Lethal effects of cocaine are reduced by the dopamine-1 receptor antagonist SCH 23390 but not by haloperidol

The prevalence of cocaine abuse has been associated with a host of medical complications and deaths. We investigated the effects of two dopamine antagonists with different affinities for dopamine-1 and dopamine-2 receptor subtypes on cocaine-induced lethality. Male Fischer-344 rats were given cocaine HCl (i.p.) and observed for lethality at 24 hrs. Cocaine was not lethal at 50 mg/kg and produced a steep dose-effect function from 60 to 100 mg/kg. Lethality was 88.9% at 100 mg/kg and the LD 50 was 79.7 mg/kg (95% CL: 74.8-84.9). Doses as high as 180 mg/kg failed to kill all rats. Lethality was often but not invariably associated with convulsions. Haloperidol (0.3-3 mg/kg i.p.) given 30 min prior to cocaine did not alter the lethal effects of cocaine but did reduce the lethality of methamphetamine. SCH 23390 (0.1-1 mg/kg i.p., 30 min prior) shifted the cocaine dose-effect function to the right at 0.3 mg/kg. Maximum protection was conferred by 0.3 mg/kg SCH 23390 where the LD 50 was increased to 100.1 mg/kg (95% CL: 91.5-109.5). Comparable protection was not observed if SCH 23390 was given 5 min after cocaine. These results suggest that dopamine receptors may play a role in the lethal effects of cocaine and that the D1 dopamine receptor subtype appears to be more relevant to lethality than the D2 subtype.     

Modulators of N-methyl-D-aspartate protect against diazepam- or phenobarbital-resistant cocaine convulsions.

The anticonvulsants diazepam (1-10 mg/kg) and phenobarbital (30-100 mg/kg) protected against lethality without altering clonic convulsions induced by 75 mg/kg cocaine (CD100) in male Swiss Webster mice. In contrast, the non-competitive N-methyl-D-aspartate (NMDA) antagonists, MK-801 (dizocilpine) and phencyclidine, produced dose-dependent protection against cocaine convulsions. The competitive NMDA antagonists, CPP and NPC 12626, were also anti-convulsant, without producing the behavioral disturbances associated with non-competitive antagonists. Diazepam and phenobarbital protected against convulsions induced by 60 mg/kg cocaine (90% convulsions alone). Compounds that act at the strychnine-insensitive glycine receptor of the NMDA receptor complex, ACPC and 7-chlorokynurinic acid, also protected against convulsions induced by 60 mg/kg cocaine. In contrast, the non-opioid antitussive anticonvulsants (dextromethorphan, caramiphen, and carbetapentane) were not active against either dose of cocaine. The efficacy of compounds as antagonists of the convulsant effects of cocaine and NMDA appear related. These results suggest a potential role for the NMDA receptor complex in the convulsant actions of cocaine and new molecular targets for drug discovery in treating cocaine toxicity.     

Increased antidepressant-like effect of desipramine combined with central stimulants (caffeine and amphetamine) in mice

Desipramine is a widely used antidepressive agent that inhibits the reuptake of noradrenaline and serotonin, and central stimulants such as caffeine and amphetamine help to release noradrenaline and serotonin. This work aimed to evaluate whether the combination of these agents could produce a stronger antidepressant-like effect than either of the drugs alone. To this end, male mice were treated with different doses of desipramine, caffeine, amphetamine, desipramine-caffeine and desipramine-amphetamine. The results showed that all drugs produced decreased immobility time in the forced swimming model. The combined treatment of desipramine (0.31, 1.0 or 3.1 mg/kg i.p.) with caffeine or amphetamine (0.31 or 1 mg/kg i.p.) reduced immobility time greater than either of those drugs alone. The combined treatment of desipramine (0.31, 1 and 3.1 mg/kg i.p.) with amphetamine or caffeine (0.1 and 1 mg/kg i.p.) did not increase the motor activity significantly compared to the control. These results also suggested that drugs which promote the release of noradrenaline and serotonin could increase antidepressant-like effect of desipramine.     

Effect of repeated treatment with desipramine in the behavioral "despair" test in rats: antagonism by "atypical" but not "classical" neuroleptics or antiadrenergic drugs

A 7-day treatment with 20 mg/kg/day desipramine reduced the immobility time in the behavioral "despair" test in rats. The effect of DMI was antagonized by sulpiride (100 mg/kg i.p.), metoclopramide (20 mg/kg i.p.) and clopazine (20 mg/kg i.p.) but not by haloperidol (0.5 mg/kg i.p.) or chlorpromazine (5 mg/kg i.p.). Alpha-adrenoreceptor blockers (prazosin 3 mg/kg s.c.; aceperone 10 mg/kg i.p.; azapetine 24 mg/kg s.c.; phentolamine 20 mg/kg i.p.), dl-propranolol (5 mg/kg i.p.) and clonidine (0.1 mg/kg i.p.) failed to modify the anti-immobility effect of DMI. The data suggest that a particular subtype of dopamine receptors is involved in the anti-immobility effect of a 7-day treatment with DMI in the behavioral "despair" test in rats.     

Nitrendipine: an antidote to cardiac and lethal toxicity of cocaine

Nitrendipine, a Ca2+ modulator was tested in the rat as an antagonist to the cardiac toxicity of cocaine and as an antidote to the acute lethal effects of this drug. In a first series of experiments, nitrendipine (1.46 X 10(-3) mg/kg/min) when simultaneously administered intraarterially with cocaine (2 mg/kg/min) suppresses the arrhythmias induced by cocaine and increases survival time from 73 +/- 33 min to 309 +/- 118 min and the lethal dose of cocaine from 146 +/- 66 mg/kg to 618 +/- 236 mg/kg (p less than 0.003). Nitrendipine also protects the heart from the acute morphological lesions induced by cocaine administration and antagonizes some of the central effects of cocaine. In a second series, 5 rats administered 60 mg/kg of cocaine intraperitoneally had a survival time of 8'06 +/- 5'20. Death was attributed to convulsions and respiratory arrest. Animals treated with nitrendipine (129 +/- 23 mg/kg) 4'30" after cocaine administration survived. Nitrendipine appears to have general protective effects against cocaine cardiac toxicity and the acute lethal effects of this alkaloid.     

Antidepressant-like effect of 17beta-estradiol: involvement of dopaminergic, serotonergic, and (or) sigma-1 receptor systems

17beta-estradiol has been reported to possess antidepressant-like activity in animal models of depression, although the mechanism for its effect is not well understood. The present study is an effort in this direction to explore the mechanism of the antidepressant-like effect of 17beta-estradiol in a mouse model(s) of behavioral depression (despair behavior). Despair behavior, expressed as helplessness to escape from a situation (immobility period), as in a forced swim test in which the animals are forced to swim for a total of 6 min, was recorded. The antiimmobility effects (antidepressant-like) of 17beta-estradiol were compared with those of standard drugs like venlafaxine (16 mg/kg, i.p.). 17beta-estradiol produced a U-shaped effect in decreasing the immobility period. It had no effect on locomotor activity of the animal. The antidepressant-like effect was comparable to that of venlafaxine (16 mg/kg, i.p.). 17beta-estradiol also exhibited a similar profile of antidepressant action in the tail suspension test. When coadministered with other antidepressant drugs, 17beta-estradiol (5 microg/kg, i.p.) potentiated the antiimmobility effect of subeffective doses of fluoxetine (5 mg/kg, i.p.), venlafaxine (2 mg/kg, i.p.), or bupropion (10 mg/kg, i.p.), but not of desipramine (5 mg/kg, i.p.) or tranylcypromine (2 mg/kg, i.p.), in the forced swim test. The reduction in the immobility period elicited by 17beta-estradiol (20 microg/kg, i.p.) was reversed by haloperidol (0.5 mg/kg, i.p.; a D(2) dopamine receptor antagonist), SCH 23390 (0.5 mg/kg, i.p.; a D(1) dopamine receptor antagonist), and sulpiride (5 mg/kg, i.p.; a specific dopamine D(2) receptor antagonist). In mice pretreated with (+)-pentazocine (2.5 mg/kg, i.p.; a high-affinity sigma-1 receptor agonist), 17beta-estradiol (5 microg/kg, i.p.) produced a synergistic effect. In contrast, pretreatment with progesterone (10 mg/kg, s.c.; a sigma-1 receptor antagonist neurosteroid), rimcazole (5 mg/kg, i.p.; another sigma-1 receptor antagonist), or BD 1047 (1 mg/kg, i.p.; a novel sigma-1 receptor antagonist) reversed the antiimmobility effects of 17beta-estradiol (20 microg/kg, i.p.). Similarly, in mice pretreated with a subthreshold dose of 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT, a 5-HT1A serotonin receptor agonist), 17beta-estradiol (5 microg/kg, i.p.) produced an antidepressant-like effect. These findings demonstrate that 17beta-estradiol exerted an antidepressant-like effect preferentially through the modulation of dopaminergic and serotonergic receptors. This action may also involve the participation of sigma-1 receptors.     

Alpha 2-adrenoceptor antagonists block the stimulant effects of cocaine in mice.

In the present study we have investigated the effects of the alpha 2-adrenoceptor antagonist idazoxan and its 2-ethoxy derivative RX811059 on the locomotor activity induced by cocaine in mice. The stimulant effects of cocaine (15 mg/kg i.p.) were significantly antagonised by idazoxan (3 mg/kg i.p.) and RX811059 (1 mg/kg i.p.) and also initially suppressed by idazoxan (1 mg/kg i.p.) and RX811059 (0.3 mg/kg i.p.). The alpha 2-adrenoceptor antagonists had no effect on locomotion when given alone. These results suggest that noradrenergic mechanisms may play a role in the stimulant effects of cocaine and that alpha 2-adrenoceptor antagonists like idazoxan may be of some benefit in the clinical management of cocaine abuse.     

Agastache mexicana may produce anxiogenic-like actions in the male rat.

Behavioral effects of a water-soluble extract of Agastache mexicana, a plant with purported anxiolytic actions, were studied in male Wistar rats. In the elevated plus-maze test, various doses of the plant extract (3.0 mg/kg body wt.; 9.0 mg/kg body wt.; 12.0 mg/kg body wt.) administered intraperitoneally (i.p.) decreased the exploration of open arms, showing an anxiogenic-like effect. Agastache mexicana (12 mg/kg body wt.; i.p.) did not change immobility in the forced swimming test (i.e., had no anti-depressant effect) but increased the anti-immobility action of 32.0 mg/kg body wt. (i.p.) of desipramine (i.e., increased the antidepressant-like effect of desipramine). A. mexicana had no effect on exploratory activity in an open field test, indicating that it had no sedative effect at the doses used. It is concluded that effects of the water extract of A. mexicana are more consistent with an anxiogenic-like property than an anxiolytic-like one.     

Anti-epileptic effect of the new calcium channel blocker IOS-1.1212]

In experiments on freely moving male Wistar rats it was shown that IOS-1.1212 (1,4-dihydropyridine) in a dose 2 and 10 mg/kg (i. p.) suppressed the penicillin-induced focal epileptic activity in cerebral cortex. Similar suppressing effect of IOS-1.1212 was shown on acute generalized tonic-clonic pentylenetetrazol (PTZ) seizures (75 mg/kg i. p.) and on chronic PTZ administration (PTZ-kindling, 30 mg/kg i. p. during 30 days): when injected 30 min before each PTZ administration it delayed the development of kindling-induced seizures susceptibility in randomized animals (series 1) and attenuated the severity of seizures in PTZ-sensitive animals (series 2). However, IOS-1.1212 had no effect on the strychnine-induced focal epileptic activity. In male Icr:Icl mice IOS-1.1212 in a dose 1.5 and 5 mg/kg also influenced the PTZ convulsions (i. v. titration of 1% solution at a rate of 0.01 ml/s) and had no effect on the strychnine convulsions (i. v. titration of 0.01% solution at a rate of 0.01 ml/s) and on maximal electroshock. In addition, IOS-1.1212 significantly increased antiepileptic effect of phenobarbital on maximal electroshock.     

Involvement of monoaminergic system in the antidepressant-like effect of the hydroalcoholic extract of Siphocampylus verticillatus

The antidepressant-like effect of the hydroalcoholic extract obtained from aerial parts of Siphocampylus verticillatus, a Brazilian medicinal plant, was investigated in two models of depression in mice and against synaptosomal uptake of serotonin, noradrenaline and dopamine. The immobility times in the forced swimming test (FST) and in the tail suspension test (TST) were significantly reduced by the extract (dose range 100-1000 mg/kg, i.p.), without accompanying changes in ambulation when assessed in an open-field. In addition when given orally the extract was also effective in reducing the immobility time in the TST. The efficacy of extract in the TST was comparable to that of the tricyclic antidepressant imipramine (15 mg/kg, i.p.) and with fluoxetine (32 mg/kg, i.p.). The anti-immobility effect of the extract (600 mg/kg, i.p.) assessed in the TST was not affected by pre-treatment with naloxone (1 mg/kg, i.p., a non-selective opioid receptor antagonist) or L-arginine (750 mg/kg, i.p., a nitric oxide precursor). In contrast, the extract (600 mg/kg, i.p.) antidepressant-like effect was significantly reduced by pre-treatment of animals with p-chlorophenylalanine (PCPA, 100 mg/kg, i.p., an inhibitor of serotonin synthesis), sulpiride (50 mg/kg, i.p., a selective D2 receptor antagonist), prazosin (62.5 microg/kg, i.p., an alpha1 adrenoreceptor antagonist) or by guanosine 5'-monophosphate (GMP, 250 mg/kg, i.p., a nucleotide known to block some actions elicited by NMDA). The biochemical data show that the extract of S. verticillatus inhibited in a graded manner the uptake of monoamines. However, at the IC50 level, the extract was approximately 3.2 to 3.4-fold more potent and also more efficacious in inhibiting the synaptosomal uptake of noradrenaline and serotonin than dopamine. Taken together these data demonstrate that the extract of S. verticillatus elicited a significant antidepressant-like effect, when assessed in the TST and FST in mice. Its action seems to involve an interaction with adrenergic, dopaminergic, glutamatergic and serotonergic systems.     

A Microdialysis Study of the Regulation of Endogenous Noradrenaline Release in the Rat Hippocampus

In vivo microdialysis was used to investigate the regulation of noradrenaline release in rat hippocampus. Idazoxan, an alpha 2-adrenoreceptor antagonist (1-10 mg/kg), increased noradrenaline release in a dose-dependent manner. Inhibition of noradrenaline uptake by desipramine (0.05-20 microM; via the probe) also increased the extracellular content of the transmitter. In the presence of this increased noradrenaline content (desipramine via the probe), the effect of a low dose of idazoxan (1 mg/kg) was potentiated. Local perfusion of idazoxan (1-500 microM) in the hippocampus also increased noradrenaline release but not to the same extent as following systemic administration. In the presence of desipramine, unlike the systemic injection of idazoxan, local perfusion did not potentiate noradrenaline release. The data are consistent with the regulation of extracellular noradrenaline content in the hippocampus by neuronal uptake and to a lesser extent by presynaptic autoreceptors.     

Reduction of pentylenetetrazol-induced convulsions by the octadecaneuropeptide ODN

Intracerebroventricular injection of the octadecaneuropeptide ODN in mouse, at doses of 12.5-1000 ng, reduced the percentage of convulsing animals and increased the latency of convulsions elicited by pentylenetetrazol (50 mg/kg, intraperitoneal [i.p.]). ODN also reduced the percentage of mortality induced by pentylenetetrazol (100 mg/kg, i.p.). The COOH-terminal octapeptide fragment of ODN was approximately equally effective but acted more rapidly than ODN to reverse the convulsant effect of pentylenetetrazol. ODN (100 ng, intracerebroventricular [i.c.v.]) increased the convulsion latency and reduced the percentage of animals that convulsed after the administration of the inverse agonist of benzodiazepine receptors DMCM (13 mg/kg, i.p.), whereas the benzodiazepine receptor antagonist flumazenil (1 mg/kg, subcutaneously) abrogated the protective effect of ODN (100 ng, i.c.v.) on pentylenetetrazol-induced convulsions. ODN (100 ng, i.c.v.) also reduced the percentage of DBA/2J mice displaying audiogenic convulsions. In contrast, ODN did not reduce the percentage of mice displaying tonic or clonic convulsions when electrical interauricular stimulations were applied. It is concluded that ODN, or more likely a proteolytic fragment derived from ODN, reduces pentylenetetrazol-induced convulsions through activation of central-type benzodiazepine receptors.     

Reduction of pentylenetetrazol-induced convulsions by the octadecaneuropeptide ODN

Intracerebroventricular injection of the octadecaneuropeptide ODN in mouse, at doses of 12.5-1000 ng, reduced the percentage of convulsing animals and increased the latency of convulsions elicited by pentylenetetrazol (50 mg/kg, intraperitoneal [i.p.]). ODN also reduced the percentage of mortality induced by pentylenetetrazol (100 mg/kg, i.p.). The COOH-terminal octapeptide fragment of ODN was approximately equally effective but acted more rapidly than ODN to reverse the convulsant effect of pentylenetetrazol. ODN (100 ng, intracerebroventricular [i.c.v.]) increased the convulsion latency and reduced the percentage of animals that convulsed after the administration of the inverse agonist of benzodiazepine receptors DMCM (13 mg/kg, i.p.), whereas the benzodiazepine receptor antagonist flumazenil (1 mg/kg, subcutaneously) abrogated the protective effect of ODN (100 ng, i.c.v.) on pentylenetetrazol-induced convulsions. ODN (100 ng, i.c.v.) also reduced the percentage of DBA/2J mice displaying audiogenic convulsions. In contrast, ODN did not reduce the percentage of mice displaying tonic or clonic convulsions when electrical interauricular stimulations were applied. It is concluded that ODN, or more likely a proteolytic fragment derived from ODN, reduces pentylenetetrazol-induced convulsions through activation of central-type benzodiazepine receptors.     

Cocaine fatalities increased by restraint stress

Laboratory rats injected daily with a moderate dose of cocaine hydrochloride (30 mg/kg, i.p.) showed increased fatalities when cocaine injections were followed by 30 min of restraint stress. The 5-day mortality rate was 58% for the cocaine-plus-stress group, while 17% of the animals receiving cocaine without restraint stress died. This finding suggests that stress can augment the toxic effect of cocaine and that minimizing stress may be an important consideration in the clinical management of cocaine overdose.     

Adrenergic stimulation of adrenocortical secretion in immature fowl

In 5-6-week-old cockerels the circulating corticosterone concentration was significantly increased in birds i.m. injected 30 and 60 min previously with adrenaline (0.33 mg/kg), noradrenaline (0.33 mg/kg) or a beta-adrenergic agonist (isoprotorenol, 1 mg/kg), but was reduced in birds pretreated with an alpha-adrenergic agonist (phenylephrine, 1 mg/kg). The stimulation of corticosterone secretion induced by a 30 min period of forced exercise (0.04 km/hr; 0 degree incline) was potentiated by noradrenaline, isoprotorenol and phentolamine pretreatment. In response to exogenous adrenocorticotrophin (ACTH, 8 i.u./kg), administered i.v., the increase in the plasma corticosterone concentration was elevated above that in the controls in birds pretreated with adrenaline, noradrenaline, isoprotorenol or phentolamine (an alpha antagonist administered at 1 mg/kg). The corticosterone response to ACTH was suppressed by phenylephrine pretreatment. These results demonstrate that both basal and stimulated levels of adrenocortical activity may be subtly regulated by adrenergic mechanisms acting at a site(s) within the hypothalamo-pituitary-adrenal axis.     

Effects of nitric oxide synthesis inhibition with or without nitric oxide inhalation on responses to systemic cocaine administration in rats

The effects of N^ω-nitro-L-arginine methyl ester (L-NAME) i.v. and nitric oxide (NO) inhalation on integrated systemic responses to cocaine were studied in lightly anesthetized, paralyzed, and mechanically ventilated rats. Cocaine (4 mg/kg/min i.v.) produced seizures then isoelectric electrocephalographic (isoEEG) activity as well as an initial increase in systolic blood pressure and heart rate, then progressive cardiovascular system depression culminating in asystole. Pretreatment with L-NAME (2 mg/kg/min i.v.) for 30 min significantly reduced the incidence of seizure as compared to saline treated animals (saline 7/8; L-NAME 3/8). Doses of cocaine that produced arrhythmias, isoEEG and asystole were significantly lower in the L-NAME treated animals as compared to the saline group. L-NAME did not affect peak systolic blood pressure and heart rate responses to cocaine. NO inhalation (80 ppm) did not affect CNS and cardiovascular responses to cocaine in control animals but enhanced the effects of L-NAME on cocaine toxicity. The results show that pretreatment with L-NAME reduces the central nervous system stimulatory effect of cocaine (reduced seizure incidence) and enhances its depressant effect on both the central nervous system (lower does for isoEEG) and the cardiovascular system (lower dose for arrhythmias and asystole), but does not affect the cardiovascular stimulatory action of cocaine. NO inhalation does not protect against any of the systemic effects of cocaine in animals with normal or suppressed NO production.     

Carbamazepine produces nonspecific effects on cocaine self-administration in rats.

Anecdotal evidence in humans suggest that carbamazepine suppresses cocaine-induced rush and craving. Such claims are unsupported in controlled trials using a placebo control. In the present study, rats were trained to self-administer i.v. cocaine in daily 2-hr sessions in which every tenth lever press delivered 1 mg/kg cocaine. After responding was stable, they were injected before each session with the vehicle for 2 days followed by carbamazepine for 2 days. At a 7 mg/kg dose, carbamazepine was without effect, whereas 15 mg/kg suppressed responding for cocaine only on the second (day 4) day of carbamazepine treatment. With 4 consecutive days of treatment, carbamazepine (15 mg/kg) reduced cocaine-maintained responding slightly, but significantly. In another group of animals trained to lever-press for food reinforcement, carbamazepine (15 mg/kg) also significantly decreased the rate of responding, suggesting that the suppression of responding was not specific to cocaine-reinforced behavior.     

Measurement of Endogenous Noradrenaline Release in the Rat Cerebral Cortex In Vivo by Transcortical Dialysis: Effects of Drugs Affecting Noradrenergic Transmission

The release of endogenous noradrenaline was measured in the cerebral cortex of the halothane-anesthetized rat by using the technique of brain dialysis coupled to a radioenzymatic assay. A thin dialysis tube was inserted transversally in the cerebral cortex (transcortical dialysis) and perfused with Ringer medium (2 microliter min-1). Under basal conditions, the cortical output of noradrenaline was stable over a period of at least 6 h and amounted to 8.7 pg/20 min (not corrected for recovery). Histological control of the perfused area revealed very little damage and normal morphology in the vicinity of the dialysis tube. Omission of calcium from the perfusion medium caused a marked drop in cortical noradrenaline output. Bilateral electrical stimulation (for 10 min) of the ascending noradrenergic pathways in the medial forebrain bundle caused a frequency-dependent increase in cortical noradrenaline output over the range 5-20 Hz. Stimulation at a higher frequency (50 Hz) resulted in a levelling off of the increase in cortical noradrenaline release. Systemic administration of the dopamine-beta-hydroxylase inhibitor bis-(4-methyl-1-homopiperazinylthiocarbonyl) disulfide (FLA 63) (25 mg/kg i.p.) markedly reduced, whereas injection of the monoamine oxidase inhibitor pargyline (75 mg/kg i.p.) resulted in a progressive increase in, cortical noradrenaline output. d-Amphetamine (2 mg/kg i.p.) provoked a sharp increase in cortical noradrenaline release (+450% over basal values within 40 min). Desmethylimipramine (10 mg/kg i.p.) produced a twofold increase of cortical noradrenaline release. Finally, idazoxan (20 mg/kg i.p.) and clonidine (0.3 mg/kg i.p.), respectively, increased and decreased the release of noradrenaline from the cerebral cortex.(ABSTRACT TRUNCATED AT 250 WORDS)     

Naloxone attenuation of the effect of cocaine on rewarding brain stimulation

Antagonism of the threshold lowering effect of cocaine for brain stimulation reward by naloxone was investigated. Rats with bipolar electrodes implanted in either the median forebrain bundle (MFB) or the ventral tegmental area (VTA) were trained on a rate-independent threshold procedure. Effective threshold lowering doses of cocaine (10-15 mg/kg i.p.) were determined for each subject. A moderate dose of naloxone (4 mg/kg i.p.) effectively blocked the threshold lowering action of the cocaine. Lower (2 mg/kg) and higher (8 mg/kg) doses of naloxone attenuated but did not completely block the cocaine effect. These results provide further evidence for a catecholamine/endogenous opioid interaction in central reward.     

PI3 kinase is involved in cocaine behavioral sensitization and its reversal with brain area specificity

Phosphatidylinositol 3-kinase (PI3K) is an important signaling molecule involved in cell differentiation, proliferation, survival, and phagocytosis, and may participate in various brain functions. To determine whether it is also involved in cocaine sensitization, we measured the p85alpha/p110 PI3K activity in the nuclear accumbens (NAc) shell, NAc core, and prefrontal cortex (PFC) following establishment of cocaine sensitization and its subsequent reversal. Na?ve rats were rank-ordered and split into either daily cocaine or saline pretreatment group based on their locomotor responses to an acute cocaine injection (7.5 mg/kg, i.p.). These two groups were then injected with cocaine (40 mg/kg, s.c.) or saline for 4 consecutive days followed by 9-day withdrawal. Cocaine sensitization was subsequently reversed by 5 daily injections of the D1/D2 agonist pergolide (0.1 mg/kg, s.c.) in combination with the 5-HT3 antagonist ondansetron (0.2 mg/kg, s.c., 3.5h after pergolide injection). After another 9-day withdrawal, behavioral cocaine sensitization and its reversal were confirmed with an acute cocaine challenge (7.5 mg/kg, i.p.), and animals were sacrificed the next day for measurement of p85alpha/p110 PI3K activity. Cocaine-sensitized animals exhibited increased PI3K activity in the NAc shell, and this increase was reversed by combined pergolide/ondansetron treatment, which also reversed behavioral sensitization. In the NAc core and PFC, cocaine sensitization decreased and increased the PI3K activity, respectively. These changes, in contrast to that in the NAc shell, were not normalized following the reversal of cocaine-sensitization. Interestingly, daily injections of pergolide alone in saline-pretreated animals induced PI3K changes that were similar to the cocaine sensitization-associated changes in the NAc core and PFC but not the NAc shell; furthermore, these changes in saline-pretreated animals were prevented by ondansetron given 3.5h after pergolide. The present study suggests that selective enhancement of the PI3K activity in the NAc shell may be one of key alterations underlying the long-term cocaine sensitization. To the extent cocaine sensitization is an important factor in human cocaine abuse, pharmacological interventions targeted toward the NAc shell PI3K alteration may be useful in cocaine abuse treatment.