NMR-based metabolomic characterisation of the immortalised human liver cell line HepG2 as a model system for toxicology studies

Maren Jinks

Modern toxicology calls for in vitro solutions that reduce the need for animal testing, are more human-relevant, high throughput, and cost effective. Immortalised human cell lines are becoming increasingly popular for these reasons. As the liver is the primary organ of detoxification, human liver cell lines are extensively used for toxicology studies due to their characteristic liver-distinctive functions. However, very little is known about their baseline biochemical composition and the changes that occur within these cells upon toxin exposure. Better characterisation of baseline composition would enable more subtle biochemical changes in cell systems to be identified, for example upon chronic low-level toxin exposure. In this thesis, untargeted metabolic profiling, and multivariate statistical analyses of in vitro samples, including whole live cells, cell extracts, and their spent media supernatants, were used to characterise the biochemical composition of the most used, immortalised human liver cell line HepG2, both at baseline and post exposure to the model toxin hydrogen peroxide (H2O2). Cells were routinely cultured and treated with varying acutely cytotoxic concentrations of H2O2. As HepG2 cells are known to have poor expression of cytochrome P450 (CYP450) liver enzymes, which are crucial for phase I metabolism, the effects of hepatic enzyme activation were also investigated. This was done by inducing CYP450 activity with rifampicin in HepG2 cells prior to toxin exposure and comparing metabolic profiles with those of cells that were not pre-treated. For whole cell metabolic profiling, cells were harvested and analysed using 1H High-Resolution Magic Angle Nuclear Magnetic Resonance (HR-MAS NMR) spectroscopy. Cellular extracts and spent medium supernatant samples were profiled by standard 1D 1H and 2D 1H J-resolved (j-RES) NMR experiments. In addition to standard analyses on all samples, 2D 1H-1H COrrelated SpectroscopY (COSY), 1H-1H TOtal Correlation SpectroscopY (TOCSY), 1H-13C Heteronuclear Single Quantum Correlation (HSQC), and 1H-13C Heteronuclear Multiple Bond Correlation (HMBC) NMR experiments were performed to confirm the identification of metabolites. Spectral resonances were assigned, and key metabolites identified. Multivariate principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (O-PLS-DA) of the acquired spectral data sets showed a clear separation between H2O2 treated cells and controls as well as between rifampicin treated and non-treated cells and their supernatants. Significantly altered metabolites between the groups highlighted major perturbations to central metabolic pathways upon toxin exposure because of cellular oxidative stress, including those involved in energy, amino acid, and lipid metabolism. In this project, we demonstrated that NMR-based metabolomics is suitable, efficient, and cost-effective for metabolic profiling of cells and for identifying and comparing metabolite classes involved in central metabolic pathways before and after toxin exposure. Results from this study highlighted the importance of metabolic activation of CYP450 enzymes in HepG2 cells for physiologically relevant results that can predict metabolism-mediated hepatotoxicity in humans. The work presented here provides crucial baseline metabolomic data and demonstrates a useful NMR-based methodological tool kit for continued studies of toxins using HepG2 cells and other in vitro systems. This will enable future research to develop cell systems that are reflective of in vivo liver metabolism, which can be used in pre-clinical drug development and toxicity assessment pipelines to help lower compound attrition and improve xenobiotic safety overall.

NMR-based metabolomic characterisation of the immortalised human liver cell line HepG2 as a model system for toxicology studies

Files and links (1)

Abstract

Details

Metrics