Comparative Proteomic Characterization of 4 Human Liver-Derived Single Cell Culture Models Reveals Significant Variation in the Capacity for Drug Disposition, Bioactivation, and Detoxication

Abstract

In vitro preclinical models for the assessment of drug-induced liver injury (DILI) are usually based on cryopreserved primary human hepatocytes (cPHH) or human hepatic tumor-derived cell lines; however, it is unclear how well such cell models reflect the normal function of liver cells. The physiological, pharmacological, and toxicological phenotyping of available cell-based systems is necessary in order to decide the testing purpose for which they are fit. We have therefore undertaken a global proteomic analysis of 3 human-derived hepatic cell lines (HepG2, Upcyte, and HepaRG) in comparison with cPHH with a focus on drug metabolizing enzymes and transport proteins (DMETs), as well as Nrf2-regulated proteins. In total, 4946 proteins were identified, of which 2722 proteins were common across all cell models, including 128 DMETs. Approximately 90% reduction in expression of cytochromes P450 was observed in HepG2 and Upcyte cells, and approximately 60% in HepaRG cells relative to cPHH. Drug transporter expression was also lower compared with cPHH with the exception of MRP3 and P-gp (MDR1) which appeared to be significantly expressed in HepaRG cells. In contrast, a high proportion of Nrf2-regulated proteins were more highly expressed in the cell lines compared with cPHH. The proteomic database derived here will provide a rational basis for the context-specific selection of the most appropriate ‘hepatocyte-like’ cell for the evaluation of particular cellular functions associated with DILI and, at the same time, assist in the construction of a testing paradigm which takes into account the in vivo disposition of a new drug.

Inter-individual differences in the susceptibility of primary human hepatocytes towards drug-induced cholestasis are compound and time dependent

“…Though these cell lines are wellestablished, cheap, easy to handle, and allow generation of reproducible data in a highthroughput setting, they are limited by their immature phenotypes. CYP activities and expression of DMEs and hepatic transporters are drastically lower in these cell lines compared to primary human hepatocytes (PHH) (198)(199)(200), which is also reflected in their reduced sensitivity towards drug hepatotoxicity compared to PHH and the lack of CYP inducibility in the case of HepG2 cells (200,201). To overcome the limited liver-specific functionalities, cell lines overexpressing specific DMEs or NRs have been established for improved toxicity prediction and assessments of CYP enzyme induction (202)(203)(204), yet the overall molecular phenotypes still remain drastically different from the human liver in vivo.…”

Evolution of Novel 3D Culture Systems for Studies of Human Liver Function and Assessments of the Hepatotoxicity of Drugs and Drug Candidates

“…Primary human hepatocytes retain their metabolic capacity in suspension for some hours, but when plated, the expression of drug‐metabolizing enzymes and drug uptake transporters will be reduced to very low levels . Recently, a large multi‐centre study compared the toxicity using a set of nine liver‐toxic drugs and four compounds not associated with any liver toxicity in HepG2, HepaRG and Upcyte cells in conventional 2D cultures . It was concluded that none of the cell models could distinguish between liver‐toxic and non‐liver‐toxic compounds and thus not useful to investigate new chemical entities.…”

Evidence-based selection of training compounds for use in the mechanism-based integrated prediction of drug-induced liver injury in man

“…Several publications have recently appeared in which the here selected training compounds and negative controls have been investigated in order to substantiate their utility as training compounds to profile and elucidate further DILI mechanisms and to interrogate existing and novel in vitro systems for the prediction of human DILI (Sison-Young et al 2015; Bell et al 2016; Sison-Young et al 2016; den Braver-Sewradj et al 2016; den Braver et al 2016). …”