Complete inhibition of mTORC1 cascade is required to suppress hepatocarcinogenesis induced by AKT and Ras co-expression

Abstract

Introduction: AKT and Ras pathways are frequently concurrently hyperactivated in HCC. Previously, we have shown that co-expression of constitutively active forms of AKT (Myr-AKT) and activated Ras (N-RasV12) rapidly induces liver tumors in mice. mTORC1 is the main downstream effector of AKT, acting by phosphorylation of RPS-6 and 4E-BP1, and thus regulating cell metabolism, protein translation and other cellular processes. However, in which manner mTORC1 and its key downstream effectors RPS-6 and 4E-BP1/eIF4E do contribute to hepatocarcinogenesis remains unknown. Methods: Plasmids encoding Myr-AKT, RasV12, 4EBP1A4 (dominant negative form of 4E-BP1), or 4EBP1WT (wild type) were delivered into the mouse liver by hydrodynamic injection. The mTORC1 inhibitor Rapamycin or vehicle were intraperitoneally administrated into the mice. Western blotting and immunohistochemistry were performed to analyze the expression levels of proteins in the tissues. Results: To determine whether the mTORC1 downstream effector RPS6 is required for hepatocarcinogenesis in AKT/Ras mice, we treated AKT/Ras mice with rapamycin (AKT/RAS/Rapa) or vehicle (AKT/RAS/Veh) daily for 7 weeks. All of the AKT/RAS/Veh mice developed large HCCs and were required to be euthanized. By contrast, none of the AKT/RAS/Rapa mice showed HCC, but microscopically small regressive tumor-like nodules and preneoplastic hepatocellular foci were still visible. In addition, withdrawal of rapamycin treatment was followed by a relapse in HCC development. Biochemical analysis demonstrated that Rapamycin inhibited the phosphorylation of RPS6, but had no effect on phosphorylated/inactivated levels of 4E-BP1. To define the role of 4EBP1/eIF4E in AKT/Ras-driven hepatocarcinogenesis, we overexpressed 4EBP1A4 or 4EBP1WT along with AKT and Ras into the mouse liver. Seven weeks after hydrodynamic injection, AKT/Ras/4EBP1WT mice developed large HCC which were equivalent to those developed in AKT/RAS/Veh mice. However, only few very small hepatocellular adenomas (HCA) developed in the livers of AKT/Ras/4EBP1A4 mice. These lesions, together with preneoplastic hepatocytes, occupied 40–60% of the liver parenchyma. 20 weeks post injection, the AKT/Ras/4EBP1A4 mice eventually developed large liver tumors, including HCC. These results indicated that inhibition of 4E-BP1/eIF4E axis efficiently delayed the AKT/RAS induced liver tumor, yet was unable to completely block the tumorigenesis process. Finally, we co-injected AKT/Ras mice with 4EBP1A4 and treated the mice with Rapamycin. We found that simultaneously blocking p-RPS6 and p-4EBP1 completely inhibited AKT/Ras induced hepatic carcinogenesis. Conclusion: Our experiments demonstrate the critical role of mTORC1 in mediating activated AKT and Ras induced liver tumor development in vivo. The two major downstream effectors of mTORC1, RPS6 and 4E-BP1/eIF4E, are both required for tumorigenesis. Yet they are likely to have distinct roles within the neoplastic process.