5-Hydroxytryptamine and avoidance performance in the rat. Antagonism of the acute effect of p-chloroamphetamine by zimelidine, an inhibitor of 5-hydroxytryptamine uptake

Impulsivity as a confounding factor in certain animal tests of cognitive function

Performance in cognitive tasks which require the subject to wait and/or to process a large amount of information can be disrupted by an increase in impulsive-like behaviour. Accordingly, a decrease in impulsive-like behaviour can improve performance in such tasks. Conversely, impulsive-like behaviour may improve performance in cognitive tasks where simple and fast responses and/or only little information processing is required. Thus, impulsivity constitutes a confounding factor in studies of cognitive function. Impulsive-like behaviour may be modified by serotonergic (5-HT) activity, with underactivity in 5-HT neurotransmission increasing impulsivity and vice versa. Drug- or lesion-induced alteration in 5-HT neurotransmission may, therefore, constitute suitable tools to investigate the role of impulsivity in animal tests of cognitive function. Benzodiazepines also increase impulsive-like behaviour, possibly by decreasing 5-HT neurotransmission. Hence, the effects of modulation of 5-HT systems and of the benzodiazepine-binding site on performance in animals tests of cognitive function will be discussed. It is predicted that the effects of manipulations of serotonergic activity or of benzodiazepine administration depend upon the nature of the response required, and that these effects may be mediated through changes in impulse control.

Differential effects of d-fenfluramine and p-chloroamphetamine on H75/12-induced depletion of 5-hydroxytryptamine and dopamine in the rat brain

The effects of the two 5-HT-releasing drugs, p-chloroamphetamine and d-fenfluramine, on central serotoninergic and dopaminergic systems were compared in adult rats. Both drugs (0.5-5.0 mg/kg i.p., 2 hr before death) produced a dose-dependent reduction in levels of 5-HT, but only p-chloroamphetamine decreased the levels of 5-hydroxyindoleacetic acid (5-HIAA) in the hippocampus, striatum and cerebral cortex. Within the dose range tested, d-fenfluramine did not affect the levels of DA and of its metabolites dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) in brain. By contrast, p-chloroamphetamine significantly increased the levels of DA and HVA and decreased the levels of DOPAC, notably in the striatum. As expected of a 5-HT uptake inhibitor, d-fenfluramine at small doses (0.2-0.5 mg/kg) prevented the depletion of 5-HT due to 4-methyl-alpha-ethyl-meta-tyramine (H75/12, 40 mg/kg i.p.), whereas at large doses (1.0-5.0 mg/kg) d-fenfluramine, like p-chloroamphetamine (0.2-1.0 mg/kg), slightly enhanced the effect of H75/12. Neither d-fenfluramine (0.5 mg/kg) nor p-chloroamphetamine (0.5 mg/kg) affected the depletion of DA due to H75/12. These data indicate that p-chloroamphetamine is a 5-HT-releasing drug, at any dose between 0.2 and 5.0 mg/kg, whereas d-fenfluramine acts as a 5-HT uptake inhibitor at 0.2-0.5 mg/kg and as a 5-HT releasing drug at larger doses. On account of the potential neurotoxicity of 5-HT-releasing drugs but not 5-HT uptake inhibitors, it can be inferred that d-fenfluramine is very probably devoid of any neurotoxic action in the dose range (less than 1.0 mg/kg) required for its anorectic action.

Tianeptine stimulates uptake of 5-hydroxytryptamine in vivo in the rat brain

The neurochemical effects of the atypical tricyclic antidepressant, tianeptine, were further assessed on central serotoninergic and dopaminergic systems in the rat. Acute treatment with tianeptine (10 mg/kg i.p.) significantly enhanced the levels of metabolites of 5-HT and DA, 5-hydroxyindole acetic acid and dihydroxyphenylacetic acid respectively, in the brain stem, striatum and cerebral cortex. These effects could be prevented by the administration of drugs acting selectively (or preferentially) on serotoninergic systems such as d,l-fenfluramine and 4-methyl-alpha-ethyl-metatyramine (H75/12), suggesting that the increased metabolism of DA was secondary to a modification of serotoninergic systems in tianeptine-treated rats. In contrast to that found with inhibitors of the uptake of 5-HT, treatment with tianeptine markedly enhanced depletion of 5-HT due to administration of H75/12. However, depletion of DA induced by H75/12, was not altered by tianeptine. In vitro measurement of the uptake of [3H]5-HT also confirmed that tianeptine exerted opposite effects to those of classical tricyclic antidepressants, since the in vivo administration of tianeptine (2 x 10 mg/kg i.p.) induced a significant increase in the uptake of [3H]5-HT in cortical synaptosomes. The fact that both inhibitors of the uptake of 5-HT and tianeptine which, in contrast, enhanced the in vivo uptake of 5-HT, are potent antidepressants, challenges the current hypothesis on the central mechanisms of action of these drugs.

Portacaval anastomosis attenuates the impairing effect of cyproheptadine on avoidance learning in rats--an involvement of the serotonergic system

Portacaval-anastomized (PCA) rats were used to demonstrate the involvement of the serotonergic system in long-term memory formation. Significant increases in the concentration of 5-hydroxyindoleacetic acid, a metabolite of 5-hydroxytryptamine (5-HT), in all regions examined and the turnover rate of this indoleamine transmitter in the hippocampus, hypothalamus, midbrain and medulla oblongata were observed in PCA rats in comparison with sham-operated controls. Cyproheptadine, a 5-HT receptor blocking agent, impaired the retention of two-way avoidance learning reinforced by light stimuli when the drug was intraperitoneally injected immediately after the completion of training. PCA treatment attenuated the impairing effect of cyproheptadine. When cyproheptadine was injected 2 h after the completion of training, the correct response in the retention test period was not decreased. The present results suggest that memory formation is a time-requiring process and is mediated by the central serotonergic mechanism.

The inhibition of the cage-leaving response--a model for studies of the serotonergic neurotransmission in the rat

It was observed that rats that had been given drugs that enhance serotonergic neurotransmission, e.g. the serotonin releasing compounds p-chloroamphetamine (PCA) and fenfluramine, the MAO-A inhibitors and serotonin releasing agents amiflamine and alpha-ethyltryptamine and the serotonin agonists 5-methoxy-N, N-dimethyltryptamine (5-MeODMT), 8-hydroxy-2-(di-n-propylamino) tetraline (8-OH-DPAT), m-chlorophenyl piperazine (m-CPP) and 5-methoxy-3 (1,2,3,6-tetrahydropyridin-4-yl)1H-indole (RU 24969), did not leave their home-cages when the grid-covers were removed in contrast to normal rats who almost immediately left the cages. The association between the serotonin neurotransmission and the inhibitory effect of PCA on the cage-leaving response was indicated by the findings that 1. Serotonin uptake inhibitors (alaproclate and citalopram) antagonized the effect of PCA. 2. High, neurotoxic doses of PCA antagonized the effect of PCA when tested one week after the former administration. The serotonin uptake inhibitor zimeldine counteracted the effect of neurotoxic PCA. 3. Depletion of brain serotonin with p-chlorophenylalanine counteracted the effect of acute PCA. 4. Repeated treatment of rats for 7 days with zimeldine, amiflamine, alpha-ethyltryptamine or clorgyline plus a low dose of PCA counteracted the effect of acute PCA probably due to a functional down-regulation at postsynaptic receptors. Clorgyline or a low dose of PCA by themselves had no effect. 5. Compounds interacting with dopamine or noradrenaline mechanisms, e.g. alpha-methyltyrosine, N-2-chloroethyl-N-ethyl-2-bromobenzylamine (DSP 4), pimozide, remoxipride and prazosin did not antagonize the effect of PCA nor did (+)-amphetamine inhibit the cage-leaving response. None of the serotonin receptor antagonists (cinanserin, ketanserin, metergoline, methysergide, metitepine, mianserin, pirenperone) blocked the inhibition of the cage-leaving response produced by PCA, indicating that the receptors involved may not be of the S1- and S2- types. Observation of the cage-leaving response may be a valuable technique in studies of drugs that enhance the serotonin neurotransmission in the rat brain.

Analysis of the avoidance learning deficit induced by the serotonin releasing compound p-chloroamphetamine

The effects of the serotonin-releasing compound p-chloroamphetamine (PCA, 2.5 mg/kg) on avoidance acquisition, retention and memory retrieval were examined in male Sprague-Dawley rats using a one-way active avoidance and a one-trial passive avoidance task. The drug was injected IP prior to training, following acquisition and prior to the retention test 24 hr after training using a state-dependent design. In the normal context situation pretraining administration of PCA markedly impaired active avoidance acquisition, but PCA-treated rats did not differ from controls in their retention performance when tested 24 hr after training. In the dark/light box test pretraining administration of PCA caused a dose-dependent impairment of both active and passive avoidance retention which could not be explained in terms of changes in locomotor activity or behavioural disinhibition at the time of testing or state-dependent retention. Post-training administration of PCA failed to affect avoidance retention in both tasks. The drug was found to impair memory retrieval in a dose- and time-dependent fashion in the one-way active but not in the passive avoidance test. Pretraining administration of PCA produced a progressive loss of passive and active avoidance performance at increasingly longer retention intervals. The present results suggest that serotonin has dual effects on processes underlying learning and memory involving effects on both associative and non-associative learning processes in the rat. The time-dependent loss of memory retention following 5-HT release indicates that serotonin has a role in the way information is processed in the brain.

Serotonin receptor involvement in the avoidance learning deficit caused by p-chloroamphetamine-induced serotonin release

The receptor involvement in the p-chloramphetamine (PCA, 2.5 mg kg-1) induced impairment of active avoidance acquisition was examined in the male rat. The avoidance deficit was blocked at low doses by serotonergic (5-HT)-receptor blocking agents but not by alpha-adrenergic-, beta-adrenergic-, opiate-, muscarinic- or dopamine D2-receptor antagonists. The potency of the 5-HT antagonists to block the PCA-induced deficit correlated with their affinity in displacing [3H]ketanserin but not [3H]5-HT binding in the frontal cortex. The potencies of the 5-HT antagonists to block the action of PCA could not be related to their action on muscarinic-, histaminergic H1- or dopaminergic D2-receptor binding in vitro. It is concluded that the avoidance learning deficit caused by PCA-induced 5-HT release is related to activation of 5-HT receptors in the frontal cortex having the characteristics of a 5-HT2 receptor.

Ejaculations induced by p-chloroamphetamine in the rat

The ejaculatory response following acute injections of p-chloroamphetamine (PCA) and several other drugs was measured by weighing the compact seminal material accumulated over 2 hr. p-Chloroamphetamine caused a dose-dependent ejaculatory response that was inhibited by the inhibitor of the synthesis of 5-hydroxytryptamine (5-HT), p-chlorophenylalanine (PCPA), neurotoxic doses of PCA, reserpine, DSP 4 a selective noradrenergic neurotoxin given 48 hr before PCA, the inhibitor of synthesis of noradrenaline (NA) FLA 63, the specific inhibitors of uptake of 5-HT, alaproclate, fluoxetine and norzimeldine and the selective inhibitor of the uptake of NA, CPP 199, the E form of norzimeldine. The doses of several receptor antagonists producing a 50% decrease in the weight of seminal material were determined. The non-selective 5-HT receptor antagonists, metitepine and methergoline, the selective alpha 1-adrenoreceptor antagonists, prazosin and phenoxybenzamine and the non-selective alpha-adrenoreceptor antagonist, phentolamine, had strong effects, followed by the selective 5-HT2 antagonists, ketanserin and pirenperone. Yohimbine, an alpha 2-adrenoreceptor antagonist and atropine, a muscarinic receptor antagonist, only produced a partial blockade. The rank order of potency for some dopamine (DA) receptor antagonists was chlorpromazine, domperidone, haloperidol, pimozide. Remoxipride, a selective DA2 receptor antagonist and the selective DA1 antagonist, Sch 23390, had no effect. The following drugs had no effect: propranolol, naloxone, picrotoxin, cimetidine and mepyramine. The 5-HT receptor agonist 5-methoxy-N,N-dimethyltryptamine (5-MeODMT 3 mg/kg, i.p.) produced a small effect on the weight of seminal material, although 72% of the rats ejaculated. d-Amphetamine did not induce ejaculation at 5 mg/kg but had a marked effect at 15 mg/kg.(ABSTRACT TRUNCATED AT 250 WORDS)

Central serotonin neurones in avoidance learning: interactions with noradrenaline and dopamine neurones

The effects of p-chloroamphetamine (PCA), a serotonin (5-HT) releaser, on acquisition and retention were examined in male Sprague-Dawley rats using a one-way active avoidance task. PCA was found to impair avoidance acquisition and retention in a time dependent fashion which followed closely the temporal effects of the drug on 5-HT release in the brain. Thus, the avoidance deficit is related to the rate of change and not to the steady-state levels of 5-HT. The 5-HT releasing effect was most pronounced in the forebrain with less effect in the spinal cord. PCA caused time dependent, regional variations in catecholamine content, which was not related to avoidance performance. The avoidance and retention impairment induced by PCA was blocked by the 5-HT synthesis inhibitor p-chlorophenylalanine (PCPA) but not by depletion of catecholamines with alpha-methyl-p-tyrosine (H44/68) or by the noradrenergic-selective neurotoxin DSP4. Analysis of the time dependent effects of PCA on monoamine content in saline or PCPA-treated rats indicated that the temporal effects of PCA on avoidance performance is not due to a direct or indirect action on catecholamine neurones. The present experiments support the view that the ascending serotonergic pathways play a significant role in aversive learning in the rat.

Involvement of spinal serotonergic pathways in nociception but not in avoidance learning

The effects of selective lesions of the descending serotonergic (5-HT) pathways on analgesia and avoidance deficit induced by the 5-HT releasing compound p-chloroamphetamine (PCA, 2.5 mg/kg) were investigated in male rats. Intrathecal injection of 5,6-DHT (20 micrograms/rat) reduced the uptake of labelled 5-HT into spinal synaptosomes by approximately 85% but did not significantly affect the uptake of noradrenaline. The lesions produced a significant hyperalgesia and strongly attenuated the analgesic effect of PCA in the hot-plate test. In the flinch-jump test 5,6-DHT lesioned rats receiving PCA did not differ from the saline control group. Spinal lesioning did not, however, affect one-way active avoidance performance and did not prevent the marked impairment of avoidance performance induced by PCA. Thus, the avoidance deficit caused by PCA is independent of the descending serotonergic pathways and of the analgesia induced by PCA. These results support the view of a differential involvement of the ascending and descending serotonergic projections in behavioural processes controlled by aversive stimuli.

Test-dependent variations in the antinociceptive effect of p-chloroamphetamine-induced release of 5-hydroxytryptamine

p-Chloroamphetamine (PCA), in doses that did not significantly impair motor performance in a rotating-wheel task, induced marked analgesia in rats tested with the hot-plate and flinch-jump methods. In the tail-flick test, moderate hyper- or hypo-analgesia was found to be dependent on dose. In hot-plate experiments the analgesia was attenuated by inhibition of uptake of 5-HT (with zimelidine), depletion of stores of 5-HT (with PCPA) and by lesioning of 5-HT-containing terminals (long-term PCA treatment). Blockade of serotonin receptors by metergoline produced hyperalgesia, but failed to reduce the analgesia induced by p-chloroamphetamine. Manipulation of catecholaminergic and opioid systems did not reduce the effect of p-chloramphetamine. It is concluded that induction of release of 5-HT by chloroamphetamine induces antinociception which varies in magnitude between tests, suggesting that different serotonergic mechanisms modulate complex and reflex responses to noxious stimulation. The failure of metergoline to antagonize the analgesia induced by p-chloroamphetamine suggests an involvement of 5-HT receptors different from the ones implicated in other types of behaviour mediated by 5-HT.

Role of serotonin in memory: facilitation by alaproclate and zimeldine

The effects of alaproclate and zimeldine on memory retrieval were examined in male Swiss-Webster mice using a one-trial inhibitory avoidance task. All drugs were administered IP prior to the retention test 24 h after training. Both drugs were found to facilitate memory retrieval significantly in a dose- and time-dependent fashion that could not be explained in terms of non-specific effects of the drug (illness, lack of motility, etc.) at the time of the test. The temporal effects of alaproclate and zimeldine on memory closely followed their course of concentration of the drug within the blood stream. The facilitation of retrieval induced by alaproclate and zimeldine was blocked by the putative serotonergic receptor agonist quipazine but not blocked by the antagonist cyproheptadine. Pretreatment with quipazine alone in a group of animals trained to a shock level which normally results in high levels of suppression was not sufficient to produce memory impairment, suggesting that quipazine was probably antagonizing the facilitative effects of alaproclate and zimeldine directly, rather than overriding the facilitation through an indirect action on retrieval in general. The present results lend further support to the suggestion that serotonin plays a significant role in memory.

Retention deficits induced by acute p-chloroamphetamine following fear conditioning in the rat

Rats were given four inescapable shocks (1.0 mA) when confined to the right-hand corner of a modified shuttlebox. p-Chloroamphetamine (PCA) injected just before the retention test 24 h later completely blocked the immobile posture that was observed after saline injections. This retention deficit was shown to be selectively associated with 5-hydroxytryptamine (5-HT) release, since the administration of the 5-HT uptake inhibitors zimelidine and citalopram 60 min prior to PCA antagonized this effect. The 5-HT specificity of the deficit was further established by the findings that 5-HT-depleted rats (PCA, 2 x 10 mg/kg, and fenfluramine, 2 x 25 mg/kg), but not NA-depleted rats (DSP4, 1 x 50 mg/kg), or rats treated with zimelidine (2 x 20 mg/kg) 60 min before PCA (2 x 10 mg/kg), showed an almost complete blockade of the retention failure. The data presented may provide a useful experimental model for investigating the efficacy of functional 5-HT activity in the treatment of phobic anxiety.

Serotonin involvement in aversive conditioning: reversal of the fear retention deficit by long-term p-chloroamphetamine but not p-chlorophenylalanine

5-Methoxy-N,N-dimethyltryptamine (5-MeO-DMT), a serotonin (5-HT) agonist, fenfluramine and p-chloroamphetamine (PCA), which are 5-HT releasers, produce deficits in fear retention as indicated by a notable lack of the immobility resulting from inescapable shocks. Depletion of central 5-HT neurones after long-term PCA treatment (2 X 10 mg/kg) completely blocked the retention impairment resulting from acute PCA (2.5 mg/kg) and fenfluramine (5 mg/kg), and partially blocked the deficit produced by 5-MeO-DMT (4 mg/kg). 5-HT depletion after p-chlorophenylalanine (PCPA) treatment (200, 100, 100 mg/kg, 72, 48 and 24 h before) did not do so; this is in agreement with other findings which suggest the involvement of different 5-HT stores in the action of PCA and PCPA. These data further underline the importance of the ascending 5-HT pathway in aversive conditioning in the rat.

Forebrain serotonin and avoidance learning: behavioural and biochemical studies on the acute effect of p-chloroamphetamine on one-way active avoidance learning in the male rat

The acute effects of p-chloroamphetamine (PCA) on one-way active avoidance learning and on central monoamine concentrations were examined in the male rat. The 5-HT specificity of the acute behavioural effect of PCA was examined in several experiments. PCA (0.08-5 mg/kg IP) injected 30-60 min before testing produced a dose-related impairment of both avoidance acquisition and retention. Pretreatment with the selective serotonin (5-HT) uptake inhibitors fluoxetine and zimelidine, but not the noradrenaline (NA) uptake inhibitor desipramine, resulted in a blockade of the avoidance deficit. Degeneration of brain 5-HT neurons by a high neurotoxic dose of PCA (2 x 10 mg/kg IP) 7 days prior to the administration of PCA also blocked the avoidance deficit. There was also a complete blockade of the PCA-induced avoidance deficit by pretreatment with metergoline, a central 5-HT receptor blocking agent. A 2.,5 mg/kg dose of PCA examined 60 min after injection produced regional changes in the 5-HT-levels preferentially in the forebrain region with significant reductions in the cerebral cortex, hippocampus and striatum while marginal effects were observed in the hypothalamus, midbrain and spinal cord. PCA failed to reduce dopamine and noradrenaline concentrations in the time- and dose-range of the avoidance deficit. Thus, the avoidance learning impairment appears to be specifically related to the acute release of endogenous 5-HT from presynaptic nerve endings possibly in the forebrain resulting in stimulation of postsynaptic 5-HT receptors. These findings indicate that 5-HT neurons in the forebrain play a role in active avoidance learning possibly by an involvement in memorial and/or retrieval processes.

The acute effect of p-chloroamphetamine on the retention of fear conditioning in the rat: evidence for a role of serotonin in memory consolidation

The acute effect of the 5-hydroxytryptamine (5-HT) releasing compound p-chloroamphetamine (PCA) on the acquisition and retention of shock-elicited fear conditioning to the contextual cues of a normal two-compartment shuttlebox in the rat was studied in three experiments. PCA (5 mg/kg) did not impair the acquisition of fear conditioning (Experiment 1). PCA, administered either 30 or 60 min before fear conditioning, caused a total blockade of fear retention when tested 24 h after acquisition. This retrograde amnesic effect was blocked by the 5-HT uptake blocker zimelidine (10 mg/kg) when PCA was injected 60 min before shock. These findings indicate that 5-HT neurones, possibly in the forebrain, may exert an inhibitory influence upon the long-term aspects of information consolidation in memory.

Test-specific effects of the 5-HT reuptake inhibitors alaproclate and zimelidine on pain sensitivity and morphine analgesia

The effects of the specific 5-HT uptake inhibitors alaproclate and zimelidine, the 5-HT releasing compound p-chloroamphetamine (PCA) and the specific NA uptake inhibitor desipramine on pain sensitivity were examined in male rats using the hot-plate and tail-flick methods. The effects of alaproclate and zimelidine on 5-HT uptake mechanisms in the hypothalamus and spinal cord were also studied. Alaproclate, zimelidine, PCA and desipramine produced hypoalgesia in the hot-plate but not in the tail-flick test. Naloxone (1 mg/kg) failed to block the hypoalgesia produced by alaproclate and PCA in the hot-plate test. Zimelidine but not desipramine pretreatment blocked the analgetic action of PCA in the hot-plate test. Alaproclate significantly enhanced morphine analgesia in the hot-plate test but did not affect morphine analgesia in the tail-flick test. In contrast, zimelidine tended to enhance and significantly prolonged morphine analgesia in the tail-flick test but did not affect morphine analgesia in the hot-plate test. Zimelidine inhibited 5-HT uptake with equal potency in the hypothalamus and spinal cord, while alaproclate produced a greater inhibition of 5-HT uptake in the hypothalamus. These findings show test-specific effects after enhancement of central 5-HT neurotransmission. It is suggested that various aspects of pain sensitivity and morphine analgesia may involve different 5-HT pathways in the brain and spinal cord. Moreover, 5-HT pathways in the forebrain may mediate analgesia of a non-opiate type.

Para-halogenated phenethylamines: similar serotonergic effects in rats by different mechanisms

The serotonin (5-HT) behavioral syndrome in rats served as a model to test for possible acute serotonergic effects of para-halogenated phenethylamines. p-Chloro-, p-chloro-beta-methyl-, p-fluoro-, p-bromo-, and p-iodophenethylamine produced the same 5-HT behavioral syndrome as did p-chloroamphetamine, but unlike the latter did not deplete brain 5-HT 3 days after injection. Pretreatment of rats with the 5-HT depletor p-chlorophenylalanine (pCPA)( prevented the serotonergic effects of both chloro-derivatives, and partially prevented the effects of bromo- and iodophenethylamine. 5-hydroxytryptophan restored the behavioral responses to these compounds in pCPA-pretreated rats. pCPA treatment did not prevent the behavioral effects of p-fluorophenethylamine. Similarly, zimelidine, a 5-HT uptake inhibitor, prevented the serotonergic behavioral effects of all compounds tested except p-fluorophenethylamine. Taken as a group, para-halogenated phenethylamines are short-acting serotonergic compounds which, unlike p-chloroamphetamine, do not produce long-lasting depletion of brain 5-HT. p-Chlorophenethylamine and its beta-methyl analog apparently activate central 5-HT receptors indirectly, i.e., by 5-HT release; p-fluorophenethylamine is a direct 5-HT agonist. The p-bromo- and p-iodo-derivatives apparently possess both properties.

Zimelidine and clomipramine: different influence on fenfluramine but not p-chloroamphetamine-induced pharmacological effects

The influence of zimelidine and clomipramine on two p-chloroamphetamine (PCA)- or fenfluramine-induced pharmacological effects, regarded as resulting from the serotoninergic stimulation, was studied. In the flexor reflex test in the spinal rat zimelidine prevented potentiation induced by PCA but not that induced by fenfluramine. Clomipramine antagonized both the PCA- and fenfluramine-induced effects. Zimelidine counteracted also the hyperthermia in rats induced by PCA at a high ambient temperature but not that induced by fenfluramine. These findings seem to indicate that: (1) zimelidine and clomipramine differ in their mechanism of presynaptic action on serotonin neurons; (2) fenfluramine and PCA are taken up to serotonin neurons via different mechanisms.