By pharmacological manipulation of endogenous adenosine, using chemically distinct methods, we tested the hypothesis that endogenous adenosine tempers proinflammatory cytokine responses and oxyradical-mediated tissue damage during endotoxemia and sepsis. Rats were pretreated with varying doses of pentostatin (PNT; adenosine deaminase inhibitor) or 8-sulfophenyltheophylline (8-SPT; adenosine receptor antagonist) and then received either E. coli endotoxin (lipopolysaccharide; 0.01 or 2.0 mg/kg) or a slurry of cecal matter in 5% dextrose in water (200 mg/kg). Resultant levels of tumor necrosis factor (TNF)-alpha, interleukin (IL)-1beta, and IL-10 were measured in serum and in liver and spleen. Untreated, 2 mg/kg lipopolysaccharide elevated serum TNF-alpha, IL-1beta, and IL-10. PNT dose dependently attenuated, without ablating, the elevation in serum TNF-alpha and IL-1beta and raised liver and spleen IL-10. PNT also attenuated elevation of TNF-alpha in serum, liver, and spleen at 4 and 24 h after sepsis induction, and 8-SPT resulted in higher proinflammatory cytokines. Modulating endogenous adenosine was also effective in exacerbated (8-SPT) or diminished (PNT) tissue peroxidation. Survival from sepsis was also improved when PNT was used as a posttreatment. These data indicate that endogenous adenosine is an important modulatory component of systemic inflammatory response syndromes. These data also indicate that inhibition of adenosine deaminase may be a novel and viable therapeutic approach to managing the systemic inflammatory response syndrome without ablating important physiological functions.