The effects of dextrometorphan and its metabolite dextrorphan on nicotine-induced antinociception in two acute thermal pain assays after systematic administration were evaluated in mice and compared with that of mecamylamine. Dextrometorphan and dextrorphan were found to block nicotine's antinociception in the tail-flick and hot-plate tests with different potencies (dextrometorphan is 10 times more potent than its metabolite). This blockade was not due to antagonism of N-methyl-d-aspartate receptors and/or interaction with opiate receptors, since selective drugs of these receptors failed to block nicotine's analgesic effects. Our results with the tail-flick and hot-plate tests showed an interesting in vivo functional selectivity for dextrometorphan over dextrorphan. In oocytes expressing various neuronal acetylcholine nicotinic receptors (nAChR), dextrometorphan and dextrorphan blocked nicotine activation of expressed alpha(3)beta(4), alpha(4)beta(2), and alpha(7) subtypes with a small degree of selectivity. However, the in vivo antagonistic potency of dextrometorphan and dextrorphan in the pain tests does not correlate well with their in vitro blockade potency at expressed nAChR subtypes. Furthermore, the apparent in vivo selectivity of dextrometorphan over dextrorphan is not related to its in vitro potency and does suggest the involvement of other mechanisms. In that respect, dextrometorphan seems to behave as another mecamylamine, a noncompetitive nicotinic receptor antagonist with a preferential activity to alpha(3)beta(4)(*) neuronal nAChR subtypes.