Phylogenetic conservation of the benzodiazepine binding sites: Pharmacological evidence

OOZS-3908jSS$3.00 + 0.00 Copyright 0 1988 Pergamon Journals Ltd

Neuropharmacology Vol. 27, No. 2, PP. 163-170, 1988 Printed in Great Britain. All rights reserved

PHYLOGENETIC CONSERVATION OF THE BENZODIAZEPINE BINDING SITES: PHARMACOLOGICAL EVIDENCE W. FRIEDL, J. HEBEBRAND, S. RABE and P. PROPPING Institut

Fiir Humangenetik

der Universitgt

Bonn, WilhelmstraDe 31, D-5300 Bonn 1, F.R.G.

Summary-Phylogenetic research can help to elucidate the structure of the GABA/benzodiazepine receptor complex. In this study the evolution of the j?-carboline binding site was traced to see whether it paralleled that of the benzodiazepine binding site. The ratio of [‘H]ethyl-P-carboline-3-carboxylate (/?-CCE) to [3H}flunitmzepam (FNZ) binding sites was determined in several nonmammalian species. The results further substantiate the tight link between these two binding sites. Photoafinity labelling of the benzodiazepine receptor (BZR) has revealed phylogenetic variation of the molecular weight of the benzodiazepine binding proteins. The IC, values for inhibition of [3H]FNZ by various compounds which are active at the central benzodiazepine receptors were determined in three phylogenetically distant species that each showed distinct subunit patterns. In these species, the respective affinities of the compounds were remarkably similar, suggesting that the binding sites for benzodiazepines are conserved in higher bony fishes and tetrapods. The conserved binding sites, in addition to recent immunological results obtained in other research groups, provide further evidence for the existence of the

GABA/BZR as an isoreceptor complex. Key words: Benzodiazepine receptor, phylogeny, fi-carbolines, photoaffinity labelling, pharmacological characterization.

The central benzodiazepine receptor (BZR) has been well characterized pha~acolo~cally in mammals. The potency of various benzodiazepines in displacing diazepam closely parallels their therapeutic and pharmacological potencies (Squires and Braestrup, 1977; Mahler, Okada, Heitz and Ulrich, 1978). The interaction between ~nzodi~epines and non-benzodiazepinoid compounds, that are able to potently displace benzodiazepines from membranes of the mammalian brain has been reviewed extensively (Martin, 1986). Binding studies, using the t~azolopyrid~ine 3”methyl-6-[3-trifluoromethylphenyl]-1,2,4-t~azolo[4,3-b] pyridazine (Cl 218872) and esters of p-carboline-carboxylic acid revealed regional heterogeneity of the binding sites (Klepner, Lippa, Benson, Sano and Beer, 1979; Nielsen and Braestrup, 1980). Photoaffinity labelling with 13H]flunitrazepam ([‘HIFNZ), followed by SDSpolyacrylamide gel electrophoresis and fluorography, also showed regional heterogeneity in rats (Sieghart and Karobath, 1980). Using this method, a qualitative variation of the benzodi~epine binding proteins was previously observed in different species (Hebebrand, Friedl, Lentes and Propping, 1986a; Hebebrand, Fried], Breidenbach and Propping, 1987a) and regional heterogeneity in avians confirmed: whereas in all avians investigated two benzodiazepine binding proteins of 53 and 54 K were photoaffinity labelled with a similar intensity in the telencephalon, the 54 K band was hardly detectable

in the cerebellum (Hebebrand, Friedl, Unverzagt and Propping, 1986b). A quantitative phylogenetic comparison revealed that the central benzodiazepine receptor appeared late during evolution, being detectable only in bony fishes and tetrapods. Slight differences in the affinity of the binding of [3H]diazepam to membrane preparations from brain were apparent among the species investigated (Nielsen, Braestrup and Squires, 1978; Femholm, Nielsen and Braestrup, 1979). Similar Kd values were found using [3H]FNZ as a ligand and suggested that differences in affinity partially correspond to specific subunit patterns (Hebebrand et al., 1987a). At present it is still controversial whether benzodiazepines and /&carbolines bind to the same recognition site or whether they recognize different sites that are adjoining or interact by allosteric modulation of the y-aminobutyric acid/benzodiazepine receptor (GABA/BZR). The results of studies on the relative proportion, the regional distribution or synaptic localization of benzodiazepine and #?-carboline binding sites suggest a rather complex interaction between these sites (Tietz, Chiu and Rosenberg, 1985; Skolnick, Schweri, Kutter, Williams and Paul, 1982; Medina, Novas and De Robertis, 1983). Phylogenetic studies may be a useful approach in answering this question. In order to see whether the phylogenetic appearance of fi-carboline binding sites parallels that of benzodiazepine binding sites, the binding of 163

W.

164

FRIEDLet

[3H]fi-CCE was evaluated in several different vertebrate species. The ratio of [3H]FNZ to [3H]p-CCE binding sites was determined in these species in order to examine whether nonmammalian animals showed ratios comparable to those described in mammals. In addition, the potency of ligands for benzodiazepine receptors for inhibiting the binding of [3H]FNZ was compared in three vertebrates that possess distinct photolabelhng patterns. METHODS

Fishes (excluding the lungfish) were kindly provided by the Meeresbiologische Anstalt Helgoland, other animals were obtained locally. Diazepam, clonazepan, flunitrazepam, nitrazepam, ethyl-&fluoro5,6-dihydro-5-methyl-6-oxo-4H-imidazo[l,5-a][1,4] benzodiazepine-3-carboxylate (Ro 15-1788) and 7-chloro-l,3-dihydro-l-methyl-5-(p-chlorophenyl)2H-1,4-benzodiazepine-2-one (Ro5-4864) were a kind gift from Hoffmann-La Roche, Base]. 2-Phenylpyrazolo-[4,3-cl-quinoline-3(5H)-one (CGS 8216) and 2-(p-chlorophenyl)-pyrazolo[4,3-clquinoline3(5H)-one (CGS 9896) were kindly provided by Ciba-Geigy, Basel. Ethyl-P-carboline-3-carboxylate (/?-CCE) was purchased from Research Biochemicals Inc. [3H]Flunitrazepam (spec. act. 83.9 Ci/mmol) and [3H]fl-CCE (spec. act. 60.3 Ci/mmol) were from NEN. Brains were removed immediately after death and stored at -70°C. Whole brains of smaller animals were pooled; in mammals and avians the cortices and cerebella were investigated separately. Membrane preparations and binding assays were performed essentially as previously described (Hebebrand et al., 1987a). Crude synaptic membranes were prepared according to Sieghart and Karobath (1980) by homogenizing and washing the membranes 4 times in 50 mM Tris-citrate buffer, pH 7.1, containing protease inhibitors (1 mM EDTA, 1 mM benzamidinefluoride, phenylmethylsulfonyl 0.1 mM HCl, 10 p g/ml leupeptin, 10 p g/ml pepstatin, 200 pgbl bacitracin, 20 lg/ml soybean trypsin inhibitor), according to Lang, Barnard, Chang and Dolly (1979). Binding parameters for [3H]FNZ were determined according to Sieghart and Drexler (1983) using 6 [‘H]FNZ (range different concentrations of 0.2-10 nM). Membranes were washed twice with the homogenizing buffer and resuspended in the same buffer, containing 150 mM NaCl. Incubations were performed in an icebath for 90 min, in a final volume of 500 ~1. The binding of [3H]p-CCE was determined in parallel experiments for a given species, using the same procedure, except that the incubation time was only 15 min. Specific binding was defined as the difference in amount of [3H]FNZ or [3H]fi-CCE, respectively, bound in the absence and presence of 3 PM diazepam (a similar nonspecific binding was obtained in the presence of 3 PM cold fl-CCE). At a radioligand concentration of 5 nM, nonspecific bind-

al.

ing was about 15% of total binding of [3H]FNZ and about 40% of total binding of [3H]p-CCE, respectively, except in shark and ray, where nonspecific binding was more than 90% for both ligands. For inhibition experiments, membranes from brain were incubated with 3 nM [3H]FNZ and 7-10 different concentrations of compounds active at the benzodiazepine binding site. The IC,, values were calculated from Hill’s plots. Concentrations of protein were determined using a modified Lowry method (Peterson, 1977). RESULTS

The binding of [3H]p-CCE and [3H]FNZ were determined together for each species in parallel experiments. The binding parameters are presented in Table l(a). In both Chondrichthyes (shark and ray) specific binding of [3H]FNZ and [3H]fi-CCE were so low as to preclude an estimation of Kd and B,,, values. In contrast, the lungfish, which is phylogenetically the oldest species with a benzodiazepine receptor comparable to the mammlian benzodiazepine receptor of the central type (Hebebrand et al., 1987a), revealed high-affinity binding of [3H]j?-CCE. However, in this species the evaluation of both the binding parameters of [3H]FNZ and [3H]fl-CCE was complicated by the fact that in contrast to all other species, including Chondrichthyes, the membrane preparations from brain of the lungfish are not readily homogenized. This accounts for the low linear correlation coefficients of Scatchard plots in the lungfish (0.85 for [3H]FNZ and 0.80 for [3H]fi-CCE). With the exception of shark and ray, the correlation coefficients of Scatchard plots in the other species were > 0.95. Accordingly, the absolute Kd and B,,, values for the lungfish must be viewed with caution. However, the fact that high affinity binding sites for /?-CCE were detectable, suggests that binding sites for fi-CCE and FNZ, comparable to those in mammals, evolved simultaneously. In the other higher evolved vertebrates K,-values for the binding of [3H]P-CCE varied between 0.21 and 0.81 nM in different species. Whilst a systematic phylogenetic relationship was not apparent, the lowest K,-values applied to the two investigated Teleostean fishes, which both had a specifically labelled band of 55 K (comparable to the herring in Fig. 1). The Kd values for the binding of [3H]FNZ varied between 1.7 and 3.5 nM, with the highest values in Teleostean fishes, as described previously (Hebebrand et al., 1987a). In all the species investigated, the density of binding sites for [3H]P-CCE was lower as compared to binding sites for [3H]FNZ (Table la). There is a slight tendency towards a larger p-CCE/FNZ ratio in the cerebellum than in the telencephalon or cortex in both duck and pig, respectively, but this regional variation was not so pronounced as that described in

54K

=

53K

la

lb

55K

,m

2a

2b

-

3

Fig. I. Pattern of photoaffrnity fabelled benzodiazepine binding proteins in brain membranes. Lane I: guinea-pig (a-cortex, b-cerebellum); lane 2: duck (a-telencephafon, b--cerebellum); lane 3: herring. Brain membranes were incubated with 10 nM [sH]flunitrazepam for 90 min at 0°C irradiated with UV at 366nm and subjected to SDS-polyacrylamide gel electrophoresis and fluorography, as previously described (Hebebrand et al., 1987a).

165

167

Conservation of benzodiazepine binding sites Table

I(a). Binding

parameters

Saecies

of [‘HID-CCE and [‘H]FNZ different suecies

[‘H]FNZ B msr fmol/mg Kd orotein nM

[‘HI/?-CCE B max fmol/mg Kd vrotein nM

n

from brain of

B-CCEIFNZ B_.. ratio

559 937 1737 856 1110

0.26 0.21 1.7 0.81 0.51

n.d. n.d. 1009 I666 2343 1233 1688

856 525

0.57 0.56

1312 702

2.1 2.5

0.65 0.78

1429 939

0.32 0.30

2465 1246

I.8 1.9

0.58 0.75

n.d.* n.d.

Shark Ray Mackerel Sprat Lungfish Frog Leguan Duck Telencephalon Cerebellum Pig Cortex Cerebellum

in membranes

3.5 3.2 2.5 1.9 2.5

0.55 0.57 0.73 0.70 0.66

Binding experiments for the two ligands were performed in parallel. The figures represent the means of ?I” independent experiments, analyzed in duplicate. The standard deviation of independent experiments was below 40% for binding of p-CCE and below 15% for binding of FNZ (excluding the lungfish). *n.d.-binding parameters could not be determined because of the very low and inconsistent specific binding. Table I(b). Binding parameters of [‘HI/?-carbolines and [“Hlbenzodiazepines of mammalian suecies (selected literature data)

Brain regions Rat Cortex Hippocampus Cerebellum

[3H]/3-carbolines BIllax fmol/mg K,, protein nM

[‘Hlbenzodiazepines B max fmol/mg K,, protein nM

86.1’ 53.3’ 57.31

0.8 1.2 0.9

120.7’ 93.9* 62.Y*

700 620

0.80 1.13

1200 680

Cortex Cerebellum

1406 834

I.1 0.95

1774 875

Cortex Hippocampus Cerebellum

1129 658 483

HU?PlClfl Cortex Hippocampus Cerebellum

690 500 510

Hippocampus Cerebellum

II80 II00 1000

Reference

1.5 1.3 1.4

0.71 0.57 0.91

Braestrup and Nielsen, 1981

I .73 2.18

0.58 0.92

Stapleton 1982

0.78 0.95

Skolnick 1982

1.30 I .28 1.45

Medina 1983

0.58 0.45 0.51

Montaldo 1984

8.8 10.4

870 516 333

0.70 0.64 0.58

b-carbolinel benzodiazepine B _“. ratio

in different brain regions

2.1 1.9 2.1

ef al., et al., et al.,

et al.,

The radioligands used were fl-CCE and flunitrazepam (Braestrup and Nielsen, 1981; Stapleton er al., 1982). L?-CCE and diazeuam (Skolnick et al., 1982) or p-CCE and flunitrazepam (Montaldo et al., i9i4, Medina ef al.,’ 1983j. ‘__B msx values given in pmol/g tissue.

rats (Table 1b) and other mammals, such as calf, mouse or guinea-pig (Braestrup and Nielsen, 1981). However, the lack of regional variation of the B-CCE/FNZ ratio in human tissues indicated that there may be interspecies differences in the distribution of the two binding sites (Montaldo, Serra, Concas, Corda, Mele and Biggio, 1984). It is important to note that in this study nonmammalian species presented j?-CCE/FNZ ratios similar to those of mammals. Thus, the ratios were relatively similar throughout vertebrate evolution. With ligands that belong to different structural groups, IC,, values for the inhibition of the binding of [3H]FNZ were determined in three species presenting different subunit patterns (Fig. 1). The rank order of the ligands which were primarily active at the central type of benzodiazepine receptor was remarkably similar: CGS 8216

had the highest, diazepam the lowest affinity for the benzodiazepine receptor in all three species; Ro 5-4864, the ligand for the peripheral benzodiazepine receptor, had the weakest affinity in all membrane preparations from brain (Table 2). A similar rank order of compounds was reported for membrane preparations from the rat and human (Sieghart, Eichinger, Riederer and Jellinger, 1985). The separate investigation of cerebellum and cortex or telencephalon in the guinea-pig and duck, respectively, revealed that the IC,, values for /I-CCE were smaller in the cerebellum, in accordance with previous comparisons of these two regions of the brain in mammals (Sieghart and Schuster, 1984; Sieghart et al., 1985). Hill’s coefficients for /I-CCE were similar in the cortex of the guinea-pig and telencephalon of the duck (0.70 and 0.69, re-

W. FRIEDL et al.

168

Table 2. IC,, values of different compounds active at the benzodiazepine binding site. Brain membranes were incubated with 3 nM [‘HIFNZ and 7-10 different concentrations of the ligands. ICI0 values were determined from Hill’s plots IC,, values (nM) CGS 8216

CGS 9896

B-CCE

Ro15-1788

Clonazeoam

Guinea-pig Cortex Cerebellum

0.21 0.18

0.22

1.83 0.93

0.95 1.20

3.62 2.34

7.35 10.9

14.5 13.9

31.1 26.3

> 10,000 > 10,000

Duck Telencephalon Cerebellum

0.24 0.10

0.21

2.16 1.07

I .42 1.15

2.51 2.90

12.9 16.1

16.5 13.6

53.2 45.7

> 10,000 z 10,000

Herring

0.09

0.27

0.49

I .78

2.08

21.6

28.5

1100,000

Soecies

FNZ

6.30

Nitrazeuam

Diazeoam

Ro-5-4864

provides further evidence for a tight link between the /?-carboline and the benzodiazepine binding sites. In mammals, most studies reported a slightly lower density of binding sites for b-carboline as compared to the binding of benzodiazepines (Braestrup and Nielsen, 1981; Tietz et al., 1985; Hirsch, Garrett and Beer, 1985; Stapelton, Prestwich and Horton, 1982; Skolnick et al., 1982) although a greater proportion of binding sites for b-CCE has also been reported (Medina et al., 1983). In general, the /3-carboline/FNZ ratio was smaller in the hippocampus and cortex as compared to the cerebellum in rats (Table lb), calf, mouse and guinea-pig (Braestrup and Nielsen, 1981). However, regional differences were not found in tissue from human brain (Montaldo et al., 1984), indicating that interspecies variation in the proportion of binding sites may occur. In this study the ratio of binding sites for DISCUSSION [3H]fl-CCE to [‘H]FNZ varied between 0.55 and 0.78 The lungfish is the most primitive vertebrate with in bony fishes and tetrapods. The largest ratios a high-affinity benzodiazepine binding protein of applied to avian and mammalian cerebella. However, 53 K, which apparently has the same apparent molthe regional and phylogenetic differences were small. ecular weight (M,) as the a-subunit of the GABA/ In nonmammalian species the range of the ratios was benzodiazepine receptor in mammals (Hebebrand comparable to that previously described within et al., 1987a). The fact that both bony fishes and different regions of a single mammalian species. tetrapods had high affinity binding sites for Thus, the phylogenetic results did not offer an alter[3H]/?-CCE, whereas the Chondrichthyes tested native hypothesis for the observed complex inter(shark and ray) did not, indicate that these binding action between benzodiazepine and p-CCE binding sites evolved together with the high affinity binding sites. The &-values for the binding of [3H]fi-CCE sites for [3H]diazepam and [‘HIFNZ. Whilst these were all below 1 nM, a phylogenetic relationship is findings are compatible with a simultaneous late not readily apparent. The &-values for [‘H]FNZ phylogenetic appearance of binding sites for both varied between 3.5 nM in fishes and 1.9 nM in the fi-CCE and benzodiazepines, caution is warranted. frog. It has previously been suggested that &-values Recent investigations have provided evidence for for [3H]FNZ partially corresponded to specific subspecific binding of [3H]FNZ in insects (Robinson, unit patterns as revealed by photoaffinity labelling MacAllan, Lunt and Battersby, 1986; Lees, Beadle, (Hebebrand et al., 1987a). Neumann and Benson, 1987). In addition, it was Previously, Nielsen et al. (1978) had shown that 5 found that the benzodiazepine receptor antagonist different benzodiazepines (clonazepam, lorazepam, [‘H]Ro 15 1788 binds specifically to membranes from diazepam, oxazepam and chlordiazepoxide) inhibited the shark with an affinity only slightly lower than that the binding of [3H]diazepam similarly in codfish, observed in mammals. In this species the binding of pigeon and mouse. The present results in three com[3H]Ro 15-1788 is displaceable by both cold p-CCE parable species indicate that similar I&-values apply and diazepam, but with greater B&-values than in not only to agonists of the benzodiazepine receptor higher vertebrates (unpublished data). Thus, it is but also to the antagonist Ro 15-1788, the inverse important to stress that, while binding sites with a agonist /?-CCE and the pyrazoloquinolinones CGS 8216 and 9896. Thus, the affinities and the lower affinity for both /I-CCE and benzodiazepines are apparently present in the shark, high affinity respective binding sites for these substances, which binding sites for these ligands evolved together. This are presumably identical or partially overlap, are

spectively) and smaller than the values for the cerebellum (both 0.82). The herring revealed the largest value (0.96), suggesting a single class of binding sites in this species. The regional differences apparently also reflected the fact that photolabelling of the benzodiazepine binding protein of 51 K (53 K in these determinations) present in both mammalian and avian cerebellum and cortex was preferentially inhibited by fl-CCE (Sieghart and Drexler, 1983; Hebebrand et al., 1986b). The ICso values for other ligands did not reveal a systematic variation in the regions of the brains of the two species investigated (Table 2). The small I& value for /I-CCE in the herring confirmed that bony fishes have a highaffinity binding site for b-CCE.

169

Conservation of benzodiazepine binding sites conserved throughout vertebrate evolution with the exception of boneless fishes. Considering the phylogenetic distance and the different labelling patterns of the species investigated, the similarity of the IC,,-values for various benzodiazepine receptor ligands, belonging to different classes of drugs, is remarkable. A systematic phylogenetic comparison of labelling patterns led to the conclusion that additional benzodiazepine binding proteins in higher vertebrates evolved by gene duplications and subsequent divergence (Hebebrand et al., 1987a). Based on these phylogenetic results, furthermore on the similar regional heterogeneity and ontogenetic development of the benzodiazepine binding proteins in mammals and avians, it was suggested that the GABA/ benzodiazepine receptor in higher vertebrates is an isoreceptor complex encompassing at least two different a-subunits (Hebebrand, Friedl, Propping, 1987b). This hypothesis has recently been substantiated by the development of new monoclonal antibodies directed against the GABA/benzodiazepine receptor. A monoclonal antibody was found to crossreact mainly with the major 50K protein of the receptor complex; using larger concentrations of this monoclonal antibody, two additional proteins with a slightly greater M, were detected in the cortex of the rat (Sweetnam, Gallombardo and Tallman, 1986). The pattern was similar to that described upon photoaffinity labelling of the benzodiazepine binding proteins in mammalian cortex by Hebebrand et al. (1987a). Other authors have also described a monoclonal antibody that reacted strongly with the bovine a-subunit and in addition recognized a protein with a slightly greater M,; the authors assumed it to be the B-subunit of the GABA/benzodiazepine receptor, thus indicating that both subunits are homologous (Mamalaki, Stephenson and Barnard, 1987). Alternatively, these results could also be compatible with the presence of an additional cr-subunit(s) in mammalian cortex. An immunoblot of a preparation of cerebellar membrane should help to clarify whether regional heterogeneity is detectable with these monoclonal antibodies. The p-subunit, which presumably encompasses the GABA binding site (Deng, Ransom and Olsen, 1986; Casalotti, Stephenson and Barnard, 1986), is recognized in all regions of the brain by the p-subunit specific monoclonal antibody bd-17 (Haring, Stahl, Schoch, Takacs, Staehelin and Mohler, 1985). At present, the regional heterogeneity of the benzodiazepine binding sites has only been described by photoaffinity labelling. If indeed additional cc-subunits evolved by gene in higher tetrapods, the similar duplications IC,,-values indicate that the binding sites are conserved. Whereas the I&,-values were virtually identical in telencephalon and cortex of the chicken and guinea-pig, respectively, the cerebellar values for /I-CCE were smaller. Whilst the two regions of the brain in these species showed specific labelling pat-

terns, the slight variation in affinity is of questionable biological significance. In summary, the I&,-values for the different ligands showed only slight interspecies variation, despite variation of the labelling patterns. Viewed in light of recent immunological results, it is suggested that the different bands photolabelled by [3H]FNZ represent homologous cc-subunits, on which the benzodiazepine binding site, and thus the affinities for the various ligands, are conserved. Acknowledgements-This study was supported by the Deutsche Forschungsgemeinschaft. We would like to thank Mrs Susanne Schnarr for her skillful technical assistance. Furthermore, the help of the Meeresbiologische Anstalt Helgoland is kindly acknowledged.

REFERENCES

Braestrup C. and Nielsen M. (1981) ‘H-Propyl-/?carboline-3-carboxylate as a selective radioligand for the BZl benzodiazepine receptor subclass. J. Neurochem. 37: 333-341. Casalotti S. O., Stephenson F. A. and Barnard E. A. (1986) Separate subunits for agonist and benzodiazepine binding in the y-aminobutyric acid receptor, oligomer. J. biol. Chem. 261: 15013-15016. Deng L., Ransom R. W. and Olsen R. W. (1986) ‘H-Muscimol photolabels the y-aminobutyric acid receptor binding site on a peptide subunit distinct from that labeled with benzodiazepines. Biochem. biophys. Res. Commun. 138: 1308-1315. Femholm B., Nielsen M. and Braestrup C. (1979) Absence of brain specific benzodiazepine receptors in cyclostomes and elasmobranchs. Comp. Biochem. Physiol. 62: 209-2 11. Haring P., Stahl C., Schoch P., Takacs B., Staehelin T. and Mijhler H. (1985) Monoclonal antibodies reveal structural homogeneity of y-aminobutyric acid/ benzodiazepine receptors in different brain areas. Proc. natn. Acad. Sci. U.S.A. 82: 48374841. Hebebrand J., Fried1 W., Breidenbach B. and Propping P. (1987a) Phylogenetic comparison of the _ -photoaffinity-labeled benzodiazepine receptor subunits. J. Neurochem. 48: 1103-l 108. Hebebrand J., Fried1 W., Lentes K.-U. and Propping P. (1986a) Qualitative variation of ohotolabelled benzodiazepme receptors in different species. Neurochem. Int. 8: 267-27 1. Hebebrand J., Fried1 W., Unverzagt B. and Propping P. (1986b) The benzodiazepine receptor subunits in avian brain. J. Neurochem. 47: 79&793. Hebebrand J., Fried1 W. and Propping P. (1987b) The concept of isoreceptors: application to the nicotinic acetylcholine receptor and the gamma-aminobutyric acid,/benzodiazepine receptor complex. J. Neural Trans. (In press). Hirsch J. D., Garrett K. M. and Beer B. (1985) Heterogeneity of benzodiazepine binding sites: a review of recent research. Pharmac. Biochem. Behav. 23: 681-685. Klepner C. A., Lippa A. S., Benson D. I., Sano M. C. and Beer B. (1979) Resolution of two biochemically and pharmacologically distinct benzodiazepine receptors. Pharmac. Biochem. Behav. 11: 457462. Lang B., Barnard E. A., Chang L. R. and Dolly J. 0. (1979) Putative benzodiazepine receptor: a protein solubilised from brain. FEBS Lat. 104: 149-153. Lees G., Beadle D. J., Neumann R. and Benson J. A. (1987) Responses to GABA by isolated insect neuronal somata:

170

W. FRIEDL et al.

pharmacology and modulation by a benzodiazepine and a barbiturate. Brain Res. 401: 267-278. Mamalaki C., Stephenson F. A. and Barnard E. A. (1987) The GABA,/benzodiazepine receptor is a heterotetramer of homologous a- and P-subunits. EMBO J. 6: 561-565. Martin I. L. (1986) The benzodiazepine receptor. In: Neurometho&, Series I: Neurochemistry, Vol. 4. Receptor Binding (Boulton A. A., Baker G. B. and Hrdina P. D., Eds) pp. 415-453. Humana Press, Clifton, New Jersey. Medina J. H., Novas M. L. and De Robertis E. (1983) Heterogeneity of benzodiazepine receptors: experimental differences between ‘H-flunitrazepam and 3H-ethyl-pcarboline-3-carboxylate binding sites in rat brain membranes. J. Neurochem. 41: 703-709. Mijhler H., Okada T., Heitz P. and Ulrich J. (1978) Biochemical identification of the site of action of benzodiazepines in human brain by 3H-diazepam binding. Life Sci. 22: 985-996. Montaldo S., Serra M., Concas A., Corda M. G., Mele S. and Biggio G. (1984) Evidence for the presence of benzodiazepine receptor subclasses in different areas of the human brain. Neurosci. Lett. 52: 263-268. Nielsen M. and Braestrup C. (1980) Ethyl j-carboline-3carboxylate shows differential benzodiazepine receptor interaction. Nature 286: 606607. Nielsen M., Braestrup C. and Squires R. F. (1978) Evidence for a late evolutionary appearance of brain-specific benzodiazepine receptors: an investigation of 18 vertebrate and 5 invertebrate species. Brain Res. 141: 3422346. Peterson G. L. (1977) A simplification of the protein assay method of Lowry et al. which is more generally applicable. Analyt. Biochem. 83: 346356.

Robinson T., MacAllan D., Lunt G. and Battersby M. .(1986) y -Aminobutyric acid receptor complex of insect CNS: Characterization of a benzodiazepine binding site. J. Neurochem. 47: 1955-1962. Sieghart W. and Drexler G. (1983) Irreversible binding of ‘H-flunitrazepam to different proteins in various brain regions. J. Neurochem. 41: 47-55. Sieghart W., Eichinger A., Riederer P. and Jellinger K. (1985) Comparison of benzodiazepine receptor binding in membranes from human or rat brain. Neuropharmacology 24: 751-759. Sighart W. and Karobath M. (1980) Molecular heterogeneity of benzodiazepine receptors Nature 286: 285-287. Sieghart W. and Schuster A. (1984) Affinity of various ligands for benzodiazepine receptors in rat cerebellum and hippocampus. Biochem. Pharmac. 33: 40334038. Skolnick P., Schweri M., Kutter E., Williams E. and Paul S. (1982) Inhibition of ‘H-diazepam and 3H-3-carboethoxy-fi-carboline binding by irazepine: Evidence for multiple “domains” of the benzodiazepine receptor. J. Neurochem. 39: 1142-I 146. Squires R. F. and Braestrup C. (1977) Benzodiazepine receptors in rat brain. Nature 266: 732-734. Stapleton S. R., Prestwich S. A. and Horton R. W. (1982) Regional heterogeneity of benzodiazepine binding sites in rat brain. Eur. J. Pharmac. 84: 221-224. Sweetnam P., Gallombardo P. and Tallman J. (1986) Molecular aspects of benzodiazepine receptor function. Psychopharmac. Bull. 22: 641645. Tietz E. I., Chiu T. H. and Rosenberg H. C. (1985) Preversus postsynaptic localization of benzodiazepine and fi-carboline binding sites. J. Neurochem. 44: 15241534.