1. Dipsacus saponin C (DSC) administered intracerebroventricularly (i.c.v.) showed an antinociceptive effect in a dose-dependent (from 3.75 to 30 micrograms) manner as measured by the tailflick assay. The antinociception induced by DSC at the dose of 30 micrograms was maintained at least 1 h. 2. Sulfated cholecystokinin (CCK, from 0.1 to 0.5 ng); muscimol (a GABAA receptor agonist, from 50 to 200 ng); MK-801 [(+/-)-5-methyl-10, 11-dihydro-5H-dibenzo (a,d) cyclohepten-5, 10-imine maleate, from 0.1 to 1 microgram], a noncompetitive N-methyl-D-aspartic acid (NMDA) receptor antagonist; or CNQX (6-cyano-7-nitroquinoxaline-2,3-dione, from 0.1 to 0.5 microgram), a non-NMDA receptor antagonist, injected i.c.v. significantly reduced the inhibition of the tail-flick response induced by DSC (30 micrograms) administered i.c.v. However, naloxone (an opioid receptor antagonist, 2 micrograms) or baclofen (a GABAB receptor antagonist, 10 ng) did not affect the inhibition of the tail-flick response induced by DSC. 3. The intrathecal (i.t.) injection of yohimbine (an alpha 2-adrenergic receptor antagonist, from 5 to 20 micrograms) and methysergide (a serotonin receptor antagonist, from 5 to 20 micrograms) but not naloxone (from 2 to 8 micrograms), significantly attenuated inhibition of the tail-flick response induced by DSC (30 micrograms) administered i.c.v. 4. Our results suggest that DSC has an antinociceptive effect when it is administered supraspinally and GABAA, NMDA and non-NMDA receptors, but not opioid and GABAB receptors located at the supraspinal level, may be involved in DSC-induced antinociception. Furthermore, DSC administered supraspinally may produce antinociception by stimulating descending alpha 2-adrenergic and serotonin pathways but not the opioidergic pathway.