Assessment of a cholinergic contribution to chlordiazepoxide-induced deficits of place learning in the Morris water maze.

This investigation sought to characterize the interaction between benzodiazepine and cholinergic systems in place learning in the Morris water maze. In the first experiment, rats were treated with scopolamine (1 mg/kg) alone or concomitantly with one of two doses of flumazenil (15 and 30 mg/kg) or with chlordiazepoxide (5 mg/kg) alone or concomitantly with flumazenil (15 mg/kg). Chlordiazepoxide and scopolamine severely impaired place learning but not cue learning. The low dose of flumazenil completely reversed the impairment produced by chlordiazepoxide and both high and low doses of flumazenil attenuated the place learning deficit produced by scopolamine. Neither dose of flumazenil affected place learning when administered alone. In the second experiment, rats were administered chlordiazepoxide (5 mg/kg) or scopolamine (1 mg/kg) alone or concomitantly with one of four doses of physostigmine (0.05, 0.10, 0.25, and 0.5 mg/kg). Once again, both chlordiazepoxide and scopolamine impaired place but not cue learning. Physostigmine reversed the impairment produced by scopolamine in a dose-dependent manner but failed at every dose to attenuate the impairment produced by chlordiazepoxide. The higher doses of physostigmine impaired place learning when administered alone. None of the drug treatments impaired cue learning. Together, these results suggest that the scopolamine-induced impairment of place learning is due to an increase in benzodiazepine/GABA activity, and contradict the notion that benzodiazepines impair memory by cholinergic mechanisms.