In papers in Nature Biotechnology and Current Biology, the Schier lab reports how thousands of single-cell transcriptomes can be used to define cell types and reconstruct lineage relationships in the brain.
Understanding brain development and function rests on the ability to define and map neural cell types, but the size and complexity of the brain has hindered the comprehensive discovery of cell types. Shristi Pandey in the Schier lab, with Karthik Shekhar in the Regev lab, focused on a part of the zebrafish forebrain called the habenula to systematically identify novel cell types and gene expression domains (Current Biology). The authors developed protocols for the dissociation and capture of individual neurons from the zebrafish brain and used single-cell RNA-sequencing (scRNA-seq) of several thousand habenula cells to identify 18 neuronal types, including 13 new types, and dozens of new marker genes. They then mapped the different cell types onto a reference atlas and found that 15 neuronal subsets are highly restricted to habenular sub-regions. Such regionalization contrasts with the organization in the retina where different cell types are intermixed within the same spatial location. Comparing the developing larval habenula to the mature adult habenula, Pandey et al. made the surprising discovery that cell types were highly similar, despite extensive brain growth and functional maturation. Finally, they used the reference atlas to compare molecularly defined neuronal types with those defined by neural activity, and identified a specific region that responds to noxious stimuli. Beyond its importance for the study of habenular development and function, the study offers a general framework for similarly comprehensive characterizations of other vertebrate brain regions.
The brain develops from a small number of embryonic progenitors through numerous rounds of divisions that give rise to a diverse set of differentiated neurons and glia. Lineage trees describe the divisions and specializations of cells over time, but the large size and cell type diversity of the brain has hindered the construction of large-scale brain lineage trees. Bushra Raj, James Gagnon and Shristi Pandey in the Schier lab, together with Aaron McKenna in the Shendure lab and Dan Wagner in the Klein lab, established an approach to address this challenge (Nature Biotechnology). They extended the lineage recording method GESTALT, developed by the Schier and Shendure labs (McKenna et al. Science 2016), and combined it with droplet-based single-cell RNA sequencing to provide a snapshot of the heterogeneity of cell types and developmental lineages in a vertebrate brain. They used CRISPR–Cas9 to edit transgenic lineage barcodes during early and late zebrafish development, resulting in the accumulation of >10,000 mutant barcodes over time that uniquely labeled ancestral progenitors. Single-cell RNA sequencing of edited juvenile brains simultaneously identified cellular transcriptomes and mutated lineage barcodes of thousands of cells. This approach defined >100 cell types in the juvenile brain providing a rich catalogue of cell diversity, and comparison of barcodes reconstructed lineage relationships between profiled cells. The resulting lineage trees had hundreds of branches that uncovered restrictions at the level of cell types, brain regions, and gene expression cascades during differentiation. Individual barcodes were mostly regionally enriched, consistent with the limited long-range migration of neurons. Despite regional restriction, progenitors that acquired a specific barcode gave rise to descendants that mapped to multiple different types of neurons, suggesting that most ancestral progenitors were multipotent. But close inspection of the lineage trees also revealed divergent lineage trajectories; e.g., in the hypothalamus, which contained cell types with distinct lineage relationships. These results illustrate that this technology – scGESTALT – can be used to simultaneously characterize molecular identities and lineage histories of thousands of cells during development and disease, not only in zebrafish but also in any system that allows genome editing and scRNA-seq.