Quantification of intralysosomal drug accumulation using in silico and in vitro methods

Abstract

Lysosomes are acidic intracellular organelles that can sequester weakly basic lipophilic compounds as a result of the pH gradient, membrane partitioning, and binding to acidic phospholipids (APs). Lysosomal sequestration can lead to substantially higher unbound drug concentrations inside the organelle relative to cytosol. Depending on the drug target site, this process may affect its efficacy and/or lead to toxicological concerns such as phospholipidosis (PLD). Whilst a number of indirect experimental methods currently exist to examine lysosomal sequestration, these methods cannot directly quantify intralysosomal drug concentrations. The aim of this project was to examine and enhance our understanding of lysosomal drug sequestration using both in silico and in vitro methods. Critical analysis of reported experimental data from the literature was performed to generate the most comprehensive lysosomotropic and PLD database to date. Lysosomotropic compounds in the database (n = 53) spanned a broad range of therapeutic classes, and generally showed LogP and pKa values >3.15 and >7.4, respectively. PLD-inducing compounds had similar properties, with LogP and pKa values generally >3.3 and >8.85, respectively. Despite these similarities, not all PLD-inducing compounds were lysosomotropic, highlighting that the physico-chemical properties of compounds should only be used as a preliminary guide for assessing potential lysosomotropic behaviour and induction of PLD. In silico rat alveolar macrophage (AM, lysosome rich) and rat hepatocyte (lysosome poor) cell models were developed to predict subcellular drug distribution based on cellular physiological parameters and a drug’s physico-chemical properties. The drug distribution across four subcellular compartments (medium, cytosol, lysosomes and mitochondria) was predicted for 18 drugs (selected from the database) based on passive diffusion driven by pH partitioning, membrane partitioning, binding (including AP binding), and electrostatic attractions/repulsions. The in silico cell models revealed that lysosome rich cells (i.e., AMs) were superior cellular systems for investigating lysosomal drug accumulation, with a greater ability to detect the influence of the lysosomal pH gradient on drug sequestration at the cellular level. The importance of accurate organelle AP concentration data (particularly in lysosomes) was highlighted, due to model and Kp prediction sensitivity on this parameter. In silico cell models are useful tools for exploring subcellular drug distribution; however, their predictive capabilities are currently limited by inconsistent/inadequate physiological cell parameters and lack of in vitro intralysosomal drug concentration data for model verification. In order to obtain intralysosomal drug concentration data, a novel in vitro method for the direct quantification of intralysosomal drug concentrations was developed. This method uses magnetically isolated lysosomes from the lysosome rich NR8383 (rat AM) cell line and LC MS/MS to quantify subcellular drug concentrations. The purity and functionality of the isolated lysosomes obtained was critically assessed by enzymatic assays (>4-fold enrichment of the purified fraction by lysosomal enzyme assay and

Date of Award	1 Aug 2019
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Aleksandra Galetin (Supervisor) & David Hallifax (Supervisor)