The involvement of adenosine A 1 and A 2A receptors in the motor effects of caffeine is still a matter of debate. In the present study, counteraction of the motor-depressant effects of the selective A 1 receptor agonist CPA and the A 2A receptor agonist CGS 21680 by caffeine, the selective A 1 receptor antagonist CPT, and the A 2A receptor antagonist MSX-3 was compared. CPT and MSX-3 produced motor activation at the same doses that selectively counteracted motor depression induced by CPA and CGS 21680, respectively. Caffeine also counteracted motor depression induced by CPA and CGS 21680 at doses that produced motor activation. However, caffeine was less effective than CPT at counteracting CPA and even less effective than MSX-3 at counteracting CGS 21680. On the other hand, when administered alone in habituated animals, caffeine produced stronger motor activation than CPT or MSX-3. An additive effect on motor activation was obtained when CPT and MSX-3 were coadministered. Altogether, these results suggest that the motoractivating effects of acutely administered caffeine in rats involve the central blockade of both A 1 and A 2A receptors. Chronic exposure to caffeine in the drinking water (1.0 mg/ml) resulted in tolerance to the motor effects of an acute administration of caffeine, lack of tolerance to amphetamine, apparent tolerance to MSX-3 (shift to the left of its 'bell-shaped' dose-response curve), and true crosstolerance to CPT. The present results suggest that development of tolerance to the effects of A 1 receptor blockade might be mostly responsible for the tolerance to the motor-activating effects of caffeine and that the residual motor-activating effects of caffeine in tolerant individuals might be mostly because of A 2A receptor blockade.

The aim of the present study was to evaluate whether, and by means of which mechanisms, the adenosine A2A receptor antagonist SCH 58261 [5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine] exerted neuroprotective effects in a rat model of Huntington's disease. In a first set of experiments, SCH 58261 (0.01 and 1 mg/kg) was administered intraperitoneally to Wistar rats 20 min before the bilateral striatal injection of quinolinic acid (QA) (300 nmol/1 microl). SCH 58261 (0.01 but not 1 mg/kg, i.p.) did reduce significantly the effects of QA on motor activity, electroencephalographic changes, and striatal gliosis. Because QA acts by both increasing glutamate outflow and directly stimulating NMDA receptors, a second set of experiments was performed to evaluate whether SCH 58261 acted by preventing the presynaptic and/or the postsynaptic effects of QA. In microdialysis experiments in naive rats, striatal perfusion with QA (5 mm) enhanced glutamate levels by approximately 500%. Such an effect of QA was completely antagonized by pretreatment with SCH 58261 (0.01 but not 1 mg/kg, i.p.). In primary striatal cultures, bath application of QA (900 microm) significantly increased intracellular calcium levels, an effect prevented by the NMDA receptor antagonist MK-801 [(+)-5-methyl-10,11-dihydro-5H-dibenzo [a,d] cyclohepten-5,10-imine maleate]. In this model, bath application of SCH 58261 (15-200 nm) tended to potentiate QA-induced calcium increase. We conclude the following: (1) the adenosine A2A receptor antagonist SCH 58261 has neuroprotective effects, although only at low doses, in an excitotoxic rat model of HD, and (2) the inhibition of QA-evoked glutamate outflow seems to be the major mechanism underlying the neuroprotective effects of SCH 58261.

Previous studies have demonstrated opposing roles for adenosine A 1 and A 2A receptors in the modulation of extracellular levels of glutamate and dopamine in the striatum. In the present study, acute systemic administration of motoractivating doses of the A 2A receptor antagonist MSX-3 significantly decreased extracellular levels of dopamine and glutamate in the shell of the rat nucleus accumbens (NAc) and counteracted both dopamine and glutamate release induced by systemic administration of motor-activating doses of either the A 1 receptor antagonist CPT or caffeine. Furthermore, exposure to caffeine in the drinking water (1 mg/mL, 14 days) resulted in tolerance to the effects of systemic injection of CPT or caffeine, but not MSX-3, on extracellular levels of dopamine and glutamate in the NAc shell. The present results show: first, the existence of opposite tonic effects of adenosine on extracellular levels of dopamine and glutamate in the shell of the NAc mediated by A 1 and A 2A receptors; second, that complete tolerance to caffeine's dopamine-and glutamatereleasing effects which develops after chronic caffeine exposure is attributable to an A 1 receptor-mediated mechanism. Development of tolerance to the dopamine-releasing effects of caffeine in the shell of the NAc may explain its weak addictive properties and atypical psychostimulant profile.

Adenosine, by acting on adenosine A 1 and A 2A receptors, exerts opposite modulatory roles on striatal extracellular levels of glutamate and dopamine, with activation of A 1 inhibiting and activation of A 2A receptors stimulating glutamate and dopamine release. Adenosine-mediated modulation of striatal dopaminergic neurotransmission could be secondary to changes in glutamate neurotransmission, in view of evidence for a preferential colocalization of A 1 and A 2A receptors in glutamatergic nerve terminals. By using in vivo microdialysis techniques, local perfusion of NMDA (3, 10 lM), the selective A 2A receptor agonist 2-p-(2-carboxyethyl)phenethylamino-5¢-N-ethylcarboxamidoadenosine (CGS 21680; 3, 10 lM), the selective A 1 receptor antagonist 8-cyclopentyl-1,3-dimethylxanthine (CPT; 300, 1000 lM), or the non-selective A 1 -A 2A receptor antagonist in vitro caffeine (300, 1000 lM) elicited significant increases in extracellular levels of dopamine in the shell of the nucleus accumbens (NAc). Significant glutamate release was also observed with local perfusion of CGS 21680, CPT and caffeine, but not NMDA. Co-perfusion with the competitive NMDA receptor antagonist DL-2-amino-5-phosphonovaleric acid (APV; 100 lM) counteracted dopamine release induced by NMDA, CGS 21680, CPT and caffeine. Co-perfusion with the selective A 2A receptor antagonist MSX-3 (1 lM) counteracted dopamine and glutamate release induced by CGS 21680, CPT and caffeine and did not modify dopamine release induced by NMDA. These results indicate that modulation of dopamine release in the shell of the NAc by A 1 and A 2A receptors is mostly secondary to their opposite modulatory role on glutamatergic neurotransmission and depends on stimulation of NMDA receptors. Furthermore, these results underscore the role of A 1 vs. A 2A receptor antagonism in the central effects of caffeine.