Brain chimeras in birds: application to the study of a genetic form of reflex epilepsy.

A strain of chicken, called here FEpi (for Fayoumi epileptic), bearing an autosomal recessive mutation, exhibits a form of reflex epilepsy with EEG interictal paroxysmal manifestations and generalized seizures in response to either light or sound stimulations. By using the brain chimera technology, we demonstrate here that the epileptic phenotype can be partially or totally transferred from an FEpi to a normal chick by grafting specific regions of the embryonic brain. The mesencephalon contains the generator of all epileptic manifestations whether they involve visual or auditory neuronal circuits, with the exception of the abnormal EEG which is transmitted exclusively by telencephalic grafts. This analysis supports the hypothesis that certain forms of human and mammalian epilepsies have a brainstem origin.

Cortical hypometabolism and its recovery following nucleus basalis lesions in baboons: a PET study.

The cerebral metabolic rate for glucose was measured serially with positron emission tomography and [18F]fluorodeoxyglucose in five baboons with stereotactic electrocoagulation of the left nucleus basalis of Meynert (NbM). Four days after lesion, a significant metabolic depression was present in the ipsilateral cerebral cortex, most marked in the frontotemporal region, and which recovered progressively within 6-13 weeks. These data demonstrate that adaptive mechanisms efficiently compensate for the cortical metabolic effects of NbM-lesion-induced cholinergic deafferentation. Moreover, unilateral NbM lesions also induced a transient reduction in contralateral cortical metabolic rate, the mechanisms of which are discussed. Explanation of these effects of cholinergic deafferentation in the primate could further our understanding of the metabolic deficits observed in dementia of the Alzheimer's type.

Effects of unilateral lesion of the nucleus basalis of Meynert on brain glucose utilization in callosotomized baboons: a PET study.

Prior work has demonstrated that unilateral lesions of the nucleus basalis of Meynert (NbM) in baboons induce a marked reduction in glucose utilization of the ipsilateral cerebral cortex, linearly proportional to the depression in cortical choline acetyltransferase (ChAT) activity achieved. Unexpectedly, there was also marked hypometabolism of the contralateral cerebral cortex, and glucose utilization recovered gradually on both sides despite persistent deficit in cortical ChAT activity. To investigate the role of the corpus callosum (CC) in this bilateral metabolic effect and subsequent recovery, three baboons were subjected to unilateral electrolytic NbM lesion greater than 3 months following section of the anterior CC. Brain glucose utilization was sequentially studied by positron emission tomography; ChAT activity was measured and histological sections obtained after death. In these animals, the NbM lesion also induced significant metabolic depression over the ipsilateral cortex, proportional to the reduction in ChAT activity. Corpus callosotomy did not prevent the contralateral metabolic effects, suggesting that the latter do not normally operate through the CC. However, there was no significant recovery of glucose utilization, suggesting that, following unilateral NbM lesion, the CC normally mediates, at least in part, the recovery of cortical glucose utilization.

Anticonvulsant activity of the diaryltriazine, LY81067: studies using electroencephalographic recording and positron emission tomography.

It is reported that LY81067, a new diaryltriazine, possesses anticonvulsant properties against grand mal status epilepticus induced by intravenous administration of picrotoxin binding site ligands (Ro 5-4864 and pentylenetetrazole) in the baboon. Intravenous administration of LY81067 during the seizures blocked grand mal type electroencephalographic (EEG) paroxysmal discharges and led to a long electrical silence, progressively replaced by spike-and-wave discharges of low frequency (2 c/sec). A transient blocking effect was also observed when LY81067 was injected during grand mal status epilepticus induced by the benzodiazepine inverse agonist methyl beta-carboline-3-carboxylate; however, the long electrical silence observed after administration of LY81067 was rapidly followed by grand mal type paroxysmal discharges in the EEG, which could be stopped by a subsequent injection of Ro 15-1788. However, LY81067 also displayed intrinsic epileptogenic properties. Administration of this drug alone led to the appearance of rhythmic EEG (2-3 c/sec) associated with myoclonia. Concomitantly with the EEG studies, interactions of all these drugs with benzodiazepine receptors were observed in vivo using [11C]Ro 15-1788 as radioligand and positron emission tomography (PET) as a non-invasive technique to measure the binding of the [11C]benzodiazepine antagonist in brain, in vivo. The [11C]Ro 15-1788 bound in the brain could not be displaced by the administration of LY81067 but rather, the [11C]antagonist binding in the brain was somewhat enhanced. Administration of pentylenetetrazole or Ro 5-4864 decreased the rate of wash-out of the radioligand. This fast effect of these two convulsant drugs was partially inhibited by the subsequent administration of LY81067. The concomitant blocking of the grand mal status epilepticus was also observed.(ABSTRACT TRUNCATED AT 250 WORDS)

Pentylenetetrazol-induced seizure is not mediated by benzodiazepine receptors in vivo.

The selective benzodiazepine antagonist RO 15-1788, labelled with carbon 11 [11C] RO 15-1788, as a specific marker, together with positron emission tomography, allows the in vivo study of benzodiazepine receptors in primates. In addition, when coupled with recordings of electroencephalographic activity, this method offers the feasibility of studying the correlation between occupancy of benzodiazepine receptors and the convulsant action of drugs acting at the benzodiazepine-GABA receptor complex in vivo. The present study showed that convulsant doses of pentylenetetrazol (PTZ) could affect the binding of [11C] RO 15-1788 in vivo in two ways, depending on the doses tested: at concentrations of 20 and 30 mg/kg, pentylenetetrazol increased the binding of [11C] RO 15-1788 whereas larger concentrations displaced the binding of [11C] RO 15-1788. The direct correlation between the occupancy of respective benzodiazepine receptors, afforded by increasing convulsant doses of pentylenetetrazol, revealed that competitive interaction with benzodiazepine receptors was not necessary for pentylenetetrazol to induce the appearance of seizures in vivo.

Continuous determination of the cerebrovascular changes induced by bicuculline and kainic acid in unanaesthetized spontaneously breathing rats.

Cerebral blood flow was sequentially determined (every 2-3 min) with helium clearance in two "vulnerable" structures: the hippocampus and the frontoparietal cortex during bicuculline (n = 11) and kainic acid (n = 9)-induced seizures in unanaesthetized, spontaneously breathing rats. Tissue partial pressures of oxygen and carbon dioxide were continuously and simultaneously evaluated in the same brain areas. All these variables were measured by mass spectrometry with a single gas sampling cannula previously implanted in each structure. The systemic variables, arterial blood pressure, arterial partial pressures of oxygen and carbon dioxide, pH, and bicarbonate concentration were also determined. Arterial and venous catheters were chronically implanted several days prior to the definitive experiments. Bicuculline induced short (about 15 min), recurrent, generalized seizures, with an abrupt rise in arterial blood pressure, an arterial metabolic acidosis and comparable blood flow increases (4-fold) in the hippocampus and the neocortex. A marked increase in tissue partial pressure of oxygen was always preceded by an increase in tissue partial pressure of carbon dioxide. After the seizures, in the 5 rats that survived, cerebral blood flow was significantly lowered; tissue partial pressure of oxygen and partial pressure of carbon dioxide also decreased, but to a lesser extent. Histological examination revealed two types of lesions: predominantly selective chromatolysis but also ischaemic cell change. Kainic acid first induced a decrease in arterial pressure and then hypertension during status epilepticus, with a return of arterial pressure towards basal levels during the recovery period (4 h after the injection). Respiratory alkalosis occurred throughout the experiment. Cerebral blood flow increased progressively to become maximal during status epilepticus. This vasodilatation was greater in the hippocampus (x 8) than in the neocortex (x 4). During recovery, cerebral blood flow tended to decrease but remained significantly elevated. In both structures, tissue partial pressure of oxygen was first lowered while tissue partial pressure of carbon dioxide was elevated; with the occurrence of the wet dog shakes, tissue partial pressure of O2 increased and tissue partial pressure of CO2 decreased. The changes in tissue gases were maximal during status epilepticus and tended to return to their basal levels thereafter, but no decrease in tissue partial pressure of O2 was observed, even 4 h after kainic acid administration. Histological analysis demonstrated ischaemic cell changes, particularly in the limbic system.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of exponential compression curves with nitrogen injection in humans.

Two series of experiments were carried out on humans to study the effects of fast and slow exponential compression curves with N2 additions. Eight subjects in the first series and 13 subjects in the second series were analyzed up to the depths of 400-450 m of seawater (msw). The data indicated that injections of N2 in He-O2 mixture reduced or suppressed the hyperbaric tremor in the two series. Electroencephalographic (EEG) changes were recorded with the two types of compression, but these changes (increase in slow waves, decrease in alpha-activity, appearance of microsleep EEG traces) were more important with the fast exponential compression curves between 200 and 300 msw than with the slow exponential curves. The effects of the fast rates of compression on EEG activities were not compensated by addition of 4-5% N2. Consequently, the fast exponential compression curves, even with N2 injections, cannot be used without risk and must be avoided; the slow exponential compression curve with N2 injection allowed a human subject to reach 450 msw in satisfactory condition, i.e., without tremor and with light EEG changes.

Occurrence of high-pressure nervous syndrome at constant pressure during change of mixture.

Three professional divers have performed a dive to 450 msw. From 200 msw and during the first 64 h on the bottom, they breathed a H2-He-O2 mixture with 54-56% H2. At this time a switch was performed to a mixture with 30% H2, and 8 h later a second switch was performed to 0% H2. In the H2-He-O2 mixture the clinical symptoms of high-pressure nervous syndrome (HPNS) were not present and the electroencephalogram changes were slight. The switch of the mixture induced an isobaric HPNS of high intensity. Twenty-four hours later the HPNS decreased, but the clinical symptoms persisted throughout the stay in the He-O2 mixture. The appearance of isobaric HPNS during the switch might be due to the disappearance of the narcotic substance which suppressed or masked the clinical symptoms; it might also be due to the sudden increase in the partial pressure of He, which was equivalent to a fast compression.

EEG and sleep disturbances during dives at 450 msw in helium-nitrogen-oxygen mixture.

To study the effects of nitrogen addition to the breathing mixture on sleep disturbances at pressure, two dives were performed in which helium-nitrogen-oxygen mixture was used up to 450 m sea water (msw). In total, sleep of 12 professional divers was analyzed (i.e., 184 night records). Sleep was disrupted by compression and by stay at 450 msw: we observed an increase in awake periods and in sleep stages I and II and a decrease in stages III and IV and in rapid-eye-movement sleep periods. These changes, which were more intense at the beginning of the stay, began to decrease from the seventh day of the stay, but the return to control values was recorded only during the decompression at depths below 200 msw. These changes were equivalent to those recorded in other experiments with helium-oxygen mixture in the same range of depths and were independent of the intensity of changes recorded in electroencephalographic activities in awake subjects.

Absence of seizure activity following focal cerebral injection of enkephalins in a primate.

Baboons (Papio papio) with photosensitive have been chronically prepared with guide cannulae and deep electrodes to study the effects of focal injections of opioids. In the hippocampus, amygdala and thalamus (centre median) 50--100 micrograms morphine, 20--100 micrograms Met-enkephalin or 2--10 micrograms FK 33,824 do not induce local or general electrographic or motor signs of epilepsy. The acute epileptogenic effect of morphine and enkephalins observed in rats is not a general phenomenon whereas the anticonvulsant action of opioids acting on mu-receptors is seen in rodents and primates.

[Evolution of visual evoked responses during various states of vigilance in Papio papio (author's transl)]. / Evolution des réponses évoquées le long de la voie visuelle au cours du sommeil nocturne chez Papio papio

Averaged evoked responses (AER) to light flashes were recorded in baboons (Papio papio) during wakefulness, slow-wave sleep and rapid eye movement (REM) periods, at the visual cortex, retrocalcarine sulcus, optic tract (OT), lateral geniculate (LG) and pulvinar. Waking AERs were composed: in the OT, of a negative, low amplitude wave at 13.3 msec (I), a high amplitude wave at 34.8 msec (II), a negative wave at 72 msec (III) and a late component at 151 msec; in the LG, a small positive wave (II) with a peak latency of 40 msec, a high amplitude negative wave (III) with a latency of 70 msec and a late component; in the pulvinar of two low amplitude short latency waves (I and II), respectively negative and positive at 25 msec and 40 msec, then a high amplitude negative wave (III) at 75 msec and a late component; in the retrocalcarine sulcus 3 positive waves (I, II and III) were recorded at 25, 45 and 100 msec and a late component; in the visual cortex, 3 low amplitude negative waves (II, III and IV at 40, 50 and 54 msec, then a positive wave at 80 msec and some late components. In slow-wave sleep, AERs did not change in the OT, but in the LG and pulvinar, they showed an increase in the amplitude of wave II from stage 1 to stage 3. At the cortical level, early waves (II for the retrocalcarine sulcus, II and IV for the visual cortex) presented a marked increase in amplitude during stages 2 and 3, but only a slight increase for stage 1. Peak latency increase of each wave in cortical and subcortical AERs was seen during slow-wave sleep. REM AERs resembled, in amplitude and peak latency, those recorded in the LG and pulvinar during wakefulness; in the visual cortex and retrocalcarine sulcus, they were similar to those obtained during wakefulness and stage 1. In conclusion, a different evoked response was found between visual cortex and deep structures (except for the OT): firstly, during slow-wave sleep (the AERs showed a difference for stage 1 between the visual cortex or the retrocalcarine sulcus and the LG or the pulvinar), secondly, in REM (on the cortex, REM AERs looked like wakefulness and stage 1 responses); on the contrary, in the LG and pulvinar, REM AERs were similar only to those recorded during waking. Finally, it can be said that for Papio papio the differentiation and structural responses between the various stages of sleep (particularly light sleep and REM) were greater in the cortex than in the thalamic structures.

Audiogenic seizures evoked in DBA/2 mice induce c-fos oncogene expression into subcortical auditory nuclei.

Recently the nuclear proto-oncogene c-fos has been shown to be rapidly and transiently expressed following seizures in many types of epilepsies. Until now, immunohistochemical as well as in situ hybridization studies have reported that the dentate gyrus of the hippocampus and most of the cortical areas were invariably heavily labeled. In order to see whether this distribution was reproduced or not in a model of epilepsy which has been proved to not involve these structures, a study was performed on genetically epilepsy-prone DBA/2 mice. Here we show that following audiogenic seizures, c-fos oncoprotein is not expressed in cortical and limbic structures but rather mapped the subcortical auditory nuclei.

High expression of noradrenaline, choline acetyltransferase and glial fibrillary acidic protein in the epileptic focus consecutive to GABA withdrawal. An immunocytochemical study.

Interruption of a chronic GABA infusion into the rat somatosensory cortex induces the appearance of focal epileptic manifestations, known as the 'GABA withdrawal syndrome' (GWS). The aim of the present study was to determine, by immunocytochemistry, if neurotransmitters other than GABA are involved in GWS, namely: noradrenaline (NA), serotonin, choline acetyltransferase (CAT), cholecystokinin, neuropeptide Y, somatostatin and glial fibrillary acid protein (GFAP). Immunocytochemical data were compared in three animal groups: GABA-, saline- and L-aspartate (L-Asp)-infused rats. Only GABA-infused rats presented epileptic manifestations after interruption of the infusion. Saline- and L-Asp-infused rats served as controls. Observations were limited to the region surrounding the cortical infusion site. GABA-infused rats showed in the zone of the epileptic focus a number of cell bodies strongly immunoreactive to NA antibodies much larger than control rats. In addition, NA-immunoreactive fibers formed a dense plexus and some of them were observed around perikarya. In saline- and L-Asp-infused rats, the NA-immunolabelled fibers were sparse and NA immunolabelling was rarely observed in cell bodies. These results contrast to those obtained for the serotonergic system, where no significant difference was observed among the three groups of rats. CAT immunolabelling was observed in cell bodies, but not in nerve terminals in rats of the three groups. The number of CAT-immunoreactive cell bodies was much greater in GABA-infused rats than in the control animals. GFAP immunolabelling showed an important number of astrocytes throughout the cortex of the GABA-infused hemisphere, whereas, astrocytic reaction was limited to the infusion site in controls. Immunocytochemical data concerning peptides revealed cortical neuronal elements labelled similarly in the three groups of rats. Noradrenergic, cholinergic and glial modifications observed mainly in GABA-infused rats can result from lesion and from a specific action of GABA in chronic infusion. These modifications may contribute to the epileptogenesis of GWS, as recently demonstrated by electrophysiological recordings that show a modulating action of NA on firing activity of neurons involved in the epileptic focus.

Effects of opiate-like peptides, morphine, and naloxone in the photosensitive baboon, Papio papio.

The effects of intracerebroventricular (i.c.v.) or systemic injections of Met- or Leu-enkephalin, beta-endorphin, FK 33.824 (D-Ala2, MePhe4, Met(O5)-ol-enkephalin) and of morphine and naloxone have been studied in baboons, Papio papio, which spontaneously show photically induced epileptic responses. Animals were chronically implanted with epidural or deep recording electrodes and a cannula in one lateral ventricle, and tested whilst seated in a primate chair. In some animals the natural syndrome was enhanced by the prior administration of DL-allylglycine, 100--200 mg/kg, i.v. Met- or Leu-enkephalin, 1--10 mg, i.c.v., did not lead to any manifest focal or generalized seizure discharges. Nor did it lead to any consistent enhancement or reduction of photically induced myoclonic responses (as tested 5--10 min after injection). beta-Endorphin, 0.1--0.5 mg, i.c.v., did not enhance or impair photically induced myoclonic responses. FK 33.824, 0.1--0.5 mg, i.c.v., depressed respiration and slowed EEG background rhythms for 9--15 h. This was associated with a loss of myoclonic responses to photic stimulation. These effects were reversed for 20--40 min following the injection of naloxone, 1 mg/kg i.m. A depression of respiration and a slowing of EEG rhythms was seen beginning 5--20 min after FK 33.824, 2 or 4 mg/kg, i.v. The higher dose also abolished photically induced myoclonic responses. Naloxone, 1 mg/kg, definitively reversed these effects. Morphine, 5--10 mg i.c.v., tended to increase the latency to onset of generalized myoclonus during photic stimulation. Myoclonic responses were delayed or diminished after morphine, 5 mg/kg, i.m. Naloxone, 1--2 mg/kg i.m., reversed this effect. Naloxone, 0.2--5.0 mg/kg i.m., alone, did not significantly modify photically induced myoclonus, either in animals of low or high initial responsiveness, or in those pretreated with allylglycine.

Effects of pharmacological manipulation on neurotransmitter and other amino acid levels in the CSF of the Senegalese baboon Papio papio.

The concentrations of GABA, glutamate, aspartate, glycine, taurine, glutamine, asparagine and alanine were determined in the CSF of 10 Senegalese baboons (Papio papio) following initial ketamine anaesthesia and subsequent administration (4 h later) of different compounds known to alter either inhibitory or excitatory neurotransmission. Ketamine itself was apparently without effect as the administration of a second dose of ketamine did not significantly alter the levels of any of the amino acids studied, although GABA levels tended to decrease. The presence of haemolysed material in occasional samples was associated with high GABA, glutamate, aspartate, taurine and asparagine levels. Therefore only haemolysate-free samples were included for analysis. Of the compounds administered, gamma-vinyl GABA had the most evident effect on CSF amino acid levels, increasing GABA (greater than 5-fold) and decreasing glutamate (greater than 50%), aspartate (40-50%), asparagine (20%) and alanine (30-35%) levels. The changes in GABA, glutamate and aspartate were still apparent 24 h post-gamma-vinyl GABA administration. In contrast, sodium valproate did not significantly alter the CSF levels of any of the amino acids studied. Upon acute administration allylglycine decreased the CSF concentrations of GABA and alanine, but not glutamate. These alterations are unlikely related to the occurrence of allylglycine-induced convulsions (in 2 of 4 experiments) as electroconvulsive shock did not alter CSF amino acid levels. During the experimental period encompassing the allylglycine injection (8 weeks), basal (initial post-ketamine, pre-drug sample) amino acid levels were abnormal with large increases in glutamate, GABA, aspartate and taurine whereas asparagine levels were below the limit of detection. Diazepam administration was followed by a significant increase in taurine and a decrease in aspartate levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Relationship between tolerance to GABAA agonist and bursting properties in neocortical neurons during GABA-withdrawal syndrome.

The interruption of intracortical, chronic GABA infusion is known to give rise to 'GABA withdrawal syndrome' (GWS) consisting of electroencephalographic paroxysmal focal activities, associated with behavioral epileptic signs. Neocortical slices were obtained from rats presenting the GWS (GWS slices), and intracellular recordings were performed in the vicinity of the gamma-aminobutyric acid (GABA)-infused site. Electrical stimulation of the underlying white matter induced paroxysmal depolarization shifts (PDSs) in virtually all neurons. Bath-applied GABA (1-10 microM) had no effect on these neurons, while the same dose range was found effective in blocking action potentials in saline-infused cortex slices obtained from control rats. In the GWS slices a population of neurons presented, in addition to synaptically induced PDSs, voltage-dependent and cobalt-sensitive PDSs and bursts of action potentials induced by depolarizing current injections. These intrinsic bursting neurons were unresponsive to high doses of GABA (100 microM). Dose-response curves of isoguvacine, a specific GABAA agonist, showed a shift to the right for the intrinsic bursting cells whatever the parameter measured (depolarization or conductance increase): the ED50 was 50-100 times higher for intrinsic bursting cells than for other non-intrinsic bursting cells, thus indicating that intrinsic bursting cells are tolerant to GABAA agonist. This tolerance may result from a decreased number of receptors or from a change in their properties as a consequence of the previous prolonged GABA infusion. The decrease in the GABA efficacy could lead to disinhibition and could thus give the appearance of epileptic events.

The GABA-withdrawal syndrome: a new model of focal epileptogenesis.

A novel model of focal, cortical epilepsy is described. Chronic (6 h to 14 days), localized application of gamma-aminobutyric acid (GABA) into the somatomotor cortex of rats induces, upon withdrawal, the appearance of epileptogenic activity with maximal electrographic expression circumscribed to the infused site. This GABA-withdrawal syndrome (tested for a 100 micrograms/microliter/h dose) lasted from 24 to 168 h (mean values). A significant correlation was found between infusion time and duration of the excitability rebound, with the longer duration corresponding to the shorter infusion time. A distant lesion effect was observed in the thalamic area of cortical projection. The potential use of this neurotransmitter-induced phenomenon in the study of brain plasticity in general, and of epilepsy in particular, is discussed.

Cerebrospinal fluid amino acid and monoamine metabolite levels of Papio papio: correlation with photosensitivity.

Several putative neurotransmitter amino acids and monoamine metabolites were measured in the cerebrospinal fluid of spontaneously photosensitive baboons (Papio papio) at different periods with varying degrees of photosensitivity in the same animals. At maximum photosensitivity the inhibitory amino acids gamma-aminobutyric acid and taurine were lower, and those of asparagine (metabolite of the excitatory amino acid aspartate) were higher, than when the animals were not photosensitive. Thus a decreased inhibition and perhaps increased excitation correlates with the level of photosensitivity.

Brain chimeras for the study of an avian model of genetic epilepsy: structures involved in sound and light-induced seizures.

The epileptic homozygotes of the Fayoumi strain of chickens (Fepi) are affected by photogenic reflex epilepsy with complete penetrance. Here we demonstrate that they are equally affected by audiogenic reflex epilepsy induced by intense sound stimulation. All the Fepi display sound-induced seizures from hatching to adulthood consisting of initial 'ictal arousal' and running fits usually followed by generalized clonico-tonic convulsions. A running fit is the preconvulsive motor symptom specifically induced by auditory stimulation while neck myoclonus is the preconvulsive motor symptom specifically induced by photic stimulation. The EEG interictal spikes and spike and waves are suppressed and replaced by a desynchronized trace during the seizures of both kinds. Viable neural chimeras were obtained by graft of embryonic brain vesicles from Fepi donors into normal chick embryos. Transfer of the complete audiogenic and photogenic phenotypes was obtained in chimeras resulting from embryonic substitution of both the prosencephalon and mesencephalon. The substitution of the prosencephalon alone resulted in transfer of interictal paroxysmal EEG activity accompanied by the sound and light-induced desynchronization and 'ictal arousal' with no motor seizures. Chimeras with embryonic substitution of the mesencephalon alone displayed running fits and convulsions induced by sound stimulation but only neck myoclonus following light stimulation. The conclusions are reached that: (i) the Fepi is a model of audiogenic and photogenic reflex epilepsy; (ii) in both types, the seizure initiator and the convulsion generator are localized in the brainstem, although reinforcement from telencephalic visual structures is needed to trigger photogenic generalized convulsions.

The kinetics and displacement of [11C]flunitrazepam in the brain of the living baboon.

The distribution and kinetics of [11C]flunitrazepam in the brain were studied by positron emission tomography in the living baboon. Flunitrazepam was labelled on the methyl group with the 20 min positron emitter carbon 11. Fifteen to 25 mCi corresponding to 15-30 nmol were injected i.v. and sequential tomographic pictures of the brain were obtained. In some experiments, therapeutic doses of various benzodiazepines were injected i.v. subsequently in order to study the displacement of the radioactive ligand from brain structures. Lorazepam was shown to displace [11C]flunitrazepam from brain tissue, although other benzodiazepines (chlordiazepoxide, Ro 116896 and Ro 116893) led to a redistribution of the radioactive ligand in the body accompanied by an increase of brain radioactivity.