Computational modeling of in vivo COMT manipulations reveals dopamine clearance-dependent synaptic weight modifications in the mPFC during learning

RAPANELLI MAXIMILIANO; LEW SERGIO; FRICK LUCIANA ROMINA; ZANUTTO SILVANO

New Orleans, LA, USA

Congreso; Annual Meeting of the Society for Neuroscience; 2012

Society for Neuroscience.

Dopamine (DA) codifies reward and its prediction in reinforcement learning. Catechol-o-methyl Transferase (COMT) activity in the medial Prefrontal Cortex (mPFC) has been shown to influence cognition by modifying DA clearance. Nevertheless, it is unknown how DA clearance influences synaptic plasticity in the mPFC and learning an operant conditioning task. Systemic Entacapone (50 mg/kg), as well as local Entacapone (10 nM) and recombinant COMT (17 µg) in the mPFC were administered to Long Evans rats prior to training in the operant conditioning task. We found that Entacapone significantly improves learning performance. Conversely, recombinant COMT administration showed to totally impair learning. These data were included in a computational model that uses the temporal difference (TD) to simulate the phasic firing of dopaminergic neurons. A gating window lasting T ms models the DA postsynaptic effects. Gating window of 800 ms, 200 ms and 500 ms were selected for Entacapone, recombinant COMT and control groups respectively, and these values were used to simulate learning and synaptic modifications within the mPFC. The model predicted an improved performance when Entacapone effects were simulated, whereas for recombinant COMT, a detrimental effect on learning performance was observed. Synaptic weight changes in the mPFC account for the effects of DA on learning performance, changing faster in the Entacapone group than in the recombinant COMT, which revealed diminished synaptic weights changes.  Our findings provide evidence that support DA clearance in the mPFC as a mechanism that controls learning by modulating the synaptic plasticity window during acquisition of an operant task.