The effect of d-lysergic acid diethylamide (LSD), 2,5-dimethoxy-4-methylamphetamine (DOM), pentobarbital and methaqualone on punished responding in control and 5,7-dihydroxytryptamine-treated rats

The neurobiology and control of anxious states

Fear is an adaptive component of the acute "stress" response to potentially-dangerous (external and internal) stimuli which threaten to perturb homeostasis. However, when disproportional in intensity, chronic and/or irreversible, or not associated with any genuine risk, it may be symptomatic of a debilitating anxious state: for example, social phobia, panic attacks or generalized anxiety disorder. In view of the importance of guaranteeing an appropriate emotional response to aversive events, it is not surprising that a diversity of mechanisms are involved in the induction and inhibition of anxious states. Apart from conventional neurotransmitters, such as monoamines, gamma-amino-butyric acid (GABA) and glutamate, many other modulators have been implicated, including: adenosine, cannabinoids, numerous neuropeptides, hormones, neurotrophins, cytokines and several cellular mediators. Accordingly, though benzodiazepines (which reinforce transmission at GABA(A) receptors), serotonin (5-HT)(1A) receptor agonists and 5-HT reuptake inhibitors are currently the principle drugs employed in the management of anxiety disorders, there is considerable scope for the development of alternative therapies. In addition to cellular, anatomical and neurochemical strategies, behavioral models are indispensable for the characterization of anxious states and their modulation. Amongst diverse paradigms, conflict procedures--in which subjects experience opposing impulses of desire and fear--are of especial conceptual and therapeutic pertinence. For example, in the Vogel Conflict Test (VCT), the ability of drugs to release punishment-suppressed drinking behavior is evaluated. In reviewing the neurobiology of anxious states, the present article focuses in particular upon: the multifarious and complex roles of individual modulators, often as a function of the specific receptor type and neuronal substrate involved in their actions; novel targets for the management of anxiety disorders; the influence of neurotransmitters and other agents upon performance in the VCT; data acquired from complementary pharmacological and genetic strategies and, finally, several open questions likely to orientate future experimental- and clinical-research. In view of the recent proliferation of mechanisms implicated in the pathogenesis, modulation and, potentially, treatment of anxiety disorders, this is an opportune moment to survey their functional and pathophysiological significance, and to assess their influence upon performance in the VCT and other models of potential anxiolytic properties.

The Vogel conflict test: procedural aspects, gamma-aminobutyric acid, glutamate and monoamines

A multitude of mechanisms are involved in the control of emotion and in the response to stress. These incorporate mediators/targets as diverse as gamma-aminobutyric acid (GABA), excitatory amino acids, monoamines, hormones, neurotrophins and various neuropeptides. Behavioural models are indispensable for characterization of the neuronal substrates underlying their implication in the etiology of anxiety, and of their potential therapeutic pertinence to its management. Of considerable significance in this regard are conflict paradigms in which the influence of drugs upon conditioned (trained) behaviours is examined. For example, the Vogel conflict test, which was introduced some 30 years ago, measures the ability of drugs to release the drinking behaviour of water-deprived rats exposed to a mild aversive stimulus ("punishment"). This model, of which numerous procedural variants are discussed herein, has been widely used in the evaluation of potential anxiolytic agents. In particular, it has been exploited in the characterization of drugs interacting with GABAergic, glutamatergic and monoaminergic networks, the actions of which in the Vogel conflict test are summarized in this article. More recently, the effects of drugs acting at neuropeptide receptors have been examined with this model. It is concluded that the Vogel conflict test is of considerable utility for rapid exploration of the actions of anxiolytic (and anxiogenic) drugs. Indeed, in view of its clinical relevance, broader exploitation of the Vogel conflict test in the identification of novel classes of anxiolytic agents, and in the determination of their mechanisms of action, would prove instructive.

5-Hydroxytryptamine pathways in anxiety and its treatment

The effects of manipulating 5-hydroxytryptamine (5HT) neuronal function in humans and in animals are reviewed. 5HT pathways do not have a unitary function in modulating anxiety. It is proposed that, rather than acting as input or output channels for brain aversive systems, these pathways provide information concerning waking/motor status, which is crucial to the organisation of appropriate responses to threat. Each terminal region can make use of this information in different ways. Globally, the influence of 5HT neurones on higher centres appears predominantly to facilitate information processing relevant to threat, while their major influence on brainstem centres may be a restraining one.

Conflict behavior in Maudsley reactive and nonreactive rats: effects of noradrenergic neuronal destruction

The present studies were designed to examine the effects of treatment with the noradrenergic neurotoxin N-(2-chloroethyl)-n-ethyl-2-bromobenzylamine HCl (DSP4; 65 mg/kg, IP) on conflict behavior in the Maudsley reactive (MR) and nonreactive (MNRA) rat strains. In daily 10-min sessions, water-restricted rats were trained to drink water from a tube that was occasionally electrified; electrification was signaled by the presence of a tone (7-s duration; ISI = 30 s). Consistent with previous reports, the number of shocks accepted by rats of the MR and MNRA strains did not differ initially, but MNRA rats exhibited a dramatic increase in punished responding relative to their MR counterparts over the course of several weeks of conflict testing. This MR vs. MNRA strain difference in punished responding did not exhibit extinction following discontinuation of CSD conflict behavior testing for a period of 6 weeks. Whether it was administered after conflict training or before, DSP4 treatment did not reduce the MR vs. MNRA strain difference in conflict behavior; rather, DSP4 treatment tended to increase the magnitude of the MR vs. MNRA difference in conflict behavior. The effects of DSP4 on norepinephrine (NE) and 5-hydroxytrypamine (5-HT) concentrations in the pons medulla region were determined in one group of conflict-experienced MR and MNRA rats (35 weeks after administration) and in a second group of naive MR and MNRA rats (3 weeks after administration). There were no MR vs. MNRA strain differences in NE or 5-HT concentrations in vehicle-treated rats. DSP4 treatment significantly reduced NE, but not 5-HT, concentrations when compared to control values; rats that were sacrificed 3 weeks following DSP4 administration exhibited a greater NE depletion than did rats sacrificed 35 weeks after DSP4 administration. Finally, there were no significant correlations between pons medulla region NE concentrations and conflict behavior in either strain alone or when the data from the two strains were combined. The present results are not consistent with the hypothesis that the MR vs. MNRA strain difference in conflict behavior is the result of strain differences in brain NE function.

Anxiolytic effects of serotonergic interventions in the shock-probe burying test and the elevated plus-maze test

Although serotonergic neural systems have been implicated in the control of anxiety for a number of years, evidence in favour of this role is controversial. The present experiments were designed to further characterize the putative role of serotonin (5-HT) in anxiety, using two pharmacologically validated animal models: the elevated plus-maze and the shock-probe burying tests. If the integrity of 5-HT neural systems is necessary for the expression of 'anxious' behaviors, then disruption of 5-HT systems should produce effects in the plus-maze and shock-probe tests that are similar to those of anxiolytic drugs. In the present experiments, serotonergic function was disrupted in rats, either by chemical depletion using the synthesis inhibitor p-CPA, by inhibitory autoreceptor activation using the selective 5-HT1A receptor ligand 8-OH-DPAT, or by electrolytic lesions of the serotonin-containing, dorsal raphe nucleus. p-CPA and dorsal raphe lesions produced robust anxiolytic effects in the elevated plus-maze and the shock-probe burying tests, whereas 8-OH-DPAT produced anxiolytic effects only in the shock-probe burying test, and 'anxiogenic' effects in the elevated plus-maze test. Although these results generally support the view that serotonin plays a role in the expression of 'anxious' behavior, the opposite effects of 8-OH-DPAT in the two behavioral paradigms suggest that the 5-HT1A receptor subtype exerts differential control over different types of experimental anxiety.

Serotonergic Mechanisms in Anxiolytic Effect of Tandospirone in the Vogel Conflict Test

To clarify which 5-HT1A receptors, autoreceptors located in the raphe nuclei or post-synaptic receptors in the forebrain areas receiving a 5-HT input, mediate the anticonflict action of tandospirone (a 5-HT1A receptor-related anxiolytics), the behavioral effects of tandospirone were studied in 5,7-dihydroxytryptamine (5,7-DHT) treated rats. By measuring both monoamines and their metabolite levels and densities of [3H]8-OH-DPAT binding in 5,7-DHT-treated rat brain, we confirmed that pretreatment with 5,7-DHT destroyed 5-HT neurons selectively without affecting postsynaptic 5-HT1A receptors located on the postsynaptic neurons. This selective destruction produced no significant changes in the drinking behavior of rats in either punished or unpunished sessions of the Vogel conflict test. Furthermore, this destruction altered neither the effect of tandospirone on punished responding in this procedure nor the potency of tandospirone to induce a flat body posture in rats, which is known as the "serotonin behavioral syndrome". These results suggested that the anticonflict action of tandospirone may be produced, at least in part, by binding to postsynaptic 5-HT1A receptors and activating them as agonists, and not to 5-HT1A autoreceptors located on the cell bodies of 5-HT neurons.

Cyproheptadine prevents the initial occurrence of successive negative contrast

Rats shifted from 32% to 4% sucrose make fewer licks for 4% sucrose than rats having only experienced the lower reward. In Experiment 1, the occurrence of this contrast effect was prevented by the administration of the nonspecific serotonin antagonist cyproheptadine (3.0 or 6.0 mg/kg). These results of Experiments 2 and 3 demonstrated that the contrast-reducing action of cyproheptadine was not mediated by the antiserotonergic properties of the drug since systemic administration of the serotonin synthesis inhibitor, PCPA (150 or 300 mg/kg), failed to influence either the occurrence of contrast or the attenuation of contrast by cyproheptadine. The results of Experiment 4 indicated that the contrast-reducing action of cyproheptadine was not mediated by the antihistaminergic properties of the drug since the antihistamine, pyrilamine (6 or 12 mg/kg), also failed to prevent the occurrence of contrast. Finally, the contrast-reducing action of cyproheptadine was not due to rate-dependent and/or appetite stimulating effects since cyproheptadine did not serve to increase lick frequency in rate-dependent controls.

MR/Har and MNRA/Har Maudsley rat strains: differential response to chlordiazepoxide in a conflict task

The Maudsley Reactive (MR/Har) and Non-Reactive (MNRA/Har) rat strains, selectively bred for differences in open field defecation, have also been shown to differ in their baseline behavior in the Conditioned Suppression of Drinking (CSD) procedure, a second "model" behavior for the study of anxiety and/or emotionality in rats. The present studies were designed to compare the responsiveness of these two strains to the typical antianxiety agent chlordiazepoxide in the CSD paradigm. In daily 10-minute sessions, water-deprived rats were trained to drink from a tube that was occasionally electrified (0.5 mA), electrification being signaled by a tone. Consistent with previous reports, after several weeks of CSD testing, MNRA/Har rats accepted significantly more shocks than did MR/Har rats during control (nondrug) sessions. In both strains, the number of shocks accepted was inversely related to the intensity of the shock used (0.25-1.0 mA), with MNRA/Har rats accepting significantly more shocks than MR/Har rats at all intensities examined. The effects of various doses (1.25-28.4 mg/kg, IP) of chlordiazepoxide were determined in subjects of the MNRA/Har strain at the original training intensity (0.5 mA), while a lower intensity (0.25 mA) was utilized in MR/Har rats. Although punished responding in control (i.e., nondrug) CSD sessions did not differ under these conditions, MNRA/Har rats were found to be more responsive to the anticonflict effects of chlordiazepoxide than rats of the MR/Har strain. This strain difference in anticonflict efficacy of chlordiazepoxide was quite dramatic, with MNRA/Har rats accepting twice as many shocks as MR/Har rats following maximally effective doses of chlordiazepoxide. Low doses of chlordiazepoxide increased water intake slightly, while higher doses decreased water intake. Surprisingly, the chlordiazepoxide-induced depression of water intake was greater in rats of the MR/Har strain. Thus, these Maudsley Reactive and Non-Reactive rat strains, bred originally for their differences in open field behavior, also differ markedly in their responsiveness to chlordiazepoxide in the CSD paradigm. These findings further support the hypothesis that the MR/Har and MNRAHar rat strains may represent a genetically-based "animal model" for the study of emotionality and/or anxiety.

The effects of beta-antagonists and anxiolytics on conflict behavior in the rat

The present studies were designed to evaluate the effects of beta-adrenoceptor antagonists and traditional anxiolytics (phenobarbital and diazepam), alone and in combination, on behavior in the Conditioned Suppressioned of Drinking (CSD) conflict paradigm, an "animal model" for the study of anxiety and antianxiety agents. In daily 10-minute sessions, water-deprived rats were trained to drink from a tube which was occasionally electrified (0.5 mA), electrification being signalled by the presence of a tone. Within 2-3 weeks, control responding had stabilized (10-15 shocks/session and 10-15 ml water/session); drug tests were then conducted at weekly intervals. As expected, diazepam (0.6-10 mg/kg) and phenobarbital (10-40 mg/kg) administration resulted in a marked and dose-dependent increase in punished responding at doses which did not markedly alter background responding (water intake). Neither propranolol (0.5-8 mg/kg) nor the beta-1-selective antagonist atenolol (1-16 mg/kg) significantly affected punished responding in the CSD. Both propranolol and atenolol produced significant beta-1-adrenoceptor blockade, as evidenced by the production of significant bradycardic effects in conscious rats at the doses employed. Pretreatment with 2.0 mg/kg propranolol did not alter the anticonflict effects of diazepam (0.6-10 mg/kg) or phenobarbital (10-40 mg/kg). Further, reduction of the shock intensity to 0.125 mA (i.e., decreased suppression) failed to alter the behavioral response to propranolol (1.5-5 mg/kg) or the interaction of 2.0 mg/kg propranolol with diazepam. Finally, chronic administration of propranolol (2.0 mg/kg, twice daily) did not affect punished responding over the course of 5 weeks of treatment. These data suggest that the CSD paradigm, although an effective "animal model" for the study of benzodiazepine and barbiturate anticonflict effects, cannot serve as an "animal model" for the study of the situation-specific (i.e., phobic) anxiety for which propranolol and related agents are presently used.

Blockade of potentiated startle responding in rats by 5-hydroxytryptamine1A receptor ligands

A potentiated whole-body startle response was produced in rats by pairing intense acoustic stimuli with a light (conditional stimulus) formerly presented contiguously with electric shock. Administration of the selective serotonin (5-HT) 1A ligand 8-hydroxy-2-(di-n-propylamino)tetralin (0.125 and 0.5 mg/kg) blocked the potentiation of startle induced by the conditional stimulus. 1-[3-Chlorophenyl]piperazine, a non-anxiolytic 5-HT1B/1C agonist, did not block potentiated startle, even at a dose (1.0 mg/kg) that induced significant overall decreases in startle amplitude. The non-benzodiazepine anxiolytics buspirone (1.25-5.0 mg/kg), gepirone (3.0-10.0 mg/kg) and the related 5-HT1A ligand ipsapirone (1.0-10.0 mg/kg) blocked potentiated startle, though effects of the non-selective serotonin antagonist methysergide (0.3-10.0 mg/kg) fell short of significance. These data support a role for the 5-HT1A binding site in the anti-anxiety effects of buspirone and related compounds over a range of behavioral procedures.

Convulsant versus typical barbiturates: effects on conflict behavior in the rat

Typical barbiturates produce a spectrum of behavioral effects, including anti-convulsant, muscle relaxant, sedative hypnotic and anti-anxiety actions. In contrast to these typical barbiturates, there exists a group of barbiturates which are pro-, rather than anti-convulsant. The effects of these convulsant barbiturates on anxiety-related behaviors have not been examined. Therefore, the present studies were designed to compare the effects of the convulsant barbiturate CHEB to those of a number of typical barbiturates in the Conditioned Suppression of Drinking (CSD) paradigm, an "animal model" for the study of anxiety and anti-anxiety agents. In daily 10-minute sessions, water-deprived rats were trained to drink from a tube which was occasionally electrified (0.5 mA), electrification being signalled by a tone. Within 3-4 weeks control responding had stabilized (10-15 shocks and 10-15 ml water/session); drug tests were then conducted at weekly intervals. Consistent with previous reports, typical barbiturates (pentobarbital, secobarbital, phenobarbital) produced dose-dependent increases in the number of shocks received at doses which did not depress background responding (water intake). In contrast, sub-convulsant doses of CHEB (0.3-2.5 mg/kg) produced a dose-dependent depression of both punished responding and background responding. Finally, it was found that pre-treatment with 1.25 mg/kg CHEB did not alter the anti-conflict effects of pentobarbital. These results suggest that (1) convulsant and typical barbiturates have markedly different effects on conflict behavior in the rat and (2) CHEB appears not to possess any "barbiturate antagonist" qualities.

Buspirone effects in an animal conflict procedure: comparison with diazepam and phenobarbital

Buspirone has been introduced as a novel non-benzodiazepine anti-anxiety agent. The Conditioned Suppression of Drinking (CSD) paradigm is an "animal model" for anxiety which provides information on both the relative potency and relative efficacy of anti-conflict agents. The present study compared the anti-conflict effects of buspirone to those of more "classical" anti-anxiety agents, diazepam and phenobarbital. In daily 10-minute sessions, water-deprived rats were trained to drink from a tube which was occasionally electrified (0.5 mA), electrification being signalled by a tone. Within 2-3 weeks control CSD responding had stabilized (approximately 15-20 shocks/session and 10-15 ml water/session); drug tests were conducted at weekly intervals. Diazepam and phenobarbital markedly (400-500%) increased the number of shocks received at doses which did not depress background responding (i.e., water intake). A number of agents, most notably morphine and ethanol, did not reliably affect punished responding in the CSD. Administered IP, low doses (0.25-1 mg/kg) of buspirone increased punished responding only slightly (less than 100% increase); higher doses (2, 4 mg/kg) depressed background responding. Administered SC, buspirone (0.125-1.0 mg/kg) had more potent effects on both punished and unpunished responding; again, anti-conflict efficacy was only marginal. These results suggest that buspirone might be less effective than the benzodiazepines in the management of anxiety.

Recent developments in 5HT-related pharmacology of animal models of anxiety

The proposed anxioselective drug, buspirone, interacts with 5HT1 receptors. An analogue, MJ 13805, produces a 5HT behavioural syndrome blocked by central 5HT pathway lesion. Both compounds inhibit 5HT neurone firing. An association of any such action with models of anxiety is not yet possible. Several compounds selective for 5HT receptor sub-types have been tested in models of anxiety. Ritanserin, a selective 5HT2 antagonist, shows activity in an emergence test but not conflict models. Preliminary clinical reports indicate qualitatively different anxiolytic activity from that of benzodiazepines. TVXQ 7821 is selective for 5HT1 receptors and has shown activity in several models of anxiety. 8OHDPAT and RU 24969 are 5HT1 agonists, selective for 5HT1A and 5HT1B sites respectively. 8OHDPAT released punished drinking but reversed a similar effect of PCPA. Its mode of action remains unclear. RU 24969 has shown no marked anxiolytic-like activity in food or water-motivated conflicts. Further studies are necessary before associating modulation of central 5HT systems with anxiolytic activity, either in animal models or patients.

5-HT and anxiety: promises and pitfalls

This review examines the evidence implicating serotonergic systems in the control of anxiety. The effects of manipulations of 5-HT, including 5-HT lesions, pharmacological reduction and enhancement of serotonergic function in animal tests of anxiety are reviewed. Biochemical and behavioral evidence implicating serotonergic pathways in the anxiolytic action of benzodiazepines is presented. In each section the promises of a serotonergic involvement, as indicated by positive findings, and the pitfalls, evidenced by inconsistent and conflicting results, are discussed. Finally the dangers of a superficial interpretation of the behavioral and biochemical findings is stressed.

Are serotonergic neurons involved in the control of anxiety and in the anxiolytic activity of benzodiazepines?

Several studies have shown that, like benzodiazepines (BZP), treatments able to reduce or block the activity of CNS serotonergic (5-HT) neurons released punished behavior. Therefore, 5-HT mechanisms have been tentatively implicated in the anti-punishment (anxiolytic?) activity of BZP. Numerous data, however, are not in keeping with this hypothesis. Since not responding enables the animals to avoid punishment but also delays the receipt of food-reward, one of these factors could be an alteration of waiting capacities. Indeed, we have shown that diazepam released behavioral suppression in conflict schedules only when the duration of the punished periods exceeded 1 minute. Moreover, in rats allowed to choose in a T-maze between immediate-but-small vs. delayed-but-large reward, BZP significantly decreased the frequency with which the delayed reward was chosen, with 5-HT uptake blockers producing opposite effects. Therefore, one can hypothesize that BZP render the animals less prone than controls to tolerate delay of reward and that 5-HT mechanisms may be involved in this phenomenon. An altered tolerance to delay of reward should be taken into account when interpreting the BZP-induced release of behavioral inhibition in classical conflict procedures.

Maudsley reactive and non-reactive rat strains: differential performance in a conflict task

Based on open field defecation rates, distinct strains of Maudsley rats have been identified. The Maudsley Reactive strains exhibit high levels of open field defecation, while the Maudsley Non-Reactive strains exhibit low levels of open field defecation. This difference in open field behavior has been interpreted as an indicator of "emotionality." The present studies were designed to compare the performance of two Maudsley strains, MR/Har and MNRA/Har, in the conditioned suppression of drinking (CSD) paradigm, another "model" behavior for the study of emotionality and/or anxiety in animals. In daily 10-minute sessions, water-deprived rats were trained to drink from a tube that was occasionally electrified (0.5 mA), electrification being signalled by a tone. In this procedure, MNRA/Har rats accepted significantly more shocks than did MR/Har rats. This difference between MR/Har and MNRA/Har strains, initially somewhat slight, became quite impressive over weeks of repeated testing. Although present in both male and female subjects, this strain difference was more prominent in the males. Thus, the MR/Har and MNRA/Har rat strains, bred originally for their differences in open field behavior, also differ markedly in their performance in the CSD paradigm. These findings suggest that the Maudsley rat strains may represent a useful genetically-based "animal model" for the study of emotionality and/or anxiety.

Brain cholinergic, behavioral, and morphological development in rats exposed in utero to methylparathion

The purpose of this study was to determine the effects of subchronic administration of the organophosphate methylparathion (MPTH) during gestation on behavior and development of brain cholinergic neurons in the offspring. Pregnant rats received daily po doses of MPTH from Day 6 through Day 20 of gestation at doses causing no (1.0 mg/kg) or minimal (1.5 mg/kg) visible signs of maternal toxicity. Acetylcholinesterase (AChE) and choline acetyltransferase (CAT) activities, and [3H]quinuclidinyl benzilate (QNB) binding to muscarinic receptors, were determined in several brain regions at 1, 7, 14, 21, and 28 days postnatal age and in maternal brain at Day 19 of gestation. Prenatal exposure to 1.5 mg MPTH/kg reduced AChE and increased CAT activity in all brain regions at each developmental period and in maternal brain. Similar exposure to 1.0 mg MPTH/kg caused a significant but smaller and less persistent reduction in AChE activity but no change in brain CAT activity of the offspring. Both doses of MPTH decreased the Bmax of 3H-QNB binding in maternal frontal cortex but did not alter the postnatal pattern of 3H-QNB binding. In parallel studies, prenatal exposure to MPTH did not affect a variety of behaviors. However, cage emergence, accommodated locomotor activity, and operant behavior in a mixed paradigm were impaired in rats exposed to 1.0 but not to 1.5 mg/kg MPTH. No morphological changes were observed in hippocampal or cerebellar tissue. Thus, subchronic prenatal exposure to MPTH altered postnatal development of cholinergic neurons and caused subtle alterations in selected behaviors of the offspring.

Behavioral and biochemical evidence for serotonergic actions of tetrahydro-beta-carbolines

In two groups of rats trained to discriminate LSD (0.1 mg/kg or 0.24 mg/kg) from saline, tetrahydro-beta-carboline (THBC; 1-12 mg/kg as free base) and its derivative 6-methoxy-THBC (1-12 mg/kg as free base) substituted partially for LSD. The substitution of THBC for 0.1 mg/kg of LSD was analyzed further with antagonism tests in 16 animals and was attenuated by the serotonin (5-HT) antagonist BC-105 (pizotifen; 3 mg/kg) but not by the dopamine (DA) antagonist haloperidol (0.1 mg/kg). It was abolished by pretreatment with the 5-HT synthesis inhibitor p-chlorophenylalanine (100 mg/kg/day for 3 days). In addition, THBC was found to inhibit 3H-LSD binding to homogenates of rat frontal cortex with an IC50 value of 4 microM which is similar to that previously reported for other 5-HT agonists. These data indicate that THBCs exert potent 5-HT agonist actions. Since THBCs have recently been found in mammalian brain and other tissues, the present results are of interest in relation to a possible role of these substances in endogenous psychosis.

Lack of central 5-hydroxytryptamine influence on the anticonflict activity of diazepam

This study examined the effects of various drug treatments (IP injections) proposed to modify central 5-hydroxytryptamine (5-HT) activity on a conditioned suppression of drinking behavior in water-deprived rats. The subjects were trained to drink their daily water requirement during a 10-min session. Intermittent tone periods of 7 s were then introduced, the last 5 s of which the drinking tube was electrified. The animals gradually suppressed tube contacts during the tone to a low constant level within 2 weeks of training. Diazepam increased punished responding dramatically. The 5-HT antagonists methysergide (1 - 18 mg/kg), cyproheptadine (1 - 18 mg/kg), metergoline (0.25 - 2.0 mg/kg) and cinanserin (10 - mg/kg) failed to induce large, reliable increases in punished responding. When a low dose of diazepam was combined with 5-HT antagonists, only one treatment, methysergide at 3 mg/kg, potentiated the anticonflict activity of diazepam. Acute or chronic treatment with PCPA increased behavior suppressed by punishment, but this effect was weak, brief, and poorly related to the depletion of brain 5-HT. LSD (0.3 - 100 microgram/kg) administered 1, 10, or 30 min before the test was ineffective in overcoming suppression by punishment. Mescaline (6 - 30 mg/kg) had no significant effect on punished responding. 5-HTP (18 mg/kg) decreased the number of shocks accepted, but not after pretreating with carbidopa. Pretreatment with carbidopa plus 5-HTP potentiated the anticonflict effect of diazepam. The 5-HT agonist mCPP (0.25 - 2.0 mg/kg) enhanced suppression due to punishment, but only in doses that interfered with unpunished responding. The 5-HT-releasing agent fenfluramine (0.25 - 1.0 mg/kg) did not affect this behavior. Amitriptyline pretreatment in a dose not affecting unpunished behavior (5.6 mg/kg) potentiated the diazepam-induced increase in punished responding. These results are difficult to reconcile with the proposal that suppression of behavior consequent to punishment is related to brain 5-HT activity.