Highly efficient methods to obtain homogeneous dorsal neural progenitor cells from human and mouse embryonic stem cells and induced pluripotent stem cells.

BACKGROUND: Embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs) have been widely used to generate cellular models harboring specific disease-related genotypes. Of particular importance are ESC and iPSC applications capable of producing dorsal telencephalic neural progenitor cells (NPCs) that are representative of the cerebral cortex and overcome the challenges of maintaining a homogeneous population of cortical progenitors over several passages in vitro. While previous studies were able to derive NPCs from pluripotent cell types, the fraction of dorsal NPCs in this population is small and decreases over several passages. Here, we present three protocols that are highly efficient in differentiating mouse and human ESCs, as well as human iPSCs, into a homogeneous and stable population of dorsal NPCs. These protocols will be useful for modeling cerebral cortical neurological and neurodegenerative disorders in both mouse and human as well as for high-throughput drug screening for therapeutic development.

METHODS: We optimized three different strategies for generating dorsal telencephalic NPCs from mouse and human pluripotent cell types through single or double inhibition of bone morphogenetic protein (BMP) and/or SMAD pathways. Mouse and human pluripotent cells were aggregated to form embryoid bodies in suspension and were treated with dorsomorphin alone (BMP inhibition) or combined with SB431542 (double BMP/SMAD inhibition) during neural induction. Neural rosettes were then selected from plated embryoid bodies to purify the population of dorsal NPCs. We tested the expression of key dorsal NPC markers as well as nonectodermal markers to confirm the efficiency of our three methods in comparison to published and commercial protocols.

RESULTS: Single and double inhibition of BMP and/or SMAD during neural induction led to the efficient differentiation of dorsal NPCs, based on the high percentage of PAX6-positive cells and the NPC gene expression profile. There were no statistically significant differences in the variation of PAX6 and SOX1-positive NPCs between the two human pluripotent cell-derived methods; therefore, both methods are suitable for producing stable dorsal NPCs. When further differentiated into mature neurons, NPCs gave rise to a population of almost exclusively forebrain cortical neurons, confirming the dorsal fate commitment of the progenitors.

CONCLUSIONS: The methods described in this study show improvements over previously published studies and are highly efficient at differentiating human and mouse pluripotent cell types into dorsal PAX6-positive NPCs and eventually into forebrain cortical neurons.