Zhonggang Hou

Genome engineering in human pluripotent stem cells (hPSCs) holds great promise for biomedical research and regenerative medicine. Recently, an RNA-guided, DNA-cleaving interference pathway from bacteria [the type II clustered, regularly interspaced, short palindromic repeats (CRISPR)-CRISPR-associated (Cas) pathway] has been adapted for use in eukaryotic cells, greatly facilitating genome editing. Only two CRISPR-Cas systems (from Streptococcus pyogenes and Streptococcus thermophilus), each with their own distinct targeting requirements and limitations, have been developed for genome editing thus far. Furthermore, limited information exists about homology-directed repair (HDR)-mediated gene targeting using long donor DNA templates in hPSCs with these systems. Here, using a distinct CRISPR-Cas system from Neisseria meningitidis, we demonstrate efficient targeting of an endogenous gene in three hPSC lines using HDR. The Cas9 RNA-guided endonuclease from N. meningitidis (NmCas9) recognizes a 5'-NNNNGATT-3' protospacer adjacent motif (PAM) different from those recognized by Cas9 proteins from S. pyogenes and S. thermophilus (SpCas9 and StCas9, respectively). Similar to SpCas9, NmCas9 is able to use a single-guide RNA (sgRNA) to direct its activity. Because of its distinct protospacer adjacent motif, the N. meningitidis CRISPR-Cas machinery increases the sequence contexts amenable to RNA-directed genome editing.

BACKGROUND: Human pluripotent stem cells offer the best available model to study the underlying cellular and molecular mechanisms of human embryonic lineage specification. However, it is not fully understood how individual stem cells exit the pluripotent state and transition towards their respective progenitor states. RESULTS: Here, we analyze the transcriptomes of human embryonic stem cell-derived lineage-specific progenitors by single-cell RNA-sequencing (scRNA-seq). We identify a definitive endoderm (DE) transcriptomic signature that leads us to pinpoint a critical time window when DE differentiation is enhanced by hypoxia. The molecular mechanisms governing the emergence of DE are further examined by time course scRNA-seq experiments, employing two new statistical tools to identify stage-specific genes over time (SCPattern) and to reconstruct the differentiation trajectory from the pluripotent state through mesendoderm to DE (Wave-Crest). Importantly, presumptive DE cells can be detected during the transitory phase from Brachyury (T) (+) mesendoderm toward a CXCR4 (+) DE state. Novel regulators are identified within this time window and are functionally validated on a screening platform with a T-2A-EGFP knock-in reporter engineered by CRISPR/Cas9. Through loss-of-function and gain-of-function experiments, we demonstrate that KLF8 plays a pivotal role modulating mesendoderm to DE differentiation. CONCLUSIONS: We report the analysis of 1776 cells by scRNA-seq covering distinct human embryonic stem cell-derived progenitor states. By reconstructing a differentiation trajectory at single-cell resolution, novel regulators of the mesendoderm transition to DE are elucidated and validated. Our strategy of combining single-cell analysis and genetic approaches can be applied to uncover novel regulators governing cell fate decisions in a variety of systems.

Human pluripotent stem cell-based in vitro models that reflect human physiology have the potential to reduce the number of drug failures in clinical trials and offer a cost-effective approach for assessing chemical safety. Here, human embryonic stem (ES) cell-derived neural progenitor cells, endothelial cells, mesenchymal stem cells, and microglia/macrophage precursors were combined on chemically defined polyethylene glycol hydrogels and cultured in serum-free medium to model cellular interactions within the developing brain. The precursors self-assembled into 3D neural constructs with diverse neuronal and glial populations, interconnected vascular networks, and ramified microglia. Replicate constructs were reproducible by RNA sequencing (RNA-Seq) and expressed neurogenesis, vasculature development, and microglia genes. Linear support vector machines were used to construct a predictive model from RNA-Seq data for 240 neural constructs treated with 34 toxic and 26 nontoxic chemicals. The predictive model was evaluated using two standard hold-out testing methods: a nearly unbiased leave-one-out cross-validation for the 60 training compounds and an unbiased blinded trial using a single hold-out set of 10 additional chemicals. The linear support vector produced an estimate for future data of 0.91 in the cross-validation experiment and correctly classified 9 of 10 chemicals in the blinded trial.