Distinct Role of Dopamine in the PFC and NAc During Exposure to Cocaine-Associated Cues

Abstract

Background: Dopamine neurotransmission plays a critical role in reward in drug abuse and drug addiction. However, the role of dopamine in the recognition of drug-associated environmental stimuli, retrieval of drug-associated memory, and drug-seeking behaviors is not fully understood.

Methods: Roles of dopamine neurotransmission in the prefrontal cortex (PFC) and nucleus accumbens (NAc) in the cocaine-conditioned place preference (CPP) paradigm were evaluated using in vivo microdialysis.

Results: In mice that had acquired cocaine CPP, dopamine levels in the PFC, but not in the NAc, increased in response to cocaine-associated cues when mice were placed in the cocaine chamber of an apparatus with 2 separated chambers. The induction of the dopamine response and the development of cocaine CPP were mediated through activation of glutamate NMDA (N-methyl-D-aspartate)/AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptor signaling in the PFC during conditioning. Activation of dopamine D1 or D2 receptor signaling in the PFC was required for cocaine-induced locomotion, but not for the induction of the dopamine response or the development of cocaine CPP. Interestingly, dopamine levels in the NAc increased in response to cocaine-associated cues when mice were placed at the center of an apparatus with 2 connected chambers, which requires motivated exploration associated with cocaine reward.

Conclusions: Dopamine neurotransmission in the PFC is activated by the exposure to the cocaine-associated cues, whereas dopamine neurotransmission in the NAc is activated in a process of motivated exploration of cues associated with cocaine reward. Furthermore, the glutamate signaling cascade in the PFC is suggested to be a potential therapeutic target to prevent the progression of drug addiction.

Keywords: Conditioned place preference; D1 receptor; D2 receptor; glutamate; microdialysis.

Figure 1.

Dopamine (DA) responses to saline and cocaine injections in additional place conditioning (day 7) and to saline- and cocaine-associated cues (day 8) in the nucleus accumbens (NAc) and prefrontal cortex (PFC). (A) Protocols for the conditioned place preference (CPP) to cocaine and subsequent measurements of DA. Following the pretest of chamber preference (day 0), saline and cocaine (7.5 mg/kg i.p.) were injected daily, and mice were placed in saline- and cocaine-paired chambers, respectively, for cocaine conditioning for 3 days (day 1–3) using the 2-connected chamber apparatus with a shielding plate between chambers. CPP was tested on day 4 using the 2-connected chamber apparatus without a shielding plate between chambers, and only the mice showing a preference for the cocaine-associated chamber were used for the following analyses. Additional cocaine conditioning for the developed CPP to cocaine was conducted on day 7 using the 2-separated chamber apparatus consisting of saline and cocaine chambers. The extracellular DA level was recorded from 2 to 3 hours before saline conditioning until 2 hours after the cocaine conditioning. Mice were kept in their home cage throughout the dopamine recording except for when they were placed in the saline or cocaine chamber for 20 minutes, where mice received injections. The saline or cocaine chamber has the same contexts as each paired chamber used for conditioning. On day 8, extracellular DA was subsequently recorded when mice were exposed to the saline or cocaine chambers without injections for 20 minutes, which was considered a cue. (B) CPP scores recorded on day 0 and day 4. CPP scores on day 4 of the mice used and unused for microdialysis experiments were shown as closed and open circles, respectively. (C–J) Extracellular levels of DA were measured in the NAc (C–F) and PFC (G–J) using in vivo microdialysis. Effects of the injections of saline (C, G) and cocaine (7.5 mg/kg, i.p.) (E, I) for additional cocaine conditioning (day 7) and the saline (D, H)- and cocaine (F, J)-associated cues (day 8) on DA levels were determined. Exposure of mice to the cues is indicated with black bars. Data represent mean ± SEM. The number of mice is indicated in each graph. *P < .05, **P < .01, ***P < .001 vs the basal levels of DA; mixed linear models.

Figure 2.

Effects of antagonist infusion into the prefrontal cortex (PFC) during cocaine conditioning on the locomotor response and conditioned place preference (CPP) to cocaine. (A) Daily injections of saline and cocaine (7.5 mg/kg) for conditioning were conducted as described in Figure 1A. Surgery for microdialysis probe implantation was conducted on day 0 following the pretest of chamber preference. Pharmacological and chemogenetic manipulation was performed during conditioning (day 1–3), and the CPP test was performed on day 4. (B–C) Pharmacological ligands (a D1 receptor antagonist, SCH23390 [10 µM]; a D2 receptor antagonist, raclopride [10 µM]; ionotropic glutamate antagonists, CNQX [10 µM] plus 3-2-Cpp [30 µM]) were infused into the PFC during conditioning (day 1–3). The locomotor response to saline or cocaine injection was evaluated in paired chambers using the 2-connected chamber apparatus with a shielding plate on day 1 and day 3 (B), and cocaine CPP was evaluated using the 2-connected chamber apparatus without a shielding plate on day 4 (C). (D, E) For chemogenetic inhibition of D1 receptor-expressing neurons in the PFC, saline and CNO (1 mg/kg, i.p.) were administered 30 minutes before the saline-paired place conditioning and cocaine-paired place conditioning (day 1–3), respectively, in [Drd1]-Cre mice injected with Gi-DREADD virus (rAAV-hsyn-DIO-rM4D(Gi)-mCherry) or control virus (rAAV2-Ef1a-DIO-mCherry) into the PFC. The locomotor response to saline injection or cocaine injection after CNO administration (C/CNO) was evaluated on day 1 and day 3 (D), and cocaine CPP was evaluated on day 4 (E). Data represent mean ± SEM. The number of mice is indicated under each experimental group. *P < .05, **P < .01, ***P < .001 as indicated; 1-way ANOVA and Bonferroni post hoc test. ^###P < .001 vs Ringer’s control; mixed effect models. ^§P < .05, ^§§§P < .001 vs CPP score in pretest, Welch’s t test.

Figure 3.

Dopamine (DA) responses to the saline- and cocaine-associated cues in the PFC on day 4 after cocaine place conditioning with pharmacological manipulation. In another group of animals, DA levels were determined on day 4 by in vivo microdialysis following cocaine conditioning as shown in Figure 2A. DA responses to the saline (A–D)- and cocaine (E–H)-associated cues in mice that received infusion of Ringer’s solution (A, E), the D1 receptor antagonist SCH23390 (10 µM) (B, F), the D2 receptor antagonist raclopride (10 µM) (C, G), or the ionotropic glutamate receptor antagonists CNQX (10 µM) plus 3-2-Cpp (30 µM) (D, H) into the PFC during conditioning (day 1–3). Exposure of mice to the cues is indicated with black bars. Data represent mean ± SEM. The number of mice is indicated in each graph. *P < .05, **P < .01, ***P < .001 vs the basal levels of DA; mixed linear models.

Figure 4.

The dopamine (DA) response to the cocaine-associated cues in the NAc and PFC using the 2-connected chamber apparatus. (A) Daily injections of saline (am) and saline or cocaine (7.5 mg/kg) (pm) for conditioning were conducted. Surgery for microdialysis probe implantation was conducted on day 0 following the pretest of chamber preference. DA levels were determined on day 4 by in vivo microdialysis. Mice were conditioned with saline in 1 chamber and cocaine in the other chamber (the cocaine CPP group: black circles) or with saline in both chambers (the control group: open circles) using the 2-connected chamber apparatus with a shielding plate between chambers for 3 days (day 1–3). (B) In different experimental groups subjected to saline (am) and saline (pm) for conditioning, CPP was tested on day 4. The cocaine conditioning procedure was shown to induce the cocaine CPP (Figure 1B). The DA levels in the NAc (C) or PFC (D) were determined on day 4, when mice were placed at the center of the 2-connected chamber apparatus without a shielding plate and allowed to explore both the saline- and cocaine-associated cues. Exposure of mice to the cues is indicated with a black bar. Data represent mean ± SEM. The number of mice is indicated in the graph. *P < .05, **P < .01, ***P < .001 vs the basal levels of DA; mixed linear models.

Figure 5.

Schematic representation of the distinct role of the dopamine pathways in the prefrontal cortex (PFC) and nucleus accumbens (NAc) in the cocaine conditioned place preference (CPP). This study demonstrates that the dopamine (DA) response in the PFC plays an important role in the recognition of cocaine-associated cues in the cocaine CPP, whereas the DA response in the NAc is related to motivated exploration of cocaine reward. Ionotropic glutamate receptor signaling in the PFC is activated during the cocaine conditioning, leading to the development of cocaine CPP and the induction of the DA response to the cocaine-associated cues in the PFC. The DA responses in the PFC did not differ between 2-separated and 2-connected chambers. In contrast, dopamine in the NAc is responsive to the cocaine-associated cues when 2-connected chambers, but not 2-separated chambers, are used, suggesting that DA in the NAc plays important roles in the exploratory process of the cocaine-associated cues. DA in the PFC also responded to the saline-associated cues via mechanisms involving ionotropic glutamate receptor and DA D1 receptor signaling. In addition, the locomotor response to cocaine is mediated through activation of DA D1 and D2 receptor signaling in the PFC.