Cocaine withdrawal reduces group I mGluR-mediated long-term potentiation via decreased GABAergic transmission in the amygdala

Cocaine relapse can occur when cocaine-associated environmental cues induce craving. Conditioned place preference (CPP) is a behavioral paradigm modeling the association between cocaine exposure and environmental cues. The amygdala is involved in cocaine cue associations with the basolateral amygdala (BLA) and central amygdala (CeA) acting differentially in cue-induced relapse. Activation of metabotropic glutamate receptors induces synaptic plasticity, the mechanism of which is thought to underlie learning, memory and drug-cue associations. The goal of this study was to examine the neural alterations in responses to group I metabotropic glutamate receptor (mGluR) agonists in the BLA to lateral capsula of CeA (BLA-CeLc) pathway in slices from rats exposed to cocaine-CPP conditioning and withdrawn for 14 days. mGluR1, but not mGluR5, agonist-induced long-term potentiation (mGluR1-LTP) in the BLA-CeLc pathway was reduced in rats withdrawal from cocaine for 2 and 14 days, and exhibited an altered concentration response to picrotoxin. Cocaine withdrawal also reduced γ-aminobutyric acid (GABA)ergic synaptic inhibition in CeLc neurons. Blocking cannabinoid receptor 1 (CB(1) ) reduced mGluR1-LTP in the saline-treated but not cocaine-withdrawn group. Response to CB(1) but not CB(2) agonist was altered after cocaine. Additionally, increasing endocannabinoid (eCB) levels abolished mGluR1-LTP in the saline but not cocaine-withdrawn group. However, CB(1) and CB(2) protein levels were increased in the amygdala of cocaine-withdrawn rats while mGluR1 and mGluR5 remained unchanged. These data suggested that the mechanisms underlying the diminished mGluR1-LTP in cocaine-withdrawn rats involve an altered GABAergic synaptic inhibition mediated by modulation of downstream eCB signaling. These changes may ultimately result in potentiated responses to environmental cues that would bias behavior toward drug-seeking.