This study investigates the mechanisms of absorption and the role of intestinally localized purine salvage pathway enzymes on the ileal availabilities of 2',3'-dideoxyinosine (ddI), a substrate for purine nucleoside phosphorylase (PNP); 2'-fluoro-2',3'-dideoxyinosine (F-ddI), a non-PNP substrate; and 6-chloro-2',3'-dideoxypurine (6-Cl-ddP), an adenosine deaminase (ADA) activated prodrug of ddI. The potential for increasing the intestinal availability of 6-Cl-ddP through the use of ADA inhibitors, namely, 2'-deoxycoformycin (DCF) and erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA), is also explored. Drug permeability coefficients across the intestinal epithelium were determined in in situ perfusions in the mesenteric vein cannulated rat ileum based on both drug appearance in blood (Pblood) and disappearance from the lumen (Plumen) and their paracellular and transcellular components were estimated by comparison to the permeabilities of two paracellular markers, mannitol and urea. Values of Pblood for ddI were determined to be (1.1 +/- 0.3) x 10(-6) cm/s, in close agreement with the value of (1.0 +/- 0.3) x 10(-6) cm/s obtained for F-ddI, a PNP resistant analogue of ddI having virtually the same molecular size and lipophilicity as ddI. This indicates that PNP may not play an important role in the low intestinal absorption of ddI. The Pblood for 6-Cl-ddP, (19 +/- 2) x 10(-6) cm/s, was 4.5-fold lower than Plumen, (84 +/- 12) x 10(-6) cm/s, which means that 77 +/- 6% of 6-Cl-ddP was metabolized during its intestinal transport, thus qualitatively accounting for the low oral bioavailability (7%) of 6-Cl-ddP observed in vivo in rats. Extensive intracellular metabolism of 6-Cl-ddP by ADA was confirmed by the high concentrations of ddI found both in the intestinal lumen and blood during 6-Cl-ddP perfusions and by a rate of ddI appearance in blood which was approximately 10-fold higher than ddI controls. Co-perfusion of the potent, hydrophilic ADA inhibitor DCF (Ki = 0. 001-0.05 nM) with 6-Cl-ddP led to only partial inhibition of intestinal ADA, while complete inhibition was obtained using the less potent but more lipophilic inhibitor EHNA (Ki = 1-20 nM). Hence, EHNA may be used to improve intestinal absorption of 6-Cl-ddP in vivo.