RNA splicing is the process of removal of intronic sequences from the primary RNA transcript before the final mRNA is generated. Unlike lower organisms, the vast majority of mammalian genes are spliced. Most genes give rise to multiple mRNAs resulting from differential promoters or termination sequences, or the use of alternative exons. In the broader clinical picture, alternative splicing is an important new area of investigation. Alterations in RNA splicing have now been found in many cancers. As cancer is a leading cause of death among the VA population, understanding how these oncogenic splice variants occur and how they contribute to carcinogenesis is paramount to our understanding of the pathogenesis of the disease and our discovery of novel therapeutic approaches. We have recent data in mice that eliminating a particular RNA splicing factor causes chronic liver damage due to hepatocyte apoptosis and compensatory proliferation. As the mice age, their livers show steatohepatitis and fibrosis, and eventually 100% of the mice succumb to hepatocellular carcinoma (HCC). The loss of the splicing factor does not cause tumors but rather creates a pre-disposition to cancer, similar to a tumor suppressor gene. We believe this model has many parallels to human HCC and can provide great insights into the pathogenesis of this disease that could guide therapeutic development. Based on extensive preliminary data on our splicing factor knockout mouse, we are proposing a comprehensive series of experiments to understand how loss of a single splicing factor in the liver can lead to metastatic hepatocellular carcinoma. These studies will address key questions concerning the fundamental biological process of carcinogenesis and will integrate cell and molecular biological experiments with physiological studies in mice lacking specific splicing factors in liver. We will establish the relevance of this model for human liver cancer by assessing how frequently SRSF3 loss occurs in human HCC, whether loss occurs in all HCC or particular subtypes, and whether loss occurs in early-stage liver disease such as NAFLD, NASH or cirrhosis. In the mouse, loss of SRSF3 predisposes to HCC but not every cell becomes transformed, suggesting that somatic mutations or epigenetic changes must be acquired to produce the carcinoma phenotype. We will attempt to identify these secondary genetic changes by sequencing somatic mutations in the mouse tumors, or assessing changes in CpG methylation in differentially-methylated regions. We will also test whether these genetic changes allow the tumors to become independent of SRSF3 status by restoring SRSF3 expression in the tumor cells. We investigate the molecular mechanisms underlying the aberrant RNA splicing and the associated cellular changes using a combination of CLIP-seq and transcriptome sequencing and splice junction arrays to identify the direct targets for SRSF3 in the hepatocyte. These targets will be tested for SRSF3-dependence using in vitro cultured human hepatoma cells. We will also test a number of SRSF3 mutations have been identified in a variety of cancers including HCC. We will test two potential mechanisms that may contribute to apoptosis, fibrosis, and cell proliferation. We are attempting to connect changes in splicing of the fibronectin and insulin receptor genes with specific intermediate phenotypes, and with tumor growth. These two genes are direct targets for SRSF3 and their splicing is altered in the knockout mouse. We will test whether miss-splicing of the fibronectin gene causes oxidative stress and fibrosis, and whether miss-splicing of the insulin receptor gene allows IGF2 to stimulate tumor cell proliferation and prevent the normal terminal endo-replication of the hepatocyte. These experiments will combine new genetic knockouts with pharmacological stimulation or inhibition of individual pathways. Ultimately, we will test whether these two pathways contribute to hepatocellular carcinogenesis.
More than 700,000 people are diagnosed with liver cancer and 600,000 people die each year worldwide. The incidence of liver cancer is increasing in the US and the American Cancer Society estimates that 30,460 new cases will be diagnosed and 21,670 people will die of liver cancer in 2013. Part of the increased incidence in the US is thought to be due to HCV infections acquired during the 1960s and 1970s. Aside from HBV and HCV infection, non-alcoholic fatty liver disease, non-alcoholic steatohepatitis and cirrhosis are major risk factors for liver cancer. Liver cancer is of particular concern to the VA as HCV infection rates are higher than the general population and are estimated to be between 7 and 20% of all VA patients. A detailed understanding of the mechanisms involved in hepatocellular carcinoma would greatly help the treatment of Veterans with liver cancer.
