British Journal of Anaesthesia 1993; 71: 127-133

GABA RECEPTORS AND BENZODIAZEPINES

C. S. GOODCHILD

From an historical standpoint, the development ofour understanding of basic mechanisms of the actionsof gamma aminobutyric acid (GABA) and benzo-diazepines began on separate paths. Classical phar-macology revealed similarity between the actions ofGABA and benzodiazepines in the central nervoussystem. Furthermore, pharmacological experimentswith selective agonists and antagonists indicated thatthere exists subtypes of both GABA and benzo-diazepine receptors and that the actions of thebenzodiazepines are produced by modulation ofGABA effects. Thus the two paths of research onGABA and benzodiazepines had crossed in ex-plaining the effects of benzodiazepines by aninteraction with GABAA receptors that were locatedon the same macromolecular complex. More re-cently, this interaction has been elucidated furtherby the cloning of subunits of GABAA receptors: theinteraction between GABA and benzodiazepines isnow being explained at a molecular level.

GABA

Hayashi first proposed GABA as an inhibitoryneurotransmitter in the mammalian central nervsystem in 1954. This was based on the obserpresence of the amino acid in the brain andanticonvulsant properties of the compound winjected into animals. Supporting evidence forconcept came when GABA was extracted fromcentral nervous system, identified chemicallyshown to have inhibitory actions on the nervsystem [3, 25]. In mammals, GABA is present alnexclusively in the central nervous system, althosmall amounts may be found in other areas sue]autonomic ganglia. It is a simple amino acid mole<which is formed by decarboxylation of L-glutaacid by the enzyme glutamic acid decarboxy(GAD). This enzyme is found only in the cytopliof neurones containing GABA.

Although distributed widely in the central nervsystem, the concentration of GABA varies inferent regions with the greatest concentrations foiin basal ganglia, hippocampus, cerebellumhypothalamus in the brain, and in the substaigelatinosa of the dorsal horn of the spinal cord.

(Br. J. Anaesth. 1993; 71: 127-133)

KEY WORDSHynotics: benzodiazepines, diazepam, midazolam. Receptors:GABA.

Large concentrations are found also in the cere-bellum and retina. GABA in the retina is localizedmostly in the horizonal cell layer where it isresponsible for lateral inhibition.

Most GABAergic neurones are interneurones,with short axons and local connections. Generallythe actions of GABA within the central nervoussystem are inhibitory. This effect is achieved by anincrease in chloride conductance [4, 24, 30, 31]. Pre-and postsynaptic effects have been described. Forexample, GABA has long been known to producepresynaptic inhibition by hyperpolarization of af-ferent terminals in the spinal cord [16]. This effectis shown in figure 1. The GABAergic interneuronereleases GABA onto a primary afferent fibre before itmakes a synaptic contact. The effect of this is tocause hyperpolarization of the primary afferent fibre.This effect may be recorded from the cut centralends of these primary afferent fibres as a " dorsal rootpotential". The hyperpolarization decreases thecentral propagation of the action potential. Thus lessexcitatory primary afferent neurone transmitter isreleased onto the postsynaptic neurone.

Stim.

DR

VR

REC

AMP

DRP DRP

FIG. 1. GABAergic presynaptic inhibition. The shaded inter-neurone releases GABA onto the presynaptic terminal (P). DRP= Dorsal root potential; VR = ventral root; Ps = postsynaptic

cell; Stim = stimulation; REC = recording electrodes.

C. S. GOODCHILD, M.A., M.B., B.CHIR., PH.D., F.R.C.A., MonashMedical Centre, Clayton Road, Melbourne, Victoria 3168,Australia.

128 BRITISH JOURNAL OF ANAESTHESIA

Pharmacological studies have indicated functionsfor GABA within the central nervous system in theinduction of sleep, control of neuronal excitability/epilepsy, anxiety, memory and hypnosis. Evidencefor the involvement of GABA in these widely varyingfunctions comes from the effects of the compounditself when injected into animals, and also from theeffects of drugs that either alter the synthesis ormetabolism of GABA or act at postsynaptic GABAreceptors. Thus glutamic acid decarboxylase inhibi-tors lead to reduction in brain concentrations ofGABA. This ultimately leads to convulsions. Drugswhich block the reuptake of GABA from thesubsynaptic cleft back into the presynaptic terminal,such as diaminobutyric acid, lead to increasedconcentrations of GABA. Such drugs possess anti-convulsant and sedative properties. Bicuculline andpicrotoxin, which bind to the postsynaptic GABAreceptor, are convulsants in large doses and insmaller doses alter memory.

HypnosisThe sedative and hypnotic effects of GABA were

the main properties of interest to early researchers.Sedation was found to be of anaesthetic interest. Forexample, Galindo [26] showed that halothane poten-tiates the inhibitory action of GABA in the cuneatenucleus. These observations led to one of the theoriesof anaesthetic action, that is potentiation of centralnervous system depression by GABA. Other an-aesthetic drugs are known to interact with GABA inthis respect: propofol [50], barbiturates [1,23,69]and neurosteroids [52].

EpilepsyThe role of GABA in the prevention of epilepsy is

now well established. Many of the conventionaldrugs used in the management of epilepsy are knownto interact with GABA receptors (vide infra forbenzodiazepines, barbiturates and neurosteroids).Sodium valproate is thought to produce its anti-convulsant action by reducing the metabolism ofGABA. More recently introduced drugs includeGABA analogues such as vigabatrin [32].

AnxietyAnti-anxiety effects of GABA in animals are

reduction in aggression and increase in socialbehaviour, thus allowing the animals to be handledmore easily. This action of GABA is thought to be aresult of interaction with 5-hydroxytrytamine (5-HT) in the limbic system. These interactions areknown to occur at a number of sites within thenervous system [48] and also explain the anxiolyticeffects of 5-HT3 receptor antagonists [14, 15]. It isinteresting to note that GABAA receptor subunitmRNA (vide infra) are increased by social stress inanimals [35]. This observation indicates that this is areal physiological role for GABA.

MemoryThe first evidence for involvement of GABA

mechanisms in memory storage and retention camefrom the observation that systemic picrotoxin, whichis an antagonist at the postsynaptic GABA receptor,

enhanced both these functions [9,13,41,42]. In-volvement of the GABAB subtype receptor in thelaying down of memory has also been indicated bythe observation that microinjections of the GABABreceptor agonist baclofen, impair memory [12].

Two subtypes of GABA receptors have beensuggested by classical pharmacological experimentsand these have been denoted GABAA and GABAB.They are characterized by different rank orders ofpotencies of available agonists for each of these tworeceptors and selective action of antagonists. GABAbinds to both types of receptor. Baclofen is a potentGABA mimetic selective for the GABAB receptor; itis inactive at the GABAA receptor. Muscimol andisoguavacine are potent GABA mimetics selectivefor the GABAA subtype. Although the GABAAsubtype is of primary interest to this review becausethis is responsible for benzodiazepine actions,GABAB receptors will be considered briefly.

GABAB receptorsGABAB binding sites were described first in

peripheral tissue preparations in which they werelocated presynaptically on autonomic nerve terminals[5]. However, further studies indicated a similarpresynaptic location in central nervous systemneurones, with functions being diminished release ofamines, neuropeptides, hormones and excitatoryamino acids, as well as GABA via autoreceptors (i.e.GABAB receptor binding presynaptically on aGABAergic neurone leads to diminished release ofgamma aminobutyric acid from that neurone) [68].

GABAB and GABAA binding sites may be demon-strated in the central nervous system by auto-radiography. The distribution of the two receptorsubtypes is quite different. GABAB receptors arelocated both presynaptically on nerve terminals andpostsynaptically in many brain regions, as well as inthe dorsal horn of the spinal cord. In the spinal corddorsal horn, 50 % of GABAB binding sites have beenreported to be associated with small diameterprimary afferent fibres [56].

The functional role of GABAB receptors in thecentral nervous system is still understood poorly.The wide distribution in the nervous system suggeststhey are likely to have many effects. One therapeuticbenefit of GABAB agonists is control of spasticity.Baclofen (GABAB agonist) given orally or intra-thecally, reduces spasticity and prevents flexorspasms in a number of different neurological dis-orders. This is thought to be a result of interruptionof mono- and polysynaptic spinal reflexes [17].

GABAA receptorsGABAA receptors are stimulated by GABA,

muscimol and isoguavacine. Molecular biologicalstudies performed in the past decade indicate thatthe A subtype of GABA receptor may be a group ofdifferent receptors rather than a single entity. Fourtypes of subunits have been described: a, (J, y and 8[2, 28, 75]. Combinations of these subunits producemacromolecular complexes which include a chlorideion channel. It has been noted that there are manysimilarities with the nicotinic acetylcholine receptor.

GABA RECEPTORS AND BENZODIAZEPINES 129

It has been proposed that the GABAA receptor has asimilar structure (i.e. a pentameric form). Com-bination of GABA and a GABAA subtype receptorleads to an increase in chloride conductance andsubsequent hyperpolarization of the postsynapticmembrane.

Multiple variants exist within each of these classes:six a, three (3, two y and one 5 [36,44,58,73].Analysis of GABA receptors within the retina haveyielded two other subunits, pi and p2. Each of thesubunits contains an N-terminal segment of approxi-mately 200 amino acids and four hydrophobicmembrane-spanning segments. Each of these foursegments is connected to the others by shorthydrophilic segments of amino acids, except for thethird and fourth which have a longer segment thatmay be phosphorylated by intracellular cyclic AMP-dependant protein kinase (protein kinase A). Thisopens up the possibility for a large number ofinteractions with other drugs that may alter in-tracellular cyclic AMP levels, such as: opioids (u and5); purinoceptors; alpha2 adrenoceptors (subtypesA and B); beta adrenoceptors; dopamine Dl andD2; GABAB, 5-HT (subtypes 1A, IB andID); muscarinic; cholinergic (M2); prostanoid DPreceptors; and neurophysin and oxytocin receptors(subtype V2) [47].

Co-expression studies using cloned GABAA sub-unit variants in mammalian cells have led tocombinations of subunit that respond to GABA, andthis response may be modulated by benzodiazepines[43,57], The presence of a y subunit confersbenzodiazepine sensitivity [37]. Less well dennedactions at GABAA receptors have also been describedfor barbiturates and neurosteroids (e.g. alphaxalone,alphadolone, both constituents of Althesin, andpregnanolone, currently in phase I trials as an i.v.anaesthetic) [11,21,33,55,59,74]. Although it ispossible to make a number of different combinationsof these subunits, apart from the alteration inbenzodiazepine specificity and sensitivity (vide infra),the functional significance of this diversity has yet to beelucidated [49].

BENZODIAZEPINES

The first clinically useful benzodiazepine, chlor-diazepoxide hydrochloride, was discovered in 1957after a search for compounds with sedative andanxiolytic properties related to 4,5-benzo-hept-1,2,6-oxdiazines, compounds known since 1891 [66].These chemicals lend themselves readily to modifi-cation by organic chemists so that a large number ofrelated compounds were synthesized. Three basictypes of structure exit: basic 1—4 benzodiazepines,such as diazepam and lorazepam (see fig. 2);heterocyclic 1—4 benzodiazepines, such as triazolam(fig. 2); 1-5 benzodiazepines, such as clobazam(fig. 2).

All of these compounds possess similar properties,that is reduction of anxiety, sedation and anti-convulsant properties. However, the 1-5 benzo-diazepines have been found to be less sedative.These drugs also have significant muscle relaxantproperties, without any effect on coordination. This

FIG. 2. Molecular structure of three benzodiazepines.

latter effect may be because of an antianxiety action,but is probably mediated also by potentiation of theeffects of GABA on spinal and supraspinal motorreflexes. As the dose of benzodiazepine is increased,anxiolytic effects are produced first followed byanticonvulsant effects and then a reduction in muscletone, followed by sedation and hypnosis. Thesimilarity between these effects and those of GABAwas obvious. It was logical to look to GABA toexplain the therapeutic effects of this group ofcompounds.

GABA-BENZODIAZEPINE INTERACTION

Before Davidoff s suggestion of a role for GABA inmediation of presynaptic inhibition [16], Schmidt,Vogel and Zimmerman [62] showed that diazepamincreased the amplitude of the dorsal root potential,and enhanced monosynaptic ventral root responses.Pole and colleagues [53] investigated this further andshowed an interaction between diazepam and GABA.Benzodiazepines alone lack GABA mimetic effects.However, they have been shown to enhance theeffects of GABA in numerous regions in the centralnervous system, for example spinal cord, cerebellum,hippocampus, cuneate nucleus, cerebral cortex,hypothalamus, amygdala and brain stem raphe nuclei[34,69].

Benzodiazepine receptors

Specific high affinity benzodiazepine binding siteswere demonstrated in the central nervous system in1977 [7, 46]. These specific binding sites are locatedon neurones within the central nervous system, butalso on non-neuronal tissues, perhaps glia. Thefunction of these is unknown [20]. Binding studieswith agonists and antagonists promoted the conceptof these recognition sites as functional receptors.Drugs acting stereospecifically with this receptormay be classified as selective agonists (e.g. flu-nitrazepam) that potentiate GABA effects, inverseagonists, such as DMCM or Ro 194603 that produceexactly opposite effects (proconvulsant) and an-

130 BRITISH JOURNAL OF ANAESTHESIA

tagonists, such as flumazenil, that stereospecificallyblock the effects of either agonists or inverse agonists.A strong correlation between binding and phar-macological activity has been reported.

Haefely suggested that a GABAA receptor, benzo-diazepine receptor and chloride ionophore formedone macromolecular complex in order to explainexperimental observations that benzodiazepineswere not GABA mimetic per se, but potentiated theeffects of binding with GABAA receptors in operat-ing a chloride channel [24, 29, 63]. Not all GABAreceptors are coupled to benzodiazepine receptors;this macromolecular complex forms a subset ofGABAA receptors. This subspecialization of GABAAreceptors with respect to the benzodiazepines goesfurther. The first indications of this came from thepharmacological profile of agonists at the benzo-diazepine receptor.

Two subtypes of benzodiazepine receptor weresuggested originally. The benzodiazepine, subtype(high affinity for triazolopyridazines) is foundthroughout the brain with large concentrations in thecerebellum. The benzodiazepine,, type was foundprincipally in the cerebral cortex, spinal cord andhippocampus. These distributions were derivedfrom the differential binding of the triazolopy-ridazine CL 218772, leading to displacement of[3H] benzodiazepine binding, particularly in thehippocampus. Beta carbolines tended to react withbenzodiazepine receptors in the cerebellum [54]. Inaddition, CL 218872 appeared to possess anxiolyticeffects with only mild sedation. Therefore, anxiolysiswas ascribed to the benzodiazepine, site [39].

Finally, multiple subunits associated with thereceptor labelled with [3H]flunitrazepam weredemonstrated [64]. These four proteins had differentmolecular masses of 51, 53, 55 and 59 kDa. Thesmaller molecular mass protein had a greater affinityfor CL 218872 and this had a different regiondistribution to the others, a distribution whichcoincided with the benzodiazepine,. Cloning experi-ments have subsequently elucidated further on thisheterogeneity of benzodiazepine—GABAA receptorcomplexes.

Molecular biology experiments

Experimental studies using specific antibodiesraised against subunit variants of GABAA receptorshave shown that the 51 kDa protein, labelled withradiolabelled benzodiazepine, is the al subunit: the53 kDa protein is the a2 subunit and the 59 kDaprotein corresponds to the oc3 subunit [10,65].Pritchett, Luddens and Seeburg [57] analysed thepharmacological characteristics of GABAA receptorsproduced by recombining subunits, each moietycontaining a single a subunit isoform. They com-bined different a subunits with pi and y2 in isolatedhuman kidney cells. They then evaluated the bindingcharacteristics of the subsequently formed GABAAreceptors. Receptors containing the al subunitbehaved like benzodiazepine, receptors, that is theyhad high affinity for CL 218872 and the betacarbolines. Those GABAA receptors which con-tained a2 or oc3 subunits behaved like benzo-diazepine,, receptors.

FIG. 3. Venn diagram showing the effect on benzodiazepinepharmacology of the a subunit in combination with 02 y2subunits. Subunits inside the same shape have similar phar-

macology (see text for explanation and details).

Further experiments with other a subunits com-bined with P2 and y2 along similar lines revealed amultiplicity of GABAA receptors and benzodiazepineaffinities. These are summarized in figure 3. Thelargest circle contains a subunits that produceGABAA receptors which bind muscimol. Polygon 2contains all those that confer high affinity fordiazepam while the circle labelled 3 contains thosethat have a low affinity for diazepam. The al subunit(circle labelled 4) confers benzodiazepine, phar-macology (i.e. high affinity for CL 218872 andzolpidem). Ellipse 5 contains receptors with benzo-diazepine,, pharmacology. Those receptors contain-ing the a5 subunit also behaved like benzodiazepine,,receptors, but were different from the benzodi-azepinen receptors containing a2 and a3 subunitsbecause these recombinant receptors had a differentrank order of affinity for zolpidem and CL 218872[53]. The a6 isoform of the GABAA receptor has anextremely small binding affinity for benzodiazepines.It is found in the cerebellar granule cell layer [40].This subtype may be involved in the pharmaco-logical effects of ethanol.

Although this new molecular biological evidencehas brought together the pharmacology of GABAand the benzodiazepines, the obvious question to askconcerns the functional significance in vivo of thesein vitro studies. The messenger RNA for the sixdifferent a subunits are known and have beenmapped within the mammalian central nervoussystem. It has been shown that the subunits havedifferent topographical distributions [38, 71, 72].The differences in these distributions may indicatedifferent types of GABA receptor associated withdifferent functions. As yet, this remains to be realizedin the development of a drug specially selected foranxiety, lacking sedative and muscle relaxant effects.

Recent pharmacological experiments with theGABAA antagonist bicuculline indicated separate

GABA RECEPTORS AND BENZODIAZEPINES 131

10CH

0.001 0.01 0.1 1Bicuculline (pmol)

10

FIG. 4. Bicuculline dose-response curves for suppression of thespinal antinociceptive effects of equieffective doses of intrathecal

midazolam ( • ) and 5-HT (#) . Values are mean, SEM.

roles for different GABAA receptors in spinalantinociception [27]. In these experiments, nocicep-tive thresholds were measured in the tail of the ratafter equipotent intrathecal injections of either 5-HT or the benzodiazepine midazolam. It is knownthat intrathecal midazolam produces spinally medi-ated antinociceptive effects by combination withspinal cord benzodiazepine receptors and this maybe antagonized by intrathecal bicuculline [22]. Thusmidazolam spinally mediated antinociception ismediated by a typical benzodiazepine-GABA in-teraction. A dose-response curve for bicucullinesuppression of midazolam antinociceptive effectrevealed that the dose of bicuculline required toproduce 50% suppression of midazolam spinalantinociception was approximately 50 pmol (fig. 4)[22].

Similar experiments with 5-HT spinal anti-nociception revealed that bicuculline could alsoblock the antinociceptive effect of intrathecal 5-HT[27]. However, the dose-response curve for bi-cuculline in producing this effect was 2.5 decades tothe left of that required to block the effects ofmidazolam (i.e. bicuculline 0.1 pmol was required toproduce approximately 50 % suppression of 5-HTspinal antinociception (fig. 4)). These widely separ-ated bicuculline dose-response curves may indicatethat there exists in the spinal cord two types ofGABAA receptor involved with spinal antinocicep-tion. This differential effect of bicuculline onGABAA receptors has hitherto not been described.Bicuculline tends to be used to define a receptor suchas the GABAA subtype, but no dose—response studieshave been performed in vitro that confirm or refutethis differential sensitivity.

ENDOGENOUS BENZODIAZEPINES

Naturally occurring substances that bind to benzo-diazepine receptors were isolated from animals andman as long ago as 1980 [6, 19]. These substances,however, did not become established as endogenousneurotransmitters active at the benzodiazepine re-ceptor because the chemistry used for extractiontended to produce artefacts. The beta carbolines,however, are probably produced in the brain and itis well established that the enzyme systems for theirproduction are present. These substances are inverseagonists that cause anxiety and convulsions wheninjected into animals. Better extraction systems,

thought not to produce artefacts, have led to theextraction of n-butyl-beta-carboline-3-carboxylate(beta CCB) from bovine brain [45, 51]. This com-pound varies in concentration in the brain accordingto stress levels [51]. These data suggest that beta-CCB is an endogenous inverse agonist active atbenzodiazepine receptors, perhaps involved in thecontrol of stress.

Agonist substances have also been extracted frombrain. These are surprisingly similar to the benzo-diazepines developed for use as drugs. For example,De Bias, Park and Friedrich purified desmethyl-diazepam from brain extracts [18]. An obviousinterpretation of these results is that the animal hadsomehow ingested benzodiazepines. However, San-gameeswaran and co-workers also detected benzo-diazepine immunoactivity in brains of animals andhumans that died long before the introduction of thebenzodiazepines [61]. These findings have beenconfirmed by many other groups using differentextraction procedures and different tissue sources.The " endozepines " have been purified from rat andhuman brain [60]. These extracts were shown tomodulate GABA activated Cl" conductance inisolated human cortical neurones. The diazepam-like substance has even been reported to be confinedto synaptic vesicles [45].

It is still uncertain if these endozepines arenaturally occurring neurotransmitters. These sub-stances have been isolated also from commonfoodstuffs [67, 70] thus throwing further doubt ontheir neurotransmitter role. However, an in vivomechanism for the formation of 1-4 benzodiazepineshas been described [8].

CONCLUSION

Benzodiazepines produce their sedative, hypnotic,anxiolytic, anticonvulsant and antinociceptive effectsby interactions with GABA. The benzodiazepinesmodulate the effects of GABA at the subset ofGABAA receptors. These are ligand-gated chlorideion channels, probably consisting of five subunits.The subunit composition varies in different parts ofthe brain, although the variations in subunit struc-ture have not been linked with different functions.The type of a subunit determines the benzodiazepinepharmacology—anxiolytic or sedative. Each sub-unit consists of a chain of amino acids with fourhydrophobic segments that span the lipid cellmembrane. The intracellular chains joining thesemembrane-spanning segments may be phosphory-lated by intracellular phospholipases. This may leadto modulation of the effect of both benzodiazepinesand GABA binding on the cell surface part of thereceptor. Many interactions of the benzodiazepineGABA complex with other drugs and neurotrans-mitters altering intracellular cyclic AMP concen-trations are possible. Among these are the opioids,that are well known by anaesthetists to potentiate thesedative effects of the benzodiazepines.

There are many possibilities for different subunitstructures of the GAB AA receptors which respond tobenzodiazepines. The existence and function ofendogenous benzodiazepines is controversial but it

132 BRITISH JOURNAL OF ANAESTHESIA

may be that some specialization of GABAA-benzodiazepine receptors provides more specifictargets for these endogenous compounds.

REFERENCES1. Ashton D, Willems R. In vitro studies on the broad spectrum

anticonvulsant loreclezole in the hippocampus. EpilepsyResearch 1992; 11: 75-88.

2. Barnard EA, Darlison MG, Seeburg PH. Molecular biologyof the GABAa receptor: the receptor/channel superfamily.Trends in Neurosciences 1987; 10: 502-509.

3. Bazemore AW, Elliott KAC, Florey E. Isolation of factor I.Journal of Neurochemistry 1957; 1: 334-339.

4. Bormann J. Electrophysiological characterization of diazepambinding inhibitor (DBI) on GABAA receptors. Neuropharma-cology 1991; 30: 1387-1389.

5. Bowery NG. Baclofen: 10 years on. Trends in PharmacologicalSciences 1982; 3: 400-403.

6. Braestrup C, Neilsen M, Olsen C. Urinary and brain beta-carboline-3-carboxylates as potent inhibitors of brain benzo-diazepine receptors. Proceedings of the National Academy ofSciences USA 1980; 77: 2288-2292.

7. Braestrup C, Squires RF. Specific benzodiazepine receptorsin rat brain characterised by high affinity 3-H diazepambinding. Proceedings of the National Academy of SciencesUSA 1977; 74: 3805-3809.

8. Bringmann G. A First biosynthetic proposal for the in vivoformation of naturally occurring diazepam-like 1,4-benzo-diazepines. Journal of Neural Transmission 1992; 88: 77-82.

9. Breen RA, McGaugh JL. Facilitation of maze learning withpoststrial injections of picrotoxin. Journal of Comparative andPhysiological Psychology 1961; 54: 494-501.

10. Buchstaller A, Fuchs K, Sieghart W. Identification of a,-, a2-and a3-subunit isoforms of the GABAA-benzodiazepinereceptor in the rat brain. Neuroscience Letters 1991; 129:237-241.

11. Burt DR, Kamatchi GL. GABAA receptor subtypes: Frompharmacology to molecular biology. FA SEB Journal 1991;5: 2916-2923.

12. Castellano C, Brioni JD, Nagahara AH, McGaugh JL. Post-training systemic and intra-amygdala administration of theGABA-B agonist baclofen impairs retention. Behavioral andNeural Biology 1989; 52: 170-179.

13. Castellano C, McGaugh JL. Retention enhancement withpost-training picrotoxin: lack of state dependency. Behavi-oral and Neural Biology 1989; 51: 165-170.

14. Costall B, Jones BJ, Kelly ME, Naylor RJ, Oakley NR,Onaivi ES, Tyers MB. The effects of ondansetron(GR38032F) in rats and mice treated subchronically withdiazepam. Pharmacology, Biochemistry and Behavior 1989;34: 769-778.

15. Costall B, Jones BJ, Kelly ME, Naylor RJ, Onaivi ES, TyersMB. Sites of action of odansetron to inhibit withdrawal fromdrugs of abuse. Pharmacology, Biochemistry and Behavior1990; 36: 97-104.

16. Davidoff RA. Gamma-aminobutyric acid antagonism andpresynaptic inhibition in the frog spinal cord. Science 1972;175: 331-333.

17. Davidoff RA, Sears ES. The effects of Lioresal on synapticactivity in the isolated spinal cord. Journal of Neurology 1974;24: 957-963.

18. De Bias AL, Park D, Friedrich P. Endogenous benzo-diazepine-like molecules in the human, rat and bovine brainsstudied with a monoclonal antibody to benzodiazepines.Brain Research 1987; 413: 275-284.

19. Deckert J, Kuhn W, Przuntek H. Endogenous benzo-diazepine ligands in human cerebrospinal fluid. Peptides1984;5: 641-644.

20. Del Zompo M, Bocchetta A, Corsini GU, Tallmann JF,Gessa GL. Properties of benzodiazepine binding sites inperipheral tissues. In: Biggio G, Costa E, eds. BenzodiazepineRecognition Site Ligands: Biochemistry and Pharmacology.New York: Raven Press, 1983; 239-248.

21. Doble A, Martin L. Multiple benzodiazepine receptors: noreason for anxiety. Trends in Pharmacological Sciences 1992;13: 76-81.

22. Edwards M, Serrao JM, Gent JP, Goodchild CS. On the

mechanism by which midazolam causes spinally mediatedanalgesia. Anesthesiology 1990; 73: 273-277.

23. Evans EB, Balster RL. CNS depressant effects of volatileorganic solvents. Neuroscience and Biobehavioral Reviews1991; 15: 233-241.

24. Facklam M, Schoch P, Haefely WE. Relationship betweenbenzodiazepine receptor occupancy and potentiation ofgamma-aminobutyric acid-stimulated chloride flux in vitro offour ligands of differing intrinsic efficacies. Journal ofPharmacology and Experimental Therapeutics 1992; 261:1106-1112.

25. Florey E. An inhibitory and excitatory effect of mammaliancentral nervous and their action on a single sensory neuron.Archives Internationales de Physiologie 1954; 62: 33-53.

26. Galindo A. Effects of procaine pentobarbitol and halothaneon synaptic transmission in the central nervous system.Journal of Pharmacology and Experimental Therapeutics 1969;169: 185.

27. Guo Z, Gent JP, Goodchild CS. Two GABA mechanisms forspinally mediated antinociception: evidence from antagonismof 5-Hydroxytryptamine (5HT) by bicuculline. BritishJournal of Anaesthesia 1990, 64: 589P-590P.

28. Hadingham KL, Harkness PC, McKernan RM, Quirk K, LeBourdelles B, Home AL, Kemp JA, Barnard EA, Ragan CI,Whiting PJ. Stable expression of mammalian type A gamma-aminobutyric acid receptors in mouse cells: Demonstration offunctional assembly of benzodiazepine-responsive sites. Pro-ceedings of the National Academy of Sciences USA 1992; 89:6378-6382.

29. Haefely W. Benzodiazepine interactions with GABA recep-tors. Neuroscience Letters 1984; 47: 101-206.

30. Haefely W, Kyburz E, Gerecke M, Mohler H. Recentadvances in molecular pharmacology of benzodiazepinereceptors and in structure-activity relationships of theiragonists and antagonists. Advances in Drug Research 1985;14: 165-322.

31. Haefely W, Pole P. Electrophysiological studies on theinteraction of anxiolytic drugs with GABAergic mechanisms.In: Malick JB, Enna SJ, Yamamura HI, eds. Anxiolytics:Neurochemical, Behavioral, and Clinical Perspectives. NewYork: Raven Press, 1983; 113-145.

32. Halonen T, Pitkanen A, Saano V, Riekkinen PJ. Effects ofvigabatrin (gamma-vinyl GABA) on neurotransmission-related amino acids and on GABA and benzodiazepinereceptor binding in rats. Epilepsia 1991; 32: 242-249.

33. Heyer EJ, Macdonald RL. Barbiturate reduction of calcium-dependent action potentials: correlation with anestheticaction. Brain Research 1982; 236: 157-171.

34. Izquierdo I, Pereira ME, Medina JH. Benzodiazepinereceptor ligand influences on acquisition: Suggestion of anendogenous modulatory mechanism mediated by benzo-diazepine receptors. Behavioral and Neural Biology 1990; 54:27^11.

35. Kang I, Thompson ML, Heller J, Miller LG. Persistentelevation in GABAA receptor subunit mRNAs followingsocial stress. Brain Research Bulletin 1991; 26: 809-812.

36. Khrestchatisky M, MacLennan AJ, Chiang MY, Xu WT,Jackson MB, Brecha N, Sternini C, Olsen RW, Tobin AJ. Anovel alpha subunit in rat brain GABAa receptors. Neuron1989; 3: 745-753.

37. Knoflach F, Rhyner T, Villa M, Kellenberger S, Drescher U,Malherbe P, Sigel E, Mohler H. The gamma3-subunit of theGABAA-receptor confers sensitivity to benzodiazepine re-ceptor ligands. FEBS Letters 1991; 293: 191-194.

38. Laurie DJ, Seeburg PH, Wisden W. The distribution of 13GABAA receptor subunit mRNAs in the rat brain. II.Olfactory bulb and cerebellum. Journal of Neurosience 1992;12: 1063-1076.

39. Lippa AS, Critchett D, Sano MC, Klepner CA, GreenblattEN, Coupet J, Beer B. Benzodiazepine receptors: cellular andbehavioral characteristics. Pharmacology, Biochemistry andBehavior 1979; 10: 831-843.

40. Luddens H, Pritchett DB, Kohler M, Killisch I, Keinanen K,Monyer H, Sprengel R, Seeburg PH. Cerebellar GABAaselective for a behavioural alcohol antagonist. Nature {London)1990; 346: 648-651.

41. McGaugh JL. Dissociating learning and performance: drugand hormone enhancement of memory storage. Brain Re-search Bulletin 1989; 23: 339-345.

GABA RECEPTORS AND BENZODIAZEPINES 133

42. McGaugh JL, Castellano C, Brioni J. Picrotoxin enhanceslatent extinction of conditioned fear. Behavioral Neuroscience1990; 104: 264-267.

43. Malherbe P, Draguhn A, Multhaup G, Beyreuther K, MohlerH. GABAA-receptor expressed from rat brain a- and p1-subunit cDNAs displays potentiation by benzodiazepinereceptor ligands. Molecular Brain Research 1990; 8: 199-208.

44. Malherbe P, Sigel E, Bauer R, Persohn E, Richards JG,Mohler H. Functional expression and sites of gene tran-scription of a novel alpha subunit of the GABAa receptor inrat brain. FEBS Letters 1990; 260: 261-265.

45. Medina JH, Pena C, Piva M, Paladini AC, De Robertis E.Presence of benzodiazepine like molecules in mammalianbrain and milk. Biochemical and Biophysical Research Com-munications 1988; 152: 534-539.

46. Mohler H, Okada T. Benzodiazepine receptor: demonstrationin the central nervous system. Science 1977; 198: 849.

47. Moss SJ, Smart TG, Blackstone CD, Huganir RL. Func-tional modulation of GABAA receptors by cAMP-dependentprotein phosphorylation. Science 1992; 257: 661-665.

48. Otero Losada ME. GABAergic influence on serotonergicfunction. Biogenic Amines 1990; 7: 289-297.

49. Paredes RG, Agmo A. GABA and behavior: The role ofreceptor subtypes. Neuroscience and Biobehavioral Reviews1992; 16: 145-170.

50. Peduto VA, Concas A, Santoro G, Biggio G, Gessa GL.Biochemical and electrophysiologic evidence that propofolenhances GABAergic transmission in the rat brain. Anesthes-iology 1991; 75: 1000-1009.

51. Pena C, Medina JH, Novas ML, Paladini AC, De Robertis E.Isolation and identification in bovine cerebral cortex of n-butyl beta carboline-3-carboxylate, a potent benzodiazepinebinding inhibitor. Proceedings of the National Academy ofSciences USA 1986; 83: 4952^1956.

52. Peters JA, Kirkness EF, Callachan H, Lambert JJ, TurnerAJ. British Journal of Pharmacology 1988; 94: 1257-1269.

53. Pole P, Bonetti EP, Schaffner R, Haefely W. The effect ofdiazepam on spinal cord activities: possible sites andmechanisms of action. Naunyn Schmiedebergs Archives ofPharmacology 1974; 284: 319-337.

54. Potier MC, Prado de Carvalho L, Dodd RH, Besselievre R,Rossier J. In vivo binding of beta-carbolines in mice: regionaldifferences and correlation of occupancy to pharmacologicaleffects. Molecular Pharmacology 1988; 34: 124-128.

55. Powell H, Morgan M, Sear JW. Pregnanolone: A new steroidintravenous anaesthetic. Dose finding study. Anaesthesia1992; 47: 287-290.

56. Price GW, Kelly JS, Bowery NG. The location of GABAbreceptor binding sites in mammalian spinal cord. Synapse1987; 1: 530-538.

57. Pritchett DB, Luddens H, Seeburg PH. Type I and type IIGABA-benzodiazepine receptors produced in transfectedcells. Science 1989; 245: 1389-1392.

58. Pritchett DB, Seeburg PH. Gamma-aminobutyric acid Areceptor alpha 5-subunit creates novel type II benzodiazepinereceptor pharmacology. Journal of Neurochemistry 1990; 54:1802-1804.

59. Puia G, Santi MR, Vicini S, Pritchett DB, Purdy RH, PaulSM, Seeburg PH, Costa E. Neurosteroids act on recombinanthuman GABAa receptors. Neuron 1990; 4; 759-765.

60. Rothstein JD, Garland W, Puia G, Guidotti A, Weber RJ,Costa E. Purification and characterization of naturallyoccurring benzodiazepine receptor ligands in rat and humanbrain. Journal of Neurochemistry 1992; 58: 2102-2115.

61. Sangameeswaran L, Fales HM, Friedrich P, De Bias AL.Purification of a benzodiazepine from bovine brain anddetection of benzodiazepine-like immunoreactivity in humanbrain. Proceedings of the National Academy of Sciences USA1986; 83: 9236-9240.

62. Schmidt RF, Vogel ME, Zimmerman M. Die wirkung vonDiazepam auf die prasynaptische hemmung und andereruckenmarkreflexe. Naunyn Schmiedebergs Archives ofPharmacology 1967; 258: 69-82.

63. Schoch P, Richards JG, Haring P, Takacs B, Stahli C,Staehelin T, Haefely W, Mohler H. Co-localisation of GABAreceptors and benzodiazepine receptors in the brain bymonoclonal antibodies. Nature (London) 1985; 314: 168.

64. Sieghart W, Karobath M. Molecular heterogeneity of benzo-diazepine receptors. Nature (London) 1980; 286: 285-287.

65. Stephenson FA, Duggan MJ, Casalotti SO. Identification ofthe alpha 3-subunit in the GABAa receptor purified frombovine brain. FEBS Letters 1989; 243: 358-362.

66. Sternbach LH, Randall LO, Gustafson SR. 1,4-Benzo-diazepines (chlordiazepoxide and related compounds). In:Gordon M, ed. Psychopharmacological Agents. New York:Academic Press, 1964; 137-224.

67. Unseld E, Krishna DR, Fischer C, Klotz U. Endogenousbenzodiazepines in brain: right or wrong? Trends in Neuro-sciences 1988; 11: 490.

68. Waldmeier PC, Wicki P, Feldtrauer JJ, Baumann PA.Potential involvement of a baclofen-sensitive autoreceptor inthe modulation of release of endogenous GABA from ratbrain slices in vitro. Naunyn Schmiedebergs Archives ofPharmacology 1988; 337: 289-295.

69. Wesselman JPM, Van Wilgenburg H, Long SK. The effectsof pentobarbital and benzodiazepines on GABA-responses inthe periphery and spinal cord m vitro. Neuroscience Letters1991; 128: 261-264.

70. Wildmann J, Mohler H, Vetter W, Ranalder U, Schmidt K,Maurer R. Diazepam and N-desmethyldiazepam are found inthe rat brain and adrenal and may be of plant origin. Journalof Neural Transmission 1987; 70: 383-398.

71. Wisden W, Laurie DJ, Monyer H, Seeburg PH. Thedistribution of 13 GABAA receptor subunit mRNAs in the ratbrain. I. Telencephalon, diencephalon, mesencephalon.Journal of Neuroscience 1992; 12: 1040-1062.

72. Wisden W, Seeburg PH. GABAA receptor channels: Fromsubunits to functional entities. Current Opinion in Neuro-biology 1992; 2: 263-269.

73. Ymer S, Draguhn A, Kohler M, Schofield PR, Seeburg PH.Sequence and expression of a novel GABAa receptor alphasubunit. FEBS Letters 1989; 258: 119-122.

74. Yu S, Ho IK. Effects of acute pentobarbital administration onGABAA receptor-regulated chloride uptake in rat brainsynaptoneurosomes. Brain Research Bulletin 1992; 28:703-708.

75. Zaman SH, Shingai R, Harvey RJ, Darlison MG, BarnardEA. Effects of subunit types of the recombinant GABAAreceptor on the response to a neurosteroid. European Journalof Pharmacology 1992; 225: 321-330.