Contribution of extracellular adenine nucleotide degradation to adenosine formation and internal salvage of adenosine via adenosine kinase were quantified in macrovascular porcine endothelial cells. Microcarrier beads covered with endothelial cells were kept in a perfusion column at a flow rate of 2 ml/min. Total adenine nucleotide (AN) release was quantified with a sensitive firefly luciferin-luciferase assay after enzymatic rephosphorylation of AMP and ADP to ATP. Adenosine (ADO) was measured by radioimmunoassay or high-pressure liquid chromatography (HPLC) techniques. Basal AN and ADO release under steady-state conditions were 2.2 and 13.8 pmol.min-1 x ml column volume (CV)-1, respectively. Inhibition of adenosine deaminase with erythro-9-(2-hydroxy-3-nonyl)adenine (5 x 10(-6) M) enhanced ADO release by 3.3 pmol.min-1 x ml CV-1, and AN release remained unchanged (2.8 pmol.min-1 x ml CV-1). Inhibition of adenosine kinase by 5-iodotubercidine (10(-5) M) greatly enhanced ADO release by 97.7 pmol.min-1 x ml CV-1, while AN release was unaffected. Inhibition of ecto-5'-nucleotidase by alpha,beta-methylene-ADP (5 x 10(-5) M) enhanced AN release from 2.6 to 8.2 pmol.min-1 x ml CV-1 and reduced ADO release by an equivalent extent. Stimulation of endothelial cells with Ca ionophore A23187 dose dependently augmented AN and ADO release to 2,013.2 and 92.5 pmol.min-1 x ml CV-1, respectively. Thrombin (1 U/ml) enhanced AN release from 5.0 to 8.7 pmol.min-1 x ml CV-1, whereas several other endothelium-dependent and -independent vasodilators including acetylcholine, bradykinin, isoproterenol, and norepinephrine were proven to have no significant effect.(ABSTRACT TRUNCATED AT 250 WORDS)