Abstract. The teratocarcinoma-derived C1 clone behaves as a mesodermal tripotential progenitor cell whose choice of fate, either osteoblast, chondroblast, or adipoblast, is strictly dependent on the spatial organization of the cells and the nature of the induction. In the absence of cell contact before the addition of inducers, the C1 cells maintain a stable undifferentiated phenotype while expressing potential regulators of embryonic mesodermal stem cell fate such a M-twist and Idl. Upon establishment of cell contacts before the induction of differentiation, the early genes characteristic of the three fates become expressed. In the presence of [3 glycerophosphate and ascorbate, provided the cells have formed aggregates, 95% of the C1 cells mineralize with a kinetics of gene expression close to that of osteoblasts (Poliard, A., D. Lamblin, P. J. Marie, M. H. Buc, and O.Kellermann. 1993. J. Cell Sci. 106:503-512). With 10-6M dexamethasone, 80% of the same aggregates differentiate into loci of chondroblast-like cells. The kinetics of expression of the genes encoding type II, IX, X, and XI collagens, aggrecan and link protein during the conversion towards cartilage hypertrophy resembles that accompanying in vivo chondrogenesis. The synergistic action of dexamethasone and insulin convert most confluent C1 cells into functional adipocytes and induce a pattern of gene expression close to that reported for adipoblast cell lines. The C1 clone with its capacity to differentiate along three alternative pathways with high frequency, therefore appears as a valid in vitro model for deciphering the molecular basis of mesoblast ontogeny.

We report the isolation and characterization of a serotoninergic cell line, iC1l, derived from a mouse teratocarcinoma. The clone iCli was immortalized through the expression of the simian virus 40 oncogenes. iCli presents two states: an immature epithelial-like state (1Cli precursor) and a more differentiated state (1C01*). After induction by

The inhibitor of DNA-methylation, 5-azacytidine (5-AzaC) induced the appearance of cytokeratin-containing cells in several mesenchymal cell lines such as teratocarcinomaderived fibroblasts, preadipocytes and myoblasts, NIH-3T3 fibroblasts and human embryonic fibroblasts. At optimal 5-AzaC concentrations the proportion of such cells was in the range of 10-1 compared with 10-6-10-4 in non-treated cultures. Dose-response curves indicated that the induction of cytokeratin was the result of an interaction of the drug with few targets. Stable, mature, keratinocyte cell lines, as well as lines of myoblasts and astrocytes, could be isolated from a teratocarcinoma-derived preadipocyte line, showing that 5-AzaC is able to provoke a wide range of complete phenotypic conversions. In these cell lines, the intermediate filaments corresponded to the morphological phenotype. Altogether, the results suggest that 5-AzaC preferentially activates certain genes.

Abstract. The hybrid plasmid pK4 containing the early genes of the simian virus SV-40, under the control of the adenovirus type 5 Ela promoter, was introduced into the multipotent embryonal carcinoma (EC) 1003. Expression of the SV-40 oncogenes was observed at the EC cell stage, and this allowed the derivation of immortalized cells corresponding to early stages of differentiation.Among the immortalized mesodermal derivatives obtained, one clone, C1, is committed to the osteogenic pathway. C1 cells have a stable phenotype, synthesize type I collagen, and express alkaline phosphatase activity. Although immortalized and expressing the SV-40 T antigen, the cells continue to be able to differentiate in vivo and in vitro. In vivo, after injection into syngeneic mice, they produce osteosarcomas. In vitro, the cells form nodules and deposit a collagenous matrix that mineralizes, going to hydroxyapatite crystal formation, in the presence of/~-glycerophosphate. This clonal cell line, which originates from an embryonal carcinoma, therefore differentiates into osteogenic cells in vivo and in vitro.This immortalized cell line will be useful in identifying specific molecular markers of the osteogenic pathway, to investigate gene regulation during osteogenesis and to study the ontogeny of osteoblasts.N~ current view of the development of bone cells is that osteocytes derive from primitive mesoblastic cell precursors through a cascade of events. These involve, in the case of intramembranous ossification, (a) proliferation of primitive mesoblasts, (b) differentiation of these cells into an osteoprogenitor cell, then into osteoblasts, and (c) maturation of osteoblasts with synthesis of a collagen matrix and mineralization (5,10,39,44). Up until now, models for following bone differentiation in vitro have been of two types: (a) osteoblast-like clones explanted from normal bone, and (b) clones derived from osteogenic osteosarcomas.(a) The former are normal explanted cells capable of progressively synthesizing a bone-like tissue in the presence of organic phosphate and ascorbic acid (1,3,8,9,28,35,46). This process and its hormonal regulation can therefore be studied in vitro. Collagen I maturation, in particular, has been extensively studied (11). However, the potential for division of such cells remains limited, and this precludes molecular studies. In addition, phenotypic changes with loss of osteoblastic properties often occurs in culture (47).(b) Clones derived from rat osteosarcomas have also been useful in investigating the effects of hormones (parathyroid hormone; prostaglandins) and vitamins (retinoids and vitamin D3) on bone development (17,25,29,30,41). The formation of a mineralized matrix has, however, been shown to require either the implantation of the tumoral cells in diffusion chambers within the animal (43), or the growth of the ceils in agar (37).In both model systems, the major problems encountered are the heterogeneity of the cell population within a clone, and the lack of specific markers allowing formal identi...

No abstract