Background and Purpose Adenosine transport inhibitors attenuate ischemic central neuronal damage in vivo, but the locus of this protective action is presently unknown. To help address the question of whether adenosine transport inhibitors have a protective effect directly on brain parenchyma, we tested the effect of the adenosine transport inhibitor dipyridamole on neuronal loss induced by oxygen-glucose deprivation in vitro. Methods Murine cortical cultures were exposed to combined oxygen and glucose deprivation, N-methyl-D-aspartate, or kainate. The extracellular concentrations of glutamate and adenosine were measured by high-performance liquid chromatography; neuronal cell death was assessed by morphological examination and measurement of lactate dehydrogenase release. Results Cultures exposed to oxygen-glucose deprivation for 30 to 75 minutes exhibited an insult-dependent increase in extracellular adenosine, followed shortly by an increase in extracellular glutamate and 24 hours later by neuronal death. Addition of the A1 receptor antagonist 8-cyclopentyltheophylline during oxygen-glucose deprivation enhanced both glutamate release and neuronal damage. Addition of 10 mu mol/L dipyridamole decreased extracellular adenosine and also enhanced extracellular glutamate and neuronal death. In contrast, dipyridamole increased the levels of extracellular adenosine stimulated by N-methyl-D-aspartate or kainate. Conclusions These results are consistent with the idea that endogenous adenosine has a neuroprotective effect directly on cortical cells exposed to oxygen-glucose deprivation. However, inhibition of adenosine transport with dipyridamole was surprisingly not an effective strategy for enhancing this protective effect. The beneficial effects of adenosine transport inhibitors observed in vivo may be mediated indirectly - for example, by effects on the vasculature.