The transplantation of stem cells capable of generating various descendants is useful for compensating for a loss of tissues after the chemotherapy of malignant diseases. However, it is limited to isolate somatic stem cells from human tissues due to the lack of accessibility and difficulty in obtaining sufficient cell numbers for the replacement of adult tissues. Instead of somatic stem cells, pluripotent stem cells such as embryonic stem (ES) or induced pluripotent stem (iPS) cells have been expected as a source for cell replacement therapy. ES cell is a pluripotent cell line which can give rise to various cell lineages including three germ layers. When cultured in the presence of LIF and serum, ES cell undergo unlimited symmetrical self-renewal divisions. Thus, it is reasonable to use these cells because of their characters containing the pluripotency and unlimited growth.

We are especially interested in the question of how we manipulate pluripotent stem cells to achieve the induction of the descendants that sufficiently compensate for the defect of human tissues. However, in vitro ES cell differentiation culture is still very complex as various cell types are simultaneously generated in the culture. This is one of disadvantages of this system. In order to overcome this problem, we have attempted to establish molecular markers for defining the cell lineages generated in cultures. Using such markers for identifying the point of lineage divergence, we were able to show that Vascular endothelial growth factor receptor 2 (VEGFR2, FLK1) and Platelet-derived growth factor receptor- α( PDGFRα) are useful surface markers that prospectively distinguish lateral and paraxial mesoderm, respectively (Fig.1). More recently, we showed that another set of mesoderm cells expressing goosecoid⁺(Gsc⁺), PDGFRα ⁺( α R ⁺ ), E-cadherin ^– (ECD ^– ) was generated from Gsc ⁺ , α R ⁺ , ECD ⁺ mesendoderm and was able to be induced from ES cells under an Activin-containing defined culture condition (Fig.1) .

Figure 1. Differentiation pathways from ES to germ layer cells.

Based on these results, we wish to establish the culture condition that efficiently provides us to generate the cells for the purpose and to investigate the molecular and cellular mechanisms underlying the cell specification during ES cell development.

Project 1. “Induction and specification of various mesoderms from ES cells “

Using mouse ES cells, we study molecular mechanims of mesodermal induction and specification. We have recently established efficient in vitro mesoderm differentiation systems of ES cells. We are also interested in developing serum-free mesoderm induction systems. Using these systems, we wish to understand signals that define the identity of cells in the embryonic mesoderm systems.

Project 2. “Establishment of in vitro endodermal Differentiation Systems Using ES Cells”

We wish to understand the molecular and cellular bases of mesendoderm induction and endoderm specification following the primary mesendoderm induction. Recently, we established an efficient in vitro system for ES cells to differentiate into mesendoderm that can differentiate into both mesoderm and definitive endoderm. To gain insights into molecular mechanism underlying endoderm development, we isolate ES cell-derived intermediates in the way of endoderm development and analyze genes expression profiles in these cells.

Project 3. “Studies on the origin of the mesenchymal stem cells”

Mesenchymal stem cells (MSCs) are defined by their ability both to undergo sustained proliferation in vitro and to give rise to multiple mesenchymal cell lineages including bone, cartilage, and fat cells. We wish to define the differentiation pathways of MSCs. Using ES cell culture, we show that Sox1 ⁺ neuroepithelial cells generate MSCs at the highest efficiency. In early embryos, we can induce MSCs from Sox1 ⁺ cells but not from PDGFRα⁺ mesoderm. However, as the development proceeds, neuroepithelium-derived MSCs gradually decrease and most MSCs in postnatal bone marrow are derived from other origins, which are also enriched in the PDGFR α⁺ population. Thus, at least, MSCs are generated from two sources, neuronal and non-neuronal, with those derived from neuroepithelium constituting the earliest wave (Fig.2).

Figure 2. New pathway from neuroepithelium to mesenchymal stem cell.

Project 4. “Application of new concepts to the investigation of clinical medicine”

One of our major aims is to apply the new concepts obtained from the studies on ES cell culture to clinical medicine. Very recently, we isolated ARID3B, a member of the AT-rich interaction domain (ARID) family, from the analysis of ES cell culture. The null mice have shown embryonic lethality due to the severe defect of neural crest. ARID3B was specifically expressed in advanced stage of neuroblastoma (NB) that is one of major tumors in childhood. ARID3B by itself immortalized mouse embryonic fibroblasts (MEF) in vitro and confers malignancy to MEF in combination with MYCN, the best characterized oncogene for NB. In vitro growth of NB cell lines was significantly inhibited by the suppression of ARID3B expression. Our study demonstrates that in vitro ES cell culture is powerful for identifying novel molecules to play an essential role in oncogenesis as well as embryogenesis (Fig.3).

Figure 3. ARID3B is involved in the embryogenesis (left) and tumorgenesis (right).