Toxicogenomic analysis of risk factors that predict sensitivity to acetaminophen-induced liver injury using a mouse model of the human population

Recent advances in the field of genomics have led to an improved understanding of genomic structure and function in humans and model organisms. Effective utilization of genomic information in the toxicology field has the potential to significantly improve risk assessment; however, a major limitation is a lack of animal models that can identify genetic variants underlying inter-individual differences in toxicity. Current testing strategies fail to capture sufficient genetic diversity, leading to over-generalization of the results from single-strain studies when extrapolating risk to human populations. We hypothesized that using a panel of genetically diverse inbred mouse strains (or Mouse Model of the Human Population; MMHP), would enable detection of genetic loci that affect individual toxicity responses to a model toxicant, acetaminophen. In Aim 1, we demonstrated that MMHP mouse strains experienced a range of toxicity outcomes following equal acetaminophen doses, similar to the range of toxicity observed in human subjects. Haplotype-associated mapping and genetic sequencing within the MMHP yielded a genetic variant within the gene encoding CD44 that correlated toxicity sensitivity in both mouse and man. The results of this study indicated that use of the MMHP facilitates detection of genetic variants affecting chemical toxicity. In Aim 2, population-based biomarkers of liver injury were determined by analyzing gene expression. Identified liver injury biomarkers included several genes involved in known cell death pathways. The signature also included genes that had not been previously linked to acetaminophen-induced liver injury, indicating that the model may provide a means for discovery of novel therapeutic targets. In Aim 3, the metabolism of acetaminophen was examined across selected mouse strains. Acetaminophen toxicity requires bioactivation of acetaminophen to a quinone radical; it is therefore necessary to demonstrate whether metabolic differences potentially affect genetically pre-determined injury outcomes within the MMHP. Strain differences in acetaminophen metabolism were not a determining factor in the overall liver injury outcome, further demonstrating the need for a genetically diverse mouse model to identify therapeutic targets. Overall, the data confirm that using the Mouse Model of the Human Population as a research paradigm has the potential to improve both toxicity risk assessment and mechanistic research.