Ž .Brain Research 773 1997 8–14

Research report

Influence of a GABA receptor antagonist on the sleep–waking cycle in theB

rat

Pierre Gauthier, Christian Arnaud, Gabriel Gandolfo, Claude Gottesmann )

Laboratoire de Psychophysiologie, Faculte des Sciences 06108, UniÕersite de Nice-Sophia Antipolis, Nice Cedex 2, France´ ´

Accepted 6 May 1997

Abstract

Ž .The influence of CGP 35348 a GABA receptor antagonist on the sleep–waking cycle was studied in rats. The animals wereBŽ .injected i.p. at the beginning of the light period and the data expressed by 2-h periods and total duration 6 h . At 100 mgrkg, slow-wave

Ž .sleep SWS was decreased during the 6-h recording with a peculiar decrease during the first 2 h. SWS was subdivided into three stages:Ž .slow-waves; spindles occurring as SWS deepens; and intermediate stage appearing prior to paradoxical sleep PS . Only the slow-wave

stage and intermediate stage were decreased. Waking was increased during the 6-h recording. It was subdivided into waking withŽ . Ž .hippocampal theta rhythm psychomotor active waking and waking without theta activity quiet waking . Both were increased during the

first 2 h. However, quiet waking was increased throughout the recording duration. At 300 mgrkg, SWS was decreased during the three2-h periods. This decrease was principally related to a decrease of the slow-wave stage. PS was increased over the 6-h recording with amarked increase during the second 2-h period. Consequently, under the influence of the GABA receptor antagonist, the SWS wasB

Ž .decreased at the expense of behavioral stages with cortical low-voltage activity waking and PS . GABAergic neurons are present in themesopontine structures responsible for these two stages. We can conclude that endogenous GABA acting at the GABA receptor levelB

participates in the regulation of waking and PS. q 1997 Elsevier Science B.V.

Keywords: Sleep; Waking; Intermediate stage; GABA antagonist, type B; Theta rhythm; Spindle; Rat

1. Introduction

Ž .g-Aminobutyric acid GABA appears to be the mostabundant neurotransmitter that mediates inhibition in themammalian brain since 20–50% of the central neurons are

w xestimated to be GABAergic 42 . Its ubiquitous distributionw xinvolves it in numerous behavioral functions 42 , espe-

w xcially in the circadian timing system 13 .GABA influences on sleep–waking behavior have been

the subject of study since the 1960s. In 1965, Mandel andw xGodin 33 demonstrated an increase of brain GABA level

of about 15% during sleep, a similar increase occurringafter a 15- to 20-h sleep deprivation. In the same years, the

Ž .GABA derivative g-hydroxybutyric acid GBH was shownto have hypnotic influences, particularly inducing paradox-

Ž w x.ical-like sleep up to 1972 23 . However, it progressivelyappeared that GBH has distinct neurophysiological and

w xpharmacological actions from GABA 8,57 . Since GABAdoes not cross the blood–brain barrier, its central level was

) Ž .Corresponding author. Fax: q33 4 9207-6162.

first modulated by inhibition of the catabolic enzyme, theGABA transaminase, which was generally found to in-

Ž .crease sleep, particularly slow-wave sleep SWSw x24,26,36 .

As soon as the GABA receptor structure was eluci-AŽ w x.dated see 32 , many experimental data could be inter-

preted since it appeared that several compounds, like barbi-turates and benzodiazepines, were agonists of the differentbinding sites of the receptor. From this viewpoint, we

w xpreviously showed that barbiturates 15,16 , first-genera-w xtion hypnotics, and benzodiazepines 12 , second-genera-

tion hypnotics, increase the transitional stage which pre-Ž .cedes paradoxical sleep PS at the expense of this last

w xstage. This is not the case with zolpidem 21 and zopi-w xclone 14 , which are third-generation hypnotics.

Nearby these compounds, acting on GABA receptorA

complex binding sites, the GABA site was also studiedusing agonists and antagonists. Muscimol, an agonist, when

w xgiven intraperitoneally, increased SWS and PS in rats 28 .In cats, it increased SWS andror PS when injected in the

w xposterior hypothalamus of normal animals 31 and in

0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.Ž .PII S0006-8993 97 00643-4

( )P. Gauthier et al.rBrain Research 773 1997 8–14 9

animals with preoptic lesions inducing insomnia by them-w xselves 46 . In the same way, its injection in the ventral

periaqueductal gray increased PS, whereas bicuculline, anw xantagonist, decreased PS 47 . Otherwise, injection of mus-

cimol in the pontine reticular formation of rats increasedwaking at the expense of SWS, whilst bicuculline had no

w xeffect 4 .w xWhen Hill and Bowery 22 found that baclofen and

GABA bind to bicuculline-insensitive sites, it became clearŽ .that another receptor GABA existed. It was furtherB

shown that the GABA receptor functions on the ionotropicA

mode, opening a Cly ion channel, while the GABA Bw xreceptor uses the metabotropic mode 61 . This fact clearly

explained that in the thalamic geniculate relay nucleus,where GABA and GABA receptors coexist, the stimu-A B

lation of the afferents first induced a probably glutamateŽ .excitatory postsynaptic potential EPSP followed by a

Ž .double hyperpolarizing potential IPSP . The first one, ofshort latency and duration, was mediated by GABAA

receptors, the second one, late and long-lasting, was medi-w xated by GABA receptors 5,6,59 . At least a part of thisB

late IPSP originated in the thalamus reticular nucleusŽ .NRT . Indeed, this nucleus contains GABAergic neuronsŽ .which project on the relay nuclei responsible for thesleep EEG spindles. Steriade’s group has shown that thala-mic spindles disappear after a transection eliminating the

w xNRT influence 53 , while the spindles persist in thew xdeafferented NRT 54 . The implication of GABA recep-B

tors in the sleep electrophysiological patterns was con-w xfirmed by Juhasz et al. 25 who showed that thalamic

ventroposterolateral injection with GABA receptor antag-B

onists did not change the sleeprwakefulness ratio butdecreased EEG synchronization. In other respects, theGABA receptor agonist, baclofen, injected in the peri-B

w xaqueductal gray was without effect on PS 47 . The interestin the role of GABA receptors in sleep mechanisms wasB

enhanced by the fact that a lethargic mutant mouse strainshowed an increased number of GABA receptors in theB

w x w xneocortex 29 and thalamus 30 , whereas GABA recep-A

tors were unchanged.Consequently, we decided to explore the influence of

GABA acting at the GABA receptor level on sleep–wak-BŽ . w xing cycle by using a GABA antagonist CGP 35348 40B

already known to be an anti-absence-epilepsy compoundw x34 .

2. Materials and methods

2.1. Surgery

Ž .Under thiopental anesthesia 55 mgrkg i.p. , 7 adultmale Wistar rats, weighing 250–320 g, were bilaterally

Ž .implanted with silver ball 1 mm diameter electrodes onw xthe frontal cortex area 10 27 and with a bipolar electrode

Ž .made up of coated except at the tip stainless steel wiresŽ . w x10r100 mm stereotaxically 43 placed in the CA1 areaŽ .A, 5.4; L, 2.6; D, 7.4 to record the hippocampal theta

Ž .activity. Two twisted stainless steel wires 25r100 mmwere bilaterally inserted in the dorsal neck muscles torecord the electromyogram. A ground electrode wasscrewed in front of the olfactory bulb in the middle plane.All recording electrodes were soldered to a connectorŽ .Connectral, France and secured to the skull of the animal

Ž .with dental cement Texton, UK . At the end of surgery,for prophylactic antibiotic therapy, each rat received an

Ž . Ži.m. injection 50 000 IU of benzylpenicilline Specia,.France . Postoperative analgesia was performed by an s.c.

Fig. 1. Sleep–waking stages in the rat. From left to right: AW, psychomotor active waking, characterized by frontal cortex low-voltage activity and thetarhythm in the dorsal hippocampus; NA, non-active waking, without theta rhythm; SW, cortical slow waves appearing from sleep onset; Sp, anterior cortexspindles occurring as sleep deepens; IS, intermediate stage, characterized by anterior cortex high-amplitude spindles and low-frequency hippocampal thetarhythm; PS, paradoxical sleep; F.Cx, frontal cortex; HPC, dorsal hippocampus; EOG, electro-oculogram; EMG, electromyogram. Calibration: 1 s, 100 mV.

( )P. Gauthier et al.rBrain Research 773 1997 8–1410

Ž .injection 5 mgrkg of 6-chloro-a-methylcarbozole-2-Ž .acetic acid Carprofen, ICN, USA . The recovery with

cables for habituation lasted at least 1 week under artificialŽ .12:12 h lighting lights on 9.00 h , the animals being

individually housed in the recording room with free accessto commercial rat chow and water. The ambient tempera-ture was maintained constant at 238C.

2.2. Experimental procedure

At the beginning of light period, the rats received an i.p.Žinjection of either vehicle or CGP 35348 kindly supplied

.by Ciba-Geigy, Basel at the dose of 100 or 300 mgrkg,and were recorded during 6 h. The animals were injectedin a randomized order of administration and recorded witha wash-out period of at least 10 days between two experi-mental sessions. The doses have been chosen because of

w x w xprevious studies on sleep 45 and absence-epilepsy 34 .

2.3. Scoring of the recordings

Ž .Six behavioral stages were distinguished Fig. 1 . Wak-Ž .ing was subdivided into: 1 psychomotor active andror

attentive waking, characterized by frontal cortex low-volt-age activity, dorsal hippocampal theta activity and high

Ž .muscular tone; and 2 non-active waking without thetarhythm, which corresponds to non-motivated motor activ-ity and principally to quiet waking. During SWS three

Ž .stages were differentiated: 3 slow waves occurring at theŽ .cortical level from sleep onset; 4 spindles of increasing

Ž .number, amplitude and duration as SWS deepens; and 5intermediate stage occurring prior to and sometimes justafter PS and characterized by a short-lasting association ofhigh-amplitude anterior spindles and low-frequency theta

Ž .rhythm. It corresponds to a brief stage a few secondsŽduring which the forebrain structures cortex, hippocampus

.and thalamus seem to be disconnected from the brainstem,as reflected by the spontaneous and evoked EEG field

w xactivities 16 . The sixth distinguished stage was paradoxi-cal sleep. Electrophysiological data were analyzed in realtime with our fourth generation scoring systemw x11,18,19,44 . The analysis was performed second by sec-ond because of the polyphasic sleep–waking cycle whichconsisted of several short-lasting stages. Slow waves andspindles were differentiated by their respective frequenciesŽ .2–7 vs. 7–16 cyclesrs and energy. Scored data were

Fig. 2. Influence of CGP 35348 on slow-wave sleep in the rat. The total slow-wave sleep was decreased at 100 and 300 mgrkg during the 6-h recordingperiod. At 100 mgrkg, it was principally related to a decrease of the slow waves and of the intermediate stage during the first 2 h. However, at both doses,all stages with low-frequency EEG synchronization were decreased, although the differences were not significant for the spindles.

( )P. Gauthier et al.rBrain Research 773 1997 8–14 11

Fig. 3. Influence of CGP 35348 on paradoxical sleep in the rat. Paradoxi-cal sleep was increased at 300 mgrkg during the 6-h recording, with amarked increase during the second 2-h period.

expressed by 2 h periods and the total recording durationŽ .6 h . A paired-sample Student’s t-test was used.

3. Results

3.1. Slow-waÕe sleep

At 100 mgrkg, the global SWS was decreased duringŽ .the total recording duration P-0.01 with a markedŽ . Ž .decrease during the first 2 h P-0.0001 Fig. 2 . This

decrease was the consequence of slow-wave stage lower-

Ž .ing during all the recording duration P-0.02 with anŽ .important decrease during the first 2 h P-0.0004 . A

decrease of the intermediate stage also occurred during theŽ .first 2 h P-0.04 . The spindles were not significantly

decreased at any time point.At 300 mgrkg, SWS was decreased during the total

Ž .recording duration P-0.03 and the three 2-h periodsŽ .P-0.02, P-0.04, P-0.03, respectively . This lower-ing was related to a decrease of the slow-wave stage

Ž .during the 6-h recording P-0.02 and during the succes-Žsive three 2-h periods P-0.01, P-0.04, P-0.02,

.respectively . The spindles and the intermediate stage werenot significantly decreased.

3.2. Paradoxical sleep

Only at 300 mgrkg was there an increase during theŽ . Žsecond 2-h period P-0.04 and the 6-h recording P-

. Ž .0.03 Fig. 3 .

3.3. Waking

Total waking was increased at 100 mgrkg during theŽ .6-h recording P-0.01 with a marked increase duringŽ . Ž .the first 2 h P-0.0001 Fig. 4 . It was related to an

Žincrease of waking stages during the first 2 h active.waking, P-0.02; non-active waking, P-0.001 and of

Fig. 4. Influence of CGP 35348 on waking in the rat. At 100 mgrkg, waking was increased during the 6-h recording, with a marked increase of wakingŽ . Ž .with hippocampal theta rhythm active waking and waking without theta rhythm non-active waking during the first 2 h. Non-active waking was

increased during the 6 h recording.

( )P. Gauthier et al.rBrain Research 773 1997 8–1412

Ž .non-active waking P-0.003 during the total recordingduration. There was no increase in any waking stage at 300mgrkg.

4. Discussion

4.1. Slow-waÕe sleep

One major result is the global decrease of SWS, princi-pally consecutive to a lowering of the slow-wave stage.First, this effect is long-lasting since the anti-absence

w xepilepsy influence already disappears after 2.5 h 48 .Second, after injection of GABA receptor antagonists inB

w xthe cat ventroposterolateral nuclei, Juhasz et al. 25 alsoobserved a ‘decrease of synchronization’ which in fact

Ž .corresponded to a decrease of delta waves 0–5 cyclesrs ,the equivalent of our slow waves. However, these authorscharacterized delta waves as an index of deepest SWS. Inthe cat, SWS is characterized first by low-amplitude spin-dles at sleep onset, then by slow waves as sleep deepens,finally by slow waves interspersed with spindles of in-creasing amplitude reaching maximal amplitude just prior

w xto PS 17,58 . In the rat, SWS is first characterized by slowwaves then by slow waves interspersed with spindles ofincreasing amplitude, number and duration up to PS en-

w xtrance 1,16,55 . Consequently, both in cats and rats, thedecrease of slow waves and the increase of spindles inadvanced SWS cannot be considered as a lightening ofsleep. Indeed, the spindles of advanced SWS clearly re-

w xsemble those of the ‘cerveau isole’ preparation 2,17,20 .´w xMoreover, the thalamic transmission level in the rat 10

w xand the thalamocortical responsiveness in the cat 17 ,which are controlled by brainstem activating influencesw x7,49 , progressively decrease during SWS and are thelowest when the spindles are of highest amplitude, i.e.,during the intermediate stage.

The group of Steriade has shown that the thalamus isnot only concerned in the spindle generation but also in

w xslow wave appearance 39 . The basic phenomenon wouldbe a hyperpolarization of thalamocortical neurons byGABAergic neurons principally situated from the NRTw x51 . According to these authors, the slow waves would becharacterized by a neuron higher hyperpolarization than

w xduring the spindles of sleep onset 39 . In our experiment,the decrease of slow waves under CGP 35348 seems toimply a lowering of the NRT GABAergic influences onthe GABA receptors of the relay nuclei neurons. TheB

slight, but not significant, decrease of spindles is in agree-w xment with Von Krosigk et al.’s 59 results showing that

the pharmacological block of GABA receptors abolishesB

absence-epilepsy spike waves, but not physiological spin-dles.

w xIn a recent study, Puigcever et al. 45 injected old ratsŽ .up to 24 months during the night psychomotor activeperiod, with CGP 35348, and demonstrated absence-epi-

lepsy seizures. By visual scoring they observed an increaseof SWS. Probably the difference with our experimentalconditions explains this discrepancy which, partly, couldbe indirectly related to the anti-absence-epilepsy effects of

w xthis compound. Indeed, as shown by Lin et al. 29,30 , anincrease of GABA receptor functioning induces lethargyB

and the thalamic block of these receptors decreases EEGw xsynchronization 25 . In other respects, since GABA re-B

ceptor antagonists suppress the late hyperpolarizing poten-tial induced in the relay nuclei neurons by afferent stimula-

w xtion 6 , the neuron mode of functioning should have atendency to shift from the burst firing mode of SWS to the

w xsingle-spike activity of EEG activated stages 35 .

4.2. Desynchronized EEG stages

The total waking was increased at 100 mgrkg of CGP35348, with a global increase of non-active waking and asignificant increase of psychomotor active waking duringthe first 2 h.

Only the 300-mgrkg dose increased PS during the 6-hrecording, with a marked increase during the second 2-hperiod. The same result was shown by Puigcever et al.w x45 .

Consequently, in our experiment, the global decrease ofSWS is counterbalanced by an increase of the stages withcortical low-voltage activity. It is difficult to assume thatthis result is only related to the intrathalamic phenomenaresponsible for EEG activities as suggested by the work of

w xJuhasz et al. 25 , i.e., the influence of NRT neurons on thethalamocortical relay cells. Indeed, the increase of wakingand PS implies brainstem processes. It is well establishedŽ w x.see 51 that the cholinergic mesopontine reticular nucleiŽ .pedunculopontine and laterodorsal tegmental are in-

Ž .volved: 1 in the facilitatory influences directed towardŽ .the thalamic relay nuclei and the basal forebrain; 2 in the

inhibiting influences directed toward the NRT. These in-Žfluences directly or indirectly particularly for the basal

w x.forebrain influences 3 favor cortical EEG desynchroniza-w xtion 37 . Our scoring system computed this desynchro-

nization up to 25 cyclesrs thus outside the g-range syn-w xchronization 38,52 . However, it is obvious that different

neurophysiological processes regulate waking and PS, al-though a final common mechanism leads to almost identi-cal cortical EEG activities. While the mesencephalic retic-

w xular formation is involved in waking processes 50 , theŽ . w xperi-locus coeruleus-a peri-LCa 41 andror the pedun-

w xculopontine and dorsolateral tegmental nuclei 51,56,60Ž .could be the generator s of PS.

Studying the distribution of GABAergic neurons in thew xpontomesencephalic tegmentum, Ford et al. 9 found that

the ascending reticular activating system may be controlledby local GABAergic neurons. The increased waking andthe correlative SWS decrease we observed could be relatedto a disinhibition of the waking system. In other respects,GABAergic neurons are also located in the peri-LCa and

( )P. Gauthier et al.rBrain Research 773 1997 8–14 13

w xmesopontine structures 9 . The block of local GABA B

receptors by only 300 mgrkg of CGP 35348 could explainthe increase of PS by a disinhibitory process. This hypoth-esis implies that this compound acts at the postsynapticlevel in the two cases. At any rate, our automatic scoring

Žsystem used three electrophysiological criteria cortex, hip-.pocampus, EMG to detect sleep–waking stages. Conse-

quently, the modifications induced by the compound arerelated to brainstem influences regulating these stages. Itcannot result from a dissociation of thalamic and brainstemprocesses.

To conclude, these results show that not only GABAA

receptors, but also GABA receptors, are implicated inB

sleep–waking cycle regulation. The GABA receptorsB

seem to be particularly involved in an inhibitory process ofbehavioral stages with cortical EEG low-voltage activitycorresponding to central activation.

Acknowledgements

This research was supported by DRET Grant 93r164.

References

w x1 B. Bjorvatn, D. Neckelmann, R. Ursin, Sleep spindle activity andnumber of spindle increase as sleep deepens in rats, Sleep Res. 24AŽ .1995 14.

w x2 F. Bremer, Cerveau ‘isole’ et physiologie du sommeil, C.R. Soc.´Ž .Biol. Paris 118 1935 1235–1241.

w x3 G. Buzsaki, R.G. Bickford, G. Ponomareff, L.J. Thal, R. Mandel,F.H. Gage, Nucleus basalis and thalamic control of neocortical

Ž .activity in the freely moving rat, J. Neurosci. 8 1988 4007–4026.w x4 I. Camacho-Arroyo, R. Alvarado, J. Manjarrez, R. Tapia, Microin-

jections of muscimol and bicuculline into the pontine reticularformation modify the sleep–waking cycle in the rat, Neurosci. Lett.

Ž .129 1991 95–97.w x5 D. Contreras, I. Timofeev, M. Steriade, Mechanisms of long-lasting

hyperpolarizations underlying slow sleep oscillations in at corti-Ž .cothalamic networks, J. Physiol. 494 1996 251–264.

w x6 V. Crunelli, N. Leresche, A role for GABA receptors in excitationBŽ .and inhibition of thalamocortical cells, Trends Neurosci. 14 1991

16–21.w x7 S. Dumont, P. Dell, Facilitations specifiques et non specifiques des´ ´

Ž . Ž .reponses visuelles corticales, J. Physiol. Paris 50 1958 261–264.´w x8 J.J. Feigenbaum, S.G. Howard, Gamma hydroxybutyrate is not a

Ž .GABA agonist, Prog. Neurobiol. 50 1996 1–7.w x9 B. Ford, C.J. Holmes, L. Mainville, B.E. Jones, GABAergic neurons

in the rat pontomesencephalic tegmentum: codistribution withcholinergic and other tegmental neurons projecting to the posterior

Ž .lateral hypothalamus, J. Comp. Neurol. 363 1995 177–196.w x10 G. Gandolfo, C. Arnaud, C. Gottesmann, Transmission in the ven-

trobasal complex during the sleep–waking cycle, Brain Res. Bull. 5Ž .1980 553–562.

w x11 G. Gandolfo, L. Glin, G. Lacoste, M. Rodi, C. Gottesmann, Auto-matic sleep–wake scoring in the rat on microcomputer Apple II, Int.

Ž .J. Biomed. Comput. 23 1988 83–95.w x12 G. Gandolfo, R. Scherschlicht, C. Gottesmann, Benzodiazepines

promote the intermediate stage at the expense of paradoxical sleep inŽ .the rat, Pharmacol. Biochem. Behav. 49 1994 921–927.

w x13 B. Gao, J.M. Fritschy, R.Y. Moore, GABA -receptor subunit com-AŽ .position in the circadian timing system, Brain Res. 700 1995

142–156.w x14 P. Gauthier, C. Arnaud, J.M. Stutzmann, C. Gottesmann, Influence

of zopiclone, a new generation hypnotic, on the intermediate stageŽ .and paradoxical sleep in the rat, Psychopharmacology 130 1997

139–143.w x15 C. Gottesmann, Donnees sur l’activite corticale au cours du sommeil´ ´

Ž . Ž .profond chez le Rat, C.R. Soc. Biol. Paris 158 1964 1829–1834.w x16 C. Gottesmann, The transition from slow wave sleep to paradoxical

sleep: evolving facts and concepts of the neurophysiological pro-cesses underlying the intermediate stage of sleep, Neurosci. Biobe-

Ž .hav. Rev. 20 1996 367–387.w x17 C. Gottesmann, G. Gandolfo, B. Zernicki, Intermediate stage of

Ž . Ž .sleep in the cat, J. Physiol. Paris 79 1984 365–372.w x18 C. Gottesmann, J.L. Juan de Mendoza, G. Lacoste, B. Lallement, L.

Rodriguez, M. Tasset, Etude sur l’analyse et la quantification au-tomatiques des differents etats de veille et de sommeil chez le Rat,´ ´

Ž . Ž .C.R. Acad. Sci. Paris 272 1971 301–302.w x19 C. Gottesmann, P.A. Kirkham, G. Lacoste, L. Rodrigues, C. Arnaud,

Automatic analysis of the sleep–waking cycle in the rat recorded byŽ .miniature telemetry, Brain Res. 132 1977 562–568.

w x20 C. Gottesmann, P. User, H. Gioanni, Sleep: a physiological cerveauŽ .isole stage?, Waking Sleep 4 1980 111–117.´

w x21 C. Gottesmann, S. Trefouret, H. Depoortere, Influence of zolpidem,a novel hypnotic, on the intermediate stage and paradoxical sleep in

Ž .the rat, Pharmacol. Biochem. Behav. 47 1994 359–362.w x w3 x w3 x22 D.R. Hill, N.G. Bowery, H Baclofen and H GABA bind to

Ž .bicuculline-insensitive GABA sites in rat brain, Nature 290 1981B

149–152.w x23 M. Jouvet, The role of monoamines and acetylcholine-containing

neurons in the regulation of the sleep–waking cycle, in: Ergibnisseder Physiologie, vol. 64, Springer Verlag, Berlin, 1972, pp. 166–307.

w x24 J.L. Juan de Mendoza, M. Tasset, A. Alves-Sanches, C. Gottesmann,Influence de l’inhibition de la GABA transaminase sur le sommeil

Ž . Ž .chez le Rat, C.R. Soc. Biol. Paris 164 1970 2324–2328.w x25 G. Juhasz, Z. Emri, K.A. Kekesi, O. Salfay, V. Crunelli, Blockade

of thalamic GABA receptors decreases EEG synchronization, Neu-BŽ .rosci. Lett. 172 1994 155–158.

w x26 V.T. Karadzic, Efekat povisenih nivoa g-aminobuterne kiseline ucentralnom nervnom sistemu na faze sna kod macke, Acta Med. 20Ž .1966 282–289.

w x27 W.J.S. Krieg, Connections of the cerebral cortex, J. Comp. Neurol.Ž .84 1946 221–275.

w x28 M. Lancel, T.A.M. Cronlein, J. Faulhaber, Role of GABA recep-¨ A

tors in sleep regulation. Differential effects of muscimol and midazo-Ž .lam on sleep in rats, Neuropsychopharmacology 15 1996 63–74.

w x29 F.H. Lin, Z. Cao, D.A. Hosford, Increased number of GABA BŽ .receptors in the lethargic lhr lh mouse model of absence epilepsy,

Ž .Brain Res. 608 1993 101–106.w x30 F.H. Lin, Y. Wang, Z. Cao, D.A. Hosford, GABA receptor-media-B

Ž .ted effects in synaptosomes of lethargic lhr lh mice, J. Neu-Ž .rochem. 65 1995 2087–2095.

w x31 J.S. Lin, K. Sakai, G. Vanni-Mercier, M. Jouvet, A critical role ofthe posterior hypothalamus in the mechanisms of wakefulness deter-mined by microinjections of muscimol in the freely moving cats,

Ž .Brain Res. 479 1988 225–240.w x32 R.L. Macdonald, R.W. Olsen, GABA receptor channels, Annu.A

Ž .Rev. Neurosci. 17 1994 569–602.w x33 P. Mandel, Y. Godin, Approches biochimiques au probleme du`

sommeil, in: Neurophysiologie des Etats de Sommeil, CNRS, Paris,1965, pp. 13–36.

w x34 C. Marescaux, M. Vergnes, R. Bernasconi, GABA antagonists:BŽ .potential new anti-absence drugs, J. Neural Transm. 35 Suppl.

Ž .1992 179–188.w x35 D.A. McCormick, Cellular mechanisms of cholinergic control of

neocortical and thalamic neuronal excitability, in: M. Steriade, D.

( )P. Gauthier et al.rBrain Research 773 1997 8–1414

Ž .Biesold Eds. , Brain Cholinergic Systems, Oxford Science Publica-tions, Oxford, 1990, pp. 236–264.

w x36 T.D. McIntyre, H.P. Alpern, GABAergic drugs can enhance orattenuate chlordiazepoxide-induced sleep time in a heterogenous

Ž .strain of mice, Pharmacol. Biochem. Behav. 25 1986 1077–1081.w x37 G. Moruzzi, H.W. Magoun, Brainstem reticular formation and acti-

Ž .vation of the EEG, Electroencephalogr. Clin. Neurophysiol. 1 1949455–473.

w x38 M.H.J. Munk, P.R. Roelfsema, P. Konig, A.K. Engel, W. Singer,¨Role of reticular activation in the modulation of intracortical syn-

Ž .chronization, Science 272 1996 271–274.w x39 A. Nunez, R. Curro Dossi, D. Contreras, M. Steriade, Intracellular

evidence for incompatibility between spindle and delta oscillationsŽ .in thalamocortical neurons of cat, Neuroscience 48 1992 75–85.

w x40 H.R. Olpe, G. Karlsson, M.F. Pozza, F. Brugger, M. Steinmann, H.Van Riezen, G. Fagg, R.G. Hall, W. Froestl, H. Bittiger, 35348: acentrally active blocker of GABA receptors, Eur. J. Pharmacol.B

Ž .187 1990 27–38.w x41 H. Onoe, K. Sakai, Kainate receptors: a novel mechanism in para-

Ž . Ž .doxical REM sleep generation, NeuroReport 6 1995 353–356.w x42 R.G. Paredes, A. Agmo, GABA and behavior: the role of receptor

Ž .subtypes, Neurosci. Biobehav. Rev. 16 1992 145–170.w x43 G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates,

Academic Press, New York, 1986, 148 pp.w x44 J.C. Pazzaglia, P. Gauthier, C. Gottesmann, A new automatic

Ž .sleep–waking scoring system in the rat, J. Sleep Res. 3 Suppl. 1Ž .1994 195.

w x45 A. Puigcever, E.L.J.M. Van Luijtelaar, W.H.I.M. Drinkenburg,A.L.M. Coenen, Effects of the GABA antagonist CGP 35348 onB

sleep–wake states, behaviour and spike-wave discharges in old rats,Ž .Brain Res. Bull. 40 1996 157–162.

w x46 M. Sallanon, M. Denoyer, K. Kitahama, C. Aubert, N. Gay, M.Jouvet, Long-lasting insomnia induced by preoptic neuron lesionsand its transient reversal by muscimol injection in the posterior

Ž .hypothalamus in the cat, Neuroscience 32 1989 669–683.w x47 J.P. Sastre, C. Buda, K. Kitahama, M. Jouvet, Importance of the

ventrolateral region of the periaqueductal gray and adjacent tegmen-tum in the control of paradoxical sleep as studied by muscimol

Ž .microinjections in the cat, Neuroscience 74 1996 415–426.

w x48 O.C. Snead, Evidence for GABA mediated mechanisms in experi-BŽ .mental generalized absence epilepsy, Eur. J. Pharmacol. 213 1992

343–349.w x49 M. Steriade, Ascending control of thalamic and cortical responsive-

Ž .ness, Int. Rev. Neurobiol. 12 1970 87–144.w x50 M. Steriade, Arousal: revisiting the reticular activating system,

Ž .Science 272 1996 225–226.w x51 M. Steriade, R.W. McCarley, Brain Stem Control of Wakefulness

and Sleep, Plenum press, New York, 1990, 499 pp.w x52 M. Steriade, F. Amzica, D. Contreras, Synchronization of fast

Ž .30–40 Hz spontaneous cortical rhythms during brain activation, J.Ž .Neurosci. 16 1996 392–417.

w x53 M. Steriade, M. Deschenes, L. Domich, C. Mulle, Abolition ofspindle oscillations in thalamic neurons disconnected from the nu-

Ž .cleus reticularis thalami, J. Neurophysiol. 54 1985 1473–1497.w x54 M. Steriade, L. Domich, G. Oakson, M. Deschenes, The deaffer-

ented reticular thalamic nucleus generates spindle rhythmicity, J.Ž .Neurophysiol. 57 1987 260–273.

w x55 G. Terrier, C. Gottesmann, Study of cortical spindles during sleep inŽ .the rat, Brain Res. Bull. 3 1978 701–706.

w x56 M. Thakkar, C. Portas, R.W. McCarley, Chronic low-amplitudeelectrical stimulation of the dorsolateral tegmental nucleus of freely

Ž .moving cats increases REM sleep, Brain Res. 723 1996 223–227.w x57 G. Tunnicliff, Significance of g-hydroxybutyric acid in the brain,

Ž .Gen. Pharmacol. 23 1992 1027–1034.w x58 R. Ursin, The two stages of slow wave sleep in the cat and their

Ž .relation to REM sleep, Brain Res. 11 1968 347–356.w x59 M. Von Krosigk, T. Bal, D.A. McCormick, Cellular mechanisms of

Ž .a synchronized oscillation in the thalamus, Science 261 199361–364.

w x60 H.H. Webster, B.E. Jones, Neurotoxic lesions of the dorsolateralpontomesencephalic tegmentum-cholinergic cell area in the cat

Ž .II Effects upon sleep–waking states, Brain Res. 458 1988 285–302.w x61 W.J. Wojcik, I. Holopainen, Role of central GABA receptors inB

Ž .physiology and pathology, Neuropsychopharmacology 6 1992201–214.