2023 FDA Science Forum
Gene expression changes predict the severity of NAFLD-like liver injury in male Collaborative Cross mice
- Authors:
- Center:
-
Contributing OfficeNational Center for Toxicological Research
Abstract
Nonalcoholic fatty liver disease (NAFLD), a prevalent chronic liver disease, is characterized by substantial variations in severity. In this study, we used a genetically diverse Collaborative Cross (CC) mouse population model to analyze the global transcriptome and clarify the molecular mechanisms involved in hepatic fat accumulation that determine the level and severity of NAFLD. Twenty-four strains of male CC mice were maintained on a high fat/high sucrose (HF/HS) diet for 12 weeks and their hepatic gene expression profiles were determined by next-generation RNA-sequencing. The development of the nonalcoholic fatty liver (NAFL) phenotype in CC mice coincided with significant changes in the expression of hepatic genes at the population level, as evidenced by the presence of 724 differentially expressed genes involved in lipid and carbohydrate metabolism, cell morphology, vitamin and mineral metabolism, energy production, and DNA replication, recombination, and repair. Importantly, the expression of 68 of these genes strongly correlated with the extent of hepatic lipid accumulation in the overall population of HF/HS diet-fed male CC mice. Among these genes, the Perilipin 2 (Plin2) and Aldehyde dehydrogenase 3 family member A2 (Aldh3a2) genes, which encode proteins involved in neutral lipid storage (PLIN2) and lipid detoxification (ALDH3A2), exhibited the highest correlation with osmium tetroxide staining, a marker for lipid accumulation. Partial Least Squares (PLS) was performed to identify a set of genes that could be utilized to estimate NAFL severity. PLS modeling using 247 NFL-related genes showed the highest accuracy for predicting NAFL-like liver injury and identifying the individual mouse strains that are highly susceptible to the development of NAFLD induced by a HF/HS diet. A Uniform Manifold Approximation and Projection dimension reduction analysis of the 247 gene set showed a distinct clustering of CC mouse strains characterized by different levels of lipid accumulation in the livers. These findings imply that gene expression profiling combined with a PLS modeling approach may be useful tools to predict NAFLD severity in genetically diverse patient populations and to determine key potential biomarkers of NAFL and potential targets for treatment.
