To examine the role of mu-opioid receptor subtypes, we assessed the antinociceptive effect of H-Tyr-D-Arg-Phe-beta-Ala-OH (TAPA), an analogue of dermorphin N-terminal peptide in mice, using the tail-flick test. Intracerebroventricularly (i.c.v.) or intrathecally (i.t.) injected TAPA produced potent antinociception with tail-flick as a thermal noxious stimulus. The selective mu(1)-opioid receptor antagonist, naloxonazine (35 mg/kg, s.c.), or the selective mu-opioid receptor antagonist, beta-funaltrexamine, 24 h before testing antagonized the antinociceptive effect of i.t. or i.c.v. TAPA on the response to noxious stimuli. Pretreatment with beta-funaltrexamine completely antagonized the antinociception by both i.c.v. and i.t. administered TAPA and [D-Ala(2), Me-Phe(4), Gly(ol)(5)]enkephalin (DAMGO). Especially in the tail-flick test, pretreatment with naloxonazine produced a marked rightward displacement of the i.t. TAPA dose-response curve for antinociception. Though DAMGO is a highly selective mu-opioid receptor agonist, pretreatment with naloxonazine partially blocked the antinociceptive response to DAMGO after i.c.v., but not after i. t. injection. These results indicate that TAPA can act as a highly selective mu(1)-opioid receptor agonist (notable naloxonazine-sensitive receptor agonist) at not only the supraspinal level, but also the spinal level. These data also reveal different antinociceptive mechanisms for DAMGO and for TAPA.