A Caco-2 cell model with extrinsic radioiron was used to evaluate the effect of dietary factors on nonheme iron bioavailability from a semipurified meal. Study 1 was conducted to evaluate the effect of enhancers (ascorbic acid) and inhibitors (bran, phytate and tea) on iron bioavailability when added to semipurified meal containing egg albumen as a protein source. The effect of various proteins [bovine serum albumin (BSA), casein, beef and soy] on iron bioavailability was evaluated in Study 2 by substituting the above protein sources for egg albumen. Protein solubilization following in vitro digestion for individual test meals was not significantly different from the control. On the other hand, nonheme iron solubilization (0.8+/-0.0 to 5.9+/-0.3 vs. 4.9+/-0.8 mg/L) varied significantly. The total iron uptake for each meal was calculated based on the percentage of radioiron taken up and transported by Caco-2 cells and the amount of nonheme iron present in uptake solutions. Iron uptake ratios represent test/control values. With the exception of BSA and ascorbic acid, the effect of dietary factors was similar to that found in humans reported in the literature. A significant correlation (r = 0.97; P<0.0001) was found between the published human absorption data and the iron uptake by the Caco-2 cells. The results of our study indicate the usefulness of Caco-2 cells in assessing human iron absorption and the feasibility of this cell model in studying iron bioavailability from various food combinations, otherwise not easily performed in humans.