Based on these encouraging results, we further validated this method in the context of endogenous genes by generating hESCs carrying inducible shRNAs against OCT4 (POU5F1) or B2M (Fig. S3F). Remarkably, all the sublines analyzed (six for each gene) showed robust inducible knockdown with no significant shRNA leakiness (Fig. S3G,H). Tetracycline titration identified optimal concentrations to partially or fully knockdown OCT4 (Fig. 2G, Fig. S3I,J). As expected, a strong decrease in OCT4 specifically resulted in loss of pluripotency and induction of neuroectoderm and definitive endoderm markers (Fig. 2H, Fig. S3I,J) (Thomson et al., 2011; Wang et al., 2012). Similar results were obtained with 20 additional OCT4 inducible knockdown hESC sublines, confirming the robustness and reproducibility of this method (Fig. S3K). Importantly, the generation of hESCs with strong and tightly regulated knockdown was so efficient that phenotypic analyses could be performed immediately after antibiotic selection on a mixed population of cells, thereby entirely bypassing the need to pick individual colonies for clonal isolation (Fig. S3K).
