Ginkgo biloba leaf extract (EGb 761®) and its specific acylated flavonol constituents increase dopamine and acetylcholine levels in the rat medial prefrontal cortex: possible implications for the cognitive enhancing properties of EGb 761®.

Experimental and clinical data suggest that the Ginkgo biloba standardized extract EGb 761® exerts beneficial effects in conditions which are associated with impaired cognitive function. However, the neurochemical correlates of these memory enhancing effects are not yet fully clarified. The aim of this study was to examine the effect of repeated oral administration of EGb 761® and some of its characteristic constituents on extracellular levels of dopamine (DA), noradrenaline (NA), serotonin (5-HT), acetylcholine (ACh) and the metabolites 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) in the medial prefrontal cortex (mPFC) of awake rats by use of in vivo microdialysis technique. Subacute (14 days, once daily), but not acute, oral treatment with EGb 761® (100 and 300 mg/kg) or the flavonoid fraction, which represents about 24% of the whole extract caused a significant and dose-dependent increase in extracellular DA levels in the mPFC. Repeated administration of EGb 761® also caused a modest but significant increase in the NA levels, whereas the concentrations of 5-HT and those of the metabolites DOPAC, HVA and 5-HIAA were not affected. The same treatment regimen was used in a subsequent study with the aim of investigating the effects of two Ginkgo-specific acylated flavonols, 3-O-(2''-O-(6'''-O-(p-hydroxy-trans-cinnamoyl)-ß-D-glucosyl)-α-L-rhamnosyl)quercetin (Q-ag) and 3-O-(2''-O-(6'''-O-(p-hydroxy-trans-cinnamoyl)-ß-D-glucosyl)-α-L-rhamnosyl)kaempferol (K-ag). Both compounds together represent about 4.5% of the whole extract. Repeated oral treatment with Q-ag (10 mg/kg) for 14 days caused a significant increase in extracellular DA levels of 159% and extracellular acetylcholine (ACh) levels of 151% compared to controls. Similarly, administration of K-ag (10 mg/kg) induced a significant rise of DA levels to 142% and ACh levels to 165% of controls, whereas treatment with isorhamnetin, an O-methylated aglycon component of EGb 761® flavonol glycosides had no effect. None of the tested flavonoids had a significant effect on extracellular DOPAC and HVA levels. The present findings provide evidence that the subacute treatment with EGb 761® and its flavonol constituents increases DA and ACh release in the rat mPFC, and suggest that the two Ginkgo-specific acylated flavonol glycosides Q-ag and K-ag are active constituents contributing to these effects. As seen for isorhamnetin, the effect on neurotransmitter levels seems not to be a general effect of flavonols but rather to be a specific action of acylated flavonol glycosides which are present in EGb 761®. The direct involvement of these two flavonol derivatives in the increase of dopaminergic and cholinergic neurotransmission in the prefrontal cortex may be one of the underlying mechanisms behind the reported effects of EGb 761® on the improvement of cognitive function.

St. John's wort flavonoids and their metabolites show antidepressant activity and accumulate in brain after multiple oral doses.

Over the last few years many data have been published suggesting a participation of quercetin flavonoids in the antidepressive effect of St. John's wort (SJW) extract. To elucidate these data more deeply we performed two animal behavioural studies examining the antidepressant effects of SJW extract, rutin and, in addition, the quercetin metabolite isorhamnetin. The substances were in all cases compared to imipramine using Porsolt's forced swimming test (FST) after oral gavage of the substances over a 9 day period. All three compounds were found to be effective, with isorhamnetin exhibiting the strongest effect. In addition to this pharmacological study, we carried out two pharmacokinetic studies to examine the CNS level time-curve of the quercetin flavonoids after a single oral dose of SJW extract (1600 mg/kg) and isoquercitrin (100 mg/kg), respectively, and to observe the cumulative effects after daily repeated oral doses of SJW extract over 8 days. After a single dose the maximal CNS levels for quercetin (340 ng/g) and isorhamnetin/tamarixetin (50 ng/g) were found at 4 h. With repeated doses the maximal cumulation for quercetin (367 ng/g) occurred after 5 days whilst isorhamnetin/tamarixetin (640 ng/g) did not reach its maximal cumulation level within the 8 day test period.

Inhibition of lipopolysaccharid-induced sickness behavior by a dry extract from the roots of Pelargonium sidoides (EPs 7630) in mice.

The host response to infections comprise the synthesis and release of proinflammatory cytokines (e.g. IL-1ss, TNF-alpha, IL-6) which induce symptoms of sickness behavior characterised by anorexia, depressed activity, listlessness or malaise. In laboratory animals, sickness behavior can be induced by the administration of cytokines itself or by cytokine-inducers such as lipopolysaccharide (LPS), the active fragment of endotoxin from Gram-negative bacteria. Preparations from roots of Pelargonium sidoides have been traditionally used in South African folk medicine for the treatment of different diseases (e.g. diarrhea, dysmenorrhea, hepatic disorders and respiratory tract infections including tuberculosis). Today, aqueous ethanolic extracts of Pelargonium sidoides are marketed mainly for respiratory tract infections. We studied the effects of the extract EPs 7630 and different fractions separated by ultrafiltration in an animal model of sickness behavior. The results of this study demonstrate that the extract EPs 7630 and the high-molecular weight fraction (F3) alleviate the symptoms of sickness behavior.

Effect of chronic antidepressant treatment on beta-receptor coupled signal transduction cascade. Which effect matters most?

BACKGROUND: Beta-receptor down-regulation has been described as a common biochemical effect of chronic treatment with many but not all antidepressant drugs. Beta-receptor activation leads to elevated intracellular levels of cAMP followed by the activation of several protein kinases which in turn activate various transcription factors. One of those, CREP has received increasing interest as an relevant component within the antidepressant drug modulated signal cascade as it represents a down-stream signal not only of the beta-receptor but also of serotonin receptor activation. Chronic treatment with many antidpressant drugs has been shown to alter CREP levels in several brain regions. While beta-receptor down-regulation by chronic antidepressant treatment has been a consistent finding, alterations of CREP levels have been observed in both direction. Similarly divergent findings have been reported for BDNF a major gene targeted of CREB, where most but not all findings suggest up-regulation at least at the message level following chronic antidepressant treatment. METHODS: Because of these rather divergent data, we investigated the possible effects of chronic treatment (9 or 19 days) with three different antidepressant drugs (reboxetine, citalopram, imipramine) on the individual parameters of the beta-receptor coupled signal transduction cascade. All animals were also tested for possible antidepressant effects using the forced swimming test. RESULTS: While beta-receptor density was down-regulated by reboxetine and imipramine but not citalopram, CREB protein was only mildly elevated after 9 days, and not changed or slightly reduced after 19 days. BDNF protein levels were not or only slightly enhanced, but only for the 9 days treatment. Citalopram was most active. Under the conditions chosen, all three drugs were active in the forced swimming test. CONCLUSION: Taken together, the findings reported make it difficult to identify one single component of the beta-receptor coupled signal transduction cascade as common final target of chronic antidepressant treatment.

St John's wort (Hypericum perforatum) modulates evoked potentials in guinea pig hippocampal slices via AMPA and GABA receptors.

The effects of an ethanolic extract of the plant Hypericum perforatum L. (St John's wort) (HYP) and its hydrosoluble fraction (HYPWS) on electrically evoked population spikes and fEPSP were investigated in this study. Concentration dependent (10(-6) to 10(-4) g/l) excitatory effects were found. Above concentrations of 10(-3) g/l, HYP reduced the evoked responses, whereas HYPWS further increased them. Paired pulse facilitation was unaffected with HYPWS (10(-4) to 10(-2) g/l). The excitatory effects of HYPWS were amplified by the GABA(A) and GABA(B) receptor antagonists bicuculline and phaclofen, respectively. These excitations were antagonised by the AMPA receptor antagonist CNQX. Excitations caused by hypericum were not antagonised by the NMDA receptor antagonists D-APV and MK801, the metabotropic glutamate receptor (type I and II) antagonist MCPG, or the L-type calcium channel blocker verapamil. Hypericin and hyperforin, two components of H. perforatum, were found not to be responsible for the excitatory effects of the extract.

The antiepileptic drug losigamone decreases the persistent Na+ current in rat hippocampal neurons.

The tetronic acid derivative losigamone is a new anticonvulsant drug with a mechanism of action that was previously unknown. The drug decreases the frequency of spontaneous action potentials and suppresses repetitive firing of neurons. Here we tested the hypothesis that losigamone suppresses the persistent Na+ current (I(NaP)) in hippocampal neurons of rat brain slices and in cultured hippocampal neurons. Whole-cell voltage clamp recordings from neurons of juvenile rats (P15-P25) were performed with pipettes filled with Cs-gluconate or CsF. After pharmacological block of K+ and Ca2+ currents I(NaP) was revealed by applying slow depolarizing voltage ramps from -70 to 0 mV. Losigamone (100-200 microM) was dissolved in DMSO (0.1%) and was applied by bath application or local pressure application. Losigamone induced a decrease in amplitude of I(NaP) at depolarized membrane potentials which was reversible in cultured neurons. When tetrodotoxin (TTX) was added to the bath, I(NaP) was blocked and only a residual non-specific outward cation current (I(cat)) remained. Losigamone had no obvious effect on responses to voltage ramps under these conditions. Thus, losigamone did not affect I(cat) or induce any additional currents. The data suggest that losigamone decreases neuronal excitability via a decrease in I(NaP).

Effects of two different extracts of St. John's wort and some of their constituents on cytochrome P450 activities in rat liver microsomes.

We examined two commercially available St. John's wort extracts and some of their constituents for their potential in inducing rat cytochrome P450 (CYP450) enzyme activities after oral administration. None of the extracts or pure constituents tested enhanced the hepatic cytochrome content or the activity of cytochrome P450 isozymes in rat liver microsomes. Our results demonstrated that the reported interactions between St. John's Wort and various other drugs are not mediated by CYP 450 isoforms present in rat liver.

Hypericum extract and hyperforin: memory-enhancing properties in rodents.

Effects of a Hypericum extract in therapeutic use and hyperforin sodium salt were evaluated in rat and mouse avoidance tests. In a conditioned avoidance response (CAR) test on the rat, oral daily administration of hyperforin (1.25 mg/kg/day) or of the extract (50 mg/kg/day) before the training sessions considerably improved learning ability from the second day onwards until the day 7. In addition, the memory of the learned responses acquired during 7 consecutive days of administration and training was largely retained even after 9 days without further treatment or training. The observations made using different doses indicate that these learning-facilitating and/or memory-consolidating effects by the agents follow inverse U-shaped dose-response curves in dose ranges lower than (for hyperforin) or equal to (for Hypericum extract) their effective dose in the behavioral despair test for antidepressants. In a passive avoidance response test on the mouse, a single oral dose (1.25 mg/kg) of hyperforin not only improved memory acquisition and consolidation, but also almost completely reversed scopolamine-induced amnesia. The single Hypericum extract dose tested (25 mg/kg) did not reveal any significant effects in the passive avoidance response (PAR) test on the mouse. These observations suggest that the Hypericum extract could be a novel type of antidepressant with memory enhancing properties, and indicate that hyperforin is involved in its cognitive effects. Pure hyperforin seems to be a more potent antidementia agent than an antidepressant.

Effects of Hypericum perforatum L. on evoked potentials in guinea pig hippocampal slices.

Therapeutic uses of Hypericum extracts have been demonstrated as safe and effective in treating mild to moderate depression in numerous clinical trials. To date, however, no definitive statements on their mode of action can be made, and little information on their electrophysiological effects is available. The present communication summarises the results of our efforts directed towards clarifying the effects of an ethanolic Hypericum extract (HYP) and its hydrosoluble fraction (HYPWS), and two of its constituents hypericin and hyperforin on electrically evoked population spikes in guinea pig hippocampal slices. In higher concentrations (>10 microM), the two extract constituents tested revealed inhibitory effects only, whereas concentration-dependent (between 10(-6) to 10(-4) g/l) excitatory effects were observed for HYP and HYPWS. The excitatory effects were strongly amplified by the GABA(B) antagonist phaclofen, whereas the effects of bicucullin, a GABA(A) antagonist, were marginal. The excitations were completely blocked by the AMPA antagonist CNQX, but not by the NMDA antagonists APV and MK801 or the L-type calcium-channel blocker verapamil. This kind of excitatory effect on the hippocampus is unknown in other antidepressants and; indeed, many of the latter reduce neuronal excitability. We conclude, therefore, that the mechanisms involved in the antidepressant activity of Hypericum extracts are different from those of conventional antidepressants, and that identifying their excitatory components may facilitate their more rational standardisation.

Effect of Ginkgo biloba extract (EGb761) on glucose metabolism-related markers in streptozotocin-damaged rat brain.

To reveal whether an extract of Ginkgo biloba (EGb761) may affect streptozotocin (STZ)-induced impairments in brain glucose metabolism, autoradiographies of [3H]cytochalasin-B binding to the total population of glucose transporters, [125I]insulin binding to insulin receptors, [3H]glyburide binding to sulfonylurea receptors, and radioactive in situ hybridization for GLUT3 mRNA were carried out in hippocampal brain sections of adult rats that have additionally been divided into good performers (GP) and poor performers (PP) by behavioural tests before the experiments. The STZ-induced increases in hippocampal [3H]cytochalasin-B binding to (total) glucose transporters returned to almost normal values following EGb761 treatment, regardless of the experimental animal group (GP or PP) tested. Similarly, the STZ-mediated enhancements in hippocampal insulin receptor binding of GP rats were partially compensated by the treatment with EGb761. The data suggest beneficial effects of EGb671 on impaired brain glucose metabolism, at least under the experimental conditions used in the study presented.

Hyperforin modulates gating of P-type Ca2+ current in cerebellar Purkinje neurons.

Whole-cell, patch-clamp recordings from acutely isolated cerebellar Purkinje neurons demonstrate a two-stage modulation of P-type high-voltage-activated (HVA) Ca2+ current by a constituent of St. John's wort, hyperforin (0.04-0.8 microM). The first stage of modulation was voltage dependent and reversible. It comprised slow-down of the activation kinetics and a shift in the voltage dependence of P-current to more negative voltages. Hyperforin (0.8 microM) shifted the maximum of the current/voltage (I/V) relationship by -8+/-2 mV. The second, voltage-independent stage of modulation was manifested as a slowly developing inhibition of P-current that could not be reversed within the period of study. Neither form of modulation was abolished by intracellular guanosine 5'-O-(2-thiodiphosphate) (GDPPS) or guanosine 5'-O-(3-thiotriphosphate) (GTPgammaS) or by strong depolarising pre-pulses, indicating that modulation via guanine nucleotide-binding proteins (G proteins) is not involved in the observed phenomenon. Calmidazolium (0.5 microM), an antagonist of the intracellular Ca2+-binding protein calmodulin significantly inhibited the hyperforin-induced shift of the IIV curve maximum and the slow-down of the activation kinetics. It did not, however, affect the delayed inhibition of P-current, indicating that the two stages of modulation are mediated by separate mechanisms.

Hyperforin attenuates various ionic conductance mechanisms in the isolated hippocampal neurons of rat.

Effects of hyperforin, an acylphloroglucinol derivative isolated from antidepressive medicinal herb Hypericum perforatum (St. John's Wort), on voltage- and ligand-gated ionic conductances were investigated. Whole-cell patch clamp and concentration clamp techniques on acutely isolated hippocampal pyramidal neurons and on cerebellar Purkinje neurons of rat were used. At concentrations between 3 to 100 microM hyperforin induced a dose and time dependent inward current which completely stabilized within a few seconds. Although 1 microM hyperforin inhibited virtually all investigated conductances (GABA > or = I(Ca(N)) > I(Na) > I(Ca(P) > or = AMPA > or = I(K(A)) > NMDA > I(K(DR))), its effects on several of them could not be reversed by repeated washings. Dose response studies revealed that although AMPA induced current is inhibited by hyperforin in a competitive manner, these responses are not completely blocked by very high concentration of the agent. On the contrary, however, NMDA receptor-activated ionic conductance could be completely and uncompetitively inhibited by the agent. Taken together these observation not only reconfirm that hyperforin is a major neuroactive component of hypericum extracts but also demonstrate that this structurally unique and naturally abundant molecule is a potent modulation of mechanism involved in the control of neuronal ionic conductances. Various observed effects of hyperforin do not, however, seem to be mediated by one single molecular mechanism of action of the agent.

Acute effect of KA-672, a putative cognitive enhancer, on neurotransmitter receptor binding in mouse brain.

7-Methoxy-6-[3-[4-(2-methoxyphenyl)piperazin-1-yl]propoxy]-3,4-dim ethyl-2H-1-benzopyran-2-one hydro-chloride (KA-672), structurally related to naturally occurring coumarins, has been described as a potential drug for enhancing cognitive functions. However, a detailed characterization of the pharmacological profile of KA-672 in vivo is still lacking. Quantitative neurotransmitter receptor autoradiography was used as a tool to screen for KA-672-induced changes in a number of transmitter receptors including cholinergic, noradrenergic, glutamatergic, GABAergic, and serotonergic subtypes throughout the brain. Two hours following treatment of mice with 1 mg/kg KA-672 per os, slight increases of nicotinic and M1-muscarinic cholinergic receptor binding, of alpha2-and beta-adrenoceptor as well as 5-HT1A receptors in various cerebral cortical regions were observed, whereas 5-HT2A binding sites were strikingly increased throughout the brain following KA-672 treatment. In contrast, (+/-)-alphaamino-3-hydroxy-5-methylisoxazole-4-propionic acid receptor binding was significantly decreased in some cortical regions after drug treatment. No effects of KA-672 treatment on N-methyl-D-aspartate, kainate, GABA(A) and benzodiazepine receptor as well as M2-muscarinic cholinergic and high-affinity choline uptake binding were observed. As interactions between the cholinergic, noradrenergic and serotonergic neurotransmission have been stressed to play important roles in realizing learning and memory events, the cognition-enhancing effects of KA-672 may be due to this complex in vivo pharmacological profile of KA-672.

KA-672 inhibits rat brain acetylcholinesterase in vitro but not in vivo.

KA-672, a lipophilic benzopyranone derivative which is currently under development as a cognitive enhancer and antidementia drug, has previously been shown to have facilitatory effects on learning and memory in rats at doses of 0.1-1 mg/kg. We now report that KA-672 inhibited the activity of acetylcholinesterase (AChE), measured in vitro in rat brain cortical homogenate, with an IC50 value of 0.36 microM indicating that KA-672 may improve cognitive functions as a consequence of AChE inhibition. However, when we employed the microdialysis procedure to monitor acetylcholine (ACh) release from rat hippocampus, no effect of KA-672 (0.1-10 mg/kg) was found, indicating a lack of inhibition of brain AChE under in vivo-conditions. [14C]-labelled KA-672 was found to easily penetrate the blood-brain barrier, and an apparent concentration of 0.22 nmol/g brain (equivalent to 0.39 microM tissue concentration) was calculated following an i.p. injection of 1 mg/kg KA-672. However, no labelled substance could be detected in hippocampal microdialysates or in cerebrospinal fluid (CSF) taken from the cisterna magna, indicating that the concentration of KA-672 in brain extracellular fluid must have been below 0.01 microM. We conclude that KA-672 is a potent AChE inhibitor, an activity which, however, does not contribute to its behavioural effects in vivo because the lipophilic drug does not reach sufficient concentrations in the extracellular fluid, apparently due to cellular sequestration.

Damaged neuronal energy metabolism and behavior are improved by Ginkgo biloba extract (EGb 761).

The standardized extract EGb 761 from the dried green leaves of Ginkgo biloba is a complex mixture of ingredients with an uniquely broad spectrum of pharmacological activities on the central nervous system e.g. in memory enhancing properties and in the regulation of cerebral glucose/energy metabolism. To test these effects on both behavioral and metabolic brain parameters, the animal model of intracerebroventricular (i.c.v.) streptozotocin (STZ) treatment was used. To quantify the experimental data more precisely, animals that were good performers were separated from poor performers by means of the holeboard test before i.c.v. administration of STZ. Good performers only were considered for the study. After a 1-week training period on the holeboard improvement was seen in all animals in learning, memory and cognition, and the improvement was maintained over the investigation period of 12 weeks in the control group. In this group, the energy pool in the cerebral parietotemporal cortex was found to be large and the energy turnover high. After triplicate i.c.v. STZ injection, working memory (WM), reference memory (RM), and passive avoidance (PA) behavior fell off and continued to deteriorate throughout the investigation period. Otherwise there were no significant differences in locomotor activity, excluding the possibility that activity per se might have contributed to the behavioral abnormalities. These were accompanied by a permanent deficit in cerebral energy metabolism. The ongoing deterioration in behavior and the maintained deficit in cerebral energy metabolism occurring after a triplicate i.c.v. STZ injection were significantly slowed down by EGb761. The deficits in learning, memory and cognition were partially compensated, and the disturbances in cerebral energy metabolism returned to almost completely normal values. These findings underscore the beneficial effect of EGb761 that had been reported in dementia disorders.

Antidepressant activity of hypericum perforatum and hyperforin: the neglected possibility.

Efforts leading to the identification of hyperforin as an antidepressive component of therapeutically used alcoholic hypericum extracts are described and discussed. Initially, the effects of this unique and major constituent of the herb were detected in peripheral organs using in vitro models and an extract was obtained by supercritical extraction of the herb by carbon dioxide. These extracts are highly enriched in hyperforin (38.8%) and are devoid of hypericines and numerous other components of alcoholic extracts. Studies with such an extract and with isolated hyperforin indicated that this acylphloroglucinol derivative can inhibit serotonin-induced responses and uptake of this neurotransmitter in peritoneal cells. Assuming that the effects of hyperforin were due to its actions on serotoninergic 5-HT3/5-HT4 receptors, further studies were conducted to investigate its effects on the CNS. These efforts revealed its antidepressant activity in the behavioral despair test and led to the working hypothesis that hyperforin and serotoninergic mechanisms are involved in the antidepressant activities of alcoholic hypericum extracts. The observations made during this study also indicate that hyperforin is the major, but not the only antidepressive component of alcoholic extracts.

The putative cognitive enhancer KA-672.HCl is an uncompetitive voltage-dependent NMDA receptor antagonist.

KA-672.HCl (KA-672) is a new substance demonstrating anti-dementia properties. It shows modulatory effects on several neurotransmitter systems known to be affected in patients with Alzheimer's disease. In this study the action of KA-672 on the NMDA receptors was examined by applying patch clamp techniques to acutely isolated hippocampal neurons. KA-672 antagonizes NMDA responses in a voltage-dependent manner. At a holding potential of -90 mV the IC50 value for the blocking action of KA-672 was 20+/-7 microM. This action of KA-672 is independent on the concentration either of agonist or coagonist of NMDA receptor. Ketamine, which interacts with the PCP center, does not occlude the action of KA-672. Evidently, KA-672.HCl is a weak NMDA receptor-operated channel blocker. This property may account for its pharmacological profile.

Effects of the anticonvulsant losigamone and its isomers on the GABAA receptor system.

We conducted in vitro studies to clarify the possible involvement of GABAA receptor-mediated processes in the anticonvulsant effects of losigamone and its optical isomers AO-242 (+ losigamone) and AO-294 (- losigamone). In binding experiments with cortical and cerebellar membrane preparations of rat brain, < or = 100 microM losigamone did not affect the specific binding of [3H]GABA, [3H]flunitrazepam, or [35S]t-butyl-bicyclophosphorothionate (TBPS) to their receptors. Losigamone, however, in concentrations of 10(-8)-10(-5) M, stimulated 36Cl influx into spinal cord neurons in the absence of exogenous GABA. This effect was inhibited by the GABA antagonists bicuculline (BIC) and picrotoxin (PIC). Losigamone 10(-5) M potentiated the effect of a suboptimal concentration of exogenous GABA 10(-5) M on 36 Cl influx. Both isomers of losigamone likewise stimulated 36Cl influx into spinal cord neurons, and these effects were similarly antagonized by BIC and PIC. Losigamone and its optical isomers AO-294 and AO-242 antagonized potassium-induced hyperexcitability in rat hippocampal slices concentration dependently. There were no clear differences in the potencies of losigamone, AO-242, or AO-294. However, AO-294 and AO-242 differed significantly in their ability to suppress TBPS-induced hyperexcitability of hippocampal slices. Such observations demonstrate that although losigamone does not bind to GABA, benzodiazepine (BZD) or PIC binding sites of the neuronal chloride channel, it is capable of stimulating 36Cl influx in the spinal cord neurons by a GABA-sensitive mechanism and at a side distant from the GABA channel.

Anticonvulsant drug toxicity in rat brain cell aggregate cultures.

As an extension of a previous validation study, the concentration-dependent effects of a series of anticonvulsant drugs were examined in aggregating cell cultures of foetal rat telencephalon. Cultures were treated either at an early (day 5 to day 14) or at an advanced (day 20 to day 28) developmental stage, and assayed for changes in the activities of the cell type-specific enzymes choline acetyltransferase (ChAT), acetylcholinesterase (AChE), glutamic acid decarboxylase (GAD), glutamine synthetase (GS) and 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNP). Five drugs (carbamazepine, diazepam, phenobarbital, phenytoin and valproate), currently used in the treatment of epileptic patients, were tested together with losigamone, a recently developed anticonvulsant. The results show distinct, concentration-dependent patterns of biochemical changes for the different drugs. Phenytoin, carbamazepine, losigamone and diazepam greatly reduced GAD, ChAT and AChE activities, indicating a relatively high neuron-specific toxic potential. Diazepam produced a more general pattern of toxicity and, in contrast to the anticonvulsants, showed higher toxicity in less-differentiated cultures. Phenobarbital and valproate slightly but significantly increased the activities of several enzymes. The patterns of concentration-dependent effects observed in this three-dimensional cell culture system are in good agreement with the presumed neurotoxic and/or teratogenic potential of these drugs.