This study describes the disposition and excretion of indole-3-carbinol (I3C), a natural dietary tumor modulator and candidate chemopreventive agent, in male Fisher 344 rats after continuous dietary or a single oral administration. Steady-state urinary and fecal excretion were attained 40 and 112 hr, respectively, after commencing continuous exposure. These two routes accounted for approximately 75% of the administered dose, of which 77% appeared in feces. After 7 days of 2,000 ppm dietary I3C, a mean of 1,154 microM I3C eq was found in liver, of which 17% was present as extractable, unbound I3C derivatives. Total equivalents in liver decreased to 643 and 411 microM 24 and 48 hr later, respectively, for animals returned to control diet. Mean levels of I3C eq in lung decreased from 436 to 219 microM, and blood levels decreased from 320 to 208 microM over the same 48-hr period. After administration of 1 mmol/kg radioinert I3C (a comparable daily dose as in the feeding study) for 6 days, animals were given 1 mmol/kg [3H]I3C. Mean liver levels were 257, 283, and 541 microM I3C eq at 1.5, 3, and 6 hr after dosing, and these levels represented 0.97%, 1.34%, and 2.45% of the total I3C dose administered, respectively. Concentrations of I3C eq changed little in blood, kidney, tongue, or lung over this time period. HPLC analysis of ethyl acetate extracts of liver from rats given an oral dose revealed 24 distinct [3H]I3C-derived peaks. Two of the predominant peaks were identified as 3,3'-diindolylmethane (I33', a linear dimer of I3C) and [2-(indol-3-ylmethyl)-indol-3-yl]indol-3-ylmethane (LT, a linear trimer). A novel I3C metabolite was identified as 1-(3-hydroxymethyl)-indolyl-3-indolylmethane (HI-IM). Hepatic levels of these metabolites and three major, but unidentified, products were between 1.0 and 13.1 microM; highest levels were observed at 6 hr or, for HI-IM, at 1.5 hr postdosing. Parent I3C was not detected in liver extracts, whereas the potent Ah receptor agonist 3,2-b-indolocarbazole (ICZ) was estimated at 1.6 nM. These data suggest that neither I33', LT, or ICZ alone reach sufficient hepatic concentration to account for cytochrome P450IA induction by dietary I3C, or provide effective inhibition of microsomal bioactivation of the hepatocarcinogen aflatoxin B1; however, the total hepatic mixture of I3C derivatives may be sufficient to provide both modulatory responses in the rat.