D-cycloserine reduces the context specificity of pavlovian extinction of cocaine cues through actions in the nucleus accumbens.

Extinction therapy has been proposed as a method to reduce the motivational impact of drug-associated cues to prevent relapse. Cue extinction therapy, however, takes place in a novel context (e.g., treatment facility), and is unlikely to be effective due to the context specificity of extinction. We tested the hypothesis that d-cycloserine (DCS), which enhances extinction in other procedures, would enhance extinction of cocaine-associated cues in a novel context to reduce cue-induced reinstatement. Male Sprague Dawley rats were trained to self-administer cocaine associated with a cue. The cue was later extinguished in the drug-taking context (context A) or a novel context (context B) using a Pavlovian cue extinction procedure designed to mimic human cue exposure therapy. DCS was administered systemically or into a specific brain region immediately following the cue extinction sessions to enhance the consolidation of extinction learning. We demonstrate that DCS given postextinction session in context B reduces reinstatement in context A, indicating a reduction in the context specificity of extinction learning. The effect of systemic DCS was recapitulated by administration of DCS into the nucleus accumbens core, but not in the basolateral amygdala, dorsal hippocampus, infralimbic or prelimbic prefrontal cortex. DCS treatment caused a reduction in cue-induced reinstatement only when it was given after cue extinction sessions, and not when given 1) in the absence of extinction or 2) after a brief memory reactivation session. A pharmacological method that can render extinction context independent may provide an innovative method to reduce cue-induced relapse in addicts and to study the neurobiology of addiction.