From neural border epithelium to neural crest emigration: A comprehensive single cell roadmap of the timing and regulatory logic underlying cranial and vagal neural crest emergence [morpholinoRNAseq]

Neural crest cells exemplify cellular diversification from a multipotent progenitor population. However, the full sequence of molecular choices governing the emergence of neural crest heterogeneity from the ectoderm remains elusive. Gene regulatory networks govern these steps of embryonic development and cell specification towards definitive neural crest. Here, we combine ultra-dense single cell transcriptomes with machine-learning strategies and experimental validation to provide a comprehensive gene regulatory network driving vertebrate neural crest fate diversification, from induction to early migration stages. Transcription factor connectome and bifurcation analyses demonstrate emergence of early neural crest fates at the neural plate stage, alongside an unbiased multipotent neural crest lineage persisting until after epithelial-mesenchymal transition. We also define a new and transient neural border zone state, preceding choice between neural crest and placodes during gastrulation. Theis combination of experimental tests, with Machine Learning broadly applicable to single cell transcriptomics, deciphers the circuits driving cranial and vagal neural crest formation and provides a general model for investigating vertebrate GRNs in development, evolution and disease. There are 10 samples for pax3, zic1 and tfap2e MO. Two replicates for zic1 and tfap2e and one replicate for pax3. Previously validated antisense morpholino oligonucleotides (MO) were used to block pax3 and tfap2e expression (GeneTools). 20 ng of pax3 MO (Monsoro-Burq et al., 2005) or tfap2e MO (Hong et al., 2014) were injected into one blastomere of two-cell stage X. laevis embryos, according to the assay. At the desired stage, one anterior NB (stage 14) or one NC (stage 17) was dissected from the injected side and used for RNA extraction and cDNA library preparation, which were then used for small RNA-seq (Fig S8). The resulting 100 bp paired-end sequencing reads were aligned to the X. laevis genome version 9.2 using STAR and the count reads were analyzed using String Tie.