Development of a triculture based system for improved benefit/risk assessment in pharmacology and human food

Development of a triculture based system for improved benefit/risk assessment in pharmacology and human food.


Material and methods
The triculture model was adapted by combining the in vitro model of the human follicle-associated epithelium (including M-like cells) according to the protocol of des Rieux et al. (2007) and the co-culture of Caco-2 and HT29-5M1 cells from Nollevaux et al. (2006). Briefly, Caco-2 cells with or without HT29-5M1 cells were seeded on 12-wells Transwell inserts (3.0 μm pore diameter, Costar, Elscolab, Kruibeke, BE) and cultivated for 3 days in DMEM supplemented with 10 % foetal calf serum (FCS). Some inserts were then inverted and a piece of silicon rubber (Labo-Modern, Queveaucamps, BE) placed around the basolateral side. Then, inserts were transferred into a Petri dish pre-filled with culture medium and maintained for 10 days with the basolateral medium refreshed every 2 days. Raji-B cells, suspended in DMEM + 10% FCS, were added to the basolateral compartment of the inserts. The tricultures were maintained for 5 days. Co-cultures were cultured under the same conditions, but without the addition of Raji-B cells or HT29-5M1 cells.
Cell monolayer integrity, both in mono-, co-and tricultures, was controlled by measurement of the transepithelial electrical resistance (TEER, Millipore Millicell ® ERS, Billerica, MA) and by evaluation of cell permeability to Lucifer Yellow (LY; 457 Da, Sigma-Aldrich, St-Louis, MO).
After integrity experiments, cells were washed in PBS, fixed with acetone and permeabilised with 1% (v/v) Triton X-100. Cells were then washed with PBS, blocked with 1% BSA and washed again with PBS, and then incubated with mouse anti-MUC5AC at 1:100 dilution, rinsed in PBS and incubated with Alexa Fluor 488 goat anti-mouse 10µg/ml and Rhodamine-Phalloidin 4U/ml (Invitrogen, Eugene, OR). Finally, cells were mounted with Ultracruz Mounting Medium (Santa Cruz Biotechnology, Santa Cruz, CA). The monolayers were viewed using a Zeiss LSM 710 confocal laser scanning microscope (Carl Zeiss MicroImaging GmbH, Jena, DE).

Results and discussion
To validate the triculture system as an intestinal barrier model, transport studies were used to determine the culture integrity and the permeability capacity with TEER measurement and assessment of the passage of LY. Furthermore, a morphological analysis was initiated

Cell monolayer integrity and permeability properties
A decrease of TEER was observed in the different coculture systems. This should, most probably, be correlated to an increase of the barrier permeability, upon addition of HT29-MTX or/and Raji cells in the system ( Table 1).

The effect of the addition of goblet-like cells or of the conversion of Caco-2 cells into M-like cells suggests a role of these cells in the ability of the intestinal cells to increase their absorbtion of hydrophilic molecules, as
observed by the increase of the permeability of LY in the co-and tri-culture systems in comparison with the control.
The contribution of the different cell types in the triculture system further allows the integration of mechanisms such as transcytosis (M cells), the presence of mucus (goblet-like cells), and the effectiveness of the tight junctions in the culture system, as compared to a standard Caco-2 model.

Localisation of the intestinal cell types
The intestinal cell types were detected by fluorescent labelling and confocal analysis ( Figure 1). Actin (red) and MUC5AC (green) were restricted respectively on Table 1 In vitro models characterisation: Cell monolayer integrity was determined by LY transport. LY was incubated with cell monolayers for 180 min at 37°C (n=3) and it was quantified by fluorimetry. The results were expressed as % of total fluorescent. enterocyte and goblet-like cells. The M-like cells appeared within the enterocyte layer and were identified by their lack of microvilli (actin) at their apical surface. The mucus produced by goblet-like cells was on the surface (MUC5AC). In the triculture system, there was close contact between HT29-MTX, Raji and Caco-2 cells but the goblet-like cells have tendency to cluster together.

Conclusion, perspectives
These results clearly indicate that the cocultivation of 3 different cell types allows to reconstitute a cell culture system that should better mimic the intestinal barrier, specially to investigate its interaction with nanoparticles. This should facilitate a better evaluation of pharmacological or toxicological properties of these materials in food and drugs.
We are now developing additional co-culture systems involving f.i. RAW264.7 macrophages to further determine the interactions involved in inflammatory processes; HepG2 cells to estimate the intestinal and hepatic effects in presystemic biotransformations.