The present study was designed to investigate which subtypes of spinal 5-HT receptors are involved in 5-HT-induced antinociception using the mechanical pain test. Serotonin and various selective antagonists or agonists for 5-HT receptor subtypes (5-HT(1A), 5-HT(1B), 5-HT(2A), 5-HT(2C), 5-HT(3) and 5-HT(4)) were administered intrathecally (i.t.) in rats. The i.t. injection of 5-HT (1 microg) produced significant antinociceptive effects using the paw pressure test. Pretreatment with the 5-HT(2C) receptor antagonist mesulergine (1 and 10 microg) and the 5-HT(3) receptor antagonist tropisetron (1 and 10 microg) reversed totally the antinociception induced by 5-HT. Furthermore, at a dose of 10 microg, both the 5-HT(2A) receptor antagonist ketanserin and the 5-HT(1B) receptor antagonist penbutolol, but neither the 5-HT(1A) receptor antagonist WAY 100635 nor the 5-HT(4) receptor antagonist GR113808, attenuated the antinociceptive effect induced by 5-HT. In addition, an i.t. injection of the 5-HT(3) agonist mCPBG induced significant antinociceptive effects whereas the 5-HT(2) agonist DOI did not produce analgesia. These results suggest that although the precise degree of the involvement of spinal serotonergic 5-HT(3) receptors remains to be elucidated due to some differences in the effect of agonists or antagonists, these receptors seem to play a role in the antinociceptive effect of 5-HT against a mechanical acute noxious stimulus. The involvement of 5-HT(2C) is more questionable due to the observed discrepancies between the effects of the used agonist and antagonist. 5-HT(1A) and 5-HT(4) receptors do not seem to be involved. In addition, a possible functional interaction between spinal serotonergic receptors may exist.