Deleting in vivo β-catenin degradation domain in mouse hepatocytes drives hepatocellular carcinoma or hepatoblastoma-like tumors

Background and aims One-third of hepatocellular carcinomas (HCCs) have mutations that activate the β-catenin pathway with mostly CTNNB1 mutations. Mouse models using Adenomatous polyposis coli ( Apc ) loss-of-functions (LOF) are widely used to mimic β-catenin-dependent tumorigenesis. Considering the low prevalence of APC mutations in human HCCs we aimed to generate hepatic tumors through CTNNB1 exon 3 deletion (β cat Δex3 ) and to compare them to hepatic tumors with Apc LOF engineered through a frameshift in exon 15 ( Apc fs-ex15 ). Methods We used hepatic-specific and inducible Cre-lox mouse models as well as a hepatic-specific in vivo CRISPR/Cas9 approach using AAV vectors, to generate Apc fs-ex15 and βcat Δex3 hepatic tumors harboring activation of the β-catenin pathway. Tumors generated by the Cre-lox models were analyzed phenotypically using immunohistochemistry and were selected for transcriptomic analysis using RNA-sequencing. Mouse RNAseq data were compared to human RNAseq data (normal tissues (8), HCCs (48) and hepatoblastomas (9)) in an integrative analysis. Tumors generated via CRISPR were analyzed using DNA sequencing and immunohistochemistry. Results Mice with βcat Δex3 alteration in hepatocytes developed liver tumors. Generated tumors were indistinguishable from those arising in Apc fs-ex15 mice. Both Apc fs-ex15 and βcat Δex3 mouse models induced two phenotypically distinct tumors (differentiated or undifferentiated). Integrative analysis of human and mouse tumors showed that mouse differentiated tumors are close to human well differentiated CTNNB1 -mutated tumors, while undifferentiated ones are closer to human mesenchymal hepatoblastomas, and are activated for YAP signaling. Conclusion Apc fs-ex15 and βcat Δex3 mouse models similarly induce tumors transcriptionally close to either well differentiated β-Catenin activated human HCCs or mesenchymal hepatoblastomas.