Proneural genes are conserved drivers of neurogenesis across the animal kingdom. How their functions have adapted to guide human-specific neurodevelopmental features is poorly understood. Here, we mined transcriptomic data from human fetal cortices and generated from human embryonic stem cell-derived cortical organoids (COs) to show that NEUROG1 and NEUROG2 are most highly expressed in basal neural progenitor cells, with pseudotime trajectory analyses indicating that NEUROG1-derived lineages predominate early and NEUROG2 lineages later. Using ChIP-qPCR, gene silencing and overexpression studies in COs, we show that NEUROG2 is necessary and sufficient to directly transactivate known target genes (NEUROD1, EOMES, RND2). To identify new targets, we engineered NEUROG2-mCherry knock-in human embryonic stem cells for CO generation. The mCherry-high CO cell transcriptome is enriched in extracellular matrix-associated genes, and two genes associated with human-accelerated regions: PPP1R17 and FZD8. We show that NEUROG2 binds COL1A1, COL3A1 and PPP1R17 regulatory elements, and induces their ectopic expression in COs, although NEUROG2 is not required for this expression. Neurog2 similarly induces Col3a1 and Ppp1r17 in murine P19 cells. These data are consistent with a conservation of NEUROG2 function across mammalian species.

Rodent models are used to study how the six-layered neocortex (‘cortex’) develops, but do not recapitulate several features of the human brain, which has an increased size and complexity (Dehay and Huttner, 2024). In mammals, the ventricular zone (VZ) is populated by primary neural progenitor cells (NPCs), termed apical radial glia (aRG), which initially undergo direct neurogenesis to form deep-layer neurons (Moffat and Schuurmans, 2024; Taverna et al., 2014). Later on, aRG give rise to intermediate progenitor cells (IPCs) that form a more basally located subventricular zone (SVZ) and undergo indirect neurogenesis to populate upper layers (Haubensak et al., 2004; Kowalczyk et al., 2009; Miyata et al., 2004; Noctor et al., 2004, 2008). In human and non-human primate cortices, the SVZ is enlarged and divided into an inner (i) SVZ and outer (o) SVZ (Fietz et al., 2010; Hansen et al., 2010; Reillo et al., 2011). In the oSVZ, a population of basal RG (bRG) has expanded, which correlates with increased brain size and is a primary driver of cortical evolution (Dehay et al., 2015; Fietz et al., 2010; Hansen et al., 2010; Lui et al., 2011; Reillo and Borrell, 2012). In gyrencephalic cortices, IPCs are transit-amplifying cells, dividing multiple times before differentiating to produce many more upper-layer neurons than in rodents (Fernandez and Borrell, 2023; Fietz et al., 2010). To accommodate more neurons, cortical folds have developed, which increase overall surface area (Amin and Borrell, 2020; Borrell, 2018; Moffat and Schuurmans, 2024).

In humans, these prominent phenotypic changes are associated with altered neurodevelopmental gene expression, driven in part by fast-evolving cis-regulatory regions, termed human accelerated regions (HARs), to which trans-acting transcription factors (TFs) bind (Boyd et al., 2015; Capra et al., 2013; Doan et al., 2016; Girskis et al., 2021; Kamm et al., 2013; Lambert et al., 2011; Pollard et al., 2006; Prabhakar et al., 2006; Vanderhaeghen and Polleux, 2023; Wei et al., 2019; Won et al., 2019). HAR mutations in individuals with neurodevelopmental disorders is suggestive of their importance during brain development (Doan et al., 2016). For example, HARE5 is a human-specific regulatory sequence that elevates FZD8 expression to increase Wnt signaling and expand cortical NPCs (Boyd et al., 2015). PPP1R17 is a HAR-regulated gene that encodes a negative regulatory subunit for protein phosphatases such as PP1 and PP2A (Endo et al., 2003; Hall et al., 1999), which control the G1-to-S phase transition (Moura and Conde, 2019). PPP1R17 is highly expressed in cortical NPCs in primates and not other mammalian cortices, and drives a lengthened cell cycle characteristic of human corticogenesis when misexpressed in mouse cells (Girskis et al., 2021).

Proneural genes encode basic helix-loop-helix TFs that are conserved drivers of neurogenesis (Bertrand et al., 2002; Oproescu et al., 2021). In rodent cortices, Neurog2 is the main proneural gene, and is required and sufficient to specify a glutamatergic neuronal identity (Fode et al., 2000; Han et al., 2018; Mattar et al., 2008; Oproescu et al., 2021; Schuurmans et al., 2004). Neurog1 plays a more minor role in tempering the pace of early murine cortical neurogenesis (Han et al., 2018). To control fate decisions by cortical NPCs, Neurog1 and Neurog2 directly transactivate target genes, and function as pioneer TFs that open chromatin to facilitate binding by other neurogenic and lineage-specifying TFs (Aydin et al., 2019; Noack et al., 2022; Pereira et al., 2024; Smith et al., 2016). Neurog2 re-organizes the chromatin landscape at several levels, driving DNA demethylation at Neurog2 motifs, and increasing chromatin looping and accessibility (Manelli et al., 2024 preprint; Noack et al., 2022; Pereira et al., 2024). In mice, Neurog2 is initially expressed in aRG and drives the transition from aRG to IPC, and later is expressed in IPCs (Britz et al., 2006; Kovach et al., 2013; Miyata et al., 2004; Ochiai et al., 2009). In fluorescence-activated cell sorting (FACS)-enriched NPCs from human fetal cortices, NEUROG2 and its target genes are expressed at the highest levels in bRG and to a lesser extent in aRG (Johnson et al., 2015). Overexpression of NEUROG2 in the ferret cortex similarly promotes the basal translocation of aRG (Johnson et al., 2015). Here, we used human embryonic stem cell (hESC)-derived cortical organoids (COs) to compare NEUROG1 and NEUROG2 expression, and assess the function of NEUROG2 during human cortical neurogenesis.

NEUROG1 and NEUROG2 are expressed during human cortical development

In mouse cortical development, Neurog2 is expressed between embryonic day (E) 10.5 to E17.5, encompassing the neurogenic period, whereas Neurog1 expression begins at E10.5, but declines by E14.5 (Han et al., 2018; Li et al., 2012; Moffat et al., 2023 preprint). To compare mRNA levels in individual cortical cell types, we performed a pseudo-bulk analysis of single-cell RNA-sequencing (scRNA-seq) data collected from E10.5-E17.5 mouse cortices (Di Bella et al., 2021). Aggregated transcript read counts for Neurog1 and Neurog2 were higher in IPCs than in all other cell types, followed by aRG and migrating neurons (Fig. S1A). Within IPCs, Neurog1 and Neurog2 were initially expressed at roughly equivalent levels at E10.5, but by E13.5 Neurog2 expression predominated (Fig. S1A). Neurog2 transcript counts superseded Neurog1 read counts in aRG, in migrating, immature and glutamatergic neurons, and in non-neuronal cells at all stages (Fig. S1A).

To assess NEUROG2 and NEUROG1 transcript levels during human cortical development, we performed a pseudo-bulk analysis of scRNA-seq data collected from post-conception weeks (PCW) 5-14 human cortices (Braun et al., 2023). NEUROG2 and NEUROG1 transcript counts were roughly equivalent and at the highest levels in IPCs at all stages (Fig. 1A). NEUROG2 and NEUROG1 transcripts were elevated in other NPC pools, including aRG and bRG with and without a proliferative gene signature (Fig. 1A). Finally, NEUROG2 and to a lesser extent NEUROG1 transcripts were detected at lower levels in glutamatergic neurons and ‘other’ cells (i.e. erythrocytes, vascular cells, placodes, fibroblasts), and in some neuroblasts, neurons and RG clusters that could not be merged with other annotated cell types (Fig. 1A).

Fig. 1.

NEUROG1 and NEUROG2 expression in human fetal cortices. (A) Pseudo-bulk analysis of NEUROG1 and NEUROG2 transcript counts in scRNA-seq data collected from post-conception weeks (PCW) 5-14 human cortices (Braun et al., 2023), showing log2 counts per million (CPM). (B) Distribution of NEUROG1/NEUROG2 single and double-positive cells in scRNA-seq datasets from human fetal cortices between gestational week (GW) 8 and 26 (Zhong et al., 2018).

Fig. 1.

NEUROG1 and NEUROG2 expression in human fetal cortices. (A) Pseudo-bulk analysis of NEUROG1 and NEUROG2 transcript counts in scRNA-seq data collected from post-conception weeks (PCW) 5-14 human cortices (Braun et al., 2023), showing log2 counts per million (CPM). (B) Distribution of NEUROG1/NEUROG2 single and double-positive cells in scRNA-seq datasets from human fetal cortices between gestational week (GW) 8 and 26 (Zhong et al., 2018).

To determine whether the same percentage of NPCs expressed NEUROG1 and/or NEUROG2, we mined scRNA-seq data from gestational week (GW) 08 to GW26 human prefrontal cortices (Zhong et al., 2018). Of the cells assigned an NPC identity, the majority expressed NEUROG2 (72.06%), either together with NEUROG1 (38.62%) or alone (33.44%) (Fig. 1B). In contrast, NEUROG1 was expressed in fewer NPCs (47.93%), of which only 9.31% expressed NEUROG1 alone (Fig. 1B). Cortical NPCs co-expressing NEUROG1 and NEUROG2 persisted throughout the neurogenic window, from GW09 until GW26 (Fig. 1B). NEUROG2 is therefore expressed in more human cortical NPCs than NEUROG1 at the population level, even though relative transcript counts are similar within individual cells.

NEUROG1 and NEUROG2 are primarily expressed in basal NPCs in cortical organoids

To assess proneural gene function in a human model system, we used directed differentiation to generate COs with a cortical identity from hESCs (Birey et al., 2017). Neural rosettes were observed in day 30-35 COs, comprising a central zone of SOX2+ and NES+ NPCs, and an external layer of TUJ1+ (β3-tubulin; TUBB3) neurons (Fig. 2A). To analyze CO structure, day 42 COs were optically cleared, co-immunostained with SOX2 and TUJ1, and imaged in wholemount using light sheet microscopy (Fig. 2A). A network of TUJ1+ axonal fibers encased the 42-day-old COs, and, in some instances, neural rosettes comprising an inner SOX2+ NPC layer and outer TUJ1+ neuronal layer appeared as external protrusions (Fig. 2A).

Fig. 2.

Generation of COs and snRNA-seq analysis. (A) Immunolabeling of day 35 and day 30 COs with SOX2 and TUJ1 or NES and TUJ1 (left), and 3D rendering of a tissue-cleared, day 42 CO immunolabeled with SOX2 and TUJ1 imaged with light sheet microscopy (right). (B) Overlay of uniform manifold approximation and projections (UMAPs) of snRNA-seq data collected from four independent sets of pooled day 30 COs. Numbers represent identified clusters. (C) Average transcript read counts per cell in each cluster. (D) UMAP showing the cluster distribution and manually annotated cluster identities. (E,F) Feature plots showing FOXG1, PAX6, NEUROG1 and PPP1R17 transcript distributions (E) and corresponding immunolabeling in day 30 COs (F). (G,H) Feature plots showing TBR1 and BCL11B transcript distributions (G) and corresponding immunolabeling in day 30 COs (H). (I) Proportions of day 30 CO cell types expressing NEUROG1 and/or NEUROG2. (J) Monocle3 lineage trajectory analysis of NEUROG2+, NEUROG1+ and double-positive cells in 30-day COs. aRG, apical radial glia; bRG, basal radial glia; IPC, intermediate progenitor cell. Scale bars: 400 mm (top left), 100 mm (bottom left), 200 mm (right) (A); 100 µm (F,H).

Fig. 2.

Generation of COs and snRNA-seq analysis. (A) Immunolabeling of day 35 and day 30 COs with SOX2 and TUJ1 or NES and TUJ1 (left), and 3D rendering of a tissue-cleared, day 42 CO immunolabeled with SOX2 and TUJ1 imaged with light sheet microscopy (right). (B) Overlay of uniform manifold approximation and projections (UMAPs) of snRNA-seq data collected from four independent sets of pooled day 30 COs. Numbers represent identified clusters. (C) Average transcript read counts per cell in each cluster. (D) UMAP showing the cluster distribution and manually annotated cluster identities. (E,F) Feature plots showing FOXG1, PAX6, NEUROG1 and PPP1R17 transcript distributions (E) and corresponding immunolabeling in day 30 COs (F). (G,H) Feature plots showing TBR1 and BCL11B transcript distributions (G) and corresponding immunolabeling in day 30 COs (H). (I) Proportions of day 30 CO cell types expressing NEUROG1 and/or NEUROG2. (J) Monocle3 lineage trajectory analysis of NEUROG2+, NEUROG1+ and double-positive cells in 30-day COs. aRG, apical radial glia; bRG, basal radial glia; IPC, intermediate progenitor cell. Scale bars: 400 mm (top left), 100 mm (bottom left), 200 mm (right) (A); 100 µm (F,H).

To confirm a cortical identity, we performed single-nuclear (sn)RNA-seq on four independent batches of five pooled COs per sample. A total of 103,459 cells passed quality control benchmarks, with an average of 4485±506 transcript reads per cell across the four groups (Fig. S2A,B). Batch correction was used to correct for technical variance, revealing that the four CO samples were highly correlated (Fig. 2B). Unbiased Seurat clustering stratified nuclei into ten cell clusters, all with high transcript read counts (Fig. 2C). Cluster identities were inferred based on cell type-specific markers and included aRG, bRG, IPCs and neurons (Fig. 2D, Table S1). Consistent with a forebrain identity, FOXG1 was expressed in all cell clusters in day 30 COs (Fig. 2E,F). Four aRG clusters were identified (clusters 2, 4, 6, 10), of which clusters 4 and 10 expressed the highest levels of the pan-RG markers PAX6, GLI3, SLC1A3, PROM1, and PARD3, while clusters 2 and 4 expressed the highest levels of proliferation markers (TOP2A, MKI67, CENPF, NUSAP1) (Fig. 2E,F; Fig. S2C). Cluster 7 expressed the bRG marker RASGRP1, as well as NEUROG1, NEUROG2 and known proneural target genes, such as HES6, NHLH1 and CBFA2T2 (Fig. 2E,F; Fig. S2C). A relatively small number of bRG were present in the day 30 COs in accordance with previous reports indicating that COs at 1-1.5 months of age primarily include aRG, with bRG only becoming predominant after 2 months (Uzquiano et al., 2022). Cluster 8 expressed PPP1R17, an IPC marker (Pollen et al., 2015), and markers of early-born neurons, including TBR1, RELN and SLC17A6 (Fig. 2E-H; Fig. S2C). Finally, cells in clusters 0, 1, 3, 5 and 9 expressed neuronal markers, such as BCL11B, ISL1 and MEF2C (Fig. 2G,H; Fig. S2C). Thus, NEUROG1 and NEUROG2 are most highly expressed in basal NPCs in day 30 COs, including bRG and IPCs, matching observations made in human fetal samples (Fig. 1).

NEUROG1 expression predominates early and NEUROG2 later in a cortical organoid model of human cortical development

We used the snRNA-seq data from day 30 COs to interrogate proneural gene expression further. In day 30 COs, NEUROG1 and/or NEUROG2 were expressed in ∼8% of basal NPCs, including bRG and IPCs, in ∼3% of aRG, and in ∼4% of neurons (Fig. 2I). To understand lineage dynamics, we computationally isolated the 4896 cells expressing NEUROG1 and/or NEUROG2 and generated pseudotime trajectories (Fig. 2J; Fig. S3A). Pseudotime ordering of cells revealed a single main branch point and five cell states, with a relatively equal representation of cells from the four CO pools distributed within these states (Fig. 2J; Fig. S3B). State 1 and state 2 cells had the earliest pseudotime identities and predominantly included NEUROG1 single-positive cells (Fig. 2J; Fig. S3C,D). State 3 was a small population of double-positive cells that appeared to be a transition step between early pseudotime states predominated by NEUROG1 expression, and later pseudotime states (states 4 and 5) in which NEUROG2 was instead expressed (Fig. 2J; Fig. S3C,D). Markers of aRG (VIM, GLI3, PARD3) were predominant in all cell states, whereas markers of bRG (RASGRP1, EOMES), proneural target genes (HES6) and markers of early-born layer 6 neurons (FOXP2) were highest in state 1 (Fig. S3E). In contrast, BCL11B, a layer 5 marker (Du et al., 2022), was expressed at higher levels in states 2, 4, and 5, with later pseudotime identities (Fig. S3E).

We reasoned that the unexpectedly higher expression levels of NEUROG1 compared to NEUROG2 in day 30 COs may be because this is a relatively early stage in CO development. To assess lineage relationships between NEUROG1 and NEUROG2 further, we generated pseudotime trajectories from mined scRNA-seq data collected from day 90 COs that were generated using an undirected Lancaster protocol (Sivitilli et al., 2020) (Fig. S4A). In these day 90 COs, NEUROG2 expression predominated over NEUROG1 (Fig. S4B,C). The resultant pseudotime ordering of cells along a lineage trajectory revealed a single branch point and three cell states (Fig. S4D,E). State 1 cells had the earliest pseudotime identity and included the highest fraction of NEUROG1/NEUROG2 double-positive and NEUROG1 single-positive cells, and the lowest fraction of NEUROG2 single-positive cells (Fig. S4F,G). We determined that state 1 cells had the highest levels of aRG, bRG, and proliferating cell-associated transcripts (Fig. S4H). In intermediate state 2 cells, an increase in IPC marker expression coincided with an increase in the fraction of NEUROG2 single-positive cells (Fig. S4H). Finally, state 3 cells, which had the latest pseudotime identities, predominantly expressed NEUROG2 alone, with elevated levels of early neuronal marker transcripts, including deep-layer 5 and 6 markers (Fig. S4H). Thus, NEUROG1 expression is confined to early stages of neural lineage development in human COs, and is enriched in aRG and bRG, whereas NEUROG2-expression increases later, as aRG/bRG mature into IPCs and neurons.

NEUROG2 is required and sufficient to transactivate neurogenic target genes in cortical organoids

We focused on NEUROG2 since its expression predominated in human fetal cortices and later-stage COs. We confirmed that NEUROG2 was expressed in COs, initially in a small number of cells at day 18, and in an increasing number of cells by day 48 (Fig. 3A). This increase was validated using qPCR, with almost a threefold increase in NEUROG2 transcripts in day 90 versus day 30 COs (Fig. 3B). In contrast, NEUROG1 transcript levels declined over this same period (Fig. 3B), consistent with the single-cell transcriptomic data. To assess NEUROG2 function in COs, we used chromatin immunoprecipitation (ChIP) to examine whether NEUROG2 engages with known target genes (Fig. 3C). ChIP-qPCR performed on day 45 COs revealed that NEUROG2 binds to DLL3 and NEUROD4 promoter regions (Smith et al., 2016), whereas no binding was observed to an open reading frame (ORF) control sequence (Fig. 3C).

Fig. 3.

NEUROG2 is necessary and sufficient to turn on neurogenic genes in COs. (A) NEUROG2 immunolabeling of day 18 and day 48 COs. Scale bars: 100 µm. (B) qPCR of NEUROG2 and NEUROG1 in day 30 and day 90 COs. (C) NEUROG2 ChIP-qPCR (n=5), or mock control ChIP-qPCR (n=3), using day 45 COs, and qPCR amplified DLL3 and NEUROD4 promoter region binding sites in the eluted chromatin (n=5). An ORF amplified sequence was used as a negative control. (D) NEUROG2 silencing in day 60 COs using lentiviral shRNA with a scrambled control sequence (shScr) and two shRNAs targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and the expression of NEUROG2, NEUROD1, EOMES, RND2, DLL1 and DLL3 was analyzed by qPCR. (E) Neurog2 gain-of-function assay, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of NEUROG2, NEUROD1, EOMES, RND2, DLL1 and DLL3 was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

Fig. 3.

NEUROG2 is necessary and sufficient to turn on neurogenic genes in COs. (A) NEUROG2 immunolabeling of day 18 and day 48 COs. Scale bars: 100 µm. (B) qPCR of NEUROG2 and NEUROG1 in day 30 and day 90 COs. (C) NEUROG2 ChIP-qPCR (n=5), or mock control ChIP-qPCR (n=3), using day 45 COs, and qPCR amplified DLL3 and NEUROD4 promoter region binding sites in the eluted chromatin (n=5). An ORF amplified sequence was used as a negative control. (D) NEUROG2 silencing in day 60 COs using lentiviral shRNA with a scrambled control sequence (shScr) and two shRNAs targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and the expression of NEUROG2, NEUROD1, EOMES, RND2, DLL1 and DLL3 was analyzed by qPCR. (E) Neurog2 gain-of-function assay, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of NEUROG2, NEUROD1, EOMES, RND2, DLL1 and DLL3 was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

We next investigated whether NEUROG2 was required to turn on known proneural target genes in human cortical cells using a shRNA-knockdown approach (Fig. 3D). Day 60 COs were transduced with lentiviral constructs carrying a shScrambled (shScr) control sequence and two shRNAs targeting endogenous NEUROG2 (Fig. 3D). After 72 h post-transduction, we confirmed the silencing of NEUROG2 with both shRNAs and demonstrated that EOMES, NEUROD1 and RND2, known target genes, were also downregulated (Fig. 3D). In contrast, DLL1 and DLL3 transcript levels were not affected by NEUROG2 silencing (Fig. 3D), likely due to their regulation by other TFs, such as the proneural TF encoded by ASCL1 (Castro et al., 2006; Henke et al., 2009). Nevertheless, these data support the idea that NEUROG2 has an essential role in driving neurogenic gene expression in COs, mimicking its requirement during murine cortical development (Fode et al., 2000; Schuurmans et al., 2004).

When overexpressed in E12.5 murine cortical NPCs, Neurog2 is sufficient to induce the expression of downstream genes driving glutamatergic neuronal differentiation (Kovach et al., 2013). To examine whether Neurog2 can similarly induce these target genes in hESC-derived COs, we transduced day 90 COs with adeno-associated virus (AAV) 5-GFAP promoter-containing vectors driving the expression of iCre as control or Neurog2-T2A-iCre (Fig. 3E). This expression vector drives gene expression in astrocytes and neural stem cells in the ventricular-subventricular zone of the adult mouse brain (Ghazale et al., 2022). COs were harvested after 14 days in vitro, and the ectopic expression of murine Neurog2 was confirmed (Fig. 3E). Overexpression of Neurog2 in day 90 COs was sufficient to induce the expression of known neurogenic target genes, including NEUROD1, EOMES, RND2, DLL1, and DLL3 (Fig. 3E). Taken together, these data demonstrate that NEUROG2 is necessary and sufficient to turn on neurogenic gene expression in hESC-derived COs, validating the use of this model system for further investigations of NEUROG2 function.

CRISPR engineering to generate NEUROG2-mCherry-KI hESCs for cortical organoid production

To study the molecular phenotype of NEUROG2-expressing cells, we used clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 to engineer NEUROG2-mCherry knock-in (KI) reporter hESC lines (Fig. 4A). Individual hESC clones were screened for homology-directed repair (HDR) using Droplet Digital PCR, with 4/87 clones (4.6% efficiency) containing a NEUROG2-mCherry insertion (Fig. 4B). Lines 105 and 117 were expanded for further characterization, both of which lacked genomic abnormalities in the most commonly mutated regions (Fig. S5A), and expressed pluripotency genes (OCT4, SOX2, NANOG) at similar levels as the starting hESC population (Fig. S5B). Using PCR genotyping, line 117 was shown to be heterozygous with correct wild-type (577 bp) and mCherry-KI (281 bp) amplicon sizes (Fig. 4C). Sequencing of the 5′ and 3′ junctions of the Cas9 targeted site confirmed that no mutations were introduced into the NEUROG2 locus at the junction sites in line 117 (Fig. 3D). However, while mCherry was inserted into the NEUROG2 locus in line 105, smaller than expected amplicon sizes for both the wild-type and mCherry-KI allele were detected by PCR genotyping, indicative of truncations in the targeted 3′ untranslated region (UTR). Nevertheless, both line 105 and line 117 hESCs could generate COs using a directed differentiation protocol, forming neural rosettes comprising SOX2+ NPCs and TUJ1+ neurons at day 18 (Fig. S5D) and day 30 (Fig. S5E). Day 27-30 COs produced from the two modified NEUROG2-mCherry KI hESC lines had similar diameters as wild-type hESC-derived COs of the same age in culture (Fig. S5C).

Fig. 4.

Generation of NEUROG2-mCherry KI hESC-derived COs. (A) CRISPR/Cas9-strategy to generate NEUROG2-mCherry KI hESCs by homology-directed repair (HDR). (B) ddPCR analysis of genomic DNA from sorted hESC-targeted cells. Raw droplet data of ddPCR measured for a negative control and two different positive clones indicating the increase in positive droplet count (blue dots) for the HDR sequence. (C) PCR genotyping of NEUROG2 wild-type and mCherry KI alleles in line 117 and line 105, showing that both lines are heterozygous. Expected amplicon sizes were observed in line 117, but the two amplicons were smaller than expected in line 105, indicative of 3′ truncations. (D) Sanger sequencing of NEUROG2-mCherry KI targeted hESC clone 117 near the Cas9 target site. (E,F) Co-immunolabeling of day 18 COs with mCherry and NEUROG2, SOX2 or DCX (E), and associated quantification (n=3) (F). (G,H) Co-immunolabeling of day 45 COs with mCherry and NEUROG2, SOX2 or DCX (G), and associated quantification (n=3) (H). Graphs in F,H show mean±s.e.m. Scale bars: 400 µm (low-magnification images); 100 µm (high-magnification images).

Fig. 4.

Generation of NEUROG2-mCherry KI hESC-derived COs. (A) CRISPR/Cas9-strategy to generate NEUROG2-mCherry KI hESCs by homology-directed repair (HDR). (B) ddPCR analysis of genomic DNA from sorted hESC-targeted cells. Raw droplet data of ddPCR measured for a negative control and two different positive clones indicating the increase in positive droplet count (blue dots) for the HDR sequence. (C) PCR genotyping of NEUROG2 wild-type and mCherry KI alleles in line 117 and line 105, showing that both lines are heterozygous. Expected amplicon sizes were observed in line 117, but the two amplicons were smaller than expected in line 105, indicative of 3′ truncations. (D) Sanger sequencing of NEUROG2-mCherry KI targeted hESC clone 117 near the Cas9 target site. (E,F) Co-immunolabeling of day 18 COs with mCherry and NEUROG2, SOX2 or DCX (E), and associated quantification (n=3) (F). (G,H) Co-immunolabeling of day 45 COs with mCherry and NEUROG2, SOX2 or DCX (G), and associated quantification (n=3) (H). Graphs in F,H show mean±s.e.m. Scale bars: 400 µm (low-magnification images); 100 µm (high-magnification images).

To characterize the cellular composition of COs generated from NEUROG2-mCherry KI hESCs, we used the correctly targeted, heterozygous line 117. In COs cultured for 18 days in vitro, 37.3±1.4% of the NEUROG2+ cells co-expressed mCherry, whereas by day 45 only 17.9±1.9% of NEUROG2+ cells co-expressed the fluorescent reporter (Fig. 4E-H; Fig. S6A,B). The slow maturation kinetics of mCherry (∼52 min) may contribute to a delay in the onset of reporter expression (Guerra et al., 2022). Despite this delay, the persistence of mCherry expression has been exploited for short-term lineage tracing in the murine cortex using a Neurog2-mCherry-KI allele (Han et al., 2021), and in 2D neural cultures derived from NEUROG2-TagRFP-KI induced pluripotent stem cells (Park et al., 2022). We therefore characterized the co-expression of mCherry with SOX2, an NPC marker, and DCX, an immature neuronal marker, in NEUROG2-mCherry KI hESC-derived COs. In day 18 COs, 35.8±5.5% of SOX2+ NPCs co-expressed mCherry (Fig. 4E,F; Fig. S6A). In keeping with an NPC identity, the vast majority of mCherry+ cells co-expressed SOX2 (89.4±8.7%) and many had initiated DCX expression (82.5±4.4%) in day 18 COs (Fig. 4E,F; Fig. S6C,E). Notably, while DCX expression is restricted to newborn neurons in the developing murine cortex, DCX is also expressed in germinal zone NPCs during ferret cortical development (Wang et al., 2024). In contrast, by day 45, mCherry expression was only detected in 2.1±0.6% of SOX2+ NPCs, and within the overall mCherry+ population, only 3.1±0.2% of labeled cells were SOX2+ NPCs (Fig. 4G,H; Fig. S6D). Instead, 99.1±0.2% of mCherry+ cells co-expressed DCX at day 45, and since these cells are not SOX2 expressing, we infer that they are newborn neurons (Fig. 4G,H; Fig. S6F). The persistence of mCherry expression after NEUROG2 expression declines allowed us to use mCherry to profile the phenotype of CO cells derived from NEUROG2+ NPCs.

Transcriptomic analyses reveal a link between NEUROG2 expression and the extracellular matrix in cortical organoids

To characterize gene expression in the NEUROG2 lineage, we generated COs from NEUROG2-mCherry KI hESCs, using both lines 105 and 117 in two independent experiments. COs were harvested after 45-47 days in vitro, and mCherry-high and mCherry-low cells were FACS-enriched from pools of seven or eight COs. mCherry-high cells represented 3.5% of the total sorted cell pool in COs derived from hESC line 117, and 2.6% in COs from hESC line 105. The enrichment of mCherry and NEUROG2 transcripts in mCherry-high versus -low cells was verified by Droplet Digital PCR (ddPCR) on RNA isolated from the sorted cells used for transcriptomics (Fig. S7A) and by qPCR on re-sorted day 62 COs for validation (Fig. 5A). We refer to sorted cells as mCherry-high and mCherry-low to acknowledge the low level of mCherry-expressing cells collected in the ‘negative’ sorted cells, and not to suggest that mCherry expression is at variable levels within each cell pool.

Fig. 5.

Targeted transcriptomic analysis of NEUROG2-mCherry KI hESC-derived COs. (A) qPCR to validate FACS-enrichment of mCherry and NEUROG2 transcripts in mCherry-high versus mCherry-low cells (n=3 each). Data are mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05. (B) Principal component analysis of targeted transcriptomic data collected from two sets of day 45 COs generated from NEUROG2-mCherry KI hESC cell lines 105 and 117 (N1, N2), for a total of four replicate data sets. (C) Volcano plot showing enriched genes in mCherry-high versus mCherry-low day 45 CO cells. (D,E) Biological Process-Gene Ontology (GOBP) terms enriched in DEGs that were upregulated (D) or downregulated (E) in mCherry-high CO cells. (F) Bar graph showing log2FC values of DEGs encoding ECM proteins and remodelers. (G) Enrichment of NEUROG2 and ECM gene transcripts in NPC compartments from microdissected human fetal cortical zones (Fietz et al., 2012). (H,I) Pseudo-bulk analysis of NEUROG2 and COL1A1 (H) and NEUROG2 and COL3A1 (I) transcript counts in scRNA-seq data collected from PCW 5-14 human cortices (Braun et al., 2023), showing log2CPM. (J,K) Pseudo-bulk analysis of Neurog2 and Col1a1 (J) and Neurog2 and Col3a1 (K) transcript counts in scRNA-seq data collected from E10.5 to E17.5 mouse cortices (Di Bella et al., 2021), showing log2CPM. BP, biological process; CO, cortical organoid; CP, cortical plate; DEGs, differentially expressed genes; ECM, extracellular matrix; GO, gene ontology; IPCs, intermediate progenitor cells; ISVZ, inner subventricular zone; mCh, mCherry; NPCs, neural progenitor cells; OSVZ, outer subventricular zone; VZ, ventricular zone.

Fig. 5.

Targeted transcriptomic analysis of NEUROG2-mCherry KI hESC-derived COs. (A) qPCR to validate FACS-enrichment of mCherry and NEUROG2 transcripts in mCherry-high versus mCherry-low cells (n=3 each). Data are mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05. (B) Principal component analysis of targeted transcriptomic data collected from two sets of day 45 COs generated from NEUROG2-mCherry KI hESC cell lines 105 and 117 (N1, N2), for a total of four replicate data sets. (C) Volcano plot showing enriched genes in mCherry-high versus mCherry-low day 45 CO cells. (D,E) Biological Process-Gene Ontology (GOBP) terms enriched in DEGs that were upregulated (D) or downregulated (E) in mCherry-high CO cells. (F) Bar graph showing log2FC values of DEGs encoding ECM proteins and remodelers. (G) Enrichment of NEUROG2 and ECM gene transcripts in NPC compartments from microdissected human fetal cortical zones (Fietz et al., 2012). (H,I) Pseudo-bulk analysis of NEUROG2 and COL1A1 (H) and NEUROG2 and COL3A1 (I) transcript counts in scRNA-seq data collected from PCW 5-14 human cortices (Braun et al., 2023), showing log2CPM. (J,K) Pseudo-bulk analysis of Neurog2 and Col1a1 (J) and Neurog2 and Col3a1 (K) transcript counts in scRNA-seq data collected from E10.5 to E17.5 mouse cortices (Di Bella et al., 2021), showing log2CPM. BP, biological process; CO, cortical organoid; CP, cortical plate; DEGs, differentially expressed genes; ECM, extracellular matrix; GO, gene ontology; IPCs, intermediate progenitor cells; ISVZ, inner subventricular zone; mCh, mCherry; NPCs, neural progenitor cells; OSVZ, outer subventricular zone; VZ, ventricular zone.

To profile gene expression, we used targeted transcriptome analysis covering 20,802 human genes, representing >95% of the UCSC reference genome. Principal component analysis of gene expression datasets revealed that mCherry-high versus mCherry-low cells were transcriptionally divergent and segregated from each other irrespective of the initiating hESC line or experimental day (Fig. 5B). We selectively assayed transcript counts for known genes involved in cortical development, focusing on transcripts with a log2 fold change (log2FC) >1 (i.e. a doubling in the original scaling) and with an adjusted P-value <0.05, corresponding to a false discovery rate cutoff of 0.05. A comparative analysis of differentially expressed genes (DEGs) in mCherry-high versus mCherry-low cells identified 1204 genes enriched in mCherry-high cells and 263 genes enriched in mCherry-low cells (Fig. 5C; Table S2). Gene ontology (GO) analysis of DEGs revealed an enrichment of biological process (BP) terms associated with the extracellular matrix (ECM) in mCherry-high cells, including ‘extracellular matrix organization’ and ‘collagen fibril organization’ (Fig. 5D; Table S3). Within the ECM-related GO-BP category: 0030198, 76/327 genes were among the DEGs enriched in mCherry-high cells. Included were collagens, and ADAM and matrix metalloproteases, which are involved in ECM remodeling (Fig. 5F).

Assessing the transcriptional relationship between NEUROG2 and collagen genes

The enrichment of ECM-associated gene expression in the NEUROG2-mCherry lineage is also observed in human bRG (Pollen et al., 2015), with expansion of the ECM in the oSVZ providing a pro-proliferative niche for basal NPCs (Amin and Borrell, 2020; Arai et al., 2011; Fietz et al., 2012; Florio et al., 2015; Martinez-Martinez et al., 2016; Pollen et al., 2015). To validate an association between NEUROG2 and ECM gene expression, we first mined a bulk RNA-seq dataset from the human fetal cortex (Fietz et al., 2012). In this dataset, several collagen genes, such as COL1A2, COL4A1, and COL4A2, were expressed at elevated levels in the VZ, iSVZ and oSVZ, in compartments in which NEUROG2 transcript levels were also elevated (Fig. 5G). However, a pseudo-bulk analysis of PCW 5-14 human cortical scRNA-seq data (Braun et al., 2023) revealed that COL1A1 and COL3A1 expression was comparatively much lower than NEUROG2 in aRG and bRG (Fig. 5H,I), a finding also observed in E10.5 to E17.5 mouse cortices (Di Bella et al., 2021) (Fig. 5J,K).

To examine the expression of ECM proteins in the NEUROG2 lineage, we co-immunostained NEUROG2-mCherry KI hESC-derived COs with mCherry and COL4 or laminin (LAM) antibodies. In day 18 COs, robust expression of COL4 and LAM was detected in circular formations in the center of the organoid, as well as in protrusions into the organoid periphery, where mCherry+ cells were concentrated (Fig. 6A). In higher magnification images, mCherry+ cells were surrounded by COL4+ and LAM+ protrusions, but there was limited overlap in expression (Fig. 6A). Similarly, in day 45 COs (Fig. 6B) and in optically cleared day 119 COs (Fig. S7B,C), COL4+ and LAM+ fibrils invaded the patches of mCherry-expressing cells, without obvious overlap in expression. There were also several regions with abundant mCherry+ cells that were devoid of ECM expression.

Fig. 6.

Transcriptional relationship between NEUROG2 and COL1A1, COL1A2 and COL3A1. (A,B) Co-immunolabeling of day 18 (A) and day 45 (B) NEUROG2-mCherry KI hESC-derived COs with mCherry and the ECM markers collagen IV (COL4) or laminin (LAM). Scale bars: 400 µm (low-magnification images); 100 µm (high-magnification images). (C) NEUROG2 ChIP-qPCR (n=3), or mock control ChIP-qPCR (n=3), using day 45 COs. qPCR to quantify COL1A1 and COL3A1 promoter region binding sites and an ORF control sequence in the eluted chromatin. (D) Neurog2 overexpression, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of COL1A1, COL1A2 and COL3A1 was analyzed by qPCR. (E) NEUROG2 silencing in day 60 COs using lentiviral shRNA constructs, with a scrambled control sequence (shScr) or targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and the expression of COL1A1, COL1A2 and COL3A1 was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

Fig. 6.

Transcriptional relationship between NEUROG2 and COL1A1, COL1A2 and COL3A1. (A,B) Co-immunolabeling of day 18 (A) and day 45 (B) NEUROG2-mCherry KI hESC-derived COs with mCherry and the ECM markers collagen IV (COL4) or laminin (LAM). Scale bars: 400 µm (low-magnification images); 100 µm (high-magnification images). (C) NEUROG2 ChIP-qPCR (n=3), or mock control ChIP-qPCR (n=3), using day 45 COs. qPCR to quantify COL1A1 and COL3A1 promoter region binding sites and an ORF control sequence in the eluted chromatin. (D) Neurog2 overexpression, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of COL1A1, COL1A2 and COL3A1 was analyzed by qPCR. (E) NEUROG2 silencing in day 60 COs using lentiviral shRNA constructs, with a scrambled control sequence (shScr) or targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and the expression of COL1A1, COL1A2 and COL3A1 was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

Given the lack of significant overlap between mCherry and ECM proteins, we set out to decipher the transcriptional relationship between NEUROG2 and ECM-related genes. By performing ChIP-qPCR on day 45 COs, we found that NEUROG2 bound upstream promoter regions for COL1A1 and COL3A1 (Qing et al., 2022) (Fig. 6C). To assess whether NEUROG2 was sufficient to induce the expression of ECM-related genes, we transduced day 90 COs with AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-T2A-iCre. After 14 days in vitro, COL1A1, COL1A2 and COL3A1 expression increased in response to Neurog2 overexpression (Fig. 6D).

Finally, to determine whether NEUROG2 was required to turn on collagen genes in human cortical cells, day 60 COs were transduced with lentiviral constructs carrying an shScr control sequence and two shRNAs targeting endogenous NEUROG2 (Fig. 6E). After 72 h, we unexpectedly observed an increase in COL1A1, COL1A2 and COL3A1 transcripts with at least one NEUROG2-shRNA (Fig. 6E). NEUROG2 is thus required to suppress collagen gene expression, with the caveat that the lentiviral silencing vector targets dividing NPCs and post-mitotic cells, and collagen genes may be differentially regulated in both cell types. Thus, the observed correlation between collagen gene expression and mCherry-high CO cells may reflect lineage maturation, with ECM genes upregulated as NEUROG2 expression declines. However, since NEUROG2 can also induce ectopic collagen gene expression, the relationship between NEUROG2 and collagen gene expression is complex.

Transcriptomic comparisons of mCherry-high and -low cells identify PPP1R17 as a potential NEUROG2 target gene

We performed GO analysis of downregulated DEGs in mCherry-high cells and observed an over-representation of BP terms such as ‘generation of neurons’ and ‘neuron differentiation’ (Fig. 5E; Table S4). To understand why neurogenesis-related terms were downregulated in the mCherry-lineage, we performed a biased analysis of select neurogenic genes (Fig. S8). Most aRG, bRG, proliferating cell and pan-neuronal markers were expressed at roughly equivalent levels in mCherry-high and mCherry-low CO cells (Fig. S8A-E). Of the few genes enriched in mCherry-high cells, several are associated with the glutamatergic neuronal lineage, including GAP43, SLC17A6 (VGLUT2), SLC17A7 (VGLUT1), RELN and PCP4 (Fig. S8E,F,H). In contrast, GABAergic neuronal lineage markers were enriched in mCherry-low cells, such as ASCL1 and GAD2, as well as BCL11B and FOXP2, markers of GABAergic interneurons and deep-layer cortical neurons (Fig. S8G,H). Thus, a main difference between mCherry-high and mCherry-low cells is an association with glutamatergic or GABAergic neuronal lineages, respectively. These findings are consistent with the known role of Neurog2 in specifying a glutamatergic neuronal fate in the cortex, and its repression of Ascl1, a GABAergic determinant (Fode et al., 2000; Kovach et al., 2013; Schuurmans et al., 2004).

We examined the top DEGs in the mCherry-high and mCherry-low lineages more closely (Fig. 7A). FZD8 (Boyd et al., 2015) and PPP1R17 (Girskis et al., 2021) were of interest as they are controlled by known HARs (Fig. 7A). An enrichment of PPP1R17 transcripts in the NEUROG2 lineage was consistent with both of these genes being expressed in basal NPCs during human cortical development (this study; Girskis et al., 2021; Johnson et al., 2015). To compare NEUROG2 and PPP1R17 expression profiles further, we performed a pseudo-bulk comparison of scRNA-seq data collected from PCW 5-14 human cortices (Braun et al., 2023). NEUROG2 and PPP1R17 were expressed at roughly equivalent levels in all NPC pools, including IPCs, as well as in glutamatergic neurons and ‘other’ cells, especially after PCW 9 (Fig. 7B). In contrast, a pseudo-bulk comparison of Neurog2 and Ppp1r17 aggregated transcript read counts in scRNA-seq data from E10.5 to E17.5 mouse cortices (Di Bella et al., 2021) revealed that, whereas Neurog2 expression is enriched in cortical NPCs, especially in IPCs, Ppp1r17 transcripts are for the most part not detected, except for at low levels in IPCs and migrating neurons at early stages (Fig. S1B). Thus, there is a strong correlation between PPP1R17 and NEUROG2 transcript levels in human fetal cortices and a weaker correlation in embryonic murine cortices.

Fig. 7.

NEUROG2 engages with PPP1R17-regulatory elements and is sufficient to induce PPP1R17 transcription. (A) Bar graph showing log2FC values of DEGs involved in neurogenesis in mCherry-high and in mCherry-low CO cells. (B) Pseudo-bulk analysis of NEUROG2 and PPP1R17 transcript counts in scRNA-seq data collected from PCW 5-14 human cortices (Braun et al., 2023), showing log2CPM. (C) Single-cell ATAC-seq profiling of the PPP1R17 locus, showing accessible chromatin in regions and cell types in the developing human brain. Conserved accessible chromatin regions were identified in an upstream enhancer (yellow box) and surrounding the TSS (green box). A primate-specific HAR (red box) is mainly accessible in glutamatergic cortical lineages. A phyloP score was derived from multiple mammalian species, with negative scores indicative of accelerated evolution for the PPP1R17-HAR element in chimps and rhesus monkeys. (D) qPCR to validate FACS-enrichment of PPP1R17 transcripts in mCherry-high versus mCherry-low cells (n=3 each). (E) NEUROG2 ChIP-qPCR (n=3), or mock control ChIP-qPCR (n=3), using day 45 COs. qPCR was used to quantify PPP1R17-HAR and -TSS binding sites in the eluted chromatin. (F) Transcriptional reporter assay in SHSY-5Y human neuroblastoma cells using pCIG2-Neurog2 or pCIG2-GFP (negative control) expression vectors and luciferase (LUC) constructs with a minimal promoter carrying the PPP1R17-HAR or -TSS elements. (G) Neurog2 gain-of-function assay, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of PPP1R17 was analyzed by qPCR. (H) NEUROG2 silencing in day 60 COs using lentiviral shRNA constructs, with a scrambled control sequence (shScr) or targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and PPP1R17 expression was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

Fig. 7.

NEUROG2 engages with PPP1R17-regulatory elements and is sufficient to induce PPP1R17 transcription. (A) Bar graph showing log2FC values of DEGs involved in neurogenesis in mCherry-high and in mCherry-low CO cells. (B) Pseudo-bulk analysis of NEUROG2 and PPP1R17 transcript counts in scRNA-seq data collected from PCW 5-14 human cortices (Braun et al., 2023), showing log2CPM. (C) Single-cell ATAC-seq profiling of the PPP1R17 locus, showing accessible chromatin in regions and cell types in the developing human brain. Conserved accessible chromatin regions were identified in an upstream enhancer (yellow box) and surrounding the TSS (green box). A primate-specific HAR (red box) is mainly accessible in glutamatergic cortical lineages. A phyloP score was derived from multiple mammalian species, with negative scores indicative of accelerated evolution for the PPP1R17-HAR element in chimps and rhesus monkeys. (D) qPCR to validate FACS-enrichment of PPP1R17 transcripts in mCherry-high versus mCherry-low cells (n=3 each). (E) NEUROG2 ChIP-qPCR (n=3), or mock control ChIP-qPCR (n=3), using day 45 COs. qPCR was used to quantify PPP1R17-HAR and -TSS binding sites in the eluted chromatin. (F) Transcriptional reporter assay in SHSY-5Y human neuroblastoma cells using pCIG2-Neurog2 or pCIG2-GFP (negative control) expression vectors and luciferase (LUC) constructs with a minimal promoter carrying the PPP1R17-HAR or -TSS elements. (G) Neurog2 gain-of-function assay, using AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-iCre to transduce day 90 COs (n=3 each). COs were harvested after 14 days and the expression of PPP1R17 was analyzed by qPCR. (H) NEUROG2 silencing in day 60 COs using lentiviral shRNA constructs, with a scrambled control sequence (shScr) or targeting NEUROG2 (-A and -C) (n=7 each). COs were harvested after 72 h and PPP1R17 expression was analyzed by qPCR. Graphs show mean±s.e.m. Unpaired Student's t-tests were used for pairwise comparisons. Significance was defined as P<0.05.

NEUROG2 engages with PPP1R17 regulatory elements and is sufficient to induce PPP1R17 transcription

To assess the relationship between NEUROG2 and PPP1R17, we characterized the accessibility of PPP1R17 upstream regulatory elements by mining single-cell assay for transposase-accessible chromatin using sequencing (ATAC)-seq data generated across cortical cell types in the developing human brain (Ziffra et al., 2021). A comparison to enhancer peaks for Ppp1r17 in the E12.5 mouse fetal forebrain (Rhodes et al., 2022), and to an ATAC-seq profile of Ppp1r17 from mouse fetal forebrain at E15.5 (Gorkin et al., 2020) led to the identification of a conserved regulatory region (Fig. 7C, yellow box) upstream of PPP1R17, shared open chromatin peaks near the transcriptional start site (TSS; Fig. 7C, green box) and a previously identified HAR that was primarily accessible in glutamatergic neurons and to a lesser extent in IPCs (Fig. 7C). To assess evolutionary constraint, we measured phyloP scores across multiple mammalian species, with negative scores indicative of accelerated evolution for the PPP1R17-HAR element in chimps and rhesus monkeys (Fig. 7C), in line with previous analyses of PPP1R17 (Girskis et al., 2021).

We confirmed that PPP1R17 was enriched in mCherry-high cells from day 62 COs by qPCR (Fig. 7D). We then performed NEUROG2-ChIP-qPCR on day 45 COs, revealing that NEUROG2 bound both the PPP1R17-HAR and the PPP1R17-TSS element (Fig. 7E). To test whether this binding was functional, we linked the PPP1R17-HAR and -TSS elements to a luciferase reporter (Fig. 7F). Compared to baseline control values, NEUROG2 elevated luciferase activity in SHSY-5Y human neuroblastoma cells using either the PPP1R17-HAR or -TSS reporters (Fig. 7F).

To assess whether Neurog2 was sufficient to induce PPP1R17 transcription, we transduced day 90 COs with AAV5-GFAP-iCre (control) and AAV5-GFAP-Neurog2-T2A-iCre, demonstrating that after 14 days in vitro PPP1R17 transcript levels increased (Fig. 7G). Notably, similar results were obtained in murine P19 cells (Fig. S1C), suggesting that Neurog2 is sufficient to transactivate Ppp1r17 across mammalian species. Finally, to determine whether NEUROG2 was required to turn on PPP1R17 in human cortical cells, day 60 COs were transduced with lentiviral constructs carrying a shScr control sequence and two shRNAs targeting endogenous NEUROG2 (Fig. 7H). After 72 h post-transduction, PPP1R17 was not significantly affected (Fig. 7H). In summary, NEUROG2 is sufficient to turn on PPP1R17 expression, and normally engages with PPP1R17 regulatory elements in human COs, but other TFs can compensate for the loss of NEUROG2 to transcribe this HAR-associated gene.

By analyzing transcriptomic data from human fetal cortices and COs, we found that NEUROG1 and NEUROG2 are enriched in basal NPCs, similar to findings in the murine cortex (Han et al., 2018). Pseudo-bulk analyses revealed that NEUROG1 and NEUROG2 are expressed at roughly equivalent levels in individual cortical NPCs in humans, in contrast to the comparatively higher Neurog2 transcript levels during murine cortical development. However, the total number of cortical NPCs expressing NEUROG2 is higher than the number of NEUROG1+ NPCs in both human (this study) and murine (Han et al., 2018) cortices. Pseudotime trajectory analysis of day 30 (this study) and day 90 (Sivitilli et al., 2020) COs revealed that NEUROG1 lineages predominate early, whereas NEUROG2 lineages are enriched later. These findings are in line with the earlier role for Neurog1 in the developing murine cortex (Han et al., 2018). Whether NEUROG1 is required to slow down early phases of NEUROG2-driven neurogenesis in human cortices via the formation of less efficient heterodimers, as shown in mouse (Han et al., 2018), remains to be determined. These data differ from a previous study performed in human fetal cells, which found that NEUROG2 and its target genes (e.g. HES6, NEUROD4, NHLH1, NEUROD1) are expressed at the highest levels in FACS-enriched cortical bRG (CD15+, GLAST+, CD133-low) and at negligible levels in IPCs (negative for all three markers) (Johnson et al., 2015). One important difference is that Johnson et al. (2015) relied on cell-surface protein markers to identify and isolate bRG and IPCs, whereas we used transcript-based methods of cell annotation. Furthermore, Johnson et al. (2015) sorted primary human fetal cells, whereas we used a CO model. Regardless of the differences in cell type biases, both studies found that NEUROG2 is expressed in basal and apical NPCs.

By gene silencing, we showed that NEUROG2 is required to turn on known neurogenic target genes in COs, and that the same genes could be induced by ectopic Neurog2 expression. To identify additional NEUROG2-regulated genes, we engineered NEUROG2-mCherry KI reporter hESCs for CO modeling. We observed a relatively low concordance between mCherry and NEUROG2 protein expression in derivative COs, which declined between day 18 and 45. Nevertheless, NEUROG2 transcripts were enriched in mCherry-high sorted CO cells, validating the use of this system to trace and isolate NEUROG2-lineage cells. Reasons for a lack of complete concordance between NEUROG2 and mCherry expression could include the tight regulation of NEUROG2 transcription in 2-3 h oscillatory cycles (Imayoshi et al., 2008), the short-intracellular half-life of NEUROG2 protein (Li et al., 2012), and the negative regulation of NEUROG2 protein translation (Yang et al., 2014). Additionally, slow mCherry chromophore formation and maturation could delay the appearance of mCherry epifluorescence (Hebisch et al., 2013). NEUROG2-TagRFP KI induced pluripotent stem cells were used in 2D neural differentiation cultures in a separate study, and the peak overlap between RFP and NEUROG2 protein was observed by day 19, after which RFP expression similarly diminished (Park et al., 2022).

Using a targeted transcriptomic screen, we identified several genes that are differentially expressed in CO cells derived from NEUROG2-expressing NPCs, including ECM-associated genes. In rodent cortices, ECM expression levels are high in the VZ, where proliferative, aRG reside, but not in the SVZ, where IPCs have a limited proliferative potential (Arai et al., 2011; Florio et al., 2015; Pollen et al., 2015). In contrast, in gyrencephalic species, bRG and IPCs express high levels of ECM genes to support integrin signaling and to create a pro-proliferative, oSVZ niche (Amin and Borrell, 2020; Arai et al., 2011; Fietz et al., 2012; Martinez-Martinez et al., 2016). As a result, bRG and IPCs, which have lost constraining attachments to the ventricular surface, proliferate extensively to support increased neurogenesis and cortical expansion. We found that although NEUROG2 engages with collagen-gene regulatory elements and is sufficient to induce COL1A1, COL1A2 and COL3A1 transcription in COs, it is not necessary for their transcription. However, NEUROG2 silencing increases COL1A1, COL1A2 and COL3A1 expression, suggestive of a negative regulatory relationship, possibly because silencing NEUROG2 may have different effects on collagen-gene expression depending on whether the targeted cells are dividing NPCs or post-mitotic neurons, as shown for other NEUROG2 target genes (Péron et al., 2023). Further insights may be gained by assessing the relationship between NEUROG2 and other inducers of ECM gene expression, such as SOX9, which is similarly expressed in basal NPCs in the human and ferret cortex (Guven et al., 2020).

NEUROG2 is expressed in bRG, and human-gained enhancers are enriched in genes expressed in bRG, including Notch signaling genes (Reilly et al., 2015; Song et al., 2020). NEUROG2 and other basic helix-loop-helix TFs, such as ASCL1, control the expression of the Notch ligands DLL1 and DLL3 (Castro et al., 2011; Henke et al., 2009), such that these genes were not differentially expressed in our separated mCherry-high and mCherry-low cells. However, this finding does not negate the possibility that these two proneural genes may differentially regulate the expression of Notch ligands in basal NPCs, which could be tested in the future. Our analysis of the top DEGs in mCherry-high COs identified an enrichment of two HAR-associated genes, PPP1R17 and FZD8. Using ChIP-qPCR and overexpression studies, we showed that NEUROG2 binds to the PPP1R17-HAR and -TSS, and is sufficient to induce PPP1R17 expression. PPP1R17, a HAR-regulated gene encoding a phosphatase regulatory subunit is expressed in human but not mouse cortical NPCs, at least at the protein level (Girskis et al., 2021). However, based on our pseudo-bulk analyses, Ppp1r17 transcripts are detected at low level in murine IPCs, and Neurog2 is sufficient to induce Ppp1r17 expression in P19 cells. Thus, the correlation between NEUROG2 and PPP1R17 in cortical NPCs is not a uniquely human feature.

Taken together, NEUROG2 has at least some conserved gene targets in mouse cortices and human COs, although further studies will be required to compare target genes across species comprehensively, which may better explain the unique patterns of human neurogenesis.

hESC maintenance

hESCs (H1/WA01) were purchased from WiCell Research Institute, Wisconsin, USA. hESC usage for this project was approved by the Canadian Stem Cell Oversight Committee (SCOC application to C.S. and to C.S. and J.N.) as well as by the SRI's Research Ethics Board (REB Project Identification Number: 5003). Briefly, hESCs were cultured under feeder-free conditions in mTeSR Plus media (Stem Cell Technologies, 100-0276) on plates coated with Matrigel (Corning, 354277) and maintained in 5% CO2 incubators at 37°C. Versene (Thermo Fisher Scientific, 15040-066) was used to dissociate hESCs by manual pipetting every 4-5 days for maintenance. hESC cultures were monitored daily for differentiated cells, which were removed by manual scraping. Prior to generating COs, quality control tests were routinely performed on hESC cultures, using a human stem cell pluripotency detection qPCR kit (Sciencecell, 0853) and hPSC genetic analysis kit (Stem Cell Technologies, 07550), according to the manufacturer's protocols.

CRISPR/Cas9 gene editing

CRISPR genome editing was used to insert an mCherry reporter gene into the 3′UTR of the NEUROG2 locus. The vector pSpCas9(BB)-2A-GFP (PX458) (Addgene, plasmid #48138), was purchased to target NEUROG2. To promote HDR of the Cas9-cleaved NEUROG2 target locus, we co-electroporated: (1) a CRISPR plasmid containing SpCas9-2A-eGFP and a single guide RNA (sgRNA) to NEUROG2 that was a fusion of a CRISPR RNA (crRNA) and a trans-activating crRNA (tracrRNA); and (2) a repair template containing homology arms flanking an mCherry reporter cassette. Then, 1.5×106 hESCs were transfected with 1 µg DNA (500 ng Cas9 plasmid and 500 ng linearized donor plasmid) by nucleofection (pulse code CA137) using the P3 Primary Cell 4D-Nucleofector X kit (Lonza, V4XP-3024) in a 4D-Nucleofector (Lonza, AAF-1003B) and were plated in a 6-well plate containing 10 µM Rock inhibitor Y-27632 (Stem Cell Technologies, 72302). Twenty-four hours after transfection, cells were harvested for quantification of transfection efficiency by flow cytometry for EGFP expression, indicative of Cas9 transfection. To isolate individual clones, GFP-positive cells were sorted and re-plated at clonal density in multiple 10-cm plates. Individual clones were then picked and further expanded, followed by genomic DNA extraction to screen for clones that underwent HDR using ddPCR. For ddPCR, we employed a forward primer-probe upstream to the starting point of the homology arm region and a reverse primer-probe that only bound to a site inside the exogenous mCherry sequence. Two correctly targeted clones were used for downstream experiments (lines 105 and 117).

CO generation

We adapted our CO differentiation protocol according to a previously described protocol (Qian et al., 2018, 2016). For embryoid body (EB) formation on day 0, hESC colonies were dissociated with Gentle Cell Dissociation Reagent (Stem Cell Technologies, 07174) for 7 min at 37°C, and 12,000 hESCs in 100 µl STEMdiff kit EB formation media (Stem Cell Technologies, 08570) supplemented with 50 µM Rock inhibitor Y-27632 (Stem Cell Technologies, 72302) were plated in 96-well V-bottom plates (low-binding) (Greiner Bio-One, 651970). On days 1 and 3, 2 µM dorsomorphin, an inhibitor of BMP type I receptors (ALK2, ALK3, ALK6; ACVR1, BMPR1A, BMPR1B, respectively) (Stem Cell Technologies, 72102) and 2 µM A83-01, an inhibitor of TGFβ type I receptors (ALK4, ALK5, ALK7; ACVR1B, TBFBR1, ACVR1C, respectively) (Stem Cell Technologies, 72022) was added to the media, and cells were cultured for 5 days in vitro to induce EB formation. On day 5, single EBs that reached ∼400-600 µm in diameter were selected for neural induction and were transferred to individual wells of 24-well ultra-low attachment plates (Corning, 3473). The media was switched to the STEMdiff kit Induction media containing 1 µM SB431542, an ALK4, ALK5, ALK7 inhibitor (Stem Cell Technologies, 72234) and 1 µM CHIR99021, a GSK3β inhibitor that activates Wnt signaling and limits apoptosis (Delepine et al., 2021; Qian et al., 2016) (Stem Cell Technologies, 72054) and cultured for four more days. On day 9, EBs of ∼500-800 µm diameter that had translucent edges, a sign of neuroepithelial induction, were placed onto a single dimple of an embedding sheet (Stem Cell Technologies, 08579) and 15 µl of Matrigel, an undefined ECM preparation that contains collagens, laminins, other ECM molecules and growth factors (Corning, cat. 354277), was added to encapsulate each CO. Matrigel droplets were incubated at 37°C for 30 min before they were washed into 6-well ultra-low attachment plates (Stem Cell Technologies, 38071) containing STEMdiff kit Expansion media with 1 µM SB431542 and 1 µM CHIR99021. On day 13, individual EBs with clear neuroepithelial cell buds were transferred to each well of a 12-well miniature spinning bioreactor (Qian et al., 2018, 2016) containing STEMdiff kit Maturation media. From day 30, ECM proteins were supplemented in Maturation media by dissolving Matrigel at 1% (v/v) containing human recombinant brain-derived neurotrophic factor (BDNF; PeproTech, AF-450-02). The aggregated cells were referred to as COs from this stage onward and were allowed to develop further in maturation media until the experimental endpoints, as described. Once available, we switched to the STEMdiff Dorsal Forebrain Organoid Differentiation Kit (Stem Cell Technologies, 08620) following the manufacturer's protocol, except that COs were transferred to a 12-well plate with a miniaturized multiwell spinning bioreactor SpinΩ lid after day 14, Matrigel (1% v/v) was added to the media after day 20, and BDNF (20 ng/ml) was added to all media after day 30.

Cryosectioning and immunostaining

COs were rinsed with ice-cold PBS (without Ca2+ and Mg2+) (WISENT, 311-010-CL), fixed in 4% paraformaldehyde in PBS (PFA; Electron Microscopy Sciences, 19208) overnight, and immersed in 20% sucrose (Sigma-Aldrich, 84097) in 1× PBS overnight after three washes for 5 min in PBS. COs were embedded in Tissue-Tek Optimal Cutting Temperature (O.C.T.) compound (Sakura Finetek), and 10-μm-thick sections were collected with a Leica CM3050 cryostat (Leica Microsystems Canada Inc.). Samples were collected on Fisherbrand™ Superfrost™ Plus Microscope Slides (Thermo Fisher Scientific, 12-550-15). Cryosections of fixed COs were washed in 0.1% Triton X-100 (Sigma-Aldrich, T8787) in PBS (PBST), then blocked for 1 h at room temperature (RT) in 10% horse serum (WISENT, 065-150) in PBST. Primary antibodies were diluted in blocking solution as follows: SOX2 (1:500, Abcam, ab97959), PAX6 (1:500, BioLegend, 901301), NEUROG2 (1:500, Invitrogen, PA5-78556), DCX (1:500, Abcam, ab18723), COL4 (1:200, Abcam, ab6586), LAM (1:200, Sigma-Aldrich, L9393), mCherry (1:500, SICGEN, AB0040-200) and TUJ1 (1:500, BioLegend, 802001). After 1 h of blocking at RT, slides were incubated with primary antibodies at 4°C overnight. The next day, slides were washed five times for 5 min each wash in PBST, followed by incubation with 1:500 dilutions of species-specific secondary antibodies [Invitrogen: donkey anti-goat IgG, Alexa Fluor 568 (A-11057), donkey anti-rabbit IgG, Alexa Fluor 568 (A-10042) and donkey anti-rabbit IgG, Alexa Fluor 488 (A-21206)] for 1 h at RT. Slides were washed five times in PBST and counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Invitrogen, D1306) and mounted in Aqua-polymount (Polysciences Inc., 18606-20). All images were taken using a Leica DMi8 Inverted Microscope (Leica Microsystems CMS, 11889113).

Bulk RNA-seq

COs were dissociated into a single-cell suspension using the Worthington Papain System kit (Worthington Biochemical, LK003150) according to the manufacturer's instruction. We collected single-cell suspensions from day 45-47 COs, pooling seven or eight COs per sample. Briefly, prewarmed papain solution with DNase (2.5 ml) was added to the COs in a 60 mm dish. COs were minced with a sterile razor blade into smaller pieces and incubated for 30-45 min at 37°C on an orbital shaker (70 rpm). The tissue suspension was triturated eight to ten times with a P1000 pipette tip to assist the release of single cells. Cell suspensions were transferred to a 15 ml centrifuge tube and an ovomucoid protease inhibitor solution (reagent supplied in the Worthington Papain System kit) was added to stop papain activity. The cells were centrifuged at 300 g for 7 min and filtered through a 40 µm strainer to remove remaining cell aggregates. Single cells were resuspended in PBS containing FACS buffer (1 mM EDTA, 0.1% bovine serum albumin and Ca2+/Mg2+-free PBS) with DAPI before flow cytometry analysis. Each sample was sorted into mCherry-positive and mCherry-negative groups. Total RNA was extracted from FACS-isolated cells using the MagMAX-96 total RNA isolation kit (Thermo Fisher Scientific, AM1830). The extracted total RNA was quantified by Qubit 3 Fluorometer with Qubit RNA HS Assay kit. The integrity of total RNA (RIN value) was measured using an Agilent 2100 Bioanalyzer with RNA 6000 Pico kit.

Targeted transcriptome analysis

Targeted transcriptome sequencing was performed on the Ion S5XL Next Generation Sequencing system with the Ion AmpliSeq Transcriptome Human Gene Expression assay kit (Thermo Fisher Scientific). This assay covers 20,802 human RefSeq genes (>95% of UCSC refGene) with a single amplicon designed per gene target. The gDNA in the RNA sample was digested by ezDNase and the cDNA was synthesized from 10 ng of total RNA using SuperScript IV VILO Master Mix with ezDNase Enzyme kit (Thermo Fisher Scientific). The cDNA libraries were constructed with the Ion Ampliseq Library Kit Plus. The targeted areas were amplified by PCR for 12 cycles. The resulting amplicons were treated with FuPa reagent to partially digest primers. Amplicons were ligated to Ion P1 and IonCode barcode adapters and purified using Agencourt AMPure XP reagent (Beckman Coulter). Barcoded libraries were quantified using the Ion Library TaqMan Quantitation Kit (Thermo Fisher Scientific) and diluted to a final concentration of 80 pM. The sequencing template preparation was done using Ion Chef with Ion 540 Chef Kits. Sequencing was performed for 500 flows on an Ion S5XL Sequencer with Ion 540 chip.

Next-generation sequencing data analysis of bulk targeted transcriptome data

The Ion Torrent platform-specific pipeline software, Torrent Suite version 5.18.1 (Thermo Fisher Scientific) was used to separate barcoded reads and to filter and remove polyclonal and low-quality reads. Ion Torrent platform-specific plugin, ampliseqRNA (v.5.18.0.0) was used for the alignment of the raw sequencing reads and quantitation of normalized gene expression level (reads per million). DESeq2 was used to analyze differential expression. Principal component analysis, generation of scatter plots and volcano plots, hierarchical clustering and pathway analysis were performed with Transcriptome Analysis Console (TAC) 4.0 software using CHP files.

FACS and qPCR

To validate our original sort, we collected three pools of 5-day 62 NEUROG2-mCherry KI COs, each counted as an independent replicate. Cells were dissociated using papain dissociation kit (Worthington Biochemical, LK003150), following the kit protocol. Briefly, COs were minced and incubated at 37°C with 500 rpm shaking for 30 min, with pipetting to mix every 5 min. Cells were pelleted at 300 g for 5 min followed by passing through the ovomucoid density gradient and centrifugation at 100 g for 6 min. The cell pellet was resuspended in HBSS buffer with 0.15% bovine serum albumin and 1 mM EDTA. FACS was performed using BD FACSDiva 8.0.3 software to collect mCherry-positive and -negative cells. RNA was isolated using the QIAGEN RNeasy micro kit (74004) followed by cDNA preparation using SuperScript IV VILO™ Master Mix with ezDNase enzyme, following the kit protocol (Invitrogen, 11766050). qPCR was performed using RT2 SYBR green qPCR master mix. The mCherry-specific primers used were: F: 5′-GACTACTTGAAGCTGTCCTTCC-3′; R: 5′-CGCAGCTTCACCTTGTAGAT-3′ (Thermo Fisher Scientific).

Preparation of snRNA-seq libraries and sequencing

Four batches of five COs were pooled and flash-frozen. Nuclei extraction was performed on frozen tissue as per the manufacturer's instructions (Nuclei Isolation Kit, 10x Genomics). Freshly isolated nuclei were counted using a Countess® II FL Automated Cell Counter (Thermo Fisher Scientific) and immediately processed using the Chromium Next GEM Single Cell 5′ Reagent Kit v2 (10x Genomics, 1000263). For each reaction, 16,500 nuclei were loaded onto GEM Chip K for an expected recovery of 10,000 nuclei. Gel Beads-in-emulsion were generated using the Chromium Controller followed by cDNA generation and amplification (13 cycles) as per the manufacturer's instructions. For each sample, 50 ng of cDNA was used for library generation. Equal molar amounts of each library for all samples were pooled and sequenced at an expected depth of 35,000 reads/nuclei using the Illumina NovaSeq X 10B flow cell system (The Centre for Applied Genomics, The Hospital for Sick Children, Toronto, Canada).

snRNA-seq data analysis in COs

Seurat v.4.0.1 R package (Han et al., 2021) was used for scRNA-seq analysis. Cells that were of low quality or represented doublets were excluded by filtering out cells with >120,000 and <1000 RNA counts and cells with mitochondrial RNA percentage >15. The samples were integrated using the FindIntegrationAnchors and IntegrateData functions followed by SCTransform. Clustering was performed by the RunPCA, FindNeighbors and FindClusters functions using the first 30 principal components. The 2D projection of the clustering was carried out by the RunUMAP function. Proneural negative, NEUROG2 or NEUROG1 single- and double-positive cells were identified with an expression threshold >0. Monocle3 R package was used for a pseudotime analysis using DEGs at an adjusted P-value cutoff of 0.001.

Single-cell pseudo-bulk analysis of murine and human cortical datasets

An scRNA-seq dataset from the developing human cortical brain (Braun et al., 2023) was downloaded from a data repository (https://github.com/linnarsson-lab/developing-human-brain/). An scRNA-seq dataset from the developing mouse cortical brain (Di Bella et al., 2021) was downloaded from Gene Expression Omnibus (GSE153164). In each case, raw counts were CPM-normalized and aggregated by cell type at each time point. The resulting pseudo-bulk dataset (containing CPM-normalized average counts per cell type) was log-transformed for visualizing gene expression patterns across select time points (5-14 weeks post-conception in human, and E10.5 to E17.5 in mouse).

Analysis of PPP1R17 enhancer peaks

A single-cell ATAC-seq profile of PPP1R17 across broad cortical cell types in the developing human brain was downloaded from the UCSC Genome Browser (https://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&lastVirtModeType=default&lastVirtModeExtraState=&virtModeType=default&virtMode=0&nonVirtPosition=&position=chr21%3A15477990%2D16373898&hgsid=2400341615_qfAhtZ2DO1WyK9M3sZt9AQaRAw24; Ziffra et al., 2021). Enhancer peaks for Ppp1r17 from mouse fetal forebrain at E12.5 were downloaded from the UCSC Genome Browser (https://genome.ucsc.edu/cgi-bin/hgTracks?db=mm10&lastVirtModeType=default&lastVirtModeExtraState=&virtModeType=default&virtMode=0&nonVirtPosition=&position=chr12%3A52713333%2D62634435&hgsid=2400341871_UU4W4mxmItwfPNO7I8aAxgOJYqiu; Rhodes et al., 2022), and an ATAC-seq profile of Ppp1r17 from mouse fetal forebrain at E15.5 was downloaded from ENCODE 3 (Gorkin et al., 2020) regulation tracks on the UCSC Genome Browser. The conserved regulatory region (Fig. 7C, yellow box) upstream of PPP1R17 in the human fetal brain was translated to mouse Ppp1r17 using the LiftOver tool in the UCSC Genome Browser.

P19 cell transfection

P19 cells were maintained in growth media containing 1× Alpha Modification of Eagle's Medium (AMEM; WISENT, 310-010-CL), 20% fetal bovine serum and 1% penicillin/streptomycin antibiotic. The cells were transfected using Lipofectamine 3000 (Thermo Fisher Scientific, L3000001) with pCIG2 control and pCIG2-Neurog2 DNA. The cell growth media was changed to fresh media 24 h post-transfection and the cells were harvested 48 h post-transfection.

RNA isolation, cDNA preparation and qPCR

Total RNA extraction was performed from COs using the QIAGEN RNeasy micro kit (74004), followed by CDNA preparation using the RT2 First strand reverse transcription kit (330401). qPCR using specific primers was performed using RT2 SYBR green qPCR Kit (Qiagen, 330513). The QIAGEN primers used were: murine Neurog2 (PPM28944A), Dll3 (PPM25734G), Neurod1 (PPM05527D), Rnd2 (PPM33691A), Ppp1r17 (PPM28954C), Col1a1 (PPM03845F), and Col3a1 (PPM04784B). Human primers were: NEUROG2 (PPH11564A), DLL3 (PPH06025A), DLL1 (PPH06024E), EOMES (PPH12647A), RND2 (PPH05839G), NEUROD1 (PPH00039E), COL1A1 (PPH01299F), COL1A2 (PPH01918B), COL3A1 (PPH00439F), PPP1R17 (PPH14658A), NEUROD4 (PPH16515A), NEUROG1 (PPH02437A) and PDGFRA (PPH00219C). qPCR analysis was performed using 2−ΔΔCT method by normalizing to the CT value of control samples.

ddPCR

The QX200 Droplet Digital PCR (ddPCR) system (Bio-Rad) was used for all ddPCR reactions. Detailed information for all primers is in Table S5. For HDR screening, the absolute number of NEUROG2-mCherry KI gene copies per cell was quantified and normalized to RPP30 (Bio-Rad, 10031243). Twenty nanograms of genomic DNA was used in a 20 µl PCR reaction containing 900 nM of the forward and reverse NEUROG2-mCherry KI and RPP30 primers, 250 nM of NEUROG2-mCherry KI and RPP30 probes, and 10 µl of 2× ddPCR Supermix for probes (Bio-Rad). Assay mixtures were loaded into a droplet generator cartridge (Bio-Rad), followed by the addition of 70 µl of droplet generation oil for probes (Bio-Rad) into each of the eight oil wells. The cartridge was then placed inside the QX200 droplet generator (Bio-Rad). Generated droplets were transferred to a 96-well PCR plate (Eppendorf), which was heat-sealed with foil and placed in C1000 Touch Thermal Cycler (Bio-Rad). Thermal cycling conditions were as follows: 95°C for 10 min, 44 cycles of 94°C for 30 s, 53°C for 1 min, and 98°C for 10 min. FAM fluorescent signal, which labeled the NEUROG2-mCherry KI DNA sequence, and HEX fluorescent signal which labeled the RPP30 DNA sequence, were counted by a QX200 Droplet Digital reader and analyzed by QuantaSoft analysis software v.1.7.4.0917 (Bio-Rad). Identified positive clones were expanded and underwent further quality checks.

To quantify the absolute number of mCherry and COL1A2 transcripts, RNA from mCherry-high and mCherry-low cell populations were collected from COs, and 10 ng of total RNA was reverse-transcribed in a 10 µl reaction using the SuperScript VILO cDNA Synthesis Kit (Invitrogen). The resulting cDNA was diluted to either 1:5 (mCherry) or 1:1500 (COL1A2) before amplification. The ddPCR reaction was performed in a 20 µl volume containing 10 µl of 2× QX200 ddPCR EvaGreen Supermix (Bio-Rad), 5 µl of diluted cDNA, and 1 µl each of 4 µM forward and reverse primers and 3 µl of nuclease-free water. Droplet generation was completed as above but with the addition of 70 µl of droplet generation oil for EvaGreen (Bio-Rad). Thermal cycling conditions were as follows: 95°C for 5 min, then 44 cycles of 96°C for 30 s and 56°C (mCherry) or 60°C (COL1A2) for 1 min, and then 4°C for 5 min, 90°C for 5 min and 4°C for indefinite hold for dye stabilization. EvaGreen fluorescent signal in each droplet were counted and analyzed as described. The copy number of mCherry and COL1A2 transcripts were normalized to the copies per ng of total RNA. All ddPCR analyses were performed at the SRI Genomics Core Facility.

Luciferase assay

SHSY-5Y human neuroblastoma cells (ATCC, CRL-2266) were plated in 6-well plates and were transfected using Lipofectamine 3000 with 5 µg of pCIG2 NEUROG2-expression vectors (Li et al., 2012), luciferase reporter plasmids, 0.25 µg firefly luciferase and 0.125 µg Renilla plasmid (transfection control). Two luciferase reporters were designed with PPP1R17 promoter with and without primate selective element (HAR). A 1.6-kb primate-selective element-containing (4-5.6 kb upstream of human PPP1R17) or a 0.7-kb TSS-flanking (−100 to +600) fragment was synthesized and cloned (GenScript Biotech) into the KpnI and SacI sites upstream of the SV40 promoter driving luciferase in pGL3-Promoter (Promega). To generate pCIG2-NEUROG2 expression vectors, the NEUROG2 coding-domain sequence (NM_024019) with Kozak consensus was synthesized and cloned (GenScript Biotech) into the SmaI site between the CAG promoter and IRES-EGFP of pCIG2. Cells were harvested at 48 h post-transfection to measure firefly luciferase and Renilla activities using the Dual-luciferase Reporter Assay System (Promega, E1910) following the kit instructions, using a TD 20/20 Luminometer (Turner Designs). Firefly luciferase data was normalized to the corresponding Renilla values.

shRNA lentivirus transduction

Day 60 COs were transduced with two shRNA lentiviruses (Origene, SKU TL302977V) to knock down NEUROG2 and a scrambled shScr control. Briefly, COs were incubated with 1.5×105 TU lentivirus in 50 µl of maintenance media in a 96-well U-bottom plate for 2 h. COs were transferred to a 24-well plate containing 1 ml maintenance media per well and were harvested 72 h post-transduction. RNA was isolated using the QIAGEN RNeasy micro kit (74004), followed by cDNA preparation (RT2 first strand kit; Qiagen, 330401) and qPCR using RT2 SYBR Green master mix (Qiagen, 330513).

AAV transduction in COs

Day 90 COs were incubated with 9×1010 GC of each AAV in 50 µl of maintenance media in a 96-well U-bottom plate for 1 h. COs were transferred to a 24-well plate with 500 µl maintenance media and were harvested 14 days post-transduction. AAV5-packaged GFAP-Neurog2-T2A iCre or GFAP-iCre vectors were used to overexpress Neurog2 or control, respectively. pAAV-GFAP-Neurog2-T2A-iCre was created by replacing the Neurod1 in pAAV-GFAP-mNeuroD1-T2a-iCre (kind gift of Dr Maryam Faiz, Department of Surgery, University of Toronto, Canada) with that of Neurog2. The ITR-flanked region included the GFAP promoter, Neurog2 coding-domain sequence, T2A self-cleaving site, iCre sequence, Woodchuck Hepatitis Virus Post-transcriptional Regulatory Element and a bovine growth hormone polyadenylation signal. RNA isolation was carried out using the QIAGEN RNeasy micro kit (74004), followed by cDNA preparation (RT2 first strand kit; Qiagen, 330401) and qPCR using RT2 SYBR Green master mix (Qiagen, 330513).

ChIP-qPCR

Day 45 COs were fixed using 2 mM disuccinimidyl glutarate (Sigma-Aldrich, 80424) for 20 min and 1% formaldehyde for 10 min at RT, followed by a 0.125 M glycine quench and three washes in PBS with protease, proteasome and phosphatase inhibitors, including 50 mM sodium fluoride, 0.2 mM sodium orthovanadate, 0.05 mM MG132 (Sigma-Aldrich, M7449), 2 mM PMSF, and 1× Complete protease inhibitor cocktail (Roche, 04 693 116 001). COs were lysed using Tris-EDTA (TE) buffer containing 1% SDS for 20 min at 4°C, followed by sonication with a bioruptor (Diagenode Pico) with 30 s ON/30 s OFF for 20 cycles. Chromatin was centrifuged at 13,000 g for 10 min and 5% of the supernatant was used as input. Chromatin was precleared using protein G Dynabeads (Invitrogen, 10003D) for 1.5 h and was then incubated with protein G Dynabeads preincubated with 2 µg anti-Neurog2 antibody (R&D Systems, MAB3314) overnight. The beads were washed twice with 0.5 M LiCl wash buffer, twice with 1 M NaCl wash buffer and once with TE buffer. Elution was performed using 1% SDS containing TE buffer at 65°C with 1400 rpm shaking for 15 min. To the eluted sample, we added 11 µl 5 M NaCl and 0.1 µg/µl proteinase K, and then incubated the sample at 42°C for 2 h and overnight at 65°C for reverse crosslinking. A phenol-chloroform extraction was performed, followed by centrifugation at 13,000 g for 10 min to collect the clear upper aqueous layer. DNA was precipitated by adding 1 µg/µl glycogen, 50 µl 3 M sodium acetate (pH 5.2) and 900 µl isopropanol for 20 min at −20°C followed by centrifugation for 20 min at maximum speed (13,000 g) at 4°C. DNA was washed using 70% ethanol and dissolved in Tris buffer. Qubit quantification was performed and 1 ng/µl was used to perform qPCR using RT2 SYBR Green master mix. The ChIP qPCR primers are described in Table S5. ChIP-qPCR fold enrichment was determined using the 2−ΔΔCT method and by normalizing to the negative ORF target as well as to mock (no antibody) controls.

Tissue clearing and fluorescence microscopy

For imaging of immunolabeled sections, we used a Leica DMI8 fluorescent microscope or a Zeiss Axiovert 200M confocal microscope. For imaging of COs in 3D, we first performed tissue clearing. Briefly, day 42 and 119 COs were fixed in 4% PFA overnight at 4°C. To preserve the tissue protein architecture, samples were cleared using the SHIELD (Stabilization to Harsh conditions via Intramolecular Epoxide Linkages to prevent Degradation) method (Park et al., 2019). Specifically, COs were incubated in SHIELD OFF solution at 4°C with shaking for 24 h. Subsequently, they were incubated in a mixture of SHIELD ON-Buffer and the SHIELD-Epoxy solution (7:1 ratio) at 37°C with shaking for 6 h. Lastly, the samples were incubated in SHIELD ON-Buffer at 37°C with shaking overnight. To carry out tissue delipidation, the samples were passively run down in the Delipidation buffer (LifeCanvas Technologies) for 3 days at RT. Samples were washed in 0.1% PBST three times over 3 h following each incubation and PFA fixation. COs were incubated in primary antibodies diluted in 0.1% PBST at RT for 48 h. Samples were then incubated in conjugated secondary antibodies diluted in 0.1% PBST [Alexa Fluor 488 goat anti-mouse IgG2a (Invitrogen, A-21131), Alexa Fluor 568 donkey anti-rabbit (Invitrogen, A-10042), Alexa Fluor 488 donkey anti-rabbit IgG (H+L) (Invitrogen, A-21206); 1:250] at RT for 48 h. For index matching and to make samples optically transparent, samples were incubated in EasyIndex medium (LifeCanvas Technologies; RI=1.52) at RT overnight. CO images were acquired using an UltraMicroscope Blaze light sheet fluorescence microscope (Miltenyi Biotech) with a 4× objective. The samples were mounted on a small sample stage using photoactivated adhesive (Bondic CNA) and placed into a custom organic imaging medium in the microscope's chamber (Cargille Immersion Liquid; RI=1.52). Two channels were acquired with a 488 nm wavelength and 85 mW power, and 639 nm with 70 mW, for mCherry and SOX2/TUJ1/LAM or COL4, respectively. A 1.67× magnification post-objective lens was employed generating an in-plane resolution of 1.95 μm and a step size of 3.55 μm (scanning protocol parameters: laser sheet thickness=7.1 μm, NA=0.050, and laser width=30% single sided multi-angle excitation).

Quantification and statistics

Statistical analysis was performed using GraphPad Prism Software version 8.0 (GraphPad Software). For pairwise comparisons, we used unpaired Student's t-tests to calculate statistical significance. For multiple comparison between more than two groups, we used one-way ANOVAs with Tukey post-hoc analyses. In all graphs, error bars represent s.e.m. If a P-value was less than or equal to 0.05, we considered the result as statistically significant.

Key resources

A full listing of all reagents and catalog numbers is presented in Table S6.

We thank Maryam Faiz (University of Toronto) for providing reagents. We acknowledge the support of the Sunnybrook Research Institute (SRI) Genomics Core Facility for targeted transcriptomics and ddPCR, the SRI Histology Core Facility for cryosectioning (Petia Stefanova), and the Sunnybrook Centre for Cytometry and Scanning Microscopy for FACS (Kevin Conway). We also acknowledge The Imaging Facility at The Hospital for Sick Children for assistance with light sheet microscopy.

Author contributions

Conceptualization: L.V., V.C., D.Z., F.S., Y.A., A.S., O.R., J.G., S.O., C.S.; Methodology: L.V., V.C., F.S., D.Z., C.K., H.S., Y.T., S.P., M.R., Y.A., A.S., M.G., O.R., J.G., C.W., S.O., C.S.; Software: H.S., J.G., S.O.; Validation: L.V., V.C., F.S., D.Z., H.S., S.P., M.R., Y.A., A.S., M.G., O.R., J.G., C.W., S.O., C.S.; Formal analysis: L.V., V.C., F.S., D.Z., C.K., H.S., H.G., L.B., Y.T., A.-M.O., S.P., M.R., Y.A., S.H., A.M., A.S., M.G., O.R., J.G., C.W., S.O., C.S.; Investigation: L.V., V.C., F.S., D.Z., C.K., H.S., H.G., L.B., Y.T., A.-M.O., S.P., M.R., Y.A., S.H., A.M., M.G., J.G., C.W., S.O.; Resources: H.S., Y.A., A.S., M.G., O.R., J.G., C.W., S.O., C.S.; Data curation: L.V., V.C., F.S., D.Z., C.K., H.S., H.G., L.B., Y.T., A.-M.O., S.P., M.R., Y.A., S.H., A.M., A.S., M.G., J.G., C.W., S.O., C.S.; Writing - original draft: L.V., V.C., F.S., H.S., Y.A., M.G., C.W., J.G., S.O., C.S.; Writing - review & editing: L.V., V.C., F.S., D.Z., C.K., H.S., H.G., L.B., Y.T., A.-M.O., S.P., M.R., Y.A., S.H., A.M., S.E.B., J.M., J.N., A.S., M.G., O.R., J.G., C.W., S.O., C.S.; Supervision: A.S., M.G., O.R., J.G., C.W., C.S.; Project administration: C.S.; Funding acquisition: S.E.B., J.M., J.N., O.R., C.S.

Funding

This work was supported by operating grants from the Canadian Institutes of Health Research (CIHR) (PJT-162108 to C.S.; PJT-183715 to C.S. and J.N.), and by a Government of Canada New Frontiers in Research Fund Transformation grant (to C.S., J.M. and J.N.), funded through the three federal research funding agencies [CIHR, Natural Sciences and Engineering Research Council of Canada (NSERC) and Social Sciences and Humanities Research Council of Canada]. We acknowledge and are grateful for philanthropic support from the Sunnybrook Foundation (community contributors: Martha G. Billes, J. Brian Prendergast and Catherine Rogers). C.S. holds the Dixon Family Chair in Ophthalmology Research. A.-M.O. was supported by a Canada Graduate Scholarship – Master's NSERC Studentship, A.M. was supported by a Canada Graduate Scholarship – Master's Canadian Institutes of Health Research (CGS-M/CIHR) and by a University of Toronto Vision Science Research Program (VSRP) scholarship, and S.H. by a Cumming School of Medicine, University of Calgary, Ontario Graduate Scholarship (OGS), University of Toronto VSRP scholarship, the Peterborough K.M. Hunter Charitable Foundation and a Margaret and Howard Gamble Research Grant. Open Access funding provided by the University of Toronto. Deposited in PMC for immediate release.

Data availability

Targeted transcriptomic analysis from this study has been deposited in ampliSeq and in the Gene Expression Omnibus (GEO) database under accession number GSE277092. snRNA-seq data from day 30 COs in this study were deposited in GEO under accession number GSE280812. We mined scRNA seq data in human COs available from GSE137877 (Sivitilli et al., 2020) and scRNA seq data of human fetal cortices was available from https://singlecell.broadinstitute.org/single_cell/study/SCP1290/molecular-logic-of-cellular-diversification-in-the-mammalian-cerebral-cortex (available in GEO under accession number GSE153164) and downloaded the human data from https://cellxgene.cziscience.com/collections/4d8fed08-2d6d-4692-b5ea-464f1d072077.

Amin
,
S.
and
Borrell
,
V.
(
2020
).
The extracellular matrix in the evolution of cortical development and folding
.
Front. Cell Dev. Biol.
8
,
604448
.
Arai
,
Y.
,
Pulvers
,
J. N.
,
Haffner
,
C.
,
Schilling
,
B.
,
Nusslein
,
I.
,
Calegari
,
F.
and
Huttner
,
W. B.
(
2011
).
Neural stem and progenitor cells shorten S-phase on commitment to neuron production
.
Nat. Commun.
2
,
154
.
Aydin
,
B.
,
Kakumanu
,
A.
,
Rossillo
,
M.
,
Moreno-Estelles
,
M.
,
Garipler
,
G.
,
Ringstad
,
N.
,
Flames
,
N.
,
Mahony
,
S.
and
Mazzoni
,
E. O.
(
2019
).
Proneural factors Ascl1 and Neurog2 contribute to neuronal subtype identities by establishing distinct chromatin landscapes
.
Nat. Neurosci.
22
,
897
-
908
.
Bertrand
,
N.
,
Castro
,
D. S.
and
Guillemot
,
F.
(
2002
).
Proneural genes and the specification of neural cell types
.
Nat. Rev. Neurosci.
3
,
517
-
530
.
Birey
,
F.
,
Andersen
,
J.
,
Makinson
,
C. D.
,
Islam
,
S.
,
Wei
,
W.
,
Huber
,
N.
,
Fan
,
H. C.
,
Metzler
,
K. R. C.
,
Panagiotakos
,
G.
,
Thom
,
N.
et al. 
(
2017
).
Assembly of functionally integrated human forebrain spheroids
.
Nature
545
,
54
-
59
.
Borrell
,
V.
(
2018
).
How cells fold the Cerebral Cortex
.
J. Neurosci.
38
,
776
-
783
.
Boyd
,
J. L.
,
Skove
,
S. L.
,
Rouanet
,
J. P.
,
Pilaz
,
L. J.
,
Bepler
,
T.
,
Gordan
,
R.
,
Wray
,
G. A.
and
Silver
,
D. L.
(
2015
).
Human-chimpanzee differences in a FZD8 enhancer alter cell-cycle dynamics in the developing neocortex
.
Curr. Biol.
25
,
772
-
779
.
Braun
,
E.
,
Danan-Gotthold
,
M.
,
Borm
,
L. E.
,
Lee
,
K. W.
,
Vinsland
,
E.
,
Lönnerberg
,
P.
,
Hu
,
L.
,
Li
,
X.
,
He
,
X.
,
Andrusivová
,
Ž
et al. 
(
2023
).
Comprehensive cell atlas of the first-trimester developing human brain
.
Science
382
,
eadf1226
.
Britz
,
O.
,
Mattar
,
P.
,
Nguyen
,
L.
,
Langevin
,
L. M.
,
Zimmer
,
C.
,
Alam
,
S.
,
Guillemot
,
F.
and
Schuurmans
,
C.
(
2006
).
A role for proneural genes in the maturation of cortical progenitor cells
.
Cereb. Cortex
16
Suppl. 1
,
i138
-
i151
.
Capra
,
J. A.
,
Erwin
,
G. D.
,
Mckinsey
,
G.
,
Rubenstein
,
J. L.
and
Pollard
,
K. S.
(
2013
).
Many human accelerated regions are developmental enhancers
.
Philos. Trans. R. Soc. Lond. B Biol. Sci.
368
,
20130025
.
Castro
,
D. S.
,
Skowronska-Krawczyk
,
D.
,
Armant
,
O.
,
Donaldson
,
I. J.
,
Parras
,
C.
,
Hunt
,
C.
,
Critchley
,
J. A.
,
Nguyen
,
L.
,
Gossler
,
A.
,
Gottgens
,
B.
et al. 
(
2006
).
Proneural bHLH and Brn proteins coregulate a neurogenic program through cooperative binding to a conserved DNA motif
.
Dev. Cell
11
,
831
-
844
.
Castro
,
D. S.
,
Martynoga
,
B.
,
Parras
,
C.
,
Ramesh
,
V.
,
Pacary
,
E.
,
Johnston
,
C.
,
Drechsel
,
D.
,
Lebel-Potter
,
M.
,
Garcia
,
L. G.
,
Hunt
,
C.
et al. 
(
2011
).
A novel function of the proneural factor Ascl1 in progenitor proliferation identified by genome-wide characterization of its targets
.
Genes Dev.
25
,
930
-
945
.
Dehay
,
C.
and
Huttner
,
W. B.
(
2024
).
Development and evolution of the primate neocortex from a progenitor cell perspective
.
Development
151
,
dev199797
.
Dehay
,
C.
,
Kennedy
,
H.
and
Kosik
,
K. S.
(
2015
).
The outer subventricular zone and primate-specific cortical complexification
.
Neuron
85
,
683
-
694
.
Delepine
,
C.
,
Pham
,
V. A.
,
Tsang
,
H. W. S.
and
Sur
,
M.
(
2021
).
GSK3ss inhibitor CHIR 99021 modulates cerebral organoid development through dose-dependent regulation of apoptosis, proliferation, differentiation and migration
.
PLoS ONE
16
,
e0251173
.
Di Bella
,
D. J.
,
Habibi
,
E.
,
Stickels
,
R. R.
,
Scalia
,
G.
,
Brown
,
J.
,
Yadollahpour
,
P.
,
Yang
,
S. M.
,
Abbate
,
C.
,
Biancalani
,
T.
,
Macosko
,
E. Z.
et al. 
(
2021
).
Molecular logic of cellular diversification in the mouse cerebral cortex
.
Nature
595
,
554
-
559
.
Doan
,
R. N.
,
Bae
,
B. I.
,
Cubelos
,
B.
,
Chang
,
C.
,
Hossain
,
A. A.
,
Al-Saad
,
S.
,
Mukaddes
,
N. M.
,
Oner
,
O.
,
Al-Saffar
,
M.
,
Balkhy
,
S.
et al. 
(
2016
).
Mutations in human accelerated regions disrupt cognition and social behavior
.
Cell
167
,
341
-
354.e312
.
Du
,
H.
,
Wang
,
Z.
,
Guo
,
R.
,
Yang
,
L.
,
Liu
,
G.
,
Zhang
,
Z.
,
Xu
,
Z.
,
Tian
,
Y.
,
Yang
,
Z.
,
Li
,
X.
et al. 
(
2022
).
Transcription factors Bcl11a and Bcl11b are required for the production and differentiation of cortical projection neurons
.
Cereb. Cortex
32
,
3611
-
3632
.
Endo
,
S.
,
Nairn
,
A. C.
,
Greengard
,
P.
and
Ito
,
M.
(
2003
).
Thr123 of rat G-substrate contributes to its action as a protein phosphatase inhibitor
.
Neurosci. Res.
45
,
79
-
89
.
Fernandez
,
V.
and
Borrell
,
V.
(
2023
).
Developmental mechanisms of gyrification
.
Curr. Opin. Neurobiol.
80
,
102711
.
Fietz
,
S. A.
,
Kelava
,
I.
,
Vogt
,
J.
,
Wilsch-Brauninger
,
M.
,
Stenzel
,
D.
,
Fish
,
J. L.
,
Corbeil
,
D.
,
Riehn
,
A.
,
Distler
,
W.
,
Nitsch
,
R.
et al. 
(
2010
).
OSVZ progenitors of human and ferret neocortex are epithelial-like and expand by integrin signaling
.
Nat. Neurosci.
13
,
690
-
699
.
Fietz
,
S. A.
,
Lachmann
,
R.
,
Brandl
,
H.
,
Kircher
,
M.
,
Samusik
,
N.
,
Schroder
,
R.
,
Lakshmanaperumal
,
N.
,
Henry
,
I.
,
Vogt
,
J.
,
Riehn
,
A.
et al. 
(
2012
).
Transcriptomes of germinal zones of human and mouse fetal neocortex suggest a role of extracellular matrix in progenitor self-renewal
.
Proc. Natl. Acad. Sci. USA
109
,
11836
-
11841
.
Florio
,
M.
,
Albert
,
M.
,
Taverna
,
E.
,
Namba
,
T.
,
Brandl
,
H.
,
Lewitus
,
E.
,
Haffner
,
C.
,
Sykes
,
A.
,
Wong
,
F. K.
,
Peters
,
J.
et al. 
(
2015
).
Human-specific gene ARHGAP11B promotes basal progenitor amplification and neocortex expansion
.
Science
347
,
1465
-
1470
.
Fode
,
C.
,
Ma
,
Q.
,
Casarosa
,
S.
,
Ang
,
S. L.
,
Anderson
,
D. J.
and
Guillemot
,
F.
(
2000
).
A role for neural determination genes in specifying the dorsoventral identity of telencephalic neurons
.
Genes Dev.
14
,
67
-
80
.
Ghazale
,
H.
,
Park
,
E.
,
Vasan
,
L.
,
Mester
,
J.
,
Saleh
,
F.
,
Trevisiol
,
A.
,
Zinyk
,
D.
,
Chinchalongporn
,
V.
,
Liu
,
M.
,
Fleming
,
T.
et al. 
(
2022
).
Ascl1 phospho-site mutations enhance neuronal conversion of adult cortical astrocytes in vivo
.
Front. Neurosci.
16
,
917071
.
Girskis
,
K. M.
,
Stergachis
,
A. B.
,
Degennaro
,
E. M.
,
Doan
,
R. N.
,
Qian
,
X.
,
Johnson
,
M. B.
,
Wang
,
P. P.
,
Sejourne
,
G. M.
,
Nagy
,
M. A.
,
Pollina
,
E. A.
et al. 
(
2021
).
Rewiring of human neurodevelopmental gene regulatory programs by human accelerated regions
.
Neuron
109
,
3239
-
3251.e3237
.
Gorkin
,
D. U.
,
Barozzi
,
I.
,
Zhao
,
Y.
,
Zhang
,
Y.
,
Huang
,
H.
,
Lee
,
A. Y.
,
Li
,
B.
,
Chiou
,
J.
,
Wildberg
,
A.
,
Ding
,
B.
et al. 
(
2020
).
An atlas of dynamic chromatin landscapes in mouse fetal development
.
Nature
583
,
744
-
751
.
Guerra
,
P.
,
Vuillemenot
,
L.-A.
,
Rae
,
B.
,
Ladyhina
,
V.
and
Milias-Argeitis
,
A.
(
2022
).
Systematic in vivo characterization of fluorescent protein maturation in budding yeast
.
ACS Synth. Biol.
11
,
1129
-
1141
.
Guven
,
A.
,
Kalebic
,
N.
,
Long
,
K. R.
,
Florio
,
M.
,
Vaid
,
S.
,
Brandl
,
H.
,
Stenzel
,
D.
and
Huttner
,
W. B.
(
2020
).
Extracellular matrix-inducing Sox9 promotes both basal progenitor proliferation and gliogenesis in developing neocortex
.
eLife
9
,
e49808
.
Hall
,
K. U.
,
Collins
,
S. P.
,
Gamm
,
D. M.
,
Massa
,
E.
,
Depaoli-Roach
,
A. A.
and
Uhler
,
M. D.
(
1999
).
Phosphorylation-dependent inhibition of protein phosphatase-1 by G-substrate. A Purkinje cell substrate of the cyclic GMP-dependent protein kinase
.
J. Biol. Chem.
274
,
3485
-
3495
.
Han
,
S.
,
Dennis
,
D. J.
,
Balakrishnan
,
A.
,
Dixit
,
R.
,
Britz
,
O.
,
Zinyk
,
D.
,
Touahri
,
Y.
,
Olender
,
T.
,
Brand
,
M.
,
Guillemot
,
F.
et al. 
(
2018
).
A non-canonical role for the proneural gene Neurog1 as a negative regulator of neocortical neurogenesis
.
Development
145
,
dev157719
.
Han
,
S.
,
Okawa
,
S.
,
Wilkinson
,
G. A.
,
Ghazale
,
H.
,
Adnani
,
L.
,
Dixit
,
R.
,
Tavares
,
L.
,
Faisal
,
I.
,
Brooks
,
M. J.
,
Cortay
,
V.
et al. 
(
2021
).
Proneural genes define ground-state rules to regulate neurogenic patterning and cortical folding
.
Neuron
109
,
2847
-
2863.e2811
.
Hansen
,
D. V.
,
Lui
,
J. H.
,
Parker
,
P. R.
and
Kriegstein
,
A. R.
(
2010
).
Neurogenic radial glia in the outer subventricular zone of human neocortex
.
Nature
464
,
554
-
561
.
Haubensak
,
W.
,
Attardo
,
A.
,
Denk
,
W.
and
Huttner
,
W. B.
(
2004
).
Neurons arise in the basal neuroepithelium of the early mammalian telencephalon: a major site of neurogenesis
.
Proc. Natl. Acad. Sci. USA
101
,
3196
-
3201
.
Hebisch
,
E.
,
Knebel
,
J.
,
Landsberg
,
J.
,
Frey
,
E.
and
Leisner
,
M.
(
2013
).
High variation of fluorescence protein maturation times in closely related Escherichia coli strains
.
PLoS ONE
8
,
e75991
.
Henke
,
R. M.
,
Meredith
,
D. M.
,
Borromeo
,
M. D.
,
Savage
,
T. K.
and
Johnson
,
J. E.
(
2009
).
Ascl1 and Neurog2 form novel complexes and regulate Delta-like3 (Dll3) expression in the neural tube
.
Dev. Biol.
328
,
529
-
540
.
Imayoshi
,
I.
,
Shimogori
,
T.
,
Ohtsuka
,
T.
and
Kageyama
,
R.
(
2008
).
Hes genes and neurogenin regulate non-neural versus neural fate specification in the dorsal telencephalic midline
.
Development
135
,
2531
-
2541
.
Johnson
,
M. B.
,
Wang
,
P. P.
,
Atabay
,
K. D.
,
Murphy
,
E. A.
,
Doan
,
R. N.
,
Hecht
,
J. L.
and
Walsh
,
C. A.
(
2015
).
Single-cell analysis reveals transcriptional heterogeneity of neural progenitors in human cortex
.
Nat. Neurosci.
18
,
637
-
646
.
Kamm
,
G. B.
,
López-Leal
,
R.
,
Lorenzo
,
J. R.
and
Franchini
,
L. F.
(
2013
).
A fast-evolving human NPAS3 enhancer gained reporter expression in the developing forebrain of transgenic mice
.
Philos. Trans. R. Soc. Lond. B Biol. Sci.
368
,
20130019
.
Kovach
,
C.
,
Dixit
,
R.
,
Li
,
S.
,
Mattar
,
P.
,
Wilkinson
,
G.
,
Elsen
,
G. E.
,
Kurrasch
,
D. M.
,
Hevner
,
R. F.
and
Schuurmans
,
C.
(
2013
).
Neurog2 simultaneously activates and represses alternative gene expression programs in the developing neocortex
.
Cereb. Cortex
23
,
1884
-
1900
.
Kowalczyk
,
T.
,
Pontious
,
A.
,
Englund
,
C.
,
Daza
,
R. A.
,
Bedogni
,
F.
,
Hodge
,
R.
,
Attardo
,
A.
,
Bell
,
C.
,
Huttner
,
W. B.
and
Hevner
,
R. F.
(
2009
).
Intermediate neuronal progenitors (basal progenitors) produce pyramidal-projection neurons for all layers of cerebral cortex
.
Cereb. Cortex
19
,
2439
-
2450
.
Lambert
,
N.
,
Lambot
,
M. A.
,
Bilheu
,
A.
,
Albert
,
V.
,
Englert
,
Y.
,
Libert
,
F.
,
Noel
,
J. C.
,
Sotiriou
,
C.
,
Holloway
,
A. K.
,
Pollard
,
K. S.
et al. 
(
2011
).
Genes expressed in specific areas of the human fetal cerebral cortex display distinct patterns of evolution
.
PLoS ONE
6
,
e17753
.
Li
,
S.
,
Mattar
,
P.
,
Zinyk
,
D.
,
Singh
,
K.
,
Chaturvedi
,
C. P.
,
Kovach
,
C.
,
Dixit
,
R.
,
Kurrasch
,
D. M.
,
Ma
,
Y. C.
,
Chan
,
J. A.
et al. 
(
2012
).
GSK3 temporally regulates neurogenin 2 proneural activity in the neocortex
.
J. Neurosci.
32
,
7791
-
7805
.
Lui
,
J. H.
,
Hansen
,
D. V.
and
Kriegstein
,
A. R.
(
2011
).
Development and evolution of the human neocortex
.
Cell
146
,
18
-
36
.
Manelli
,
V.
,
Diwakar
,
J.
,
Beşkardeş
,
S.
,
Alonso-Gil
,
D.
,
Forné
,
I.
,
Chong
,
F.
,
Imhof
,
A.
and
Bonev
,
B.
(
2024
).
Context-dependent epigenome rewiring during neuronal differentiation
.
bioRxiv
,
2024.2010.2018.618996
.
Martinez-Martinez
,
M. A.
,
De Juan Romero
,
C.
,
Fernandez
,
V.
,
Cardenas
,
A.
,
Gotz
,
M.
and
Borrell
,
V.
(
2016
).
A restricted period for formation of outer subventricular zone defined by Cdh1 and Trnp1 levels
.
Nat. Commun.
7
,
11812
.
Mattar
,
P.
,
Langevin
,
L. M.
,
Markham
,
K.
,
Klenin
,
N.
,
Shivji
,
S.
,
Zinyk
,
D.
and
Schuurmans
,
C.
(
2008
).
Basic helix-loop-helix transcription factors cooperate to specify a cortical projection neuron identity
.
Mol. Cell. Biol.
28
,
1456
-
1469
.
Miyata
,
T.
,
Kawaguchi
,
A.
,
Saito
,
K.
,
Kawano
,
M.
,
Muto
,
T.
and
Ogawa
,
M.
(
2004
).
Asymmetric production of surface-dividing and non-surface-dividing cortical progenitor cells
.
Development
131
,
3133
-
3145
.
Moffat
,
A.
and
Schuurmans
,
C.
(
2024
).
The control of cortical folding: multiple mechanisms, multiple models
.
Neuroscientist
30
,
704
-
722
.
Moffat
,
A.
,
Oproescu
,
A.-M.
,
Okawa
,
S.
,
Han
,
S.
,
Vasan
,
L.
,
Ghazale
,
H.
,
Dennis
,
D. J.
,
Zinyk
,
D.
,
Guillemot
,
F.
,
Sol
,
A. D.
et al. 
(
2023
).
Proneural genes form a combinatorial code to diversify neocortical neural progenitor cells
.
bioRxiv
2023.2007.2029.551096
.
Moura
,
M.
and
Conde
,
C.
(
2019
).
Phosphatases in mitosis: roles and regulation
.
Biomolecules
9
,
55
.
Noack
,
F.
,
Vangelisti
,
S.
,
Raffl
,
G.
,
Carido
,
M.
,
Diwakar
,
J.
,
Chong
,
F.
and
Bonev
,
B.
(
2022
).
Multimodal profiling of the transcriptional regulatory landscape of the developing mouse cortex identifies Neurog2 as a key epigenome remodeler
.
Nat. Neurosci.
25
,
154
-
167
.
Noctor
,
S. C.
,
Martinez-Cerdeno
,
V.
,
Ivic
,
L.
and
Kriegstein
,
A. R.
(
2004
).
Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases
.
Nat. Neurosci.
7
,
136
-
144
.
Noctor
,
S. C.
,
Martinez-Cerdeno
,
V.
and
Kriegstein
,
A. R.
(
2008
).
Distinct behaviors of neural stem and progenitor cells underlie cortical neurogenesis
.
J. Comp. Neurol.
508
,
28
-
44
.
Ochiai
,
W.
,
Nakatani
,
S.
,
Takahara
,
T.
,
Kainuma
,
M.
,
Masaoka
,
M.
,
Minobe
,
S.
,
Namihira
,
M.
,
Nakashima
,
K.
,
Sakakibara
,
A.
,
Ogawa
,
M.
et al. 
(
2009
).
Periventricular notch activation and asymmetric Ngn2 and Tbr2 expression in pair-generated neocortical daughter cells
.
Mol. Cell. Neurosci.
40
,
225
-
233
.
Oproescu
,
A. M.
,
Han
,
S.
and
Schuurmans
,
C.
(
2021
).
New insights into the intricacies of proneural gene regulation in the embryonic and adult cerebral cortex
.
Front. Mol. Neurosci.
14
,
642016
.
Park
,
Y. G.
,
Sohn
,
C. H.
,
Chen
,
R.
,
Mccue
,
M.
,
Yun
,
D. H.
,
Drummond
,
G. T.
,
Ku
,
T.
,
Evans
,
N. B.
,
Oak
,
H. C.
,
Trieu
,
W.
et al. 
(
2019
).
Protection of tissue physicochemical properties using polyfunctional crosslinkers
.
Nat. Biotechnol.
37
,
73
-
83
.
Park
,
G.
,
Shin
,
M.
,
Lee
,
W.
,
Hotta
,
A.
,
Kobayashi
,
T.
and
Kosodo
,
Y.
(
2022
).
Direct visualization of the transition status during neural differentiation by dual-fluorescent reporter human pluripotent stem cells
.
Stem Cell Rep.
17
,
1903
-
1913
.
Pereira
,
A.
,
Diwakar
,
J.
,
Masserdotti
,
G.
,
Beşkardeş
,
S.
,
Simon
,
T.
,
So
,
Y.
,
Martín-Loarte
,
L.
,
Bergemann
,
F.
,
Vasan
,
L.
,
Schauer
,
T.
et al. 
(
2024
).
Direct neuronal reprogramming of mouse astrocytes is associated with multiscale epigenome remodeling and requires Yy1
.
Nat. Neurosci.
27
,
1260
-
1273
.
Péron
,
S.
,
Miyakoshi
,
L. M.
,
Brill
,
M. S.
,
Manzano-Franco
,
D.
,
Serrano-López
,
J.
,
Fan
,
W.
,
Marichal
,
N.
,
Ghanem
,
A.
,
Conzelmann
,
K. K.
,
Karow
,
M.
et al. 
(
2023
).
Programming of neural progenitors of the adult subependymal zone towards a glutamatergic neuron lineage by neurogenin 2
.
Stem Cell Rep.
18
,
2418
-
2433
.
Pollard
,
K. S.
,
Salama
,
S. R.
,
Lambert
,
N.
,
Lambot
,
M.-A.
,
Coppens
,
S.
,
Pedersen
,
J. S.
,
Katzman
,
S.
,
King
,
B.
,
Onodera
,
C.
,
Siepel
,
A.
et al. 
(
2006
).
An RNA gene expressed during cortical development evolved rapidly in humans
.
Nature
443
,
167
-
172
.
Pollen
,
A. A.
,
Nowakowski
,
T. J.
,
Chen
,
J.
,
Retallack
,
H.
,
Sandoval-Espinosa
,
C.
,
Nicholas
,
C. R.
,
Shuga
,
J.
,
Liu
,
S. J.
,
Oldham
,
M. C.
,
Diaz
,
A.
et al. 
(
2015
).
Molecular identity of human outer radial glia during cortical development
.
Cell
163
,
55
-
67
.
Prabhakar
,
S.
,
Noonan
,
J. P.
,
Pääbo
,
S.
and
Rubin
,
E. M.
(
2006
).
Accelerated evolution of conserved noncoding sequences in humans
.
Science
314
,
786
.
Qian
,
X.
,
Nguyen
,
H. N.
,
Song
,
M. M.
,
Hadiono
,
C.
,
Ogden
,
S. C.
,
Hammack
,
C.
,
Yao
,
B.
,
Hamersky
,
G. R.
,
Jacob
,
F.
,
Zhong
,
C.
et al. 
(
2016
).
Brain-region-specific organoids using mini-bioreactors for modeling ZIKV exposure
.
Cell
165
,
1238
-
1254
.
Qian
,
X.
,
Jacob
,
F.
,
Song
,
M. M.
,
Nguyen
,
H. N.
,
Song
,
H.
and
Ming
,
G. L.
(
2018
).
Generation of human brain region-specific organoids using a miniaturized spinning bioreactor
.
Nat. Protoc.
13
,
565
-
580
.
Qing
,
Z.
,
Yuan
,
W.
,
Wang
,
J.
,
Song
,
W.
,
Luo
,
J.
,
Wu
,
X.
,
Lu
,
Q.
,
Li
,
Y.
and
Zeng
,
M.
(
2022
).
Verapamil inhibited the development of ureteral stricture by blocking CaMK II-mediated STAT3 and Smad3/JunD pathways
.
Int. Urol. Nephrol.
54
,
2855
-
2866
.
Reillo
,
I.
and
Borrell
,
V.
(
2012
).
Germinal zones in the developing cerebral cortex of ferret: ontogeny, cell cycle kinetics, and diversity of progenitors
.
Cereb. Cortex
22
,
2039
-
2054
.
Reillo
,
I.
,
De Juan Romero
,
C.
,
Garcia-Cabezas
,
M. A.
and
Borrell
,
V.
(
2011
).
A role for intermediate radial glia in the tangential expansion of the mammalian cerebral cortex
.
Cereb. Cortex
21
,
1674
-
1694
.
Reilly
,
S. K.
,
Yin
,
J.
,
Ayoub
,
A. E.
,
Emera
,
D.
,
Leng
,
J.
,
Cotney
,
J.
,
Sarro
,
R.
,
Rakic
,
P.
and
Noonan
,
J. P.
(
2015
).
Evolutionary changes in promoter and enhancer activity during human corticogenesis
.
Science
347
,
1155
-
1159
.
Rhodes
,
C. T.
,
Thompson
,
J. J.
,
Mitra
,
A.
,
Asokumar
,
D.
,
Lee
,
D. R.
,
Lee
,
D. J.
,
Zhang
,
Y.
,
Jason
,
E.
,
Dale
,
R. K.
,
Rocha
,
P. P.
et al. 
(
2022
).
An epigenome atlas of neural progenitors within the embryonic mouse forebrain
.
Nat. Commun.
13
,
4196
.
Schuurmans
,
C.
,
Armant
,
O.
,
Nieto
,
M.
,
Stenman
,
J. M.
,
Britz
,
O.
,
Klenin
,
N.
,
Brown
,
C.
,
Langevin
,
L. M.
,
Seibt
,
J.
,
Tang
,
H.
et al. 
(
2004
).
Sequential phases of cortical specification involve Neurogenin-dependent and -independent pathways
.
EMBO J.
23
,
2892
-
2902
.
Sivitilli
,
A. A.
,
Gosio
,
J. T.
,
Ghoshal
,
B.
,
Evstratova
,
A.
,
Trcka
,
D.
,
Ghiasi
,
P.
,
Hernandez
,
J. J.
,
Beaulieu
,
J. M.
,
Wrana
,
J. L.
and
Attisano
,
L.
(
2020
).
Robust production of uniform human cerebral organoids from pluripotent stem cells
.
Life Sci. Alliance
3
,
e202000707
.
Smith
,
D. K.
,
Yang
,
J.
,
Liu
,
M. L.
and
Zhang
,
C. L.
(
2016
).
Small molecules modulate chromatin accessibility to promote NEUROG2-mediated fibroblast-to-neuron reprogramming
.
Stem Cell Rep.
7
,
955
-
969
.
Song
,
M.
,
Pebworth
,
M. P.
,
Yang
,
X.
,
Abnousi
,
A.
,
Fan
,
C.
,
Wen
,
J.
,
Rosen
,
J. D.
,
Choudhary
,
M. N. K.
,
Cui
,
X.
,
Jones
,
I. R.
et al. 
(
2020
).
Cell-type-specific 3D epigenomes in the developing human cortex
.
Nature
587
,
644
-
649
.
Taverna
,
E.
,
Gotz
,
M.
and
Huttner
,
W. B.
(
2014
).
The cell biology of neurogenesis: toward an understanding of the development and evolution of the neocortex
.
Annu. Rev. Cell Dev. Biol.
30
,
465
-
502
.
Uzquiano
,
A.
,
Kedaigle
,
A. J.
,
Pigoni
,
M.
,
Paulsen
,
B.
,
Adiconis
,
X.
,
Kim
,
K.
,
Faits
,
T.
,
Nagaraja
,
S.
,
Antón-Bolaños
,
N.
,
Gerhardinger
,
C.
et al. 
(
2022
).
Proper acquisition of cell class identity in organoids allows definition of fate specification programs of the human cerebral cortex
.
Cell
185
,
3770
-
3788.e3727
.
Vanderhaeghen
,
P.
and
Polleux
,
F.
(
2023
).
Developmental mechanisms underlying the evolution of human cortical circuits
.
Nat. Rev. Neurosci.
24
,
213
-
232
.
Wang
,
W.
,
Yin
,
C.
,
Wen
,
S.
,
Liu
,
Z.
,
Wang
,
B.
,
Zeng
,
B.
,
Sun
,
L.
,
Zhou
,
X.
,
Zhong
,
S.
,
Zhang
,
J.
et al. 
(
2024
).
DCX knockout ferret reveals a neurogenic mechanism in cortical development
.
Cell Rep.
43
,
114508
.
Wei
,
Y.
,
De Lange
,
S. C.
,
Scholtens
,
L. H.
,
Watanabe
,
K.
,
Ardesch
,
D. J.
,
Jansen
,
P. R.
,
Savage
,
J. E.
,
Li
,
L.
,
Preuss
,
T. M.
,
Rilling
,
J. K.
et al. 
(
2019
).
Genetic mapping and evolutionary analysis of human-expanded cognitive networks
.
Nat. Commun.
10
,
4839
.
Won
,
H.
,
Huang
,
J.
,
Opland
,
C. K.
,
Hartl
,
C. L.
and
Geschwind
,
D. H.
(
2019
).
Human evolved regulatory elements modulate genes involved in cortical expansion and neurodevelopmental disease susceptibility
.
Nat. Commun.
10
,
2396
.
Yang
,
G.
,
Smibert
,
C. A.
,
Kaplan
,
D. R.
and
Miller
,
F. D.
(
2014
).
An eIF4E1/4E-T complex determines the genesis of neurons from precursors by translationally repressing a proneurogenic transcription program
.
Neuron
84
,
723
-
739
.
Zhong
,
S.
,
Zhang
,
S.
,
Fan
,
X.
,
Wu
,
Q.
,
Yan
,
L.
,
Dong
,
J.
,
Zhang
,
H.
,
Li
,
L.
,
Sun
,
L.
,
Pan
,
N.
et al. 
(
2018
).
A single-cell RNA-seq survey of the developmental landscape of the human prefrontal cortex
.
Nature
555
,
524
-
528
.
Ziffra
,
R. S.
,
Kim
,
C. N.
,
Ross
,
J. M.
,
Wilfert
,
A.
,
Turner
,
T. N.
,
Haeussler
,
M.
,
Casella
,
A. M.
,
Przytycki
,
P. F.
,
Keough
,
K. C.
,
Shin
,
D.
et al. 
(
2021
).
Single-cell epigenomics reveals mechanisms of human cortical development
.
Nature
598
,
205
-
213
.

Competing interests

The authors declare no competing or financial interests.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.