Translational landscape in human early neural fate determination.

Abstract

Gene expression regulation in eukaryotes is a multi-level process, including transcription, mRNA translation, and protein turnover. Many studies have reported the sophisticated transcriptional regulations during neural development, but the global translational dynamics are still ambiguous. Here, we differentiate human embryonic stem cells (ESCs) into neural progenitor cells (NPCs) with high efficiency and perform ribosome sequencing and RNA sequencing on both ESCs and NPCs. Data analysis reveals that translational controls engage in many critical pathways and contribute significantly to neural fate determination regulation. Furthermore, we show that the sequence characteristics in the untranslated region (UTR) might regulate translation efficiency. Specifically, genes with short 5'UTR and intense Kozak sequence are associated with high translation efficiency in human ESCs, while genes with long 3'UTR are related to high translation efficiency in NPCs. In addition, we have identified four biasedly-used codons (GAC, GAT, AGA, and AGG) and dozens of short open reading frames during neural progenitor differentiation. Thus, our study reveals the translational landscape during early human neural differentiation and provides insights into the regulation of cell fate determination at the translational level.