The mechanisms and sites of action of epibatidine-induced antinociception and side effects are poorly understood. The present study tested the hypothesis that the locus coeruleus is a site of action of epibatidine. Behavioral responses of rats to hindpaw formalin injection were compared after direct administration of epibatidine into the locus coeruleus (LC), and after subcutaneous administration. Different groups of rats were injected with formalin into the rear paw after administration of either ACSF, epibatidine (0.01, 0.06, 0.12, and 0.3mug) into the locus coeruleus or epibatidine (2.5-5mug/kg) subcutaneously. Assessment of pain-related behavior was done by evaluating the incidence of favoring, lifting and licking of the injected paw in the different groups. Abnormal motor behavior was also recorded. Infusion of epibatidine into LC induced analgesia, which was reversed by prior infusion of mecamylamine into LC. Epibatidine into the locus coeruleus resulted in a significant lower pain score in the second phase of the formalin test compared to control rats and was as effective as subcutaneous epibatidine. The analgesic effects of epibatidine were regionally selective in that the administration of epibatidine outside the locus coeruleus area was not analgesic. The every tested dose of epibatidine administered into the locus coeruleus also produced freezing behavior immediately after injection, which was relatively short-lived compared to the analgesic effect. Freezing was inhibited by administration of mecamylamine into the LC. Together the results implicate the LC as a target for the analgesic effects of epibatidine.