Low but not high doses of buspirone reduce the anxiogenic effects of diazepam withdrawal

After 21 days of treatment with diazepam (2 mg/kg/day IP) rats were tested 24 h after the last injection in the social interaction and elevated plus-maze tests of anxiety. Compared with control-treated rats, they showed significant decreases in social interaction, in the % numbers of entries onto open arms of the plus-maze and in the % of time spent on the open arms, indicating an anxiogenic response on withdrawal from diazepam. Buspirone (200 µg/kg SC) significantly increased social interaction in diazepam withdrawn rats and in the plus-maze also this dose significantly reversed the anxiogenic effects of diazepam withdrawal. Buspirone (400 µg/kg SC) was without effect in the plus-maze, but buspirone (800 µg/kg SC) significantly decreased the % of time spent on open arms in control-treated rats, indicating an anxiogenic effect. In the social interaction test buspirone (800 µg/kg SC) was without significant effect. The contrasting effects of the 200 and 800 µg/kg doses are discussed in terms of the pre- and post-synaptic actions of buspirone. The findings are consistent with earlier proposals that the increased anxiety during benzodiazepine withdrawal is at least partly caused by an increased release of hippocampal 5-HT.     

Increased 5-HT release mediates the anxiogenic response during benzodiazepine withdrawal: a review of supporting neurochemical and behavioural evidence

This paper reviews the biochemical and behavioural evidence that the increased anxiety that occurs during benzodiazepine withdrawal is caused by increased 5-HT activity. In hippocampal slices taken from rats withdrawn for 24 h from chronic diazepam treatment (2 mg/kg/day for 21 days) there was a significant increase in K⁺-evoked release of [³H]5-HT and in⁴⁵Ca²⁺ uptake and both of these changes were reversed by the GABA_B agonist, baclofen. Baclofen also reversed the anxiogenic response that is detected on withdrawal from chronic diazepam treatment. Other drugs that reduce 5-HT function (tianeptine which increases 5-HT uptake; buspirone, a 5-HT_1A receptor agonist/partial agonist; zacopride, a 5-HT₃ receptor antagonist) also reversed this anxiogenic response. Finally, we present data from a group of rats that did not develop tolerance to the anxiolytic effects of diazepam (2 mg/kg), even after 5 weeks treatment. This group failed to show an anxiogenic response on withdrawal from diazepam, nor was there an increase in hippocampal 5-HT release. We discuss the extent to which increased hippocampal 5-HT release can be causally linked to the increased anxiety during benzodiazepine withdrawal.     

Anxiolytic-like effects of 5-HT(1A) agonists in drug-naive and in benzodiazepine-experienced rats

The efficacy of two 5-HT(1A) receptor agonists in drug-naive and benzodiazepine-experienced rats was compared in two animal tests of anxiety, the social interaction and elevated plus-maze tests. Benzodiazepine-experienced rats were tested 48h after the last of 28 daily injections of diazepam (2mg/kg/day), a time at which there was no anxiogenic withdrawal response. S20499 (0.04, 0.2 and 1mg/kg) ((+)-8(4-[N-(5-methoxychroman-3yl)N-propylamine]butyl)-8-azaspirol[4,5]decane-7,9-dione) and buspirone (0.2mg/kg) significantly increased social investigation, indicating anxiolytic-like actions, and there was no significant modification in these effects as a result of the previous diazepam treatment. Both drugs also significantly reduced aggressive behaviours in both drug-naive and diazepam-experienced rats. An anxiolytic-like action in the elevated plus-maze was also indicated for S 20499 (0.04 and 0.2mg/kg) by an increase in the percentage of time spent on the open arms; this effect was not significantly changed by prior diazepam experience. Buspirone (0.2mg/kg) had no significant effects on the plus-maze. The results provide no evidence for reduced efficacy of 5-HT(1A) receptor agonists in animal tests of anxiety as a result of prior diazepam treatment.     

Chronic diazepam treatment: Effect of dose on development of tolerance and incidence of withdrawal in an animal test of anxiety

The effect of the treatment dose of diazepam was assessed on the rate of development of tolerance to diazepam's effects on rat behaviour in the elevated plus-maze test of anxiety. Tolerance developed more rapidly when treatment was with the higher dose: after 14 days when rats were given 2·5 mg/kg/day, but not until 21 days when they were given 1 mg/kg/day. When rats were tested undrugged 36 h after the last of 14 daily doses of diazepam (2·5 mg/kg) they showed behavioural changes indicating increased anxiety. Rats tested at this time with a lower dose of diazepam (1 mg/kg) also showed changes indicating increased anxiety compared with the control scores. This indicates that a more gradual tapering of doses would be necessary to avoid withdrawal responses.     

Evidence that the median raphé nucleus – dorsal hippocampal pathway mediates diazepam withdrawal-induced anxiety

On the basis of our previous series of experiments we had postulated that the increased anxiety that occurred during diazepam withdrawal was mediated by increased 5-HT release in the hippocampus. The present series of experiments provide evidence for a major role of the median raphé nucleus (MRN) dorsal hippocampal pathway. Rats were treated once daily for 21 days with diazepam (2 mg/kg IP) and then tested after 24 h withdrawal in the social interaction test of anxiety. Relative to chronically vehicle treated animals, those withdrawn from diazepam were significantly more anxious and had significantly greater K⁺-evoked release of [³H]-5-hydroxytryptamine (5-HT) from slices of dorsal and of ventral regions of the hippocampus. Estimation of extracellular concentrations of 5-HT within the dorsal hippocampus, using in-vivo microdialysis, showed doubling in the levels of 5-HT in the rats withdrawn from chronic diazepam treatment. This just failed to reach significance, but 33% of the rats showed dramatic increases (650%). It was not possible to test these animals in the social interaction test, but it is proposed that only the diazepam-withdrawn rats with raised extracellular levels of 5-HT would have displayed increased anxiety. 5-HT_1A receptor agonists injected into the MRN decrease the MRN firing rate, and hence the release of 5-HT in the dorsal hippocampus. As a further test of our hypothesis, we examined the effects of MRN injection of the 5-HT_1A receptor agonist, 8-OH DPAT, on animals withdrawn from diazepam and tested in the low light familiar condition of the social interaction test. 8-OH DPAT (50–200 ng) dose-dependently reversed the anxiogenic effect of diazepam withdrawal, while having no effects in chronic vehicle-treated animals. These results provide clear evidence that the MRN-dorsal hippocampal 5-HT pathway is at least one of the pathways playing an important role in mediating diazepam withdrawal-induced anxiety. Received: 10 May 1996 / Final version: 15 October 1996     

The use of flumazenil in prevention of Diazepam dependence in the rat

Rats were treated with diazepam (4 mg/kg per day) or its vehicle for 21 days. Twenty-four hours after their last dose they were tested in the social interaction test of anxiety. Diazepam withdrawal significantly reduced the time spent in social interaction compared with controls, indicating an anxiogenic withdrawal response. The administration of a single dose of flumazenil (4 mg/kg) during chronic diazepam treatment, 7 or 14 days before testing, prevented the development of this withdrawal response. The clinical implications of these findings are discussed.     

Enhancement of acetylcholine release by flumazenil in the hippocampus of rats chronically treated with diazepam but not with imidazenil or abecarnil

The effects of long-term treatment (three times a day for 3 weeks) with pharmacologically active doses of the novel anxiolytics and anticovulsants abecarnil (0.5 mg/kg, IP) and imidazenil (0.5 mg/kg, IP) on basal hippocampal acetylcholine release in freely moving rats were compared with those of diazepam (3 mg/kg, IP). Challenge doses of diazepam, abecarnil, and imidazenil decreased the extracellular acetylcholine concentration in the hippocampus by the same extent in animals chronically treated with the respective drug or vehicle. Moreover, the abrupt discontinuation of long-term treatment with diazepam, abecarnil, or imidazenil failed to affect hippocampal acetylcholine release during the first 5 days of withdrawal. In contrast, the acute administration of the benzodiazepine receptor antagonist flumazenil (1 mg/kg, IP) 2 days after diazepam withdrawal elicited a marked increase (65%) in acetylcholine release in the hippocampus. Flumazenil failed to induce the same effect 5 days after diazepam withdrawal or 2 or 5 days after discontinuation of long-term treatment with abecarnil or imidazenil. These results indicate that (i) the inhibitory effects of full (diazepam), partial (imidazenil), and selective (abecarnil) benzodiazepine receptor agonists on acetylcoholine output in rat hippocampus are not affected by repeated drug administration; (ii) discontinuation of long-term treatment with each type of agonist does not affect hippocampal cholinergic mechanisms; and (iii) flumazenil increases acetylcholine release only in the hippocampus of rats chronically treated with diazepam. Together, these data further differentiate the pharmacology of benzodiazepine receptor full agonists from that of partial and selective agonists.     

The effects of chronic intraperitoneal administration of the GABA B receptor agonist baclofen on food intake in rats

This study was undertaken to examine the effects of repeated administration of the GABA(B) receptor agonist baclofen on food intake in male Wistar rats. In the 1st Experiment, the effects of daily administration of physiological saline and baclofen (2 mg/kg, i.p.) for 27 days were investigated on food intake and body weight in non-deprived rats (n=6 in each group). Baclofen significantly (P<0.05) increased cumulative food intake each day over the treatment period during the 60 min measurement period following administration. Tolerance did not develop to the short-term hyperphagic effect of baclofen over the course of the experiment. In addition, treatment with baclofen did not alter body weight of the animals over the 27 day treatment period when compared with the saline control rats. In the 2nd Experiment, the effects of acute and chronic administration of baclofen (2 mg/kg) were investigated on 24 h food intake in rats. The rats were injected daily for 21 days with either saline (n=6) or baclofen (n=6). Food intake was measured in 30 min time bins for 24 h on treatment Days 1, 12 and 21 following injection. The results showed that while baclofen produced short-term increases in food consumption following injection on treatment Days 1, 12 and 21, the daily (24 h) food intake of the animals was not significantly different from those of control rats. Thus, these data reveal that while chronic administration of baclofen (2 mg/kg) produces short-term increases in feeding without the development of tolerance, daily (24 h) food consumption is not affected. These findings are consistent with the observation that chronic administration of baclofen (2 mg/kg) had no effect on the body weight of these animals.     

Are there changes in sensitivity to 5-HT3 receptor ligands following chronic diazepam treatment?

Administration of 1-(3-chlorophenyl)-biguanide (mCPB), a 5-HT₃ receptor agonist (1 and 10 mg/kg IP), was found to be significantly anxiogenic in vehicle treated rats tested in the plus-maze, while having no significant effect in rats withdrawn for 24 h from 21 days diazepam treatment (2 mg/kg/day), suggesting a decreased agonist action at 5-HT₃ receptors following withdrawal from chronic diazepam treatment. In the social interaction test, diazepam withdrawn rats showed a significant decrease in social interaction when compared to the chronic vehicle treated group. This anxiogenic response was reversed by low doses of zacopride (0.0001–0.01 mg/kg IP); in the vehicle treated animals 0.1 mg/kg was significantly anxiogenic. The overall pattern of results with zacopride is explained by suggesting that the anxiogenic effects of high doses of zacopride are detectable at low levels of 5-HT function and are due to an agonist action of the S-isomer in the rat at 5-HT₃ receptors. The anxiolytic action of low doses is attributed to the R-isomer acting at the R-zacopride binding site and is enhanced in conditions of high 5-HT function, e.g. in the diazepam withdrawn rats. If this hypothesis is correct, then we would predict the R-isomer alone would be more effective in reversing the anxiogenic effects of diazepam withdrawal than the racemate, used here.     

Flumazenil but not nitrendipine reverses the increased anxiety during ethanol withdrawal in the rat

After 7 day's gradual introduction of ethanol, rats were maintained for a further 4 weeks on a liquid diet containing 10% ethanol (mean daily dose 11.8±0.2 g/kg/day). Control-treated rats received liquid diet alone. Pairs of rats were tested in the social interaction test of anxiety 8 h after withdrawal. Withdrawal from ethanol significantly reduced the time spent in social interaction compared with controls, indicating an anxiogenic withdrawal response. Nitrendipine (50 mg/kg) had no effect on, whereas flumazenil (4 mg/kg) significantly reversed, this withdrawal response. This reversal appeared to be long-lasting as there was still no evidence of increased anxiety when rats were again withdrawn after 3 more days of ethanol diet.     

Long-term diazepam treatment produces changes in cholecystokinin receptor binding in rat brain

This study examined the effect of chronic diazepam administration on central benzodiazepine and CCK-8 receptor binding in rat brain. After a two-week treatment with diazepam (5 mg/kg per day) tolerance developed towards the sedative but not towards the anxiolytic action of this drug as determined using elevated plus-maze and open field tests. The % entries the rats made onto open arms and % time the rats spent in open arms were markedly decreased 24 h after the last dose of diazepam, probably indicating withdrawal anxiety. There were no changes in [3H]flunitrazepam binding either 30 min or 24 h after the last diazepam dose. However, 30 min after the last diazepam administration the apparent number of sulphated [3H]CCK-8 binding sites was significantly increased in the primary olfactory cortex. Acute diazepam treatment (5 mg/kg) had no influence on [3H]flunitrazepam or sulphated [3H]CCK-8 binding in any brain region studied. Cessation of chronic diazepam treatment was followed after 24 h by an increase in the number of CCK-8 receptors in frontal cortex and hippocampus as compared to the vehicle group. These results demonstrate that certain alterations in CCK-8 receptor characteristics may be important in the anti-anxiety effect, tolerance, and withdrawal reaction reaction after benzodiazepine administration.     

Enhancement of the muscle relaxant action of baclofen by glucocorticoids

Effects of pretreatment with hydrocortisone or prednisolone on the pharmacological effects of baclofen, particularly on the muscle relaxant action, were studied in young adult rats. Muscle relaxation was determined using an inclined board (50 degrees). A single injection of hydrocortisone (10 mg/kg, s.c.) 15 min prior to the administration of baclofen (5 mg/kg, i.p.) increased significantly the muscle relaxant time by baclofen. Prednisolone pretreatment (2 mg/kg, s.c.) also prolonged baclofen-induced muscle relaxation. Muscle relaxation due to tolperisone (50 mg/kg, i.p.) and mephenesin (80 mg/kg, i.p.) administration was enhanced by the pretreatment of hydrocortisone. Hydrocortisone pretreatment enhanced the muscle relaxant time of diazepam (2 mg/kg, i.p.) but at a lesser extent and only in males. No synergistic effect of hydrocortisone on the muscle relaxation time by suxamethonium (0.01 mg/kg, i.p.) was observed. Seven daily injections of 2 mg/kg hydrocortisone as well as of 0.4 mg/kg prednisolone did not enhance the muscle relaxation by baclofen when tested 24 hr after the last injection of the two steroids. Fifteen min priming with hydrocortisone had no effect on the antinociceptive and hypothermic effects of baclofen. 3H-Baclofen was injected i.v. as a tracer of 1 mg/kg unlabeled baclofen. In hydrocortisone-treated rats, the uptake of 3H-baclofen into the hippocampus, medulla oblongata and cerebellum was significantly higher than that in the vehicle-treated rats, whereas the radioactivity in the cervical, thoracic and lumbar spinal cord tended to decrease. The results indicate that acute priming with hydrocortisone enhances rather selectively the muscle relaxant action of the centrally acting muscle relaxant drugs. It is suggested that the primary site of the synergistic action of hydrocortisone to baclofen could be at the supraspinal level of the CNS.     

Withdrawal from ingested diazepam produces a pentylenetetrazol-like stimulus in rats

An interoceptive stimulus produced by withdrawal from chronic diazepam is similar to that produced by the anxiogenic drug pentylenetetrazol (PTZ). Because the PTZ stimulus can be measured objectively in animals, its use was proposed as a bioassay for diazepam-withdrawal stimuli. In previous studies, diazepam was given chronically by the intraperitoneal route. The purpose of the present study was to determine whether withdrawal from diazepam given in the diet would similarly produce a PTZ-like subjective effect. Administering drug through the diet has the advantage that it eliminates the necessity of giving injections every 6–8 hr and more closely parallels human chronic dosing. Rats were trained with food-reward in a two-lever operant task. Presses on one lever were reinforced after injections of PTZ, 20 mg/kg, and on the other lever after saline. After rats had acquired the PTZ discrimination, training was halted, and they were given diazepam, 240 mg/kg/day, in a nutritionally complete liquid diet. Withdrawal was precipitated after 3 or 4 days of chronic diazepam with the benzodiazepine receptor blocker Ro 15–1788, either after a bolus dose of diazepam, given intraperitoneally, or during continued dietary diazepam. During precipitated withdrawal, the rats selected the PTZ lever, indicating the presence of a PTZ-like stimulus, and this stimulus was blocked by phenobarbital, 80 mg/kg. The percentage of rats selecting the PTZ lever depended on the dose of Ro 15–1788. By 3.5 days after cessation of chronic diazepam, Ro 15–1788 no longer produced a PTZ-like stimulus and by 11 days, rats recovered baseline discrimination performance, as indicated by saline-lever selection after saline injections and PTZ-lever selection after PTZ. These data indicate that a subjective effect of withdrawal similar to that obtained after parenteral diazepam can be produced after chronic diazepam given daily in the diet.     

Raised [3H]-5-HT release and 45Ca2+ uptake in diazepam withdrawal: inhibition by baclofen.

The effect of diazepam withdrawal (2 mg/kg/day) on release of [3H]-5-hydroxytryptamine(5-HT) and [14C]-GABA from rat cortical and hippocampal slices was studied. No changes in [14C]-GABA release (basal, K(+)-evoked, uptake) from slices of either region were observed. Similarly, all parameters of [3H]-5-HT release were unchanged in cortical slices. However, during diazepam withdrawal, depolarised [3H]-5-HT release from hippocampal slices was raised with no changes in basal release or uptake into the slices being found. This increase could be prevented by in vivo administration of 1 mg/kg baclofen--this dose having no significant effect on [3H]-5-HT release from hippocampal slices of control rats. To further investigate this effect, 45Ca2+ uptake into hippocampal synaptosomes was examined and found to be increased during withdrawal. This was blocked by in vitro addition of 10 microM (-)baclofen, which had no effect on 45Ca2+ uptake in controls. Inhibition of 45Ca2+ uptake by (-)baclofen was also enhanced in nonwithdrawn diazepam-treated rats, but not in rats treated acutely with diazepam. The results from both studies indicate that chronic diazepam treatment increases neuronal sensitivity to baclofen. These results are discussed with reference to the anxiogenic state during diazepam withdrawal and a recent report of reversal of this behaviour by baclofen.     

Prevention of diazepam withdrawal syndrome by nifedipine-behavioural and neurochemical studies

Our studies aimed at investigating whether the dihydropyridine calcium antagonist, nifedipine, could prevent anxiogenic-like consequences of diazepam withdrawal in rats. Animals withdrawn from chronic diazepam (2 mg/kg/day i.p. for 2 weeks) drank significantly less water than did control rats in the unfamiliar arm of a Y maze. This anxiogenic-like effect could be prevented by acute administration of nifedipine (at 10 mg/kg i.p., but not at lower doses), which, on its own, did not change water intake in naive rats. Given chronically in combination with diazepam for the second half of a 2-week treatment with this drug, nifedipine (at the daily dose of 5 mg/kg i.p.) also suppressed the reduction of water intake normally observed on diazepam withdrawal. Biochemical measurements showed that acutely, as well as chronically, administered nifedipine increased 5-HT turnover in the hippocampus of diazepam-treated rats, thereby suggesting that the prevention of diazepam withdrawal-induced anxiogenic behaviour by the calcium antagonist might be underlain by serotoninergic mechanisms.     

Physical dependence in baboons chronically treated with low and high doses of diazepam

Physical dependence on diazepam was evaluated in male baboons chronically treated with either low or high doses of diazepam. Baboons received either a single oral daily administration of a low dose (0.5 mg/kg per day) of diazepam (n=4) or continuous intragastric infusion of a high dose (20 mg/kg per day) of diazepam (n=7). Development of physical dependence during chronic dosing with 0.5 mg/kg per day diazepam was assessed at 2 and 4 weeks and then monthly, during 1-h behavioral observations, following injections of the benzodiazepine competitive antagonist flumazenil. After 3-24 months of diazepam treatment, dosing was discontinued and physical dependence assessed via observation and responding for food pellets. In baboons that received 0.5 mg/kg per day diazepam, flumazenil precipitated a mild- to intermediate-intensity benzodiazepine withdrawal syndrome, which included decreases in the number of food pellets earned per day and increases in withdrawal postures, self-directed behaviors, aggressive behaviors and retching/vomiting. Three of four baboons showed signs of precipitated withdrawal after only 2 weeks of chronic low-dose treatment. Flumazenil continued to precipitate withdrawal signs, but with no systematic increase in severity, throughout the 6-10 months of 0.5 mg/kg diazepam administration. When 0.5 mg/kg per day diazepam dosing was discontinued, the number of food pellets earned per day decreased in two of the four baboons, but no systematic changes in behavioral signs were observed. In contrast, within 7-10 days of termination of 20 mg/kg per day diazepam dosing, withdrawal signs of intermediate intensity and a decrease in the number of food pellets earned per day occurred in all baboons. In the present study, physical dependence developed after 2 weeks of a chronic low dose of diazepam administration but did not increase further over long-term exposure to diazepam.     

Dissociation of effects of chronic diazepam treatment and withdrawal on hippocampal dialysate 5-HT and mCPP-induced anxiety in rats

Acute (10mg/kg, i.p.) and chronic (10mg/kg/day, i.p. for 10 days) diazepam treatments decreased hippocampal dialysate 5-HT (but not 5-HIAA) concentrations in freely moving rats, suggesting decreased availability of 5-HT to receptors. Twenty-four hours after the last chronic diazepam injection, hippocampal dialysate 5-HT did not differ from that in vehicle-treated rats. However, although reduced 5-HT availability often increases postsynaptic 5-HT receptor-mediated responses, the anxiogenic effect of m-chlorophenylpiperazine (mCPP), which is mediated by the activation of postsynaptic 5-HT(2C) receptors, was not increased (as indicated by the elevated plus-maze test) when given 2 days after 10 days of chronic diazepam, in intact rats. Nevertheless, concurrently in that test, significantly increased anxiety occurred after withdrawal from chronic diazepam (10mg/kg/day x 10 days). The results suggest that benzodiazepine withdrawal-induced anxiety is not mediated by changes in 5-HT(2C) receptor sensitivity, and may be independent of the benzodiazepine-induced reduction of 5-HT release in the rat hippocampus.     

Effects of baclofen on maintenance and reinstatement of intravenous cocaine self-administration in rats

  Rationale: Recent studies suggest that the GABA_B receptor agonist, baclofen, may be a useful pharmacotherapy for cocaine abuse. Objectives: To investigate further the effects of baclofen on maintenance and reinstatement of cocaine-reinforced behavior in rats. Methods: Two groups of rats were trained to self-administer IV cocaine (0.2 or 0.4 mg/kg per infusion) during daily 7-h sessions under a fixed-ratio 1 schedule. Rats were pretreated with baclofen (1.25, 2.5 or 5 mg/kg IP) or saline before the session for 5 consecutive days. An additional group of rats was trained to self-administer IV cocaine (0.4 mg/kg per infusion) during the first 2 h of daily 7-h sessions. Cocaine was replaced by saline for the remaining 5 h of the session. Once behavior had stabilized over the 7-h period, priming injections of saline (IV), cocaine (3.2 mg/kg IV) or baclofen (1.25 or 2.5 mg/kg IP) were administered prior to hour 4. Injections of baclofen (1.25 or 2.5 mg/kg IP) or saline were also given before priming injections of cocaine. Results: Pretreatment with the two higher doses of baclofen (2.5 and 5 mg/kg) decreased the number of cocaine infusions in both maintenance groups (0.2 and 0.4 mg/kg) over the 5-day treatment period. Baclofen had a greater suppressant effect on responding maintained by the lower dose of cocaine. Priming injections of baclofen (1.25 and 2.5 mg/kg) or saline did not reinstate responding. However, these same doses of baclofen dose-dependently reduced the reinstatement of responding produced by priming injections of cocaine. Conclusions: 1) The magnitude of the suppressant effects of baclofen on maintenance of cocaine self-administration depends upon the maintenance dose, 2) baclofen may be useful in preventing reinstatement of cocaine-seeking behavior, and 3) compared to maintenance, reinstatement of responding is more sensitive to the suppressant effects of baclofen. Received: 10 August 1998 / Final version: 31 October 1998     

Enhancement by flumazenil of dopamine release in the nucleus accumbens of rats repeatedly exposed to diazepam or imidazenil

The effect of long-term treatment (three times daily for 3 weeks) with a behaviorally relevant dose of the benzodiazepine receptor partial agonist imidazenil (0.5 mg/kg, IP) on basal dopamine release in the nucleus accumbens of freely moving rats was compared with that of diazepam (3 mg/kg, IP), a benzodiazepine receptor full agonist. Challenge doses of imidazenil and diazepam decreased the extracellular dopamine concentration in the nucleus accumbens by approximately the same extent in animals repeatedly exposed to vehicle or to the respective drug. Moreover, the abrupt discontinuation of long-term treatment with diazepam or imidazenil failed to affect basal dopamine release in this brain area during the first 5 days of withdrawal. In contrast, administration of the benzodiazepine receptor antagonist flumazenil (4 mg/kg, IP) elicited a marked increase (95 or 60%) in dopamine release in the nucleus accumbens 6h after withdrawal of diazepam or imidazenil, respectively. Flumazenil induced a similar but smaller effect (50% increase) 5 days after diazepam withdrawal but had no effect 5 days after discontinuation of imidazenil treatment. The resultssupport an involvement of the mesoaccumbens dopaminergic neurons in the withdrawal syndrome precipitated by flumazenil and allow further differentiation of benzodiazepine receptor partial and full agonists with respect to dependence liability of dopaminergic neurons in the nucleus accumbens. Received: 20 June 1996 / Final version: 27 November 1996     

Withdrawal syndrome following subchronic treatment with anxiolytic agents

The acute administration of diazepam (0.1-2.5 mg/kg IP), sulpiride (0.5-20 mg/kg IP) and tiapride (0.5-40 mg/kg IP) to the mouse enhanced exploratory activity (rearings/line crossings) in the brightly illuminated white area of a two compartment white/black anxiety test box, with a corresponding decrease in the black, indicating an anxiolytic action. This profile of change was maintained during a twice daily administration for 7 days with diazepam (2.5 and 10 mg/kg), sulpiride (5 and 20 mg/kg) and tiapride (10 and 40 mg/kg). However, 8 and 48 hr following withdrawal of diazepam, the profile of exploratory behaviour was reversed to a preference for the black area: by 96 hr values for behaviour had returned to control levels. In contrast, an anxiolytic profile of action was maintained 8 and 48 hr following the withdrawal of sulpiride and tiapride, the values returning to control levels after 96 hr. It is concluded that a sub-chronic treatment with diazepam, sulpiride and tiapride induces an anxiolytic profile of action in the mouse model, that an anxiogenic profile follows the abrupt withdrawal of diazepam but that this is not recorded following the abrupt withdrawal of sulpiride and tiapride.