Cocaine-Induced Reinstatement Requires Endogenous Stimulation of μ-Opioid Receptors in the Ventral Pallidum

μ Receptor blockade in the ventral pallidum did not alter food-seeking or cocaine-induced locomotor activity. A, CTAP did not affect food seeking. Animals were trained to lever press for food (FR-5), and, after extinction to criterion, the reinstatement of lever pressing was initiated by reintroducing two food pellets. Each animal received two reinstatement trials, and the data were evaluated using a mixed-regression analysis (F_(3,12) = 0.41; p = 0.747). B, CTAP (10 μg) did not alter cocaine-induced locomotor activity. One group of rats was microinjected with CTAP (10 μg) or saline into the ventral pallidum just before saline (1 ml/kg, i.p.), and another group was microinjected just before cocaine (15 mg/kg, i.p.). Data were evaluated using a two-way ANOVA with repeated measures over microinjection, and the data are shown as mean ± SEM distance traveled over 2 h after injection: systemic treatment, F_(1,14) = 61.99, p < 0.001; microinjection treatment, F_(1,14) = 1.03, p = 0.331; interaction, F_(1,14) = 0.21, p = 0.656. ^*p < 0.05 compared with CTAP in A, cocaine in B, and systemic cocaine with saline in B.

Stimulation of μ receptors in the ventral pallidum induces the reinstatement of cocaine seeking. A, Morphine in the ventral pallidum reinstated lever pressing in cocaine-trained subjects. Extinguished subjects were microinjected with saline or morphine to reinstate lever pressing. Active lever extinction baseline was as follows: ventral pallidum, 13 ± 2; nucleus accumbens, 9 ± 3. Data were evaluated using a mixed-regression analysis, and each rat had a maximum of two reinstatement trials (ventral pallidum, F_(4,11) = 8.45, p = 0.002; nucleus accumbens, F_(4,11) = 1.92, p = 0.181). B, The reinstatement by morphine (MS; 10 μg) microinjection into the ventral pallidum was blocked by co-microinjection with CTAP (3 μg). Each animal received all three reinstatement trials in random order, and the data were evaluated using a one-way ANOVA with repeated measures over treatment: extinction baseline, 13 ± 2; F_(2,10) = 29.04; p < 0.001. C, The effect of intra-ventral pallidum morphine and systemic cocaine (Coc; 5 mg/kg, i.p.) in reinstating drug seeking. Each animal received a maximum of two reinstatement trials, and the data were evaluated using a mixed-regression analysis: extinction baseline, 16 ± 1; F_(3,9) = 11.699; p = 0.002. D, Intra-pallidal injection of high-dose morphine (10 μg) inhibits cocaine-primed (10 mg/kg, i.p.) reinstatement. Each rat received both treatments in random order, and the data were evaluated using a paired Student's t test. ^*p < 0.05 compared with CTAP in A, cocaine in B, vehicle (Veh) + Veh in C, and Veh + Coc in D.

Cocaine reduces extracellular GABA in the ventral pallidum. A, In drug-naive rats, the reverse dialysis of morphine (MS) through a probe in the ventral pallidum reduced extracellular GABA, an effect blocked by coperfusion with CTAP. Increasing doses of morphine or CTAP were added separately to the dialysis buffer at 60 min intervals (doses are shown above arrows as micromolar), and, for the CTAP plus morphine, CTAP (10 μm) was coperfused with increasing doses of morphine. The number of determinations is shown in parentheses, and the data were evaluated using a two-way repeated-measures ANOVA over time: treatment, F_(2,17) = 20.86, p < 0.001; time, F_(11,187) = 9.06, p < 0.001; interaction, F_(22,187) = 3.48, p < 0.001. B, A cocaine injection (15 mg/kg, i.p.; at arrow) in animals extinguished from cocaine self-administration produced a reduction in extracellular GABA (control) that was prevented by reverse dialysis of CTAP (10 μm) begun 10 min before administering cocaine. Data were evaluated using a two-way ANOVA with repeated measures over time: treatment, F_(1,16) = 25.84, p < 0.001; time, F_(14,224) = 3.15, p < 0.001; interaction, F_(14,224) = 6.14, p < 0.001. C, Cocaine reduced extracellular GABA levels compared with basal GABA in all treatment groups (50 and 60 min in yoked saline; 40-80 min in both cocaine groups), but the reduction was larger in the yoked and self-administering cocaine groups compared with the yoked saline animals. The self-administration group is the same self-administration group in B. The data were evaluated using a two-way ANOVA with repeated measures over time: treatment, F_(2,22) = 2.09, p = 0.147; time, F_(14,308) = 14.41, p < 0.001; interaction, F_(28,308) = 2.03, p = 0.002. D, Not all animals extinguished from cocaine reinstated lever pressing in response to cocaine when microdialysis was being conducted in the ventral pallidum. Active and inactive lever presses are shown for all animals and then divided between animals showing >20 active lever presses and those showing <20. Active and inactive lever presses were compared using a paired Student's t test. E, If the dialysis data corresponding to the behavioral data in D were also divided according to whether or not a behavioral response occurred, animals showing reinstatement demonstrated statistically equivalent reductions in extracellular GABA in response to an injection of cocaine. ^*p < 0.05 compared with basal glutamate values in A, between control and CTAP in B, with yoked saline in C, and between active and inactive in D.

Increased sensitivity to μ receptor stimulation contributed to the augmented decrease in GABA produced by withdrawal from cocaine. A, Lower doses of morphine reduced GABA in the ventral pallidum of rats trained to self-administer or yoked to cocaine relative to yoked saline subjects. Experiment was conducted as in Figure 3A, and the data are shown as the mean ± SEM of the average of the last two samples obtained at each dose of morphine (0 is baseline). Data were evaluated using a two-way ANOVA with repeated measures over dose, followed by a one-way ANOVA with repeated measures over time within each treatment group: treatment, F_(2,15) = 1.44, p = 0.267; time, F_(14,210) = 35.42, p < 0.001; interaction, F_(28,210) = 1.11, p = 0.327. B, Representative Western blots showing the lack of effect of cocaine self-administration on total and membrane-bound μ-opioid receptor levels in the ventral pallidum. ^*p < 0.05 comparing all doses of morphine with baseline (0) within each treatment group. C, Cocaine; Sa, saline.