UCSF

The Role of Signaling Pathways during the development of Hepatocellular Carcinoma

Susie Lee

Abstract:

The roles of genetic alterations in human hepatocellular carcinoma (HCC) are poorly understood. This dissertation's goals are to examine whether these de-regulated genes are sufficient to promote liver tumorigenesis and characterize the resulting effects on cellular processes and signaling pathways.

We first describe the role of receptor tyrosine kinase (RTK) feedback inhibitor Sprouty2 (Spry2) during hepatocarcinogenesis. We expressed dominant negative form of Spry2 (Spry2Y55F) and/or activated &beta-catenin into mouse hepatocytes and found that both factors are sufficient to induce liver tumorigenesis in vivo. Proliferation and angiogenesis as well as MAPK signaling are up-regulated in Spry2Y55F/&beta-catenin induced HCC. However, the loss of Spry2 function alone is not capable of promoting MAPK activity. A correlation between the over-expression of RTK c-Met and down-regulation of Spry2 has been identified in human HCC. To investigate the synergistic effects of c-Met and Spry2 during hepatocarcinogenesis, we injected c-Met and/or Spry2Y55F into Ink4A/ARF null mice. Co-expression of both factors resulted in the development of HCC. The tumors displayed increases in MAPK and AKT activity. These findings demonstrate the role of Spry2 as a tumor suppressor in hepatocarcinogenesis.

We next analyzed the role of cell cycle regulator cyclin D1 (CCND1) in hepatocarcinogenesis. Co-expression of CCND1 and c-Met resulted in the development of HCC in vivo. Although, this cyclin is a downstream target of &beta-catenin, the absence of CCND1 did not deter &beta-catenin/c-Met induced liver tumorigenesis in mice. These results indicate that CCND1 is sufficient, but not required for &beta-catenin/c-Met induced hepatocarcinogenesis.

Finally, we describe the role of polycomb repressor Bmi1 in liver carcinogenesis. We expressed Bmi1 and/or activated Ras and found that both factors cooperate to promote hepatocarcinogenesis in vivo. Although Bmi1 is a known inhibitor of Ink4A/ARF, we found no evidence that Bmi1 does so during tumorigenesis. Our study indicates that Bmi1 cooperates with Ras to induce liver carcinogenesis in an Ink4A/ARF independent manner.

Altogether, our studies have validated clinical findings and demonstrated that combinations of MAPK signaling factors and genes, including CCND1 and Bmi1, are sufficient to promote hepatocarcinogenesis.