Characterising select hepatocyte cultures for improved hepatotoxicity testing

Abstract

Drug-induced hepatotoxicity is a major contributor towards post-marketed drug withdrawals. Most of these liver injuries can be associated with drug metabolism, which is primarily performed by hepatic cytochrome P450 enzymes, a superfamily of drug metabolising haem enzymes. Pre-clinical in vitro models are commonly used in an attempt to predict the toxicity profile of lead drug compounds during the early development phases. These in vitro models need to accurately resemble the human liver functionality to be able to predict drug hepatotoxicity. Primary human hepatocytes are the ‘gold standard’ for mimicking liver function. However, due to their expensive culturing requirements, limited time in active culture, and limited availability, other commercially available, transformed hepatic cell lines are commonly used as substitutes for primary hepatocyte cultures. The HepG2 cell line is one of the most commonly used because they are readily available, and the culturing methods are relatively inexpensive. Conventional monolayer culturing of human-derived cell lines has shown limitations when used for drug toxicity tests and lacks sufficient resemblance to human liver tissue, compared to three-dimensional cultured cells. This has led to three-dimensional cell cultures being recommended over monolayer cultures to predict potential in vivo toxicity. The aim of this study was to characterise HepG2 cells differentially cultured as monolayer or three-dimensional spheroids, in the presence and absence of chronic enzyme-inducing drug cocktail exposure. Also, to evaluate the models’ feasibility over an extended culture time and compare their metabolic capabilities as candidates for hepatotoxicity screening platforms. This was done by generating HepG2 spheroids using the liquid overlay method for up to 21 days and culturing same origin HepG2 monolayers for 17 days. Cells were evaluated for morphology, viability, protein content, monolayer cell cycle profile, proteins mass profile differences, and the presence of hepatic markers in spheroids. The metabolic activity was assessed using liquid chromatography tandem mass spectrometry. The monolayer and drug cocktail exposed monolayer cells were viable for up to 17 days while the spheroids and drug cocktail exposed spheroids were viable up to 21 days. CYP1A2 activity was detected in all cultures with slightly more acetaminophen (μmol/μgprotein) detected in monolayer cultures. The activity confirms CYP1A2 expression detected in all spheroid cultures from Day 7 to Day 21. Furthermore, minimal hydroxybupropion, dextrorphan and hydroxymidazolam was detected in all cultures suggesting low CYP2B6, CYP2D6 and CYP3A4 activity, respectively. The metabolite levels were similar between induction and non-induction cultures suggesting that the induction drug cocktail had no significant effect on the metabolic capacity of the HepG2 cells under either of the culturing conditions used. Therefore, the metabolic activity may be due to accumulated innate metabolic capability and/or long-term culture. Furthermore, the growth plateau observed in spheroid cultures’ protein levels after Day 4 and the hepatic markers (AFP, HNF-4α, CK18 and albumin) expression observed from Day 7, would be desirable for repeated hepatotoxicity testing. Other studies have shown HepG2 spheroid cell viability and increased hepatic marker expression for more than 21 days, with a relatively consistent metabolic profile after 21 days, suggesting a stable differentiated phenotype. This is beneficial for repeated, long-term drug exposure for acute and chronic hepatotoxicity screening.