The locomotor effects in mice of selective A1 and A2 adenosine agonists, antagonists and combinations of agonists were investigated using a computerized activity monitor. The A2-selective agonist 2-[(2-aminoethylamino)carbonylethylphenylethylamino[-5'-N- ethylcarboxamidoadenosine (APEC), an amine derivative of 2-(carboxyethylphenylethylamino)adenosine-5'-carboxamide, was a more potent locomotor depressant than its amide conjugates. The rank order of potency after i.p. injection for adenosine agonists was 5'-N-ethylcarboxamidoadenosine (NECA) (ED50, 5.8 nmol/kg) greater than APEC (ED50, 25 nmol/kg) greater than N6-cyclohexyladenosine (CHA) (ED50, 270 nmol/kg). An A1-selective, centrally acting, adenosine antagonist, 8-cyclopentyltheophylline (10 mg/kg), completely reversed the locomotor depressant effects of CHA (A1-selective) and NECA (nonselective) at doses of agonists as high as twice the ED50, and shifted the dose-response curves to the right, suggesting a primary involvement of A1 receptors. 8-cyclopentyltheophylline did not affect the depressant effects of APEC at the ED50, consistent with the A2-selectivity of APEC. The locomotor effects of APEC and CHA were completely reversed by theophylline, but not by the peripherally active 8-p-sulfophenyltheophylline, indicating central action of the adenosine agonists. The depressant effects of APEC, but not of NECA or CHA, were reversed significantly by an A2-selective adenosine receptor antagonist, 4-amino-8-chloro-1-phenyl-[1,2,4]triazol[4,3-a]quinoxaline. Low or subthreshold doses of CHA potentiated the depressant effects of APEC. A subthreshold dose of CHA did not alter the depressant effect of NECA, whereas a subthreshold dose of APEC increased the depressant effects of low doses of NECA. Thus, it appears that A1- and A2-selective adenosine agonists have separate central depressant effects, which can be potentiative.(ABSTRACT TRUNCATED AT 250 WORDS)