A doxycycline-inducible CYP3A4-Caco-2 cell line as a model for evaluating safety of aflatoxin B1 in the human intestine

Exposure of humans to aflatoxin B1 (AFB1) via ingestion of contaminated agricultural products is a major concern for human health throughout the world because epoxidized AFB1, biotransformed from AFB1 by hepatic CYP3A4, is strongly hepatotoxic and hepatocarcinogenic. Intestinal epithelial cells serve as a physical and physiological barrier against xenobiotics via their intercellular tight junction (TJ) seals and the metabolizing enzyme CYP3A4. However, the effect of AFB1 on the intestinal barrier remains unclear. Here, we investigated the influence of AFB1 on these physical and physiological intestinal barriers by means of an in vitro human intestinal model utilizing doxycycline-inducible CYP3A4-expressing Caco-2 cells, in which CYP3A4 activity is comparable to that in the adult human intestine. Cellular toxicity of AFB1 in induced Caco-2 cells (i.e., cells in which expression of CYP3A4 is induced by doxycycline) was approximately 5 times that of uninduced Caco-2 cells. Exposure to 16 mu M AFB1 did not decrease the transepithelial electric resistance (TEER; a measure of TJ barrier integrity) in monolayers of uninduced Caco-2 cells to 95.8 % of that in vehicle-treated cells; in contrast, in induced Caco-2 cells, TEER was reduced to 28.8 %. Exposure to 16 mu M AFB1 increased paracellular permeation of 4-and 20-kDa dextrans (paracellular permeation markers) through monolayers of induced Caco-2 cells to 5.4 and 5.2 times that through uninduced Caco-2 cells. These results together show that ingested AFB1 can modulate the intestinal barrier, and that inducible CYP3A4-expressing Caco-2 cells are a promising tool for evaluating the safety of food contaminants in the human intestine.