1. Understanding the molecular basis of gastrointestinal cancer

Constitutive activation of the Wnt/β-catenin signaling pathway plays a crucial role in colorectal and liver carcinogenesis. Through the screening of the entire coding region of AXIN1, we earlier reported pathogenic mutations in the AXIN1 gene in hepatocellular cell lines and the primary tissues. Mutations of AXIN1, APC, or β-catenin stabilize β-catenin and transactivate the Lef/Tcf transcription factors, resulting in the induction of downstream genes such as MYC and CCND1. We have identified a number of targets genes including RNF43, SP5, CLDN1, MT1-MMP1, ENC1, APCDD1, FRMD5, and IFIT2. Understanding of the function of target genes will elucidate the precise mechanisms of human carcinogenesis.
To identify novel targets for the diagnosis and treatment of human cancer, we utilized global gene expression data. Among the genes whose expression was frequently elevated in colorectal cancer, we focused on MRG domain binding protein (MRGBP). Our study revealed that MRGBP physically binds to a bromodomain protein (BRD) and regulates its stability and subcellular localization. Currently, we are trying to clarify the role of BRD in cancer cells and evaluate the inhibition of its interaction with MRGBP as a therapeutic option.

2. Establishment of a novel carcinogenic mouse model and its application to carcinogenic mechanism analysis and treatment development

Genetically engineered cancer mouse models are useful tools for studying the process and mechanism of tumor development. Intensive analysis of these models should provide better understanding of human carcinogenesis and facilitate the development of new therapies. We have recently demonstrated that liver-specific expression of oncogenic Kras in combination with homozygous Pten deletion in mice induced liver tumors. Analysis of the liver tumors showed that they histologically resembled human intrahepatic cholangiocarcinoma. In addition, we have established other mouse models of gastrointestinal, liver, and pancreatic cancer. Using cell lines and organoid systems established from these mouse models, we are now studying the molecular mechanism of carcinogenesis.

3. Cancer genome analysis

Next-generation sequencing (NGS) allowed us to analyze the comprehensive human genome. Application of NGS has facilitated the identification of germline changes responsible for hereditary diseases and somatic alterations in human neoplasms. In collaboration with Human Genome Center and Health Intelligence Center, we have been working on the determination of germline mutations in patients suspected of hereditary diseases and the identification of somatic mutations accumulated in cancer cells. Furthermore, we have established analytic pipelines using artificial intelligence for personalized medicine. These projects are aimed to use the information of personal genome and/or cancer genome in clinics.

4. Cancer drug discovery through a large chemical library screening

Establishment of well-designed high-throughput screening system is essential for the identification of small molecules that inhibit signaling pathways or molecules of interest. Recently, we elaborated an efficient cell-based reporter assay system named “bidirectional reporter assay” (Figure). Applying this assay system, we started a high-throughput screening of Wnt inhibitors using small molecule and natural compound libraries.

Figure. Schematic diagram of a bidirectional reporter assay

Page Top