Characterisation of inducible proteins in the HepG2 cell line proteome

Abstract

Rationale: Attaining scientifically robust, clinically representative, and functionally translatable preclinical models for pharmaceutical research has become increasingly desirable. Relevant hepatic models are pivotal for the screening and accurate identification of safe new drug candidates. Currently, primary human hepatocytes are considered the gold standard for hepatic modelling However, their limited accessibility, high variability and terminal nature hamper their widespread adoption. Immortalised cell cultures, for example the human hepatoma cell line (HepG2), have served as widespread substitutes. However, the representativeness of HepG2 cells is of concern due to the relatively large phenotypic disparity when compared to primary human hepatocytes. Recent studies have postulated that the adoption of HepG2 cells into three dimensional (3D) conformations may promote a more closely correlated phenotype. Little is known regarding the extent of proteomic changes which occur when cells as cultured as spheroids, nor is there sufficient insight into the temporal dynamics which may drive these changes. The aim of this study was to investigate whether proteomic changes occur in HepG2 cells when cultured as spheroids compared to traditional monolayer culture techniques and if so whether these changes were culture time dependent. Additionally, HepG2 spheroid cultures were assessed for improved metabolic adaption following extended-time spheroid-based culture in the presence of specific drugs, a key characteristic of human hepatocytes that would be critical for hepatotoxicity testing models. Methods: HepG2 cells were cultured as spheroids using two established 3D culture models; cells were seeded into hanging drop plates (20 000 cells/spheroid) and into 81 well, 3D micro moulds (1000 cells/spheroid), then maintained under untreated control or drug exposed groups, using a 7-drug cocktail representing substrates of phase I metabolizing enzymes. Spheroid cultures were characterized for protein content, size, viability, and immunohistochemistry. Cultures from different groups and culture times were compared by means of quantitative proteomic methods using isobaric labelling proteomics. Sample proteins were reduced, alkylated, precipitated, digested with trypsin and labelled using 6-plex tandem mass tag (TMT) isobaric labels. Proteomic analysis was performed using a Dionex Ultimate 3000 RSLCnano liquid chromatography system coupled to a Thermo Scientific Fusion Lumos Orbitrap Mass Spectrometer. Protein identification was conducted using an alternating SequestHT/MS Amanda schema. Relative protein quantitation for the different spheroid groups were expressed relative to original monolayer cultured HepG2 control cells. Relative metabolic competences of spheroid cultures were assessed using a defined hepatic metabolism phenotyping cocktail using an in-house developed liquid chromatography tandem mass spectrometry method. Results: HepG2 spheroid cultures were shown to progressively upregulate expression of hepatic marker proteins over the 28-day culture time course. Quantitative proteomic methods initially identified over 5000 proteins which when filtered to include only proteins with at least three identified peptides and presence in all replicates, reduced to approximately 4800 proteins. Identified proteins showed differential expression between spheroid culture groups compared to monolayer cultures. Additionally, progressive differential proteome expression was evident between spheroid cultures according to time spent in culture and based on presence or absence of drug exposure. Thorough bioinformatics analyses revealed insights into the potential mechanisms responsible for the proteomic differences and identified that while spheroids exposed to drug appear to adapt metabolically to their culture environments, the nature of these changes may lack impact for clinical relevance. Metabolomic investigations corroborated the proteomic findings showing that, while maintaining some capacity for phase I metabolism, overall, clinically relevant metabolism for cytochrome P450 enzymes was absent. Conclusion: Quantitation of large protein cohorts demonstrated that there are adaptive changes occurring during long-term culture of HepG2 cell spheroids. While the proteomic changes observed, under all culture conditions, do not support adoption of these cultures for modelling drug metabolism, use in other bespoke model systems which concern other aspects of liver function is feasible. The high-quality proteomic data generated shows temporal dynamics in spheroid cultures which have implications for their utility. It necessitates the requirement for more comprehensive investigations into the relevance of the time-points used when adopting spheroid cultures.