The focus of the Section for Studies on Metastasis lies in the development of novel animal models, methods and strategies to study tumorigenesis and cancer metastasis. The specific activities in 2007 were as follows: 1) Atelocollagen-mediated drug discovery and oligonucleotide cancer therapy in animal models; 2) differentiation of endodermal cells (hepatocytes and pancreatic cells) from stem cells; 3) studies for establishment of rat embryonic stem (ES) cells.

RNA interference (RNAi) offers the potential of a novel therapeutic approach for treating cancer and virus infection. With the increased potential of RNAi as a therapeutic strategy, noninvasive assessment of small interfering RNA (siRNA) and micro RNA (miRNA) delivery to tissues of interestusing clinically relevant imaging paradigms is required to conceive and optimize experimental treatment strategies. The in vivo analysis of siRNA delivery to target tumor tissues was monitored by optical imaging of siRNA-mediated silencing. The other obstacle to developing RNAi-based therapies, is the systemic delivery of the RNAs to the metastatic tumors in mice. An atelocollagen- mediated nano-particle delivery system to deliver siRNAs and miRNAs to metastatic animal models of human cancers was studied and it was found that these RNAi strategies effectively inhibited metastasis of human prostate and breast tumors in mice. Thus, such a unique approach to RNAi delivery and imaging strategies creates a foundation for the development and optimization of a new class of promising therapeutics against cancer ⁽¹¹²⁾. Because nano-size particles of these complexes can be immobilized on a plate, they are amenable to high-throughput screening of nucleic-acid drugs by reverse transfection. RNAi-based reverse genetics using these complex particles can become a powerful tool to understand gene functions in cells and allow the identification of drug targets.

To elucidate the epigenetic role of RNAi in mammals, we disrupted the gene for Eif2c2 (Ago2), which works as the sole slicer of RNAi in the Argonaute family. In mice, disruption of Eif2c2 led to early embryonic lethality after the implantation stage ⁽¹¹³⁾. This phenotype was completely different from that in a previous report, but somewhat similar to the disruption of Dicer1, another important component of RNAi. Eif2c2 is not required for the maintenance of DNA methylation in imprinted genes, centromeric repeats or Xist. This suggests that developmental defects in the Eif2c2-deficient mouse were caused not at the transcriptional level, but rather at the posttranscriptional level through the miRNA- protein complex.

Many studies on stem cell plasticity are challenging the concept that stem cells contain an intrinsically predefined, unidirectional differentiation program. This means that the developmental fate of a stem cell is dependent on the general potential of the cell (pre-determined stem cell fate) as well as on microenvironmental cues, such as stimuli from growth factors (stem cell niche) ⁽¹¹⁴⁾. The hepatocyte differentiation ability of stem cells from two different sources, namely embryonic stem cells and adult stem cells, has been reported. All of those stem cells were revealed to have the ability to give rise to hepatocyte-like cells using different induction strategies. However, it is still not clear which of those stem cells would be the best source for hepatocyte replacement or which the best protocol would be ⁽¹¹⁵⁾.

Recent observations indicate that several stem cells can differentiate into hepatocytes; thus, cell-based therapy is a potential alternative to liver transplantation. The goal of the present study was to examine the in vitro hepatic differentiation potential of adipose tissue-derived mesenchymal stem cells (AT-MSCs) ⁽¹¹⁶⁾. AT-MSCs from patients of different ages were used, and it was found that after incubation with specific growth factors (hepatocyte growth factor [HGF], fibrobroblast growth factor [FGF1], FGF4), the CD105-fraction of AT-MSCs exhibited high hepatic differentiation ability in an adherent monoculture condition. CD105-AT-MSC-derived hepatocyte-like cells revealed several liver-specific markers and functions, such as albumin production, low-density lipoprotein uptake, and ammonia detoxification. More importantly, after transplantation into mice, CD105-AT-MSC- derived hepatocyte-like cells, became incorporated into the parenchyma of the liver. Thus, adipose tissue is a source of multipotent stem cells that can be easily isolated, selected, and induced to differentiate into mature, transplantable hepatocytes ⁽¹¹⁷⁾. The fact that they are easy to procure ex vivo in large numbers makes them an attractive tool for clinical studies in the context of establishing an alternative therapy for liver dysfunction.

Invasion is a hallmark of malignant tumors, such as gliomas, and its progression is followed by interaction between tumor cells and the extracellular matrix (ECM). Several ECM components and their fragments have been reported that exert regulatory functions on tumor progression and have potential to reduce invasion and tumorigenesis. Although type I collagen is the most abundant protein of the ECM components, no type I collagen molecule and/or fragments have been demonstrated to have any suppressive efficacy on the malignant phenotypes of a tumor.

This study examined the role of type I collagen in the invasiveness of the malignant human glioma cell line T98G, by the introduction of the type I collagen gene (HCOL1A1). Overexpression of HCOL1A1 in the malignant glioma cells significantly suppressed invasion and tumor growth both in vitro and in vivo ⁽¹¹⁸⁾. In addition, HCOL1A1 induced apoptosis of glioma cells, which suggested that apoptosis participates in a mechanism of suppression of malignancy of T98G glioma cells by forced expression of HCOL1A1. Therefore, the introduction of HCOL1A1 may form the basis of a novel therapeutic approach for the treatment of malignant human glioma.

The rat is a reference animal model for physiological studies and for the analysis of multigenic human diseases, such as hypertension, diabetes, neurological disorders and cancer. Rats have long been used in extensive chemical carcinogenesis studies. Thus, the rat embryonic stem (rES) cell is an important resource for the study of disease models. Attempts to derive ES cells from various mammals, including the rat, have not succeeded. Two independent rES cells were established from a Wister rat blastocyst with undifferentiated characteristics, such as Nanog and Oct-3/4 genes expression, and show stage-specific embryonic antigen (SSEA) -1, -3, -4, and TRA-1-81 expression. The cells were successfully cultured in an undifferentiated state over 18 passages with maintenance of more than 40% of the normal karyotype. Their pluripotent potential was confirmed by the differentiation into derivatives of endoderm, mesoderm and ectoderm. More importantly, the rES cells allowed the generation of chimera rats. Therefore, pluripotent rES cell lines are widely used to produce genetically modified experimental rats for the study of human diseases. Efforts to produce a germline-chimera rat is on going.