Treatment of human embryonic stem cells with different combinations of priming and inducing factors toward definitive endoderm.

Despite the enormous progress in studying definitive endoderm (DE) differentiation from human embryonic stem cells (hESCs), none of the reported protocols have produced a universal, cost-effective, and competent DE with the capability to further differentiate into endodermal derivatives. In this study, by using a 2-step differentiation strategy, we have treated hESCs for 1 day with "priming" small molecules (SM), [stauprimide, NSC-308848, rapamycin (Rapa), and/or CHIR] and for the next 3 days with "inducing" SM (LY294002, cymarin, IDE1, and/or IDE2) in conjunction with activin A. In the positive control group, we treated hESCs with Wnt3a (25 ng/mL) for 1 day and activin A (100 ng/mL; W/A100-A100) for the next 3 days. Gene expression analysis showed that treatment of hESCs with 100 nM Rapa and 50 ng/mL activin A (Rapa-A50) out of 25 combinations of factors gave rise to higher expressions of 2 DE-specific genes, SOX17 and FOXA2. Similar results were obtained after treating 2 other hESC lines with this regimen. To investigate the competency of Rapa-A50-induced DE for further differentiation into endodermal derivatives, these cells and W/A100-A100-induced DE cells (positive control) were further differentiated into pancreatic progenitors (PP), then into pancreatic endocrine (PE) cells using 5 previously described differentiation protocols. Gene analysis of differentiated cells showed that the established protocols were insufficient to enable universal differentiation into PE, whereas Rapa-A50-induced DE cells were more competent for PP differentiation in a protocol-dependent manner. Additionally, Rapa-A50-induced DE had the capability to differentiate into hepatocyte-like cells (HLCs) as efficiently as W/A100-A100-induced DE. These data have indicated that hESCs primed with Rapa, and induced by a lower concentration of activin A, could lead to DE that had the capability to further differentiate into HLCs and PP cells, but not PE cells. Thus, current protocols for the differentiation of DE into PE still need additional study.