A co-culture system of human intestinal Caco-2 cells and lymphoblastoid TK6 cells for investigating the genotoxicity of oral compounds

doi:10.1093/mutage/ges028 Mutagenesis vol. 27 no. 6 pp. 631–636, 2012 Advance Access publication 27 July 2012 A co-culture system of human intestinal Caco-2 cells and lymphoblastoid TK6 cells for investigating the genotoxicity of oral compounds Ludovic Le Hégarat*, Sylvie Huet and Valérie Fessard ANSES, Laboratoire de Fougères, Unité de Toxicologie des contaminants, BP 90203, 35302 Fougères Cedex, France *To whom correspondence should be addressed: Tel: +33 2 99 94 78 78; Fax: +33 2 99 94 78 80; Email: Received on January 04, 2012; revised on April 24, 2012; accepted on April 04, 2012 Introduction Most experts recommend improving the predictive power of mutagenicity tests by developing more accurate in vitro models (1,2). The single culture models commonly used for in vitro genotoxicity assays cannot adequately mimic the absorption and metabolic processes that occur in vivo. However, the development of in vitro methods to reduce the use of laboratory animals is encouraged in several regulations (i.e. REACH) and supported by various instances (i.e. ECVAM). However the predictivity of in vitro tests for evaluating carcinogenicity potential should be improved (1). In 2008, the International Workshop on Genotoxicity Testing (IWGT) held in Basel, Switzerland, highlighted the need for developing new in vitro models (3). Conventional in vitro genotoxicity tests entail a single cell line and include the use of induced rat liver S9 fractions to mimic in vivo metabolism. However, this rodent alternative does not fully reproduce human metabolism or the complex interactions between organs, including intestinal transport and first-pass metabolism. A three-dimensional model mimicking the properties of human skin (barrier function, metabolism) has been recently Materials and methods Chemicals Aflatoxin B1 (AFB1, Cas 1162-65-8), benzo[a]pyrene (BaP, Cas 50-328) and ketoconazole (Cas 65277-42-1) were purchased from Sigma (St. Quentin-Fallavier, France). Methylmethanesulfonate (MMS, Cas 66-27-3) was purchased from Acros Organics (Geel, Belgium). Caco-2/TK6 co-culture system The Caco-2 cell line was obtained from the American Type Culture Collection (ATCC N°HTB37) and was used between passages 30 and 40. The cells were maintained in modified Eagle’s medium (MEM) containing glutamaxTM, 10% fetal calf serum (FCS), 1% nonessential amino acids, penicillin (100 U/ml) and streptomycin (100 mg/ml) in a humidified atmosphere of 95% air and 5% CO2 at 37°C. Cells (80% confluence) were split and seeded at 0.175 × 106 cells/cm². The human TK6 lymphoblastoid cell line was a gift from the Institut Pasteur in Lille (Fabrice Nesslany). The cells were maintained in tissue culture flasks with RPMI 1640 medium (Gibco) containing glutamaxTM and supplemented with 10% FCS, 100 IU/ml of penicillin and 100 µg/ml of streptomycin. For the co-culture studies, Caco-2 cells were seeded at 6 × 104 cells/cm2 on polycarbonate membrane inserts (0.4 µm pore diameter, 12 mm insert, © The Author 2012. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. All rights reserved. For permissions, please e-mail: . 631 Here, we assessed a co-culture system of intestinal Caco-2 cells and lymphoblastoid TK6 cells for modelling the role of intestinal first-pass effects, i.e. absorption and metabolism, in the genotoxicity of oral drugs and food contaminants. Caco-2 cells were seeded onto semipermeable culture inserts for 21 days until differentiation, and then TK6 cells were added to the basal compartment. After apical loading with mutagenic compounds [methylmethanesulfonate (MMS), benzo[a]-pyrene (BaP) and aflatoxin B1 (AFB1)], comet and micronucleus assays were performed on both cell lines. MMS (10 µg/ml) showed positive results in the micronucleus assays in both cell lines, even though DNA damage was only detected in the Caco-2 cells with the comet assay. At concentrations of 0.5– 50 µM, BaP induced dose-dependent comet and micronucleus formation at 24 h in Caco-2 cells, but no DNA damage was observed in TK6 cells. Although AFB1 failed to induce comet formation, it resulted in a high level of micronuclei in both cell lines. Treatment of Caco-2 cells with the CYP3A4 inhibitor, ketoconazole, inhibited the AFB1-induced cytotoxicity and micronucleus formation in TK6 cells, suggesting that intestinal metabolism is involved in the AFB1 genotoxic response in TK6 cells. Our results suggest that the Caco-2/TK6 co-culture model is suitable for modelling the role of intestinal biotransformation and transport processes in the genotoxic potential of oral drugs and food contaminants in target blood cells. developed to study the genotoxic potential of chemicals applied to the skin (4). This co-culture system includes reconstructed human skin cultured on membrane inserts with the mouse lymphoma L5178Y cell line growing underneath. Apical treatment with mitomycin-C, cyclophosphamide or apigenin induces micronucleus formation in the L5178Y cells, suggesting that this model can be useful for studying the genotoxicity of compounds that come in contact with the epidermis. Similarly, more sophisticated models are needed to investigate the in vitro genotoxic potential of oral drugs and food contaminants. Before entering the systemic circulation, most orally administrated xenobiotics must pass through the intestinal barrier to reach the liver via the portal vein (5). Intestinal first-pass metabolism, along with cytochrome P450 3A4 and multidrug efflux transporters such as P-gp, actively participate in the presystemic elimination of oral drugs (6). Therefore, intestinal cells may bioactivate promutagen coumpounds into genotoxic metabolites, and upon systemic distribution induce DNA damage not only in enterocytes but also in blood cells and in other organs. We thus propose here a co-culture system involving differentiated human intestinal Caco-2 cells and lymphoblastoid TK6 cells. The Caco-2 cells can differentiate and form a monolayer with the same morphological and biochemical characteristics as human small intestine enterocytes (7). Because they also express various phase-1 and -2 enzymes and transporter proteins (e.g. CYP1A1, 1B1, SULT1A1 and 1A3, p-glycoprotein and MRP) (8–11), these cells may also be a suitable model to study xenobiotic metabolism. The P53-competent human TK6 cells are commonly used in genotoxicology testing as a target blood cell model for the micronucleus assay and in the thymidine kinase (TK) gene-mutation assay (12–14). In this study, we investigated the genotoxic potential of two promutagen compounds (BaP and AFB1) and a direct alkylating agent (MMS) on the Caco-2/TK6 co-culture system by using the comet and the micronucleus assays. L. Le Hégarat et al. 12-well plate; Corning, USA). The culture medium was changed three times a week. After 25–27 days post-seeding, the monolayers of differentiated Caco-2 cells were used for co-culture experiments. Only the monolayers expressing a (...truncated)