Dopamine D1 receptor agonist treatment attenuates extinction of morphine conditioned place preference while increasing dendritic complexity in the nucleus accumbens core

- D1R agonist given after extinction training attenuates extinction of morphine CPP.
- D1R agonist given after extinction increases dendritic complexity in accumbens core.
- D1R agonist effect on accumbal dendritic morphology depends on previous conditioning.

The dopamine D1 receptor (D1R) has a role in opioid reward and conditioned place preference (CPP), but its role in CPP extinction is undetermined. We examined the effect of D1R agonist SKF81297 on the extinction of opioid CPP and associated dendritic morphology in the nucleus accumbens (NAc), a region involved with reward integration and its extinction. During the acquisition of morphine CPP, mice received morphine and saline on alternate days; injections were given immediately before each of eight daily conditioning sessions. Mice subsequently underwent six days of extinction training designed to diminish the previously learned association. Mice were treated with either 0.5 mg/kg SKF81297, 0.8 mg/kg SKF81297, or saline immediately after each extinction session. There was a dose-dependent effect, with the highest dose of SKF81297 attenuating extinction, as mice treated with this dose had significantly higher CPP scores than controls. Analysis of medium spiny neuron morphology revealed that in the NAc core, but not in the shell, dendritic arbors were significantly more complex in the morphine conditioned, SKF81297-treated mice compared to controls. In separate experiments using mice conditioned with only saline, SKF81297 administration after extinction sessions had no effect on CPP and produced differing effects on dendritic morphology. At the doses used in our experiments, SKF81297 appears to maintain previously learned opioid conditioned behavior, even in the face of new information. The D1R agonist's differential, rather than unidirectional, effects on dendritic morphology in the NAc core suggests that it may be involved in encoding reward information depending on previously learned behavior.