1. The purpose of the present study was to develop an experimental strategy for the quantification of the cardiovascular effects of non-selective adenosine receptor ligands at the adenosine A1 and A2a receptor in vivo. 2-Chloroadenosine (CADO) was used as a model compound. 2. Three groups of normotensive conscious rats received an short intravenous infusion of 1.4 mg kg-1 CADO during constant infusions of the A1-selective antagonist, 8-cyclopentyltheophylline (CPT; 20 micrograms min-1 kg-1), the A2a-selective antagonist, 8-(3-chlorostyryl) caffeine (CSC; 32 micrograms min-1 kg-1) or the vehicle. The heart rate (HR) and mean arterial blood pressure (MAP) were recorded continuously during the experiment and serial arterial blood samples were taken for analysis of drug concentrations. The ratio MAP/HR was also calculated, which may reflect changes in total peripheral resistance on the assumption that no changes in stroke volume occur. 3. During the infusion of CPT, CADO produced a reduction in both blood pressure and MAP/HR by activation of the A2a receptor. The concentration-effect relationships were described according to the sigmoidal Emax model, yielding potencies based on free drug concentrations (EC50,u) of 61 and 68 ng ml-1 (202 and 225 nM) for the reduction of blood pressure and MAP/HR, respectively. During the infusion of CSC, an EC50,u value of 41 ng ml-1 (136 nM) was observed for the A1 receptor-mediated reduction in heart rate. The in vivo potencies correlated with reported receptor affinities (Ki(A1) = 300 nM and Ki(A2a) = 80 nM). The maximal reductions in MAP/HR and heart rate were comparable to those of full agonists, with the Emax values of -12 +/- 1 x 10(-2) mmHg b.p.m.-1 and -205 b.p.m. respectively. 4. It is concluded that this integrated pharmacokinetic-pharmacodynamic approach can be used to obtain quantitative information on the potency and intrinsic activity of new non-selective adenosine receptor agonists at different receptor subtypes in vivo.