With a multitude of HCC mouse models available, choosing the one that most closely resembles human HCC can be challenging. This study addresses this gap by conducting a comprehensive transcriptomic similarity analysis of widely used HCC mouse models. In this study, RNA-seq was performed on a model induced by double knockout of P53 and Pten via CRISPR/Cas9 in HBV-transgenic mice. Additionally, RNA-seq data from 2345 various other models induced by different methods were collected from GEO databases. The gene expression profiles, immune microenvironments, and metabolic pathways of these models were compared with those of human HCC. The analysis revealed distinct transcriptomic features among the different models. The HBV + P53&Pten KO model demonstrated the highest overall similarity to human HCC across various parameters. It shared a high degree of overlap in differentially expression genes (DEGs) between tumor and non-tumor tissues with human HCC, exhibited a transcriptome profile and immune cell infiltration pattern closely resembling human HCC, and showed metabolic alterations similar to those in human HCC. Conversely the DEN + CCl4-induced model showed the lowest similarity to human HCC in transcriptome profiles and DEGs and exhibited a distinct immune microenvironment with high NK cell infiltration, with minimal metabolic differences between tumor and non-tumor tissues. This study highlights the importance of selecting appropriate HCC mouse models for research. The HBV + p53&Pten KO model emerged as the most promising due to its remarkable similarity to human HCC across various aspects.
Keywords: CRISPR/Cas9; HCC; mouse model; similarity analysis; transcriptome.
© 2024 Wiley Periodicals LLC.