Effect of simulated intestinal fluid on drug permeability estimation across Caco-2 monolayers

Abstract

Presently, the Caco-2 cell culture model is widely used during drug discovery and development as a predictive tool for the oral absorption of drug candidates. For transport experiments in the Caco-2 system, HBSS-like buffered salt solutions are commonly used, although different shortcomings have been associated with the use of these buffers. In this paper, we investigated the effect of using fasted state simulated intestinal fluid (FaSSIF) as potential biorelevant medium for the drug permeability estimation across Caco-2 monolayers. The transport characteristics of 19 model compounds were determined in the Caco-2 cell culture model in the presence of FaSSIF as compared to classic transport medium. A sigmoidal relation was obtained when the estimated P_app, s of the apical to basolateral transport were plotted versus the reported values of the fraction absorbed in man. Although no effect of FaSSIF as compared to classic transport medium (TM) was observed on the total predictability of the model, an impact was demonstrated (1) on the bi-directional transport of actively transported drugs (including talinolol, digoxin and doxorubicin), (2) on recovery and (3) on the solubility and permeability estimation of poorly water-soluble drugs. The observed differences may be attributed to a P-gp inhibitory effect of sodium taurocholate (NaTC), micellar encapsulation by the NaTC/lecithin mixed micelles and/or an increase of the solubility of lipophilic drugs. As the experimental conditions should mimic the physiological in vivo conditions, the use of FaSSIF as medium during Caco-2 experiments may improve the biorelevance of the model.

Introduction

The transepithelial transport of compounds is an important characteristic, commonly assessed during the evaluation and selection of potential drug candidates. The Caco-2 cell culture model was introduced in the early 1990s and has become a widely used tool for the determination of the intestinal transport characteristics of compounds (Hidalgo et al., 1989, Hilgers et al., 1990, LeCluyse and Sutton, 1997, Gan and Thakker, 1997). Several reports have demonstrated the possibility to predict the oral absorption of drugs in man based on their permeability observed in Caco-2 monolayers (Artursson and Karlsson, 1991, Artursson et al., 2001). Within the framework of the Biopharmaceutic Classification System (BCS), the rate of mass transfer of a compound across the Caco-2 monolayer can even be considered to allow a waiver for in vivo bioequivalence studies (Yu et al., 2002). However, many discrepancies in culturing and experimental conditions can be identified in the literature related to Caco-2 maintenance and experiments.

For the realization of permeability and transport studies on the Caco-2 cell culture model, classic buffered salt solutions are commonly used (e.g. Hanks’ balanced salt solution (HBSS) buffered with HEPES (10 mM) at pH 7.4 and supplemented with glucose). Nevertheless, many shortcomings are associated with the use of such saline buffers for Caco-2 experiments, including the limited solubility of highly lipophilic drugs, the adsorption and/or non-specific binding to the device surfaces or (in)to the cells and the poor physiological relevance of the media used. To overcome these issues, several categories of media (for the apical and basolateral compartment), including plain salt solutions composed of inorganic salts and glucose, culture medium and solvents mimicking intestinal fluid have been proposed and have recently been reviewed (Ingels and Augustijns, 2003).

We investigated the possibility of using fasted state simulated intestinal fluid (FaSSIF) as apical solvent for Caco-2 experiments. FaSSIF has originally been introduced by the group of Professor Dressman in 1998 as dissolution medium to simulate the in vivo dissolution behavior of compounds (Galia et al., 1998, Dressman et al., 1998, Dressman and Reppas, 2000). The composition of FaSSIF is shown in Table 1. It has previously been shown that FaSSIF buffer was compatible with the Caco-2 cell monolayer for at least 2 h. The transport of different model compounds (i.e. theophylline (passive diffusion), phenylalanine (active transport)) and the activity of the brush border enzyme aminopeptidase were similar when using classic TM and FaSSIF. However, a concentration-dependent P-gp inhibitory activity of sodium taurocholate (NaTC) (present in FaSSIF) when assessing cyclosporin A (CsA) transport was demonstrated (Ingels et al., 2002a).

The aim of the present study was to evaluate the impact of FaSSIF as potential biorelevant medium on the permeability estimation and transport characteristics of drug compounds in the Caco-2 cell culture model.