Search

To monitor human embryonic stem cell (hESC) self-renewal without differentiation, we used quantitative RT-PCR to study a selection of hESC genes, including markers for self-renewal, commitment/differentiation, and members of the TGF-beta superfamily and DAN gene family. Indeed, low commitment/differentiation gene expression, together with a significant self-renewal gene expres sion, provides a better pluripotency index than self-renewal genes alone. We demonstrate that matrices derived from human mesenchymal stem cells (hMSCs) can advantageously replace murine embryonic fibroblasts (MEF) or hMSC feeders. Moreover, a xenofree molecularly-defined SBX medium, containing a synthetic lipid carrier instead of albumin, can replace SR medium. The number of selected differentiation genes expressed by hESCs in these culture conditions was significantly lower than those expressed on MEF feeders in SR medium. In SBX, the positive effect of a non-physiological concentration of activin A (10-30 ng/mL) to reduce differentiation during self-renewal could also be obtained by physiological concentrations of TGF-beta(100-300 pg/mL). In contrast, these TGF-beta concentrations added to activin favored differentiation as previously observed with TGF-beta concentrations of 1 ng/mL or more. Compared to SR-containing medium, SBX medium promoted down-regulation of CER1 and LEFTIES and up-regulation of GREM1. Thus these genes better control self-renewal and pluripotency and prevent differentiation. A strategy is proposed to analyze, in more physiological, xenofree, molecularly-defined media and matrices, the numerous genes with still unknown functions controlling hESCs or human-induced pluripotent stem cells (iPS).

Here we present a simple two-step in vitro model of vascularized trophoblastic tissue derived from human embryonic stem (hES) cells. The first step is the formation of cystic embryoid bodies (EBs) in suspension in a semisolid methyl cellulose medium, within which an endothelial platelet/endothelial cell adhesion molecule-1 (PECAM-1(+)) cell network develops. In a second step, deposition of these EBs on the bottom of nontreated, polystyrene tissue culture plates, leads by centrifugal outgrowth of the EB to the emergence of an adherent cell layer, with which a PECAM-1(+) network is associated. Cells of this adherent layer expressed VE-cadherin (CD144), PECAM-1 (CD31), and alpha-fetoprotein (alpha-FP). Trophoblastic differentiation was strongly suggested by the secretion of beta-human chorionic gonadotropin (beta-hCG) and by the presence of the cytotrophoblast and syncytiotrophoblast marker GB25. The INSL4 gene, a cyto and syncytio-trophoblast marker, was also highly expressed in the adherent layer, as well as other trophoblast genes such as CGA, CDX1, CDX2, and HAND1, compared to hES cell gene expression taken as reference. In contrast, expression of self-renewal genes, such as TERT, POU5F1, ZFP42, GDF3, and NODAL were decreased. No ectodermal or endodermal genes were expressed, but the mesodermal genes PECAM-1 and GATA2 were. The possibility of removing the EBs during the second step would permit analysis of their relative contribution to angiogenesis or possible hemangioblast formation, compared to that of the trophoblastic adherent layer. This primitive vascularized trophoblastic model could also provide a tool to study early steps of normal and pathological placental development.

Human umbilical cord blood CD34+ cells were purified using a two-step procedure by elimination of the soybean agglutinin-binding cells and by a positive panning selection with a CD34 monoclonal antibody. The isolated fraction was 88-97% pure CD34+ cells. A yield of 48.5% was obtained when comparing the number of cells recovered in the CD34(+)-purified fraction and the number of CD34+ cells detected in the initial mononuclear cell fraction. By flow cytometry, we observed that the CD34+ cells that were not recovered were those that had the lower expression of CD34 antigen and were therefore the more mature cells. A high recovery of CFU-GEMM progenitors (73.9%) was also observed. These data suggest the possibility of purifying CD34+ umbilical cord blood cells for clinical applications, in particular for umbilical cord blood banking.