The antinociceptive effect of caffeine was examined by using the hot-p
late, abdominal constriction tests in mice and the tail flick and paw-
pressure tests in rats. Caffeine (1-5 mg kg(-1) s.c. in mice; 2.5-5 mg
kg(-1) i.p. in rats) produced significant antinociception in both spe
cies which was prevented by atropine (5 mg kg(-1) i.p.), pirenzepine (
0.1 mu g per mouse. i.c.v.), hemicholinium-3 (1 mu g/ mouse i.c.v.) an
d N-6-cyclopentyladenosine (5 mu g/mouse i.c.v.), but not by naloxone
(1 mg kg(-1) i.p.), CGP 35348 (100 mg kg(-1) i.p.), alpha-methyl p-tyr
osine (100 mg kg(-1) i.p.) and reserpine (2 mg kg(-1) i.p.). Intracere
broventricular injection of caffeine in mice at doses (2.5-5 mu g per
mouse) which were largely ineffective by parenteral routes, induces an
antinociception whose intensity equalled that obtainable s.c. or i.p.
In the antinociceptive dose-range, caffeine did not produce any behav
ioural impairment as revealed by the rotarod and Irwing tests. On the
basis of the above data, it can be postulated that caffeine exerts an
antinociceptive effect mediated by central amplification of cholinergi
c transmission.