Caffeine antagonizes several central effects of diazepam

In spinal cats, caffeine (3–30 mg·kg^?1 i.v.) reduced the increase of dorsal root potentials (DRPs) caused by diazepam (0.1–1 mg·kg^?1 i.v.) without affecting the prolongation of DRPs evoked by phenobarbitone (10–20 mg·kg^?1 i.v.). Caffeine antagonized the depression by diazepam, but not that by phenobarbitone, of the ventral root-evoked Renshaw cell discharge. In unrestrained cats, 50 mg·kg^?1 caffeine i.p. abolished the elevation induced by 1 mg·kg^?1 diazepam i.p. of the threshold for eliciting a rage reaction by stimulation of the lateral hypothalamus, but was ineffective against the threshold increase caused by 20 mg·kg^?1 phenobarbitine i.p. In the horizontal wire test in mice, caffeine was more potent in reversing the depression of performance induced by diazepam that that by phenobarbitone (ED50 1.8 mg·kg^?1 and 139 mg·kg^?1 p.o., respectively). The reduction of skeletal muscle tone in mice produced by diazepam was antagonized by low doses of caffeine (ED50 0.53 mg·kg^?1 p.o.). While caffeine at low doses (0.3-3 mg·kg^?1 p.o.) abolished the anticonflict effect of diazepam in rats, high doses (ED50 160 mg·kg^?1 p.o.) were necessary to antagonize the anticonvulsant effect of diazepam on pentylene-tetrazole-induced seizures in mice. The interaction between caffeine and diazepam is not due to a competition at the benzodiazepine receptors but may involve purinergic mechanisms.