Targeting β-catenin in hepatocellular cancers induced by coexpression of mutant β-catenin and K-Ras in mice

Recently, we have shown that coexpression of hMet and mutant-β-catenin using sleeping beauty transposon/transposase leads to hepatocellular carcinoma (HCC) in mice that corresponds to around 10% of human HCC. In the current study, we investigate whether Ras activation, which can occur downstream of Met signaling, is sufficient to cause HCC in association with mutant-β-catenin. We also tested therapeutic efficacy of targeting β-catenin in an HCC model. We show that mutant-K-Ras (G12D), which leads to Ras activation, cooperates with β-catenin mutants (S33Y, S45Y) to yield HCC in mice. Affymetrix microarray showed > 90% similarity in gene expression in mutant-K-Ras-β-catenin and Met-β-catenin HCC. K-Ras-β-catenin tumors showed up-regulation of β-catenin targets like glutamine synthetase (GS), leukocyte cell-derived chemotaxin 2, Regucalcin, and Cyclin-D1 and of K-Ras effectors, including phosphorylated extracellular signal-regulated kinase, phosphorylated protein kinase B, phosphorylated mammalian target of rapamycin, phosphorylated eukaryotic translation initiation factor 4E, phosphorylated 4E-binding protein 1, and p-S6 ribosomal protein. Inclusion of dominant-negative transcription factor 4 at the time of K-Ras-β-catenin injection prevented HCC and downstream β-catenin and Ras signaling. To address whether targeting β-catenin has any benefit postestablishment of HCC, we administered K-Ras-β-catenin mice with EnCore lipid nanoparticles (LNP) loaded with a Dicer substrate small interfering RNA targeting catenin beta 1 (CTNNB1; CTNNB1-LNP), scrambled sequence (Scr-LNP), or phosphate-buffered saline for multiple cycles. A significant decrease in tumor burden was evident in the CTNNB1-LNP group versus all controls, which was associated with dramatic decreases in β-catenin targets and some K-Ras effectors, leading to reduced tumor cell proliferation and viability. Intriguingly, in relatively few mice, non-GS-positive tumors, which were evident as a small subset of overall tumor burden, were not affected by β-catenin suppression.

Conclusion: Ras activation downstream of c-Met is sufficient to induce clinically relevant HCC in cooperation with mutant β-catenin. β-catenin suppression by a clinically relevant modality is effective in treatment of β-catenin-positive, GS-positive HCCs. (Hepatology 2017;65:1581-1599).