A quantitative AOP of mitochondrial toxicity based on data from three cell lines

Area of Science:

• Toxicology
• Computational Toxicology
• In Vitro Assays

Background:

• Adverse Outcome Pathways (AOPs) facilitate integrating in vitro data for chemical hazard assessment.
• Quantitative AOPs (qAOPs) employ mathematical models to link key events (KEs) in toxicity pathways.

Purpose of the Study:

• To calibrate a qAOP for mitochondrial toxicity using data from three cell lines (LHUMES, HepG2, RPTEC/TERT1) for rotenone and deguelin.
• To assess the applicability of a single qAOP across different cell types and test chemical-independence via cross-validation.

Main Methods:

• Calibration of a qAOP for mitochondrial toxicity using rotenone and deguelin data across three cell lines.
• Cross-validation of the calibrated qAOP using a separate dataset of eight chemicals in LHUMES cells.
• Mathematical modeling of key event relationships within the qAOP framework.

Main Results:

• Practical difficulties arose when calibrating the qAOP across different cell lines, even with identical experimental protocols.
• Mathematical functions describing key event relationships varied between cell types.
• Cross-validation showed underestimation of toxicity for several chemicals, highlighting issues with chemical-specific potency estimation and downstream KE prediction.

Conclusions:

• A single qAOP may not be universally applicable across different cell lines due to variations in key event relationships.
• Accurate characterization of the molecular initiating event is critical for successful qAOP cross-validation.
• Careful selection of relevant readouts is essential for designing effective in vitro experiments for qAOP calibration.