header


11.GENETICS DIVISION


    The research goals of the Genetics Division are elucidation of the molecular basis of carcinogenesis for the development of new strategies for cancer control.

A New Whole Genome Amplification Method for Genome Research

    A whole genome amplification method, called SATAN, was developed based on an adaptor-ligation mediated PCR of randomly fragmented DNAs. The high fidelity of the amplification has been demonstrated by microsatellite allelotyping and array CGH, etc. SATAN could provide DNA sufficient for a variety of genome analyses even from 100-1000 laser-captured cells from paraffin-embedded tissues.

Microarray Analysis of Peritoneal Wash Samples to Predict Recurrence of Gastric Cancer

    To identify a set of sensitive and specific multiple markers for the risk prediction of recurrence, peritoneal lavage cells were analyzed by a microarray of 12,626 genes. Eleven genes specific to cytology-positive samples were identified including those which are more specific than CEA. Non-peritoneal recurrences within 2 years were also associated with a positive RT-PCR of the peritoneal wash samples.

Studies on Gene Amplification in Human Cancer

    The amplification status of 45 esophageal squamous cell carcinoma (ESCC) tissues and cell lines was examined by DNA array CGH and Southern blot analyses (153). Eight of the 51 loci examined (15.7%) were found to be amplified in at least one of the samples. The analysis of the 17q12 amplicon identified a new gene, CAB2, completing the list of the genes in the amplicon. CAB2 was shown to be a human homologue of the yeast COS16 required for the repair of DNA double-strand breaks (154).

WNT Signaling Pathway

    The WNT signaling pathway is impli-cated in carcinogenesis, embryogenesis and organogenesis (155-157). In 2002, cloning and characterization of WNT8B, Wnt14, Wnt14b, GIPC2, GIPC3, VANGL1, ARHV and SOX17 (158-166), and expression of WNT signaling molecules in various types of human cancer (167-178) have been reported. Gene finding has been significantly accelerated due to the paradigm shift from the bench-top approach to the desk-top approach (179,180). The bioinformatics approach also revealed that the WNT and FGF gene clusters are recombination hot spots associated with carcinogenesis (181). A genome-wide expression profile of WNT signaling molecules showed a dramatic change during neuronal differentiation of NT2 cells (182), and some WNT signaling molecules, such as WNT2 and WNT5A, were up-regulated in primary gastric cancer due to cancer-stromal interaction (183,184). Genetic alterations, transcriptional regulation and the function of WNT signaling molecules as well as other genes were analyzed in collaboration with other laboratories (185-187).

Other Genes and Molecules Relevant to Cancer Biology

    The researchers of the Division also reported cloning and characterization of ST7L (a paralog of the tumor suppressor gene ST7), OSR-1, WINS1, Wins2 (mammalian homologs of Drosophila genes implicated in gut development) (188-192), the expression of TFF1, TFF2 and TFF3 in gastric cancer (193), functional analyses of FOXO transcriptional factors (194), G9a histone methyltransferase (195) and the ArpNa nuclear actin-related protein (196). A study on the SYT-SSX1 fusion gene found in synovial sarcomas with t(X;18)(p11.2;q11.2) translocation suggested the induction of transcriptional dysregulation. A reporter gene assay for evaluation of tissue- or cancer cell-specific transcriptional responses to estrogen was also developed (197).

Molecular Basis of Cancer Susceptibility

    Identification of the genetic predisposition of "common" cancers, such as gastric and lung cancers or leukemia, is underway as a joint project between the of Center for Medical Genomics, and several other Divisions and Hospitals as part of the Millennium Genome Project (MGP). The 367 candidate genes have been selected from various pathways such as DNA repair, the immune system, cell cycle regulation, xenobiotics metabolisms, transcriptional factors, oncogenes and tumor suppressor genes and resequenced in more than 200 cancer patients and non-cancer volunteers. About 3,000 SNPs and other minor variations such as small insertions and deletions have been identified aided by a newly developed algorithm filed for a patent application. From the SNPs identified, 51 non-synonymous SNPs on the 37 DNA repair genes were examined for associations with lung cancer risk in a case-control study in collaboration with the Biology Division. So far, 6 SNPs showed a significant association. The biological activities of DNA repair factors with SNPs are also being investigated.
    In 2001, the MGP officially adopted a genome-wide association study as a strategy complimentary to the candidate gene approaches. The markers for the genome scan are c.a. 100,000 gene-associated SNPs collected from the Japanese population by Human Genome Variation Team of the MGP (http://snp.ims.u-tokyo. ac.jp/). Based on the technology transferred from RIKEN, high-throughput SNP typing on poorly differentiated gastric cancer was started in 2002.

Drug Design-tool Using VR Technology

    A computer aided molecular modeling system was designed using virtual reality technology. The most characteristic function of the prototype system is enabling its user to "touch" and "feel" the electrostatic potential field of a protein or a drug molecule through a globular probe controlled by a force feedback device (198).

Transcription Factor Database and Related Tools

    A transcription Factor Database (TFDB) with binding site sequence information has been established and maintained by a researcher in the Division. The website also offers tools for finding transcriptional regulatory elements by analysis of the human-mouse homology and CpG island distribution, etc.