Predicting Phospholipidosis: A Fluorescence Noncell Based in Vitro Assay for the Determination of Drug–Phospholipid Complex Formation in Early Drug Discovery

Abstract

This paper describes for the first time, a high-throughput fluorescence noncell based assay to screen for the drug-phospholipid interaction, which correlates to phospholipidosis. Anionic amphiphilic phospholipids can form complexes in aqueous solution, and its critical micelle concentration (CMC) can be determined using the fluorescence probe N,N-dimethyl-6-propionyl-2-naphthylamine (Prodan). Upon interaction with drug candidates, this CMC may shift to a lower value due to the association between lipids and drug candidates, the stronger the interaction, the greater the shift. Metabolism of a drug can change the degree of phospholipidosis depending on the rate of metabolism and the nature of the metabolite(s). Our data from 45 drugs and metabolites of 10 drugs using this fluorescence approach demonstrate a good correlation with phospholipidosis as reported with human studies, in vivo testing, and cellular assays. This assay therefore offers a fast, reliable, and cost-effective screening tool for early prediction of the phospholipidosis-inducing potential of drug candidates.

Identification of Drugs Inducing Phospholipidosis by Novel in vitro Data

“…Several prominent studies have been published, including reports by Pelletier et al[26] (who compiled experimental data for a set of 201 compounds), by Kruhlak et al[25a] (who investigated 482 substances), and most recently, by van de Waters et al[27] (who published data for a set of 56 compounds). In the past, various experimental methods have been proposed to measure PLD using cell-free,[28] in vitro,[29] and in vivo[30] approaches. Additionally, in silico prediction of PLD has been attempted in various studies using random forests,[31] support vector machines,[26], [31] nearest neighbor classifications,[32] decision trees,[32]–[33] logistic regressions,[26] Bayesian models,[26] and artificial neural networks,[32] based on the hitherto available data.…”

A cell-based assay using HepaRG cells for predicting drug-induced phospholipidosis

“…These compounds showed positive responses in various in vitro models of PLD (Morelli et al, 2006;van de Water et al, 2011;Zhou et al, 2011). The ability for predicting drug-induced PLD has been reported in various cellbased assays.…”

“…Thus, our cell-based assay system using HepaRG cells may have comparable detection sensitivity with other cell-based approaches for the PLD assay. Zhou et al (2011) reported that parent compounds as well as their metabolites have PLD potential in some compounds such as amiodarone. HepaRG cells express the major CYPs involved in xenobiotic metabolism (Aninat et al, 2006).…”

“…CYP2D6 contributes to the metabolism of some compounds such as amitriptyline (Olesen and Linnet, 1997) and imipramine (Su et al, 1993). Zhou et al (2011) reported that parent compounds as well as their metabolites have PLD potential in amitriptyline and imipramine. In the present study, positive changes for PLD assay were observed in amitriptyline and imipramine, suggesting that low CYP2D6 activity may not affect a predictive performance in HepaRG cells in the case where parent compounds have PLD potential.…”

Assessment of amiodarone-induced phospholipidosis in chimeric mice with a humanized liver

“…Although the relationship between DIPL and adverse effects in humans remains unknown, it is necessary to consider toxicity triggered by DIPL in humans. Some studies regarding in vitro screening to avoid DIPL have been reported (Sawada et al, 2005;Kasahara et al, 2006;Morelli et al, 2006;Zhou et al, 2011;Takagi et al, 2016). However, to evaluate DIPL in experimental animals, it is important to consider species differences, organ selectivity, and affinity for phospholipids.…”