We all hypothesized that highly qualified and quantified hepatocytes might be obtained if USCs are effectively removed during hepatic differentiation. To perform selective removal of USCs, YM155 was utilized in this study. Firstly, serial concentrations of YM155 had been used to treat human PSCs (QIA7 and WA01), distinguishing cells (QIA7-iHeps) and hAT-SCs as a somatic cell regarding determining the relevant treatment concentration (Fig.   3a ). Each cell type showed various dose responses against YM155 (Fig.   3a ). The inhibitory concentration 50 (IC 50 ) of QIA7, WA01, QIA7-iHeps and hAT-SCs was 8 nM, 14 nM, 140 nM and 3086 nM, respectively. Human PSCs had been affected by lower concentrations of YM155 than hAT-SCs plus QIA-iHeps. Therefore , we selected the ranges of YM155 (1– 100 nM) for later study. To determine an optimum time point of YM155 treatment, the selected levels of YM155 were treated at stage I plus stage II. In our preliminary study, the more YM155 therapy at stage I, the more cells survived (Additional document 2 : Figure S2). The majority of the differentiating cells were easily detached from plates simply by YM155 treatment at stage I. Therefore , YM155 had not been applicable at stage I. On the contrary, the differentiating tissue were more tolerant to YM155 treatment at phase II (day 6 of differentiation) than at phase I (day 4 of differentiation) (Fig.   3b ). To investigate cytotoxic effects of YM155 at stage II, we estimated changes of apoptosis-related genes on the whole differentiation (Additional file 2 : Figure S3). BIRC5 was down-regulated from day one of differentiation and maintained during the whole hepatic difference. BAX , a pro-apoptotic gene, was up-regulated from stage II to phase III. The expression of BCL-2 , an anti-apoptotic gene, was not detectable or even was at a very low level (data not shown). Right after YM155 treatment at stage II (day 6 associated with differentiation), BIRC5 and BCL-2 were dose-dependently decreased as compared with DMSO whereas BAX was improved (Fig.   3c ). Furthermore, Caspase-3 activity was significantly increased by a lower focus of YM155 in QIA7 more than QIA7-iHeps but was not really affected in hAT-SCs (Fig.   three dimensional ). These results were very similar with Annexin Sixth is v staining (Additional file 2 : Figure S4). In the change of cell phenotype, cellular clusters were effectively removed by more than 5 nM of YM155 but not by 1 nM and DMSO (Fig.   3e ). Nevertheless , all cells were dead with 100 nM associated with YM155 treatment. The cells affected by YM155 were closely in line with TRA-1-60 + cells (Additional document 2 : Figure S5). Furthermore, OCT4 + cells were considerably decreased by more than 5 nM of YM155 within FACS analysis (Fig.   3f ). In the genetic analysis, expression of NANOG and OCT4 was decreased plus expression of endodermal marker genes such as CXCR4 , SOX17 and FOXA2 was increased in a dose-dependent way (Fig.   3g ). That, the residual USCs were effectively eliminated by YM155 on stage II (day 6 of differentiation) as compared along with differentiating cells and somatic cells.

Fig. 3

Removal of residual USCs using YM155. Cytotoxicity had been measured by WST after QIA7, WA01, QIA7-iHeps plus hAT-SCs were treated with YM155 (0– 10, 000 nM) for 24  h ( a ). With the selected concentrations (from 1 to one hundred nM), cell viability was estimated at stage I actually of QIA7 (day 4) and stage II (day 6) ( b ). Appearance of apoptosis-related genes ( BIRC5 , BCL-2 and BAX ) was measured right after 24  h of YM155 treatment at stage II ( c ). To further assess YM155-induced apoptotic cell death, Caspase-3 activity was approximated after 16-h incubation with YM155 in QIA7, QIA7-iHeps and hAT-SCs ( d ). YM155 removed the cell clusters ( arrow ) in a concentration-dependent manner ( e ). The decrease of USCs had been confirmed by flow cytometer using OCT4 antibody ( f ). Pluripotent genes ( NANOG and OCT4 ) were decreased by YM155, but endodermal genes ( CXCR4 , SOX17 and FOXA2 ) were increased in the dose-dependent manner ( g ). * p   <   0. 05 (significantly different from the control). BAX bcl-2-like protein 4, BCL-2 B-cell lymphoma two, BIRC5 baculoviral inhibitor of apoptosis repeat-containing 5, CXCR4 C-X-C chemokine receptor type 4, DMSO dimethyl sulfoxide, FOXA2 forkhead box proteins A2, hAT-SC human being adipose tissue-derived stromal cell, iHep induced hepatocyte, OCT4 octamer-binding transcription factor, SOX17 sex determining region Y-Box seventeen