The purpose of this study was to test the hypothesis that circulating red blood cells (RBCs) release adenosine deaminase (ADA) when injured. This hypothesis was evaluated in rats using dimethyl sulfoxide (DMSO) to damage RBCs. Boluses and infusions of DMSO caused a reduction in urinary adenosine and a concomitant hemoglobinuria, and the ability of DMSO to reduce urinary adenosine was blocked by pretreatment with the ADA inhibitor erythro-9-(2-hydroxy-3-nonyl)adenine. Infusions of DMSO also significantly enhanced ADA activity in urine and plasma. Dimethylsulfone, an analog of DMSO that does not affect RBCs, did not cause hemoglobinuria and did not affect urinary adenosine. High concentrations of DMSO did not affect adenosine metabolism in rat kidneys perfused without RBCs, and DMSO did not decrease urinary adenosine in rats rendered severely anemic (hematocrit < 15%) by replacing whole blood with plasma. However, DMSO did decrease urinary adenosine in rats without a spleen, a major source of adenosine deaminase apart from circulating RBCs. DMSO reduced renal interstitial levels of adenosine and attenuated bradycardic responses to exogenous adenosine, and these effects were prevented by erythro-9-(2-hydroxy-3-nonyl)adenine. These results indicate that circulating damaged RBCs release significant amounts of ADA, a process that may predispose to vasoocclusive events.