It has been proposed that cannabinoids act at a Gi protein-coupled receptor to produce antinociception. One action of Gi-proteins is to decrease intracellular cAMP via inhibition of adenylyl cyclase activity. Although cannabinoid inhibition of forskolin-stimulated adenylyl cyclase is used as a confirmation of functional cannabinoid receptors, it is unknown whether this second messenger system specifically mediates cannabinoid-induced antinociception. This in vivo study was conducted using enantiomeric cAMP analogs, Rp-cAMPS (an antagonist) and Sp-cAMPS (an agonist), and the cAMP agonist Cl-cAMP to test the hypothesis that cannabinoid-induced antinociception is due to decreased adenylyl cyclase activity. None of the cAMP analogs, forskolin, or 1,9-dideoxy-forskolin affected delta 9-THC or CP-55,940-induced antinociception produced by intrathecal (i.t.) or intracerebroventricular (i.c.v.) injections in mice. Experiments were also conducted to investigate whether i.c.v. administration of Sp-cAMPS would block i.c.v. cannabinoid-induced antinociception in rats. Sp-cAMPS failed to block CP-55,940-induced antinociception. However, Sp-cAMPS produced hyper-excitability and reactive behavior indicating that it did elicit a pharmacological effect. Although, adenylyl cyclase may mediate other cannabinoid-induced actions, these results do not support the hypothesis that it is involved in cannabinoid-induced antinociception. Alternatively, other effector systems such as calcium or potassium channels coupled to cannabinoid receptors may mediate cannabinoid-induced antinociception.