Prelimbic cortex and ventral tegmental area modulate synaptic plasticity differentially in nucleus accumbens during cocaine-reinstated drug seeking

Abstract

Addictive drug use causes long-lasting changes in synaptic strength and dendritic spine morphology in the nucleus accumbens that might underlie the vulnerability to relapse. Although activity in mesocorticolimbic circuitry is required for reinstating cocaine seeking, its role in reinstatement-associated synaptic plasticity is not well characterized. Using rats extinguished from cocaine self-administration, we found potentiated synaptic strength (assessed as the AMPA/NMDA current amplitude ratio) and increased spine head diameter in medium spiny neurons in the accumbens core (NAcore). The basal changes in synaptic strength and morphology in cocaine-extinguished animals were further augmented during cocaine-induced reinstatement. Two NAcore afferents contributing to cocaine reinstatement are glutamatergic inputs from the prelimbic prefrontal cortex (PL) and dopamine from the ventral tegmental area (VTA). Pharmacological inhibition of either PL or VTA prevented cocaine-primed reinstatement. However, inhibiting the PL further potentiated AMPA/NMDA and spine head diameter, while inactivating the VTA or the combined systemic administration of dopamine D1 and D2 antagonists prevented the increase in AMPA/NMDA and spine diameter induced by cocaine priming. These data indicate that neuronal activity in the VTA and associated dopamine receptor stimulation is necessary for the synaptic potentiation in the NAcore during cocaine-induced reinstatement. Although activity in the PL was necessary for reinstatement, it inhibited synaptic potentiation initiated by an acute cocaine injection. Thus, although the PL and VTA differentially regulate the direction of synaptic plasticity induced by a cocaine-priming injection, coordinated synaptic potentiation by both NAcore afferents is necessary for cocaine-induced relapse.

Figure 1

Acute cocaine administration potentiated A/N without changing the AMPA receptor rectification. (a) Active lever pressing in the cocaine and yoked-saline administration/extinction training protocol used for all experiments. (b) Acute cocaine reinstates active lever pressing in cocaine-extinguished rats. Histogram shows the active lever pressing over 15 or 45 min after acute cocaine administration (mean±SEM). Before cocaine refers to the last day of extinction pressing in rats that were killed 24 h later but before initiating a cocaine reinstatement session, and Cocaine SA refers to rats extinguished from cocaine self-administration. N shown in panel b corresponds to the number of rats assigned in electrophysiological and morphological experiments in panels c and d and Figure 3b. *p<0.05 comparing active lever presses at 45 min after cocaine injection between the yoked-saline and cocaine groups using a Tukey's post hoc. (c) In the absence of an acute injection of cocaine, cocaine-extinguished rats exhibited higher A/N in NAcore MSNs than saline-yoked controls. Acute cocaine administration augmented the A/N in both Yoked-saline and Cocaine SA rats at 45 min after injection. *p<0.05 compared with the Yoked-saline/Before cocaine group using a Tukey's post hoc. ⁺p<0.05, compared with Before cocaine within the Cocaine SA group. Traces illustrate the dual-component EPSC (black) and the pharmacologically isolated AMPA (red) and NMDA (green) currents from MSNs in each group. N shown in bars is the number of recorded neurons that are from 4–6 rats in each group. (d) Neither cocaine self-administration nor acute cocaine administration changed the AMPA rectification index. Traces are examples of evoked AMPA-mediated currents with holding voltage from −80 to +40 mV. The insert shows the current–voltage (I–V) relationship for AMPA EPSCs in each group. N shown in bars is the number of recorded neurons that are from 4–6 rats in each group. S: saline-yoked group; C: cocaine-administered group; Pre-coc: before acute cocaine; Post-coc: 45 min after acute cocaine.

Figure 2

The cocaine-induced A/N is differentially regulated by pharmacological inactivation of the PL and VTA. (a) Previous microinjection GABA receptor agonist (B/M) into the PL or VTA did not affect active lever pressing in yoked-saline animals. N in parentheses corresponds to the number of animals used to generate the A/N data in panel b. (b) Inactivation of the VTA, but not the PL, inhibited acute cocaine-induced potentiation in A/N in yoked-saline rats. N in the bar is the number of cells recorded. (c) Inactivation of the PL or VTA or systemic administration of the dopamine antagonists SCH-23390 (SCH; 0.03 mg/kg, ip) plus raclopride (Rac; 0.3 mg/kg, ip) prevented cocaine-induced reinstatement. N in bars corresponds to the number of animals used to generate the A/N data in panel d. (d) PL inactivation facilitated but VTA inactivation or SCH/Rac injection prevented acute cocaine-induced potentiation in A/N in cocaine-extinguished rats. N in bars is the number of cells recorded. Grouped aCSF refers to a combination of control animals microinjected with aCSF in either the VTA or PL plus animals receiving systemic vehicle injections as controls for the SCH/Rac treatment. The data are shown as mean±SEM over the 45 min after acute cocaine or vehicle administration. *p<0.05 comparing the acute cocaine to acute saline within each treatment group within a given panel.⁺p<0.05 comparing the effect of acute cocaine in the PL, VTA, or SCH/Rac groups to the Grouped aCSF group.

Figure 3

Cocaine reinstatement for 45 min increased d_h in MSNs in rats extinguished from cocaine self-administration (for behavioral data, see Figure 1a and b). (a) Example of neuron and dendritic segment quantified for spine morphology. (b) Left, Cocaine SA rats showed a basal increase in mean d_h compared with Yoked-saline animals, and acute cocaine administration further increased d_h only in Cocaine SA rats. Data are shown as mean±SEM. N=number of neurons quantified shown in each bar, obtained from 4–6 rats. Right, the cumulative frequency distribution of spine d_h showing shift to the right 45 min after acute cocaine administration in cocaine-extinguished rats. (c) Left, inactivation of the VTA, but not the PL, prevented the increase in d_h at 45 min after acute cocaine administration in Cocaine SA rats. Right, cumulative distribution plots of d_h demonstrate a shift to the left by inactivation of the VTA and to the right by inactivating the PL compared with the distribution induced by acute cocaine administration in aCSF-microinjected, Cocaine SA rats. Grouped aCSF refers to the pooling of data from animals microinjected with aCSF into either the VTA and PL. ⁺p<0.05 comparing basal d_h (Before cocaine) between Yoked-saline and Cocaine SA. *p<0.05 compared with basal d_h in Cocaine SA rats.