Development of a Simple Double-layered Cell Culture System Using Caco-2 and TIG-1 Cells as a New Cytotoxicity Test

Yasuyuki Sakai 1, Tomonori Arai 1,2, Akiyoshi Sakoda1, and Motoyuki Suzuki1,3
1Institute of Industrial Science, University of Tokyo, 7-22-1 Roppongi, Minato-ku, Tokyo 106-8558, Japan
2Current address; Mitsubishi Tokyo Pharmaceutical Co. Ltd., 1,000 Kamoshida, Aoba-ku, Yokohama 227-0033, Japan
3United Nations University, 5-53-70 Jingumae, Shibuya-ku, Tokyo 150-8558, Japan

Corresponding author:Yasuyuki Sakai, Ph. D.
Institute of Industrial Science, University of Tokyo
7-22-1 Roppongi, Minato-ku, Tokyo 106-8558, Japan
Telephone: +81-3-3402-6231 Ext. 2458
Fax: +81-3-3408-0593

Running Head: Coculture of Caco-2 cells with fibroblasts

Original paper :AATEX 7(2,3):47-58,2001

We investigated the effect of a Caco-2 cell barrier on toxicity expression in cultured human diploid fibroblasts, TIG-1 cells, in a simple double-layered culture system. The experimental setup used a 12-well plate for the TIG-1 cells and a companion membrane culture insert for the Caco-2 cells. Four model chemicals, i.e., caffeine, diclofenac sodium, indomethacin, and paraquat, which have different levels of in vivo human absorbability, were employed to test the feasibility of the culture system. In transport studies using a Caco-2 cell layer, not only initial apparent permeation coefficients of the four chemicals but also in vitro absorption ratios in equilibrium phases were well correlated with previously reported in vivo human absorbability. This phenomenon was explained by the differences in the permeation coefficients between the two directions across the Caco-2 cell layer. Changes in the dose-response relationships in terms of TIG-1 cell growth on the ninth day in the double-layered culture system were largely predicted from the cell growth in the single-layered system (without a Caco-2 cell layer) and the in vitro absorbability. However, small but significant discrepancies between the observed and the predicted dose-response relationships were found in the loading of diclofenac sodium and paraquat. The first discrepancy suggests the involvement of possible enhanced detoxification of diclofenac sodium by the Caco-2 cells. The latter discrepancy reflects the specific permeation kinetic of paraquat across the Caco-2 cell layer. Final toxicity in terms of TIG-1 cell growth in the double-layered system reflected in vivo toxicity better than that of the single-layered system without the Caco-2 cell layer. These results demonstrate that a Caco-2-cells-based permeation test combined with a single-cell-population-based cytotoxicity test is more useful in estimating in vivo human oral toxicity.

Keywords: Cocultivation, Caco-2 cells, Permeation, Human diploid fibroblast, Growth inhibition.