Cross-dataset transcriptomic pathway ranking highlights MAPK signaling in diquat-induced hepatocellular toxicity

Diquat is used in agriculture as an herbicide but poses significant health risks upon exposure. Current treatment for toxic exposures to diquat focuses on supportive care, and there is a need for better understanding of the molecular mechanisms underlying diquat-induced injury in order to develop more targeted antidotes. To this end, TGF-alpha transgenic mouse hepatocyte (TAMH) cells were exposed to various concentrations of diquat to determine toxicologically relevant concentrations followed by subsequent transcriptomic analysis. Data mining from the MicroArray Quality Control (MAQC)-II dataset, which was accessed through the Gene Expression Omnibus (GEO) was also leveraged during the gene network analysis. A median lethal concentration (LC₅₀) for diquat in the TAMH line was determined to be 18 μM, with significant cell death observed at 9 h. Microarray data identified 3578 significantly altered transcripts in the TAMH model and 6554 from the MAQC-II dataset, with notable overlap in gene expression changes. Pathway analysis using Database for Annotation, Visualization and Integrated Discovery (DAVID) highlighted MAPK signaling as playing a role during diquat-induced toxicity in both models, with 11 shared transcripts suggesting a conserved molecular response across rodent species. This study aimed to investigate the molecular mechanisms behind diquat-induced toxicity using TAMH cells and identified MAPK signaling pathway as involved. By demonstrating the utility of combining GEO and DAVID for pathway analysis, this study not only highlights potential therapeutic targets for diquat toxicity but also presents a broadly applicable, cost-effective strategy for toxicogenomic research.