Caffeine treatment and withdrawal in mice: relationships between dosage, concentrations, locomotor activity and A1 adenosine receptor binding.

Relationships between caffeine dose, methylxanthine tissue concentrations, adenosine receptor binding and locomotor activity were examined in CD-1 mice. A method of caffeine infusion via s.c. pumps provided constant steady-state methylxanthine concentrations. Mice receiving caffeine doses of 97 mg/kg/day (with mean plasma concentration of 2.7 micrograms/ml) demonstrated motor activity depression for 6 days after pump implantation (vs. vehicle-treated controls). Mice receiving caffeine doses of 194 mg/kg/day (mean plasma concentration of 7.1 micrograms/ml) demonstrated motor stimulation 4 and 24 hr after implantation. Mice receiving this dose for 6 days developed motor depression. A reduction in the stimulant effects of acute caffeine (20 mg/kg i.p.) was found in mice receiving caffeine infusions (194 mg/kg/day for 6 days) as compared to those receiving vehicle infusions, suggestive of drug tolerance. These dose- and time-dependent behavioral effects during caffeine-infusion were associated with decreases between 20 and 69% in specific binding of A1 adenosine radioligand 1,3-[3H]dipropyl-8-cyclopentylxanthine in vivo. Behavioral alterations during caffeine infusion appear to be mediated by A1 adenosine receptor occupancy. Increasing motor depression developed on days 1 and 2 after pump removal with values returning to control levels by days 4 and 6. Behavioral alterations were associated with in vivo binding increases of 98 and 324%, respectively, and a return to control values on days 4 and 6. In vivo binding alterations were not associated with ex vivo A1 receptor binding changes. Caffeine tolerance and withdrawal effects in this animal model appear to be mediated by chronic occupancy of A1 adenosine receptors. The behavioral and in vivo receptor binding alterations observed after caffeine discontinuation follow a time course similar to caffeine withdrawal in humans.