Hepa-ToxMOA: a pathway-screening method for evaluating cellular stress and hepatic metabolic-dependent toxicity of natural products

Abstract

In the field of drug discovery, natural products have emerged as therapeutic agents for diseases such as cancer. However, their potential toxicity poses significant obstacles in the developing effective drug candidates. To overcome this limitation, we propose a pathway-screening method based on imaging analysis to evaluate cellular stress caused by natural products. We have established a cellular stress sensing system, named Hepa-ToxMOA, which utilizes HepG2 cells expressing green fluorescent protein (GFP) fluorescence under the control of transcription factor response elements (TREs) for transcription factors (AP1, P53, Nrf2, and NF-κB). Additionally, to augment the drug metabolic activity of the HepG2 cell line, we evaluated the cytotoxicity of 40 natural products with and without S9 fraction-based metabolic activity. Our finding revealed different activities of Hepa-ToxMOA depending on metabolic or non-metabolic activity, highlighting the involvement of specific cellular stress pathways. Our results suggest that developing a Hepa-ToxMOA system based on activity of drug metabolizing enzyme provides crucial insights into the molecular mechanisms initiating cellular stress during liver toxicity screening for natural products. The pathway-screening method addresses challenges related to the potential toxicity of natural products, advancing their translation into viable therapeutic agents.

Figure 1

Characterization of Hepa-ToxMOA cell lines for AP1, P53, Nrf2, and NF-κB. (a) Cell lines were seeded at a density of 30,000 cells/well in a 96-well plate. For each positive chemical (AP1 (PMA; 5, 10 µM), P53 (Nutlin-3; 1.25, 2.5 μM), Nrf2 (DL-Sulforaphane; 12.5, 25 µM), NF-κB (TNF-α; 5, 10 µg/ml)) for cell proliferation (AP1), DNA damage (P53), Oxidative stress (Nrf2), Inflammation response (NF-κB) cellular stress-reporter cell lines, after treatment for 24 h, GFP expression was verified by 10 × image through high-content screening. (b) The GFP intensity value derived through HCS was quantified and shown in a graph (*p < 0.05, **p < 0.01). (c) Cell lines were seeded at a density of 500,000 cells/well in a 6-well plate. After treating the positive chemical with the same concentration as HCS for 24 h, western blot analysis was performed.

Figure 2

Establishment of Hepa-ToxMOA cell lines depending on metabolic activation using S9 fraction conditions. (a) The concentration (0% and 1%) of S9 fractions was used to treat Hepa-ToxMOA cell lines, and image analysis and GFP intensity quantification were performed through HCS (*p < 0.05, **p < 0.01). (b,c) GFP intensity and cell viability of Hepa-ToxMOA cell lines were compared after treatment for 24 h using positive control (CPPA, cyclophosphamide) 6.25, 12.5, and 25 μM according to the concentration of the S9 fraction (*p < 0.05, **p < 0.01).

Figure 3

Integrated analysis of cell viability and GFP intensity expression for 40 natural products according to the non-metabolic or metabolic activity. (a) GFP intensity and cell viability analysis were performed after exposure of 40 natural products to four types of cellular stress-reporter cell lines for 24 h under conditions of non-metabolic or metabolic activity. Based on Supplementary Fig. 4, it was classified as a positive increase (Score 3), positive cause (Score 1 or Score 2), and no effect (Score 0). Results of a total of 40 different natural products are shown in the figures. (b) After scoring the expression of different toxic mechanisms in the non-metabolic or metabolic activity for 40 natural products, they were compared and analyzed with graphs and Venn diagrams (Score is a maximum of 12 points, which means a 'positive increase' in metabolic and non-metabolic conditions AP1, P53, Nrf2, and NF-κB, and of score 3 or higher, it is considered that toxicity is likely to be caused by natural product). (c) For the results of (b) above, the Venn diagram and list are shown for natural products exceeding Score 3. (d) In order to find natural products that are highly likely to cause toxicity under non-metabolic or metabolic activity conditions, natural products with increased GFP expression in each toxic mechanism of AP1, P53, Nrf2, and NF-κB were shown in a Venn diagram.

Figure 4

In vitro hepatotoxicity screening of 40 natural products according to the non-metabolic or metabolic activity. To perform (a) cell proliferation evaluation (HepG2-GFP-AP1), (b) DNA damage evaluation (HepG2-GFP-P53), (c) oxidative stress evaluation (HepG2-GFP-Nrf2), and (d) inflammation response evaluation (HepG2-GFP NF-κB) based on four types of cellular stress-reporter cell lines by 40 natural products under conditions of non-metabolic and metabolic activity was evaluated and analyzed. For each cellular stress-reporter cell line, natural products with increased GFP intensity regardless of non-metabolic or metabolic activity are shown in black, and natural products with increased intensity only under conditions of non-metabolic activity are shown in green. Natural products with increased only under conditions of metabolic activity are shown in red (Full bar mean S9- and hatched bar mean S9 +) (*p < 0.05, **p < 0.01).