Adenosine transport inhibitors as enhancers of extracellular levels of endogenous adenosine would, presumably, only be effective if, for example, (1) the inhibitors block influx to a greater degree than efflux (release) of intracellular adenosine or (2) the inhibitors block equally well the influx and efflux of adenosine, but significant amounts of adenosine are formed as a result of dephosphorylation of released adenine nucleotides. Limited information is available regarding the directional symmetry of adenosine transporters in neural cells. Using rat brain crude P2 synaptosomal preparations preloaded with L-[3H]adenosine, our objectives here were to determine (1) if L-[3H]adenosine, a substrate for adenosine transporters that is more metabolically stable than physiological D-adenosine, was being released from synaptosomal preparations, (2) the optimal conditions necessary to observe the release, and (3) the degree to which this release was mediated by efflux through bidirectional nucleoside transporters. L-[3H]Adenosine release was found to be concentration and time dependent, temperature sensitive, and linear with synaptosomal protein. L-[3H]Adenosine release was inhibited dose-dependently by dipyridamole, nitrobenzylthioinosine, and dilazep; at concentrations of 100 microM inhibition was at least 40% for dipyridamole, 52% for nitrobenzylthioinosine, and 49% for dilazep. After loading with L-[3H]adenosine alone or I-[3H]adenosine plus unlabeled L-adenosine, D-adenosine, or uridine, L-[3H]adenosine release was inhibited 42% by L-adenosine, 69% by uridine, and 81% by D-adenosine.(ABSTRACT TRUNCATED AT 250 WORDS)