Cell types in macaque prenatal and fetal brain development.

(A) Schematic diagram of sample collecting and data analysis. We collected the parietal lobe from the embryos across developmental stages from E40 to E90. (B) and (C) The transcriptome data of single cells were collected and used to do clustering using Seurat. Visualization of major types of cells using UMAP. Dots, individual cells; color, clusters. (D) Violin plot of molecular markers for annotating cell types. (E) The expressions of the classic marker genes for each cell type were plotted for UMAP visualization. Light grey, no expression; Dark blue, relative expression.

Excitatory neuron subclusters in the developing macaque cerebral cortex.

(A) Left, Clustering of excitatory neuron subclusters collected at all time points, visualized via UMAP. Cells are colored according to subcluster identities (left) and collection time points (right). (B) Differentially, the expression of deep-layer marker BCL11B and upper-layer marker CUX1 are highlighted. (C) Excitatory neuron subclusters UMAP plot shows the expression of classic markers for deep layers (BCL11B, FEZF2, SOX5) and upper layers (CUX1, SATB2) present at each time point. (D) The proportion of different excitatory neuron subclusters corresponding to excitatory neurons in each time point.

Cell diversity and regulation of progenitor cells in the macaque cortical neurogenesis.

(A) UMAP shows eight progenitor clusters and cell annotation. Left cells are colored according to Seurat clusters; right, cells are colored according to the collection time point. (B) Feature plot of outer radial glia marker genes HOPX shows higher expression in C10-C14-C12 (left). Feature plots of Intermediate progenitor cell marker gene EOMES show higher expression in C10-C22-C8 (right). (C) and (D) Pseudotime analysis by Slingshot of HOPX-positive cells (C10-C14-C12). The Slingshot result with the lines indicating the trajectories of lineages and the arrows indicating directions of the pseudotime. Cells are colored according to cell (C) and pesudotime (D). Dots: single cells; colors: cluster and subcluster identity. (E) The heatmap shows the relative expression of top150 genes displaying significant changes along the pseudotime axis of RG to oRG (C10-C14-C12). The columns represent the cells being ordered along the pseudotime axis. (F) Schema diagram of some significant genes related to E. (The depth of the color indicates the levels of gene expression)

Transcriptional regulation of excitatory neuron lineage during prenatal cortical neurogenesis.

(A) UMAP shows the alignment of macaque cortical NPCs, IPCs, and excitatory neurons. Left, cells are colored according to cell annotation. Different yellow/orange colors are used for deep-layer excitatory neuron subclusters (EN5 and EN10), and different red/pink colors are used for upper-layer excitatory neuron subclusters (EN1, EN2, EN3, EN4, EN6, EN7, EN9 and EN10). Right, cells are colored according to the time point of collection. (B) Dot plot showing the marker genes for the deep-layer excitatory neuron (FEZF2) and upper-layer excitatory neuron (DOK5). Light grey, no expression; Dark blue, relative expression. (C) Pseudotime analysis by Slingshot projected on PCA plot of RGCs, oRGCs, IPCs, and excitatory neuron subclusters. The Slingshot result indicates the trajectories of lineages, and the arrows indicate the directions of the pseudotime. Dots: single cell; colors: cluster and subcluster identity. Framed numbers marked the start point, endpoint, and essential nodes of the Slingshot inference trajectory. Framed number “1” was the excitatory neuron lineage trajectory start point (C10). Framed number “4” marked immature neurons. Framed numbers “2” and “3” marked deep-layer and upper-layer neurons. Cells are colored according to cell annotation and pseudotime. (D) The heatmap shows the relative expression of the top 100 genes displaying significant changes along the pseudotime axis of each lineage branch. The columns represent the cells being ordered along the pseudotime axis. (E) Left, Slingshot branching tree related to Slingshot pseudotime analysis in C. The root is E40 earliest RG (C10), tips are deep-layer excitatory neurons generated at the early stage (E40, E50), and upper-layer excitatory neurons are generated at the later stage (E70, E80, E90). Right, branching trees showing the expression of marker genes of apical progenitors (PAX6), outer radial glia cells (HOPX), intermediate progenitors (EOMES), and excitatory neurons (NEUROD2), as well as genes in Figure 3E, including callosal neurons (SATB2, CUX2), deeper layer neurons (SOX5, FEZF2), corticofugal neurons (FEZF2, TLE4). There is a sequential progression of radial glia cells, intermediate progenitors, and excitatory neurons.

Integration of human, macaque and mouse single-cell datasets reveals conserved and divergent progenitor cell types.

(A) Left: UMAP plot of cross-species integrated single-cell transcriptome data with liger. Colors represent different major cell types (Black: Human dataset; Dark grey: macaque dataset; Liger grey: mouse). Right: The UMAP plot of each dataset, colored by liger cluster. (B) The expressions of the classic oRG marker genes were plotted to UMAP visualization. Light grey, no expression; Dark blue, relative expression. (C) Comparison of vRG→oRG and vRG→ IPC developmental trajectories between human, macaque, and mouse.

The patterns of transcriptional regulation comparative analysis responsible in vRGs.

(A) Normalized expressions of genes that show temporal dynamics in the vRGs of human, macaque and mouse. (B) Temporal expression heatmap of homologous transcription factor genes among human, macaque, and mouse temporal expression heatmap. (The TFs genes in the dashed boxes showed similar temporal expression patterns across species. (C), (E) and (G) Regulon specificity score for each timepoint in human, macaque, and mouse vRG. Regulons with high scores in multiple species vRG cells are colored yellow. (D), (F) and (H) show a network generated with Cytoscape using the top 10 regulons in the human, macaque, and mouse vRG at each time point and their top 5 target genes identified by SCENIC as an input. The interactions between conserved TFs in more than one species are colored yellow.

Sample collection and Quality Control.

(A)Schematic diagram of sample collecting anatomical area.

(B)The cell number per sample after quality control.

(C) Bar chart of astrocytes proportion statistical at each time point.

(D) Single-cell transcriptome library information for each sample.

scRNA-Seq uncovers cell type in the developing macaque neocortex.

(A) Visualization of different dimensionality reduction of all cells. (Left, UMAP visualization with UMAP1 and UMAP2; right, 3D model of UMAP visualization with UMAP1, UMAP2, and UMAP3).

(B) Top marker genes for each of the 28 cell clusters shown in Figure 1B.

(C) Feature plots of marker gene expression. Colors represent scaled gene expression.

Additional information for excitatory neuron subcluster.

(A) UMAP visualization of excitatory neuron subcluster cell scRNA-seq data from individual time points. Cells are colored by excitatory neuron subcluster assignment.

(B) Top marker genes for each of the excitatory neuron subclusters.

Stem and Excitatory neuron subclusters mapping genes.

Developmental regulation of gene expression from RG to IPC.

(A) Pseudotime analysis by Slingshot of EOMES-positive cells (C10-C22-C8). Dots: single cells; Cells are colored by their identity.

(B) Heatmap shows the relative expression of top150 genes displaying significant changes along the pseudotime axis of RG to IPC (C10-C22-C8). The columns represent the cells being ordered along the pseudotime axis. The start point, endpoint, and important nodes of the Slingshot inference trajectory were marked by framed numbers. Framed number “2” was radial glial cells at the beginning of pseudo time (C10). Framed numbers”1” and “3” marked intermediate state nodes. Framed number “4” marked intermediate progenitor cell (C8).

Transcriptional regulation of OPC differentiation into astrocytes and oligodendrocytes.

(A) Feature plot of classic marker for astrocytes (AQP4) and oligodendrocytes (OLIG2).

(B) Slingshot branching tree related to Slingshot pseudotime analysis in (C) and (D). Root is OPC (C16), tips astrocytes (C13), and oligodendrocytes (C18).

(C and D) Pseudotime analysis by Slingshot of glial cell lineage (C16→C13, C16→C12). The Slingshot result indicated the trajectories of lineages, and the arrows indicated directions of the pseudotime. Cells are colored according to their identity (C) and pesudotime (D). Dots: single cells; colors: cluster and subcluster identity.

(E) Heatmap shows the relative expression of top150 genes displaying significant changes along the pseudotime axis of glial cell lineage (C16→C13, C16→C12). The columns represent the cells being ordered along the pseudotime axis.

Upper layer and deep layer excitatory neuron proportion analysis among species

(A) Cell type annotation of the integrated dataset.

(B and C) Feartureplot shows the expression of upper-layer marker genes CUX2, POU3F2, SATB2, and deep-layer marker genes FEZF2 and SOX5. (D)Proportion analysis of excitatory neuron subclusters in human, macaque, and mouse datasets at different developmental time points.

Temporal expression pattern of RNA binding protein and transcription factor genes in human, macaque, and mouse vRG.

(A) Temporal expression heatmap of RNA binding protein genes in human, macaque, and mouse ventricular radial glia.

(B and C) Temporal expression heatmap of Transcription factors genes in human, macaque and mouse ventricular radial glia sorted by human (B) and mouse (C) temporal pattern. (Note: each line is the same homologous gene).