073P University of Oxford
BPS 75th Anniversary Meeting December 2006

A comparison between compound potency in an in vitro phospholipidosis assay and at the herg-encoded potassium channel

BG Small, LR Barone1, J-P Valentin, CE Pollard and PJ Ciaccio1. Safety Assessment UK, AstraZeneca R&D Alderley Park, UK and 1Safety Assessment US, AstraZeneca R&D Wilmington, USA.

Cationic amphiphilic drugs (CADs) are associated with phospholipidosis (PLD), an effect characterised by intracellular accumulation of phospholipids and drug in concentric lamellar bodies. In physicochemical terms, inhibitors of the human ether-a-go-go related gene (hERG) encoded potassium channel are similar to PLD-inducing compounds. In view of this, and the observation that some PLD-inducing compounds invivo have been reported to prolong the QT interval (e.g. fluoxetine; Gonzalez-Rothi et al., 1995.), we tested whether there was any correlation between compound potency in an in vitro PLD assay (EC50) and at the hERG channel (IC50).

Thirty-nine compounds were tested in both assays. Briefly, the potency ( EC50) of PLD inducing compounds was determined by measuring accumulation of a surrogate phospholipid; N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine (NBD-PE) in a mouse spleen macrophage cell line (I.13-35) using a Cellomics ArrayScan . Cell culture and compound incubations were conducted at 37C, cell fixation, staining and measurement of fluorescence at room temperature. Cells were used between passages 18 - 50. The methods have been described in detail by Morelli et al. (2006). The potency (IC50) of these compounds at the hERG-encoded channel was determined in CHO cells using IonWorksHT . Cell culture was conducted at 37C; otherwise all experiments were performed at room temperature (22C). Cells were used between passages 10 - 30. The methods have been described in detail by Bridgland-Taylor et al. (2006). Physicochemical properties; cLogP and pKa of the compounds were estimated by computational methods.

Although 33 of the compounds were active in both assays, there was no positive correlation between their IC50s (r= -0.35). A t-test was performed on this data to assess the null hypothesis that the slope equals one. The associated probability of the test statistic (12) was < 0.0001 and therefore strongly rejects the null hypothesis. The cLogP and pKa range of these compounds was 2.0-8.9 and 6.1-11.2, respectively. The 6 other compounds (cLogP and pK a range 0.7-5.0 and 3.4-9.7, respectively) were inactive in the PLD assay but their hERG IC50s ranged from 3-210 μM). The lack of a potency correlation indicates that hERG data cannot be used as a quantitative indicator of in vitro PLD potency or vice versa. However, potency data in both assays relative to cLogP and pKa suggest that decreasing compound basicity and lipophilicity is likely to reduce the risk of both hERG inhibition and PLD in vitro.

Gonzalez-Rothi, R.J. et al. (1995). Chest. 107, 1763 - 1765.
Morelli, J.K. et al. (2006). Cell Biol and Toxicol. 22, 15-27.
Bridgland-Taylor, M. et al. (2006). J. Pharmacol and Toxicol Meth. 54, 189-99.