Evaluation of in vitro brain penetration: Optimized PAMPA and MDCKII-MDR1 assay comparison

Abstract

Parallel artificial membrane permeability assay (PAMPA) is arising in ADMET screening as a powerful tool to determine the passive permeability of new potential chemical entities. In an attempt to set up a sensitive high throughput method to assess passive blood–brain barrier (BBB) penetration we focused our attention on the effect of solvent and the influence of phospholipids on the permeability in PAMPA. Moreover, the high throughput nature of the assay was maximized by decreasing the incubation time and performing the assay in a cassette mode. UPLC system coupled with a mass spectrometer enormously reduces the analytical time, contemporaneously increasing the sensitivity of the method.

P_app values obtained from PAMPA were compared to permeability values from MDCKII-MDR1 assay. Evaluation of the two in vitro models with in vivo data was performed to test the predicting capacity of the two methods. Their contemporary assessment was shown to be an helpful tool in understanding the prevalent mechanism of penetration through the BBB.

Introduction

The blood–brain barrier (BBB) is one of the key issues in the pharmaceutical industry since central nervous system (CNS) drugs must penetrate the barrier while drugs targeting peripheral tissues should be impaired in the passage. The BBB is a complex endothelium formed by capillary endothelium cells with tight junctions and is rich in active transporters that facilitate or impair the passage. Several methods have been recently explored for the prediction of in vivo results: computational methods, physical measurements such as log P/log D and cell culture systems (Garberg et al., 2005).

Among cell cultures, prediction with primary bovine brain endothelial cells gives the best scoring to the in vivo system (Gumbleton and Audus, 2001) but difficulties in establishing and maintaining primary culture, as well the tediousness of the method, make the assay unfeasible as a high throughput screening assay. Among cell lines, MDCKII-MDR1 are the most widely used and promising (Garberg et al., 2005) but, in spite of their cultivation time reduced to 3 days, the assay still results in a higher cost than the test with artificial membranes.

Parallel artificial membrane permeability assay (PAMPA) was originally reported with 10% (w/v) egg lecithin in dodecane (Kansy et al., 1996) but variation in the phospholipid composition have been studied (Sugano et al., 2001; Seo et al., 2006). A comparison of the three most used PAMPA models, HDM, DOPC and DS-PAMPA were recently carried out by Avdeef and Tsinman (2006) explaining permeability's differences reported in literature for some standards. Because of the nature of the assay, PAMPA was used mainly for the prediction of the gastrointestinal absorption (Kerns et al., 2004). Attempts to modify the monolayer to improve the prediction of BBB penetration were done using porcine polar brain lipids (Di et al., 2003). While phospholipid composition was studied in depth, solvent's influence was not sufficiently investigated to evaluate if the use of porcine brain lipid is significantly affecting the permeability of the model. Aim of the paper was to evaluate the effective influence of phospholipids in PAMPA, by changing various solvent conditions. P_app values obtained from PAMPA were compared to permeability values from MDCKII-MDR1 assay, evaluation of the two in vitro models with in vivo data was performed to test the predicting capacity of the two methods. Simultaneous evaluation was shown to be a helpful tool in understanding the prevalent mechanism of penetration through the BBB. A significant improvement of the throughput of PAMPA was reached by performing the assay in cassette mode and the analysis by UPLC/MS.