Cell cultures as tools in biopharmacy

https://doi.org/10.1016/S0928-0987(00)00164-0Get rights and content

Abstract

A survey is given on a few selected cell culture models that are used for transport studies. They are characterised for growth, transcellular electrical resistance and cytoarchitecture. The importance of standardisation in view of their use as transport models is documented. Their potential for studies on passive permeation and P-glycoprotein-mediated transport is explored and related to published data. Transport studies are presented that were performed in a two-chamber set-up, the Costar® “vertical diffusion system”. A series of non-homologous compounds showed similar permeability data (Papp) in the different cell cultures. The origin of the cell type had no remarkable influence on passive transcellular permeation. MDCK cells, an epithelial cell line of canine kidney origin, are perfectly suited to screen for passive permeation. They have low expression of transporter proteins and low metabolic activity. In general, they probably represent the best-known epithelial cell line with respect to genetics as well as lipid and protein composition. MDCK cells are easy to handle. Transport experiments can be done between 7 and 14 days after seeding, when the stationary growth phase is reached. To screen for P-glycoprotein substrates, efflux and uptake studies were performed with mdr1-transfected MDCK cells (MDR1-MDCK) in a one-chamber system in the presence or absence of verapamil or cyclosporin A as inhibitor. Evidence is presented why the transfected cells, which express large amounts of P-glycoprotein, are not suitable for two-chamber transport studies.

Introduction

In drug design and drug development adequate model systems have to be introduced at an early moment to avoid loss of promising compounds at an advanced stage due to insufficient absorption into, and distribution throughout the body. Beside toxicity, the pharmacokinetic characteristics were a major reason for failures of compounds in clinical studies in the past. Several strategies are pursued to establish tools for the prediction of in vivo barrier passage of compounds (Krämer, 1999). Among them are computational approaches based on physicochemical parameters (e.g. Young et al., 1998, Crivori et al., 2000), as well as the characterisation of the partitioning behaviour of substances in biphasic systems (e.g. n-octanol water, liposome/buffer, Δlog P). For permeation studies, model systems have been established at various levels of complexity (Fig. 1), reaching from animals down to isotropic lipophilic phases (e.g. Kansy et al., 1998). When screening for permeation characteristics, the choice of a test system always represents a compromise between high throughput with low predictive potential and low throughput with high predictive potential. Cell cultures take an intermediate position within the pyramid of complexity for permeation studies. They permit low to medium throughput, and, as typical “added value assays”, represent an important tool in drug development (Audus et al., 1990, Artursson and Borchardt, 1997).

In the following, a few defined cell culture model systems are reviewed which have been used for transport studies in our laboratory. Their characteristics are summarised with respect to growth curves, transcellular electrical resistances (TEER), cytoarchitecture (e.g. monolayers vs. multilayers), tight junction (TJ) formation, and tightness of cell layers (mannitol permeation test) (Table 1). Transport studies are presented, which have been performed under controlled conditions with a two-chamber vertical diffusion system (Costar®) using a set of non homologous drugs with log Doctanol7.4 values between <0 and +3.5. The kinetics of transport was studied in various cell cultures in the apical-to-basal and basal-to-apical direction, and apparent permeation coefficients (Papp) were calculated. Information was collected about the reproducibility of Papp. The relevance of transport studies with multilayered cells was analysed with MDR1-MDCK cells, which originate from transfection of Madin Darby canine kidney (MDCK) cells with the mdr1 gene, the gene coding for the P-glycoprotein, P-gp (Pastan et al., 1988). As found for other transfected cells, MDR1-MDCK cells have lost contact inhibition and show only limited polarisation (see Section 3.4). The same cells were used for comparison with the Caco-2 cells with respect to their potential to screen for P-gp substrates and inhibitors.

Section snippets

Standardisation of cell cultures

Successful use of cell culture models is closely related to standardisation (Fig. 2). A comprehensive review with ample references was recently published dealing with cell culture techniques and standardisation procedures (Wunderli-Allenspach, 2000). Here, we will restrict ourselves to a brief survey of the most important issues.

GLP includes a close track of the origin of cells and of passage numbers. Growth media should not be changed without need, and if changes are made, the growth

General

Various cell types of epithelial and endothelial origin have been used for transport studies in search of in vitro models for in vivo barriers such as epithelia of the GI tract, nasal mucosa, skin, and the blood–brain barrier, BBB (Audus et al., 1990). This survey concentrates on a few selected cell culture models, the characteristics of which are summarised in Table 1.

Caco-2 cells

Caco-2, an epithelial human colon adenocarcinoma cell line, has been widely used to predict intestinal absorption of potential

General

The apparent permeability coefficient (Papp) is used to quantify transport in a two-chamber diffusion system. Drug transfer from the donor to the receiver chamber through the cell layer of interest is measured as increase in drug concentration in the receiver chamber over time, and the differential increase in the amount of drug is calculated. Papp [cm/s] values are determined as follows:Papp=dQdt 1c0Awhere dQ/dt [mol/s] is the increase in the amount of drug in the receiver chamber per time

Conclusions and perspectives

Provided that cell cultures are cultivated under controlled standardised conditions, they show cell type-specific characteristics with respect to growth, expression of proteins and cyoarchitecture. Papp values are reproducible. However, depending on the experimental set-up used for transport studies, calculated Papp values can differ widely even with standardised cell cultures. All tested cell culture models, independently of their origin, yield similar results for transcellular passive

Acknowledgements

Projects were supported by the Roche Research Foundation, Basel, Switzerland (S.H.), by a grant of ETHZ (K.S.), and by a travelling research fellowship from The Wellcome Trust, UK (S.K.).

References (33)

  • P. Artursson et al.

    Intestinal drug absorption and metabolism in cell cultures: Caco-2 and beyond

    Pharm. Res.

    (1997)
  • K.L. Audus et al.

    The use of cultured epithelial and endothelial cells for drug transport and metabolism studies

    Pharm. Res.

    (1990)
  • M. Cereijido et al.

    Polarized monolayers formed by epithelial cells on a permeable and translucent support

    J. Cell Biol.

    (1978)
  • C.L. Crespi et al.

    Development of Caco-2 cells expressing high levels of cDNA-derived cytochrome P4503A4

    Pharm. Res.

    (1996)
  • P. Crivori et al.

    Predicting blood–brain barrier permeation from three-dimensional molecular structure

    J. Med. Chem.

    (2000)
  • W.G. Dirks et al.

    ECV304 (endothelial) is really T24 (bladder carcinoma): Cell line cross-contamination at source

    In Vitro Cell Dev. Biol. Anim.

    (1999)
  • Cited by (136)

    • Crude plant extract versus single compounds for vitiligo treatment: Ex vivo intestinal permeability assessment on Brosimum gaudichaudii Trécul

      2020, Journal of Pharmaceutical and Biomedical Analysis
      Citation Excerpt :

      Moreover, PSO and 5-MOP oral pharmacokinetic profiles in rats shows bioavailability of 82.8 and 49.9 % for these psoralens in Radix Angelicae dahuricae, respectively, providing evidence that supports our assumption [25]. Also, non-specific binding to permeability apparatus and presence of DMSO as a solubilizing agent can also add some variability to permeability data [28]. Permeability of psoralens were measured and monitored by metoprolol (METO-Papp), as highly permeated reference standard substance for BCS drug classification.

    View all citing articles on Scopus
    View full text