent binding sites, and that no significant changes in the behavior of the agonist site occur during devel- opment. Extraction of tectal membranes with Triton- X-100 results in a 10-fold increase in ligand-receptor affinity, with no significant change in the total num- ber of sites labeled by [3H]muscimol, especially if we take into account that detergent extraction entails a loss of membrane protein (about 30%).

Pharmacological characterization of tectal sites for [3 H]muscimol

The displacement of the specific binding of [3H]muscimol to tectal GABAA receptors by mus- cimol and GABA, in regular and Triton X-100- extracted membranes, is illustrated in Fig. 3. Again, the overall effect of Triton X-100 can be seen as a

408 ALICIA BATUECAS el al.

25

o × 20

v

W r r ~ ]5

Z

0 5 m

I 1 1 I I +Triton X-lO0

I ~ I - - ~ I

"o. 0 0 I I 1 "oj

0 0.5 1.0 1.5 20 2.5 0 0.5 1.0 1.5 2.0 2.5 50

f I 550 ",,,o

300 | " ,

25C 'a

20(3 ",, "".,, ~,, %%

] 50 ""N. o ~ P IO

100 a , ~8'~. E16~a~ "

5 (3 %, %",

[3H] MUSCIMOL BOUND

(pmol /mg)

Fig. 2. Computer-fitted Scatchard plots of [3H]muscimol binding to regular (left) and Triton X-100- extracted (right) chick tectal membrane preparations. Membranes were prepared from batches of twelve 16-day embryos (El6) or six 10-day chicks (PI0). Plotted values are averages of two independent

experiments with duplicate tubes in each one.

10-fold increase in affinity in the ligand-receptor interaction. The curves in Fig. 3 confirm, in the case of chick neural tissues, the higher potency of mus- cimol as an agonist at the GABAA receptor, as it has been repeatedly shown in mammalian GABAergic systems (Enna, 1983).

Table 3 complements the pharmacological profiles of the muscimol binding sites in regular and detergent-treated membranes. As it is the case with mammalian receptors, muscimol is only displaced by GABA A agonists, but not by baclofen, a GABA B agonist, or by diaminobutyrate, a specific uptake blocker. While GABAergic agonists show a consist- ent behavior in the two membrane preparations used in this study, the GABAA antagonist bicuculline methiodide seems to be a less efficient displacer in Triton X-100-extracted membranes. This is more precisely confirmed by the displacement curves in Fig. 4, where twice--instead of one tenth, as ex- pected from data in Fig. 3--as much antagonist is necessary, in Triton X-100-extracted membranes, to

Table 2. Binding of [3H]muscimol to chick tectal membranes: kinetic parameters

Scatchard Hill

Membrane Kd Bma, K~l preparation Age (nM) (pmol/mg) n (riM)

No detergent El6 80 1.8 1.00 79 P10 98 2.6 1.01 91

Triton X-100-extracted El6 7.7 2.0 1.03 7.3 Pl0 8.2 2.9 1.04 8.1

All parameters have been calculated from data plotted in Fig. 2.

displace [3H]muscimol from the same number of receptors.

DISCUSSION

As a first step in our proposed studies on the GABAA receptor complex in the developing chick optic tectum we have undertaken a general character- ization of the agonist binding site, as labeled by [3H]muscimol, in 16-day embryonic and 10-day post- natal tectal membranes. Tritiated muscimol has been repeatedly used to characterize GABA receptors in mammalian systems (Beaumont et al., 1978; Shod- grass, 1978; Herschel and Baldessarini, 1979; De° Feudis, 1980; Jordan et al., 1982; Matsumoto and Fukuda, 1982), and it has been more recently recog- nized as a potent and selective agonist for the GABAA receptor (Simmonds, 1983). Membranes from chick optic lobes (from embryos and young chicks) bind [3H]muscimol in a specific and saturable manner. The efficient and reproducible binding of this radioligand is however strictly dependent on a thorough washing of the membrane preparation, being significantly enhanced by freeze-thawing cycles and by extraction of the membranes with low concen- trations of Triton X-100 (Table 1). The effects of extensive washing and detergent extraction have been previously ascribed to the removal of endogenous "inhibitors" of muscimol (and/or GABA) binding, either GABA itself (Napias et al., 1980; Gardner et al., 1981; Fisher et al., 1986), or phospholipids (Toffano et al,, 1981), specific regulatory proteins (Guidotti et al., 1978; Massotti et al., 1981) and other

Characterization of [3H]museimol binding sites 409

z

o Q3

100

8 0

60

40

2O

0 100

80

6O

40

20

0

i i I I i i i

~6ABA M U $ C I M O L ~ _ ~ I 2 6 0 nM

\QA.

P I I i "e'~'--~I ,"T "~'I I 8 - . + T r i t o n X - l O 0

MUSCIMOL It ~ 106 It M zg

\, \oa. i J i 'b"~ -.~.--=n,--.-a i

7 Io i0-~ lO -8 lO-'t i0-6 i0 -~ i0 -'4 lO-S

M O L A R I T Y OF D I S P L A C E R

Fig. 3. Displacement of specific [3H]muscimol binding to chick tectal GABA~ receptors, in regular (above) and Triton X-100-extracted (below) membrane preparations, by the agonists muscimol ( t , ©) and GABA ( I , [2). The curves shown were obtained using tectal lobes from 10-day chicks; embryonic tecta (e.g. 16-day) gave, however, identical re- suits. Taking into account the 10-fold difference in radioligand-receptor affinity between both membrane prep- arations (Fig. 2 and Table 2), the radioligand concentration used for the regular and detergent-treated membranes were 40 and 4 nm, respectively, so to label the same number of receptors in each case. % values have been calculated as means of two independent experiments, with duplicate samples in each one, pooling tecta from 6 different birds in the initial homogenates.

Dissociation constants for [3H]muscimol calculated from the above displacement data, using a simple competitive inhibition model, gave values of 80 and 8 nM for regular and detergent-extracted membranes, respectively, in close agreement with the results of Scatchard and Hill plots

(Table 2).

unknown low molecular weight substances (Yoneda

and Kuriyama, 1980; Kuroda et al., 1984). In addition, incubation conditions during the bind-

ing process must be carefully controlled (Fig. 1). Of the four variables tested, membrane protein concen- tration (Fig. 1D) had the most marked influence on the measured [3H]muscimoi binding, with lower bind- ing efficiencies associated to higher membrane con- centrations. This effect may be due to a more re- stricted diffusion of the radioligand caused by a thicker membrane suspension, since an excess of " iner t" chick liver membranes had the same in- hibitory effect on binding as a similar amount of receptor-containing tectal membranes. The use o f a standard protein concentration of 0.5 mg/ml ( + 0.1 mg) makes corrections unnecessary.

Table 3. Displacement of specific [3H]muscimol binding to chick tectal membranes by fixed concentrations of GABAergic substances

No Triton X-100- detergent* extracted membranes~"

% of displacement~

Muscimol 100 100 GABA 99 100 Imidazolacetate 98 100 //-Alanine 61 75 Taurine 47 44 Baclofen 11 16 Bicuculline methiodide 98 44 Diaminobutyrate 19 24

10-day chick tectum membranes were prepared, as described in Experimental Procedures, with or without detergent extraction. *Membranes prepared without detergent were incubated with 40 nM [3H]muscimol. tMembranes extracted with Triton X-100 were incubated with 4 nM [3H]muscimol. :~The concentration of the displacer was l04 times the radioligand concentration in all cases. % values have been calculated as means of two indepen- dent experiments, with duplicate samples in each one, pooling tecta from 6 different birds in the initial homogenates.

Our adopted experimental conditions reveal a single class of high-affinity, saturable binding sites for [3H]muscimol, both in embryonic and young chick optic lobes, in spite of the heterogeneity of the membrane preparation. Extraction of the membranes with Tri ton X-100 increases the affinity of the radio- iigand for its receptor by a factor of 10 (/~ decreases from ~ 80 nM to ~ 8 nM), without significant changes in the total number of binding sites (Fig. 2 and Table 2). Actually, a net loss of [3H]muscimol

I T I I I I I

100 ~...- - -S -, -- S=~.~...%

Q z 60 \ 'o 6 o . .

40

.o

0 I I I I I i • I

I0-9 lO -e lO -7 I0 o6 10-5 I0 -4 I0 -3

MOLARITY OF" DISPLACER

Fig. 4. Displacement of specific [3H]muscimol binding to chick tectal GABAA receptors, in regular ( 0 0 ) and Triton X-100-extracted ( O - - - © ) membrane preparations (from 10-day chicks) by the antagonist bicuculline methi- odide. As mentioned in Fig. 3, the radioligand concen- trations used for the regular and detergent-treated mem- branes were 40 and 4 nM, respectively, so to label the same number of receptors in both cases. Other details as in Fig. 3. Note the relative shift to the fight of the detergent

curve as compared with Fig. 3.

410 ALICIA BATUECAS et al.

binding sites may be the result of the detergent extraction, since the slight increase in Bm,x is more than offset by the associated loss of protein.

This difference in affinity has been taken into account in several experiments along this study to fix the radioligand concentration at either 40 nM (regu- lar membranes) or 4nM (Triton X-100-extracted membranes), so to label the same number of sites in both cases.

Previous reports on the kinetic properties of [3H]muscimol binding sites in mammalian tissues often give contradictory results. Thus, working with a rat brain membrane preparation, Snodgrass (1978) found one population of binding sites, while papers by Beaumont et al. (1978), Jordan et al. (1982) and Matsumoto and Fukuda (1982) reported the exist- ence of two seemingly independent binding modes. These discrepancies are not easily explained in terms of differences in membrane preparation and/or use of Triton X-100. Furthermore, the solubilized receptor preparation, from either bovine or rat brain, displays in nearly all cases a single [3H]muscimol binding mode (Stephenson et al., 1982; Chang and Barnard, 1982; Martini et al., 1983; Sigel et al., 1983; Sigel and Barnard, 1984; but see Asano et al., 1983). Thus, the multiplicity of muscimol receptor populations should be viewed with caution, and alternative explanations (negative cooperativity, changes in receptor con- formation, partial degradation of binding s i tes . . . ) should be considered to explain biphasic Scatchard plots.

The alleged effect of Triton X-100 on the removal of endogenous modulators or inhibitors of agonist binding remains also controversial. On the one hand, GABAmodulin seems to behave as an allosteric, noncompetitive inhibitor, and its removal by de- tergent does uncover an important number of new binding sites (higher Bma x), without changes in affinity (Guidotti et al., 1978; Massotti et al., 1981); on the other hand, other less well-defined endogenous in- hibitors (Yoneda and Kuriyama, 1980) act like classi- cal competitive inhibitors, modifying only the affinity of the radioligand for its receptor. Our results with Triton X-100 in chick tectal membranes could be explained by the removal of a competitive inhibitor or, simply, by a more efficient elimination of endo- genous GABA, as suggested by Fisher et al.,; alterna- tively, we can think in terms of structural modifications of the membrane in the vicinity of the receptor, induced by the detergent, leading to a facilitated access and/or interaction of the agonist with the recognition site. The latter hypothesis agrees better with previous results using phospholipase C-

treated membranes (Toffano et al., 1981), and may also help explain the effect of Triton X-100 on the efficiency of antagonists as displacers, which we will discuss in the next paragraph.

The pharmacological profiles defined by data in Fig. 3 and Table 3 suggest that the avian receptor for [3H]muscimol is very similar to its mammalian coun- terpart. The results are typical of a GABAA agonist site, with taurine being perhaps a more efficient displacer than in other GABAergic systems (Beau- mont et al., 1978). Membrane extraction by Triton X-100 did not modify the relative efficiency of agonists as displacers (same ICs0/Kd), but rendered bicuculline and its derivatives less efficient as blockers of the agonist site (Table 3). To measure this effect in a more precise fashion we carried out some progressive displacement studies, with regular and detergent-extracted membranes, incubating them with [3H]muscimol at concentrations of 40 nM and 4nM, respectively, to label the same number of receptors sites in both preparations. In these condi- t ions (Fig. 4), bicucuUine methiodide was a more effective displacer (by a factor of 20) in washed membranes than in detergent-extracted membranes. These results can be interpreted, again, in terms of a structural modification of the receptor environment, by Triton X-100, leading to a decrease in affinity for the antagonist, in agreement with previous obser- vations by Mrhler and Okada (1978). A more de- tailed quantitative evaluation of the effect of Triton X-100 on [3H]muscimol binding to tectal membranes, at different developmental ages and different radio- ligand concentrations, is given in the third paper in this series.

Acknowledgement--This work was supported by Comisirn Asesora de Investigacirn Cientifica y Trcnica, and Fondo de Investigaciones Sanitarias.

REFERENCES

Asano T., Yamada Y. and Ogasawara N. (1983) Character- ization of the solubilized GABA and benzodiazepine receptors from various regions of bovine brain. J. Neu- rochem. 40, 209-214.

Beaumont K,, Chilton W. S., Yamamura H. I. and Enna S. J. (1978) Muscimol binding in rat brain: association with synaptic GABA receptors. Brain Res. 148, 153-162.

Chang L-R. and Barnard E. A. (1982) The benzodiazepine/GABA receptor complex: molecular size in brain synaptic membranes and in solution. J. Neuro- chem. 39, 1507 1518.

De Feudis F. V. (1980) Binding studies with muscimol; relation to synaptic y-aminobutyrate receptors. Neuro- science 5, 675~88.

Characterization of [3H]muscimol binding sites 411

Enna S. J. (1983) Biochemical and electrophysiological characteristics of mammalian GABA receptors. Int. Rev. Neurobiol. 24, 181-212.

Fisher T. E., Tucbek J. M. and Johnson D. D. (1986) A comparison of methods for removal of endogenous GABA from brain membranes prepared for binding assays. Neurochem. Res. 11, 1-8.

Gardner C. R., Klein J. and Grove J. (1981) Endogenous GABA determines the characteristics of [3H]GABA-binding. Eur. J. Pharmac. 75, 83-92.

G6mez-Barriocanal J., Barat A. and Ramirez G. (1982) Kainic acid binding sites in the developing chick optic tectum. Neurochem. Int. 4, 157-166.

Guidotti A., Toffano G. and Costa E. (1978) An en- dogenous protein modulates the atfinity of GABA and benzodiazepine receptors in rat brain. Nature, Lond. 275, 553-555.

Harris R. A. and Allan A. M. (1985) Functional coupling of y-aminobutyric acid receptors to chloride channels in brain membranes. Science, N.Y. 228, 1108-1110.

Herschel M. and Baldessarini R. J. (1979) Evidence for two types of binding of 3H-GABA and 3H-muscimol in rat cerebral cortex and cerebellum. Life Sci. 24, 1849-1854.

Jordan C. C., Matus A. I., Piotrowski W. and Wilkinson D. (1982) Binding of [3H]~,-aminobutyric acid and [3H]muscimol in purified rat brain synaptic plasma mem- branes and the effects of bicuculline. J. Neurochem. 39, 52-58.

Kuroda H., Ogawa N., Nukina I. and Ota Z. (1984) An endogeneous inhibitor of GABA receptor binding. Neu- rochem. Res. 9, 21-27.

Leeb-Lundberg L. M. F. and Olsen R. W. (1983) Hetero- geneity of benzodiazepine receptor interactions with 3,-aminobutyric acid and barbiturate receptor sites. Molec. Pharmac. 23, 315-325.

Martini C., Rigacci T. and Lucacchini A. (1983) [3H]muscimol binding site of purified benzodiazepine receptor. J. Neurochem. 41, 1183-1185.

Massotti M., Mazzari S., Schmid R., Guidotti A. and Costa E. (1981) Endogenous inhibitors of Na+-independent [3H]GABA binding to crude synaptic membranes. Neu- rochem. Res. 6, 551-565.

Matsumoto K. and Fukuda H. (1982) Stimulatory and protective effects of benzodiazepine on GABA receptors labelled with 3H-muscimol. Life Sci. 30, 935--943.

M6hler H. and Okada T. (1978) Properties of -aminobutyric acid receptor binding with (+)

3H-bicuculline methiodide in rat cerebellum. Molec. Phar- mac. 14, 256-265.

Napias C., Bergman M. O., Van Ness P. C., Greenlee D. V. and Olsen R. W. (1980) GABA binding in mammalian brain: inhibition by endogenous GABA. Life Sci. 27, 1001-1011.

Olsen R. W. (1981) GABA-benzodiazepine-barbiturate re- ceptor interactions. 3". Neurochem. 37, 1-13.

Olsen R. W., Wong E. H. F., Stauber G. B. and King

R. G. (1984) Biochemical pharmacology of the y-aminobutyric acid receptor/ionophore protein. Fedn Proc. 43, 2773-2778.

Ramirez G., Manrique E., Barat A. and Villa S. (1980) Analysis of developmental interactions in reaggregated embryonic cell cultures. In: Tissue Culture in Neuro- biology, (Giacobini E., Vernadakis A. and Shahar A., eds), pp. 129-144. Raven Press, New York.

Ramirez G., G6mez-Barriocanal J., Escudero E., Fern~ndez-Quero S. and Barat A. (1981) Development of excitatory amino acid binding sites in the chick optic tectum. In: Amino Acid Neurotransmitters, (DeFeudis F. V. and Mandel P., eds), pp. 467-474. Raven Press, New York.

Ramirez G., Barat A., G6mez-Barriocanal J., Manrique E. and Batuecas A. (1983) Development of specific binding sites for [3H]kainic acid and [3H]muscimol in the chick optic tectum: modulation by early changes in visual input. In: CNS Receptors: From Molecular Pharmacology to Behavior, (Mandel P. and DeFeudis F. V., eds), pp. 187-198. Raven Press, New York.

Sigel E., Stephenson F. A., Mamalaki C. and Barnard E. A. (1983) A 7-aminobutyric acid/benzodiazepine receptor complex of bovine cerebral cortex. Purification and par- tial characterization. 3". biol. Chem. 258, 6965-6971.

Sigel E. and Barnard E. A. (1984) A ~,-aminobutyric acid/benzodiazepine receptor complex from bovine cere- bral cortex. Improved purification with preservation of regulatory sites and their interaction. J. bioL Chem. 259, 7219-7223.

Simmonds M. A. (1983) Multiple GABA receptors and associated regulatory sites. Trends Neurol. Sci. 6, 279-281.

Snodgrass S. R. (1978) Use of ~H-muscimol for GABA receptor studies. Nature, Lond. 273, 392-394.

Stephenson F. A., Watkins A. E. and Olsen R. W. (1982) Physicochemical characterization of detergent-solubilized 7-aminobutyric acid and benzodiazepine receptor pro- teins from bovine brain. Eur. J. Biochem. 123, 291-298.

Toffano G., Aldinio C., Balzano A., Leon A. and Savoini G. (1981) Regulation of GABA receptor binding to synaptic plasma membrane or rat cerebral cortex: the role of endogenous phospholipids. Brain Res. 222, 95-102.

Trifiletti R. R., Snowman A. M. and Snyder S. H. (1985) Barbiturate recognition site on the GABA/benzo- diazepine receptor complex is distinct from the picrotoxinin/TBPS recognition sites. Eur. J. Pharmac. 106, 441--447.

Villafruela M. J., Barat A., Manrique E., Villa S. and Ramirez G. (1981) Molecular forms of acetyl- cholinesterase in the developing chick visual system. Devl Neurosci. 4, 25-36.

Yoneda Y. and Kuriyama K. (1980) Presence of a low molecular weight endogenous inhibitor of 3H-muscimol binding in synaptic membranes. Nature, Lond. 285, 670--673.