Effects of acute cocaine or dopamine receptor agonists on AMPA receptor distribution in the rat nucleus accumbens

Abstract

Changes in α-amino-3-hydroxy-5-methylisoxazole-4-propionate receptor (AMPAR) surface expression in the rodent nucleus accumbens (NAc) are produced by cocaine exposure and implicated in addiction-related behaviors. The direction of change depends on the animal's prior drug history. However, little is known about the effect of a single exposure to cocaine on AMPAR distribution in the NAc of untreated rats. This is essential information for interpreting the literature on AMPAR trafficking after repeated cocaine exposure. In this study, we used a protein crosslinking assay to determine the effect of a single cocaine injection on surface and intracellular AMPAR subunit levels in the rat NAc. We found increased AMPAR surface expression in the NAc 24 h, but not 30 min or 2 h, after cocaine injection. A major effect of cocaine is to increase extracellular dopamine (DA) levels, leading to DA receptor activation. Therefore, we also evaluated the effects of directly acting DA receptor agonists. In contrast to the effects of cocaine, AMPAR surface expression was significantly decreased 24 h after injection of the D2-class agonist quinpirole, whereas no significant effects were produced by the D1-class agonist SKF 81297 or the mixed DA agonist apomorphine. Our results show that the effects of a single cocaine exposure in drug- and injection-naïve rats are distinct from those previously reported after repeated cocaine administration. They further suggest that cocaine exerts these effects by influencing neuronal circuits rather than simply stimulating NAc DA transmission.

Fig. 1

A single exposure to cocaine (15 mg/kg or 30 mg/kg) produced a significant increase in locomotor activity. Data are expressed as the average number of beam breaks per 5 min interval after saline or cocaine injection. Saline- and cocaine-injected rats destined for tissue collection at different time-points (30 min, 2 h and 24 h) were pooled for presentation of these locomotor activity data (see Methods for N values for each group) and the arrow indicates time of injection. Both cocaine-injected groups showed a significant increase in locomotor activity compared to saline-injected controls, but the locomotor response to 15 mg/kg and 30 mg/kg cocaine did not differ, probably due to the presence of stereotypy in the 30 mg/kg group as well as other variables discussed in Results.

Fig. 2

The GluA1 surface/intracellular ratio (S/I) in the NAc was increased 24 h after injection of 15 mg/kg or 30 mg/kg cocaine. NAc tissue was collected 30 min, 2 h or 24 h after injection of 15 mg/kg cocaine, 30 mg/kg cocaine, or saline. Data are normalized to the corresponding saline control group for each tissue collection time. The S/I ratio for GluA1 was significantly increased in both cocaine-injected groups 24 h after drug exposure (*p<0.05). Surface levels were significantly increased after 15 mg/kg cocaine, but did not reach significance for the 30 mg/kg cocaine group (#p=0.08, one-way t-test). No changes in GluA1 distribution were observed 30 min or 2 h after cocaine injection.

Fig. 3

GluA2 distribution was not significantly altered by 15 or 30 mg/kg cocaine at any time-point, although modest trends towards increased surface GluA2 were observed 24 h after cocaine injection. NAc tissue was collected 30 min, 2 h or 24 h after injection of 15 mg/kg cocaine, 30 mg/kg cocaine, or saline. Data are normalized to the corresponding saline control group at each tissue collection time. S/I, surface/intracellular.

Fig. 4

Administration of the D1-class agonist SKF 81297 (SKF, 0.3 or 3 mg/kg) did not alter GluA1 or GluA2 distribution in the NAc 24 h after drug injection. Panel A shows the average number of beam breaks per 5 min interval after saline or SKF injection. The arrow indicates time of injection. Only the higher dose of SKF produced a significant increase in locomotor activity compared to saline-injected controls. NAc tissue was collected 24 h after injection. No significant differences were found for GluA1 (B) or GluA2 (C) distribution in the NAc, although the lower dose of SFK produced small trends towards increases in all GluA1 measures as well as GluA2 surface expression.

Fig. 5

Administration of a low dose of the D2-class agonist quinpirole (Quin, 0.3 mg/kg) significantly decreased the GluA1 surface/intracellular (S/I) ratio 24 h after drug injection and produced the same trend for GluA2. Panel A shows the average number of beam breaks per 5 min interval after saline or quinpirole injection. The arrow indicates time of injection. The locomotor response to Quin did not differ significantly from the response to saline, although some behavioral changes were observed in response to the higher drug dose (see Results). A significant decrease in the S/I ratio 24 h after injection of 0.3 mg/kg Quin was observed for GluA1 (*p<0.05) and a strong trend was observed for GluA2 (C; #p=0.07). Trends towards decreased surface levels of GluA1 and GluA2 were also observed (B and C). No significant changes were found after injection of the higher dose of quinpirole (Quin, 3 mg/kg, B and C).

Fig. 6

Administration of the mixed D1/D2-class agonist apomorphine (Apo, 3 mg/kg) did not alter GluA1 or GluA2 distribution in the NAc 24 h after drug injection. Panel A shows the average number of beam breaks per 5 min interval after saline or Apo injection. The arrow indicates time of injection. The locomotor response to Apo was significantly greater than the response to saline injection. No significant differences in GluA1 (B) or GluA2 (C) distribution were found.