Little information is available regarding colonic permeability to macromolecules in health or disease states. In vivo permeability of rabbit colon to [14C]polyethylene glycol 4000 (14C-PEG) was examined in the presence of immune complex-mediated experimental colitis and compared with that of partially treated (control) and normal rabbits. Permeability was assessed by urinary 14C-PEG excretion after intrarectal administration of 0.1 mM solution of 14C-PEG (1 ml/kg, 7.5 X 10(6) cpm/ml). Experimental colitis greatly increased colonic permeability (p less than 0.001 in two-way analysis of variance) compared with control and normal groups (2.06% +/- 0.19%, 0.14% +/- 0.04%, and 0.01% +/- 0.004%, respectively, of rectally administered counts). Gel diffusion chromatography showed that absorbed 14C-PEG was excreted into urine unchanged, demonstrating its applicability as an inert, nonmetabolizable macromolecular probe. Urinary clearance after mesenteric vein administration of 14C-PEG was similar in normal animals and animals with colitis, implicating colonic absorption as the source of the group differences. Postmortem histology confirmed the acute colitis lesions in the experimental group. These findings support the hypothesis that nonspecific colonic inflammation is associated with significant alterations of mucosal permeability.