SK channels are on the move.

Small-conductance Ca(2+)-activated K(+) channels (SK channels) underlie the medium duration after hyperpolarization that follows single or trains of action potentials in many types of neurons. Three subtypes of SK subunits, SK1 (K(Ca)2.1), SK2 (K(Ca)2.2) and SK3 (K(Ca)2.3), have been cloned and are expressed differentially within the central nervous system (CNS). A paper in this issue of BJP reports the discovery of the first example of a positive modulator displaying not only selectivity for SK channels over other channels, but also a subtype selectivity among SK and analogous channels (SK3>SK2>>>SK1=IK). Together with other recent progress in the field, this finding enriches the repertoire of tools available to test the hypothesis that SK channels may be targets for future CNS drugs.

Synthesis and biological evaluation of N-methyl-laudanosine iodide analogues as potential SK channel blockers.

Neuronal action potentials are followed by an afterhyperpolarisation (AHP), which is mediated by small conductance Ca2+-activated K+ channels (SK channels or KCa2 channels). This AHP plays an important role in regulating neuronal activity and agents modulating AHP amplitude could have a potential therapeutic interest. It was previously shown that N-methyl-bicuculline iodide blocks SK channels but its GABA) activity represents a serious drawback. In view of the structural analogy between bicuculline and laudanosine 14, several N-quaternary analogues of the latter were developed. It was shown that N-methyl-laudanosine 15 (NML) and N-ethyl-laudanosine 16 induce a reversible and relatively specific blockade of the apamin sensitive AHP in dopaminergic neurones with mean IC50s of 15, and 47 microM, respectively. Laudanosine 14, N-butyl-17 and N-benzyl-18 derivatives were less potent. In order to find pharmacophore elements, modifications were performed at different positions such as C-1, C-6 and C-7. Intracellular recordings on rat midbrain dopaminergic neurones were made in order to evaluate the putative blockade of SK channels by these molecules. Simplified structures such as tetrahydroisoquinoline derivatives with H or Me at C-1 1-6 presented no significant activity at 300 microM. The presence of a 1-(3,4-dimethoxybenzyl) moiety seems an important feature. Indeed, compound 8 showed a blockade of the AHP of only 33% at 300 microM while compound 13 blocked it by 67%, respectively, at the same concentration. Binding experiments were also performed. Binding affinities for SK channels are in good agreement with electrophysiological data. These results indicate that the presence of a charged nitrogen group is an essential point for the affinity on SK channels. Finally, because of the similar activity of both enantiomers of NML 19 and 20, the interaction site may present a symmetrical configuration.

Modulation of small conductance calcium-activated potassium (SK) channels: a new challenge in medicinal chemistry.

Small conductance calcium-activated potassium (SK) channels are found in many types of neurons as well as in some other cell types. These channels are selective for K(+) and open when intracellular Ca(2+) rises to omega 500 nM. In neurons, this occurs during and after an action potential. Activation of SK channels hyperpolarizes the membrane, thus reducing cell excitability for several tens or hundreds of milliseconds. This phenomenon is called a afterhyperpolarization (AHP). Three subtypes of SK channels (SK1, SK2, SK3) have been cloned and exhibit a differential localization in the brain. SK channels may play a role in physiological and pathological conditions. They may be involved in the control of memory and cognition. Moreover, they are heavily expressed in the basal ganglia (in particular in the substantia nigra, pars compacta) and in the limbic system, suggesting that they may modulate motricity and emotional behaviour. Based on these facts, SK channel subtypes may be a suitable target for developing novel therapeutic agents, but more work is needed to validate these targets. Hence, there is a great need for selective ligands. Moreover, although the risk of peripheral side-effects for SK channel modulators appears to be low, some questions remain to be investigated. Currently, different molecules are known as SK channel modulators. Apamin is a very potent peptidic agent; it produces a strong blockade of these targets which is only very slowly reversible and it has limited selectivity. Dequalinium was found to be an effective blocker. Different chemical modulations on the dequalinium structure led to the discovery of highly potent bis-quinolinium derivatives such as UCL 1684. Other bis-(2-amino-benzimidazole) derivatives are in development. On the other hand, quaternary salts of bicuculline were reported to be effective in inhibiting AHPs. More recent developments on structurally-related molecules revealed that methyl-laudanosine is a new interesting tool for exploring SK channel pharmacology. Finally, a family of compounds has been shown to facilitate SK channel opening. Such compounds may be useful in treating disorders involving neuronal hyperexcitability.

The behavioral effects of acute and chronic JL 13, a putative antipsychotic, in Cebus non-human primates.

RATIONALE: Neuroleptic or antipsychotic drugs are the mainstay of treating acute and chronic psychosis. However, their efficacy is offset by a wide array of side effects, especially extrapyramidal syndromes (EPS). In an attempt to develop novel antipsychotic agents, several animal models have been developed to characterize the profile of new chemical entities. OBJECTIVE: To evaluate the behavioral characteristics of JL 13, a potentially unique antipsychotic agent, three separate studies across a wide dose range in Cebus monkeys were conducted and compared with two studies of oral and parenteral haloperidol. METHODS: Twelve Cebus monkeys were tested with single i.m. doses of JL 13 (0.1-2.5 mg/kg), single p.o. doses (1.0-50.0 mg/kg), and 35 days of continuous p.o. (up to 25 mg/kg) treatments and were blindly evaluated. The same twelve monkeys were also tested with haloperidol i.m. (0.01-0.25 mg/kg) and nine of the monkeys also received haloperidol p.o. (0.1-5.0 mg/kg). Behaviors scored included sedation/arousal, locomotor activity, EPS of parkinsonism and dystonia, as well as reactivity. RESULTS: JL 13 produced mild to moderate and dose-related increased sedation and decreased locomotor activity. Minimal decreases in eye blinking rates were also noted as a consequence of sedation. Mild dystonia and parkinsonian symptoms of slow movement developed at the highest dose tested of 50 mg/kg p.o. in only 6 of 12 monkeys. This is 50 times higher than oral doses of haloperidol that would be needed to produce similar EPS effects. Dose-related EPS of dystonia and bradykinesia occurred in relation to decreased locomotor activity and reactivity to stimuli with haloperidol i.m. and p.o., which are features characteristically seen with traditional neuroleptics. CONCLUSIONS: The behavioral effects of JL 13 in non-human primates suggest this compound is well tolerated and may have a favorable antipsychotic benefit/risk ratio in the clinic, especially if antipsychotic efficacy occurs at doses well below those that can cause EPS.

Effects of JL13, a pyridobenzoxazepine with potential atypical antipsychotic activity, in animal models for schizophrenia.

JL13 [5-(4-methylpiperazin-1-yl)-8-chloro-pyrido[2,3-b][1,5] benzoxazepine fumarate] is a substance with a close structural resemblance to clozapine. However, it is less sensitive to oxidation and may therefore have less hematological side effects. In the present study, JL13 was compared with clozapine and haloperidol in several animal models for schizophrenia. The paw test represents a screening model for antipsychotic drugs that can discriminate between drugs with extrapyramidal side effects and drugs without. Haloperidol increased both forelimb retraction time and hindlimb retraction time (HRT), whereas both clozapine and JL13 increased only HRT. In the prepulse inhibition paradigm, all three drugs reversed the apomorphine- and the amphetamine-induced disruption of prepulse inhibition. However, whereas haloperidol was equally effective against both dopaminergic drugs, JL13 and clozapine were more effective against amphetamine. Finally, only JL13 was able to increase prepulse inhibition in normal rats, whereas only clozapine reduced basal startle amplitude. Taken together, these data suggest that JL13 may be an effective antipsychotic drug, with a profile similar to clozapine.

Electrooxidation potential as a tool in the early screening for new safer clozapine-like analogues.

The chemical modification of clozapine (1) has permitted the finding of new analogues, e.g., olanzapine (2), quetiapine (3), 5-(4-methylpiperazin-1-yl)-8-chloropyrido[2,3-b][1,5]benzoxazepine fumarate (9), with a clinical or psychopharmacological profile similar to that of clozapine. However, when developing new derivatives, the designers are discouraged by the development of clozapine-induced agranulocytosis. Different researchers have raised the role played by the oxidizability of the molecule in such a deleterious effect. In the present paper, we examined the oxidation profile (direct scavenging abilities, efficacy in inhibiting lipid peroxidation, and electrooxidation potential) of newly developed methoxy and trifluoromethylsulfonyloxy analogues related to clozapine, some of them being described as putative antipsychotic. The oxazepine derivative 7, unlike the other diazepine derivatives (6, 10--12), was not readily oxidized. Using a statistical predictive model for hematotoxicity previously described, 7 was found in the cluster of potentially nontoxic compounds while diazepine derivatives 6 and 10-12 were classified as potentially toxic compounds. Among these original compounds, 7, which presents a preclinical clozapine-like profile and a low sensitivity to oxidation, could be a promising antipsychotic candidate with low side effects. Considering the tricyclic derivatives examined so far, some elements of structure-oxidation relationship (SOR) might be pointed out. Regarding the nature of the tricyclic ring substituent, from the most to the least sensitive to oxidation, the sequence was as follows: HO > Cl > CH(3)O > CF(3)SO(2)O. The nature of the tricyclic ring influenced also the sensitivity to oxidation; the diazepine moiety appeared to be the most reactive ring compared to oxa- and thiazepine congeners. These parameters could be advantageously integrated in the early design of new safer clozapine-like analogues.

[Control of cutaneous blood vessels]. / Le contrôle du réseau vasculaire cutané.

In man, three kinds of sympathetic neurons reach the skin. Some cholinergic neurons stimulate the sweat glands, they are excited by temperature-regulating centers. Adrenergic neurons release noradrenaline and ATP to reduce cutaneous blood flow while cholinergic neurons release acetylcholine and a co-transmitter to dilate skin blood vessels. The excitation of both latter types of nerve cells depends on influences from temperature-regulating centers, baroreceptors and exercise. Moreover, in cutaneous blood vessels, a synthesis of NO is enhanced by an increase of body temperature and overall by direct heating of the skin. Burns are associated with axon reflexes and release of inflammatory mediators. The involvement of these various influences is described.

The inclusion of fluoxetine into gamma-cyclodextrin increases its bioavailability: behavioral, electrophysiological and pharmacokinetic studies.

The inclusion of a drug into cyclodextrin generally results in the modification of its physical and chemical properties and sometimes can increase its oral bioavailability. The aim of this study was to compare the effects of the fluoxetine HCl/gamma-cyclodextrin complex to that of traditional fluoxetine HCl. In the forced swimming test in mice, fluoxetine HCl/gamma-cyclodextrin was more effective than fluoxetine HCl, the ED30s being, respectively, 9.5 and 16.9 mg/kg PO. Both compounds (10 mg/kg PO) were able to reduce the firing rate of dorsal raphe neurons in the rat. However, between-groups comparisons showed no significant differences between fluoxetine HCl treated animals and the vehicle group, while fluoxetine HCl/gamma-cyclodextrin appeared significantly more effective than vehicle from minute 25 of the measurement period. In healthy volunteers, the relative oral bioavailability, calculated as the ratio AUC 0-infinity fluoxetine HCl/gamma-cyclodextrin on AUC 0-infinity fluoxetine HCl (20 mg PO), was equal to 249.9%. The three experiments taken together suggest that the complexation of fluoxetine HCl into gamma-cyclodextrin increases its pharmacological efficacy in animals, this effect being related to an enhancement of its oral bioavailability as demonstrated in human healthy subjects.

Hypochlorous acid, a major oxidant produced by activated neutrophils, has low effect on two pyridobenzazepine derivatives, JL 3 and JL 13.

JL 13 (5-(4-methylpiperazin-1-yl)-8-chloro-pyrido[2,3-b]- [1,5]benzoxazepine fumarate) and JL 3 (10-(4-methylpiperazin-1- yl)pyrido[4,3-b][1,4]benzothiazepine), two pyridobenzazepine derivatives structurally related to clozapine, were selected for further development. Due to their structural similarity to clozapine, they are haunted by the spectre of clozapine-induced agranulocytosis. In a previous study, JL 13 was shown to be less sensitive to oxidation than clozapine. In the present paper, using an in vitro procedure, we report the effect of hypochlorous acid (HOCl), a major in vivo oxidant, on both drugs. It appears that the oxidations of JL 3 and JL 13, unlike clozapine, are very slow and little secondary product is formed. Moreover, in contrast to clozapine, the products that were formed are not reactive and thus do not react with glutathione or N-acetylcysteine. Thus, if, as postulated for clozapine, drug-induced agranulocytosis is due to a reactive metabolite formed by neutrophils or their precursors, JL 3 and JL 13 would not be expected to cause the same adverse reaction.

Oxidation sensitivity may be a useful tool for the detection of the hematotoxic potential of newly developed molecules: application to antipsychotic drugs.

Some antipsychotic agents have been found to produce agranulocytosis and aplastic anemia. The oxidation phenomena and/or the formation of free radicals has been suggested to be causally related to various hematological disorders, e.g., agranulocytosis. Using five experimental conditions, we tested the oxidative potential of compounds with and without a history of hematological side effects, e.g., agranulocytosis and aplastic anemia. A statistical analysis was undertaken for each experimental condition and a multivariate analysis combining all results was performed. Two peroxidase-induced free radical models did not successfully discriminate between drugs with and without a history of causing hematologic problems (<70%). The lipid peroxidation system provided even less satisfactory discrimination, with only 56.25% correct classification. However, an 87.5% correct classification was obtained when using the oxidation potentials of these drugs determined at pH 4.7 and at pH 7.4. A multivariate analysis taking into account the five variables provided 87.5% success in classification. The two clusters were better discriminated in terms of a "distance coefficient." In a second analysis, the putative antipsychotic pyridobenzodiazepine analogues (JL5, JL8, JL18, and JL25) were classified in the cluster of toxic compounds, while the oxa- and thiazepine analogues (JL2, JL3, and JL13) were classified as nontoxic compounds. On the other hand, a few metabolites of clozapine and fluperlapine were classified in the toxic compound group. The procedure described herein is, to our knowledge, the first which classifies molecules of different structures as well as different pharmacological profiles according to their hematotoxic potential. Such a procedure could be used to predict drug-induced hematological side effects.

The inflammatory reaction induced by formalin in the rat paw.

The involvement of bradykinin and some other inflammatory mediators in formalin-induced oedema and plasma extravasation was examined. Formalin was injected in rat paws at two doses, 1.75% or 5%. The lower dose induced the development of an immediate oedema associated with a progressive accumulation of 125I-labelled albumin in the paws. These changes were suppressed by pretreatment with capsaicin or xylocaine. They were abolished by RP67580, a NK1 receptor antagonist, and increased by phosphoramidon or diprotin A. They were not affected by HOE140, a bradykinin B2 antagonist, captopril, methysergide, mepyramine, indomethacin, ketoprofen or L-N(G)-nitroarginine. The higher dose of formalin induced a swelling of the paws which took place in two phases associated with two periods of increase in vascular permeability. This oedema was reduced by pretreatment with capsaicin but not with xylocaine. It was reduced by RP67580 injected before or 30 min after formalin. It was inhibited by mepyramine, methysergide, indomethacin and NS-398, a cyclooxygenase-2 inhibitor. It was not modified by HOE140. Its development was similar in normal and kininogen-deficient rats. We concluded that formalin administered at a low dose induces an oedema which mainly results from a neurogenic inflammation mediated by neuropeptides such as substance P. At higher doses, formalin induces an oedema which mainly depends on the release of substance P, prostanoids, 5-hydroxytryptamine and histamine. Bradykinin plays no significant role in the vascular changes whereas this peptide has been reported to participate in the stimulation of nociceptive afferent neurons. This discrepancy could be explained by a difference in the threshold of stimulation of the nociceptive neurons and that of the cells of the vascular walls, or by a formation of kinins in close contact of the neurons.

Facilitatory effects of chronically administered citicoline on learning and memory processes in the dog.

1. Citicoline (cytidine (5') diphosphocholine) has been shown to reverse aging-induced memory deficits, scopolamine-induced amnesia and nucleus basalis magnocellularis lesion-induced learning impairment. 2. This study aimed to evaluate the effects of citicoline on learning and retrieval processes in a complex differential reinforcement of response duration schedule in normal dogs. 3. The effects of citicoline on a stabilized performance were also measured in order to be able to differentiate specific memory effects from non specific influences on the motor, neuro-vegetative and motivational systems. 4. The results demonstrate that citicoline can exert facilitatory effects on learning and memory but also on retrieval processes. The complete absence of effects on the stabilized performance and on the motor, neuro-vegetative and motivational systems constitutes arguments in favour of a selectivity of action on the memory processes.

Simultaneous determination of pirlindole enantiomers and dehydropirlindole by chiral liquid chromatography.

Liquid chromatography was employed for the determination of pirlindole enantiomers and its oxidation product dehydropirlindole (DHP). The direct separation of pirlindole enantiomers and DHP was achieved on a cellulose tris-(3,5-dimethylphenylcarbamate) chiral stationary phase (Chiralcel OD-R). Acetonitrile was used as the organic modifier and sodium perchlorate was used as an ionic additive in the mobile phase. The influence of acetonitrile and sodium perchlorate concentrations on enantioselectivity and achiral selectivity towards DHP was investigated in order to find suitable conditions for the determination of low amounts of each analyte. The mobile phase selected consisted of a mixture of acetonitrile and phosphate buffer (pH 5.0) containing sodium perchlorate (0.05 M) (35:65, v/v) and the UV detector was set at 220 nm. The method developed was validated and was found to be linear in the 0.1-5 microg ml(-1) range (r2 = 0.999 for the three compounds). Repeatability and the intermediate precision for the three analytes at a concentration of 0.1 microg ml(-1) were about 3 and 4%, respectively. This concentration corresponds to the quantification of 0.1% for the minor enantiomer. Actual determinations of enantiomeric purity for single enantiomers of pirlindole were performed.

Comparative study of pirlindole, a selective RIMA, and its two enantiomers using biochemical and behavioural techniques.

The interaction with monoamine oxidase A (MAO-A) and B has been shown to be sensitive to the absolute configuration of molecules. Therefore, the aim of this study was to compare the effects of the racemic pirlindole (a selective and reversible MAO-A inhibitor) and its two enantiomers using biochemical techniques (in vitro and ex vivo determination of rat brain MAO-A and MAO-B activity) and behavioural models (forced swimming test and reserpine-induced hypothermia and palpebral ptosis test). In vitro, the MAO-A IC50 of (+/-)-pirlindole, R-(-)-pirlindole and S-(+)-pirlindole were 0.24, 0.43 and 0.18 microM, respectively. Ex vivo, their ID50 were 24.4, 37.8 and 18.7 mg/kg i.p. The differences between the three compounds were not significant, with a ratio between the two enantiomers [R-(-)/S-(+)] of 2.2 in vitro and 2.0 ex vivo. MAO-B was only slightly inhibited. In the forced swimming test and the reserpine-induced hypothermia and ptosis model, the three compounds had an antidepressant profile. In the forced swimming test, the minimal effective dose ratio between the R-(-) and the S-(+) was again around 2.0. The behavioural observations were thus clearly in accordance with the biochemical data.

Enantiomeric separation of pirlindole by liquid chromatography using different types of chiral stationary phases.

The enantioseparation of pirlindole by liquid chromatography (LC) was investigated using three different chiral stationary phases (CSPs) containing either cellulose tris-(3,5-dimethylphenylcarbamate) (Chiralcel OD-R), ovomucoid (OVM) or beta-cyclodextrin (beta-CD). The effects of the mobile phase pH on retention, enantioselectivity and resolution were studied. Methanol and acetonitrile were tested as organic modifiers while the influence of the addition to the mobile phase of sodium alkanesulfonates or sodium perchlorate was also investigated. Sodium perchlorate was only used on the Chiralcel OD-R column while sodium alkanesulfonates were tested as mobile phase additives on the three kinds of CSPs. The enantioseparation of pirlindole could be obtained on all CSPs tested, the best results with respect to chiral resolution being achieved on the Chiralcel OD-R and the OVM columns. The use of sodium octanesulfonate (NaOS) was found to improve the enantioseparation of pirlindole on the OVM column while enantioselectivity was considerably enhanced by addition of sodium perchlorate on the Chiralcel OD-R column.

JL 13, a potential successor to clozapine, is less sensitive to oxidative phenomena.

The oxidation behaviour of JL 13, a promising antipsychotic, was investigated in comparison with clozapine and loxapine, by measuring their direct "radical scavenging" abilities and their efficacies in inhibiting the lipid peroxidation. In the lipid peroxidation system, the reactivity of these compounds with free radicals produced by gamma-irradiation of linoleic acid may be presented as follows: JL 13 = loxapine < clozapine. In two enzymatic systems (HRP/GSH and HRP/H2O2/ GSH) which generate the thiyl free radicals, clozapine produces a strong enhancement of the thiyl-radical EPR signal intensity while JL 13 and loxapine exhibit no or minimal effect on this signal. The redox potential values for the three derivatives confirm the spectro-photometric and EPR results. Following this study, we show that JL 13, although presenting a preclinical clozapine-like profile, appears less sensitive to oxidation than clozapine.

Pirlindole: a selective reversible inhibitor of monoamine oxidase A. A review of its preclinical properties.

Pirlindole is a tetracyclic compound that has been characterized as a potential antidepressant drug. It has pharmacological characteristics in common with both tricyclic antidepressants and classical irreversible monoamine oxidase inhibitors. Its main mechanism of action consists of a selective and reversible inhibition of monoamine oxidase A. Secondarily, it exerts an inhibitory effect on noradrenaline and 5-hydroxytryptamine reuptakes. It has no effect on the dopaminergic and cholinergic systems. It has only a low potential for amplifying tyramine and noradrenaline pressor effect, which makes one expect that it will not be at the basis of a 'cheese effect'. Pirlindole has an absolute bioavailability of between 20 and 30% due to an extensive first-pass effect. Orally, the Tmax varies between 2.5 and 6 h in the rat and 0.8 and 2 h in the dog. Two phases of elimination (7.5 and 34-70 h) are measured in the rat and three phases in the dog (1.3, 10.8 and 185 h); it is extensively metabolized. The rat eliminates mainly unconjugated products while the dog eliminates mostly conjugated products. Acute and chronic toxicological studies have not revealed potentially dangerous effects of the drug at the usual doses. It does not present measurable mutagenic, clastogenic or carcinogenic properties. Thus, pirlindole shows pharmacological, pharmacokinetic and toxicological properties which make it suitable for the management of depression.

Involvement of 5-hydroxytryptamine and bradykinin in the hyperalgesia induced in rats by collagenase from Clostridium histolyticum.

The involvement of bradykinin, 5-hydroxytryptamine, substance P and prostanoids in the hyperalgesia elicited by collagenase in rat paw was investigated. Collagenase (100 micrograms) induced a slight hyperalgesia in kininogen deficient rats in comparison with the behavioural response obtained in normal rats. Lisinopril (10(-5) M), and angiotensin-converting enzyme inhibitor, increased the duration of the hyperalgesia elicited in normal rats. Ondansetron (0.5 to 5 mumol/kg), a 5-HT3 antagonist, suppressed the hyperalgesia as did methysergide (1.1 to 11 mumol/kg), a mixed 5-HT1 and 5-HT2 receptor antagonist. However, the hyperalgesia was not modified by RP 67580 (1.8 to 18 mumol/kg), a NK1 receptor antagonist, and was only slightly delayed by indomethacin (2 mg/kg), a cyclo-oxygenase inhibitor. The oedema-promoting effect of 5-HT (6 nmol) was inhibited by methysergide but not by ondansetron. The swelling induced by collagenase in rat paw was reduced by methysergide but not by ondansetron. We conclude that the behavioural response induced by collagenase depends on an interactions between bradykinin and 5-HT. Prostanoids play a minor role in the beginning of the reaction whereas substance P is not significantly involved in this hyperalgesia.

JL13, a pyridobenzoxazepine compound with potential atypical antipsychotic activity: a review of its behavioural properties.

The search for an improved clozapine-like compound has resulted in the selection of a new molecule: JL13 (5-(4-methylpiperazin-1-yl)-8-chloro-pyrido[2,3-b][1,5] benzoxazepine fumarate). Like clozapine, JL13 did not antagonize apomorphine-induced stereotypy and did not produce catalepsy but antagonized apomorphine-induced climbing in rodents (ID50 = 3.9 mg kg-1 s.c.). It was inactive against d-amphetamine-induced stereotypy but antagonized d-amphetamine-induced hyperactivity in the mouse (ID50 = 4.4 mg kg-1 i.p.). JL13, like clozapine, was able to antagonize (+/-)-DOI-induced head-twitches in the mouse (ID50 = 2.0 mg kg-1 i.p.). In the open-field test in the rat and forced swimming test in the mouse a high similarity was noted between the two drugs in the same range of doses. In a complex temporal regulation schedule in the dog, JL13 showed a high resemblance with clozapine without inducing sialorrhea, palpebral ptosis or any significant motor side effects. In rats trained to discriminate clozapine, JL13 (10 mg kg-1 i.p.) induced a high level of generalization (70%) to clozapine. In a drug discrimination procedure in the squirrel monkey, JL13 (3-10 mg kg-1 i.m.) produced a full substitution of clozapine. On the basis of these preclinical data, it is thus predicted that JL13 would be a promising atypical antipsychotic drug.