Evaluation of the definitive endoderm differentiation bias between individual hESC lines by standardized methods

Individual human embryonic stem cell (hESC) lines often demonstrate a differentiation bias towards a specific germ layer, which may hamper the efficiency of hESC-based biomedical applications. A better understanding of the molecular mechanisms causing this phenomenon is therefore needed. However, an accurate quantification of differentiation bias is challenging, as culture conditions also influence differentiation outcome. Here, we compared standardized methods to quantify definitive endoderm (DE) differentiation potential of four hESC lines. All lines carried a balanced chromosomal content and were cultured in identical conditions on laminin-521TM in NutristemTM medium. First, we used our in-house optimized embryoid body (EB) formation protocol to generate equal-sized EBs, followed by 21-days of spontaneous differentiation in APEL medium. Gene expression analysis did not show a lineage bias between differently sized EBs within the same line, but we detected consistent differences between individual hESC lines. Next, we performed a 3-day DE induction with a defined seeding density. DE samples from hESC line VUB14 showed a statistically significant reduction in expression levels of SOX17, FOXA2 and GATA4 in comparison to the other hESC lines. Also, immunocytochemistry analysis showed that the VUB14-derived DE population had the lowest percentage of SOX17-positive cells and the highest number of POU5F1-positive cells. These results were in concordance with the EB spontaneous differentiation experiment. Additionally, we performed a 12-day EB differentiation and evaluated gene expression levels using the TaqMan® hPSC Scorecard™ Panel. These results showed that VUB14 in general had the lowest tri-lineage differentiation potential amongst the evaluated lines. Our data shows that standardized EB and DE differentiation followed by gene expression analysis can serve as a reliable tool for evaluating DE differentiation bias. Our next step will be evaluating a potential reason for VUB14’s resistance to differentiation and establishing whether this hindrance is maintained during further differentiation towards hepatocytes.