Intracerebroventricular (i.c.v.) injection of histidyl-proline diketopiperazine [cyclo (His-Pro)], an active metabolite of thyrotropin-releasing hormone (TRH) in mice produced an antinociceptive effect in a dose-dependent manner as measured in four antinociceptive tests; tail-pressure, tail-flick, hot-plate and acetic acid writhing. The antinociceptive effect of cyclo (His-Pro) reached a maximum at 5 min and lasted for 30 min. The ED50 values for the tests were 760.0 (598.4-965.2), 540.0 (442.6-658.8), 595.0 (487.7-725.9) and 370.0 (286.8-477.3) nmol/mouse and the slope functions were 1.61, 1.56, 1.57 and 1.66, respectively. Pretreatment with naloxone (0.5, 2 and 8 mg/kg), an opioid antagonist, administered subcutaneously antagonized the antinociceptive effect of cyclo (His-Pro). TRH injected i.c.v. to mice also exerted a dose-dependent antinociceptive action as measured in three antinociceptive tests; tail-pressure, hot-plate and acetic acid writhing. The antinociceptive effect of TRH was only seen at 5 min. The ED50 values for each test were 112.0 (47.5-264.3), 19.2 (10.4-35.5) and 0.2 (0.1-0.4) nmol/mouse and the slope functions were 8.89, 4.14 and 3.94, respectively. TRH was without effect in the tail-flick test. In contrast to cyclo (His-Pro), TRH-induced antinociception was not antagonized by pretreatment with naloxone (0.5, 2 and 8 mg/kg). The data suggest that the two peptides have a different mechanism of antinociceptive action in relation to the involvement of the opioid system in the central nervous system.