RNA binding proteins (RNABP) represent critical controllers of development (Parra and Johnston, 2022). The RNABP Quaking 5 (Qki5) has been found to decrease in neuronal progenitors and shown to regulate the transcription of cell adhesion molecule genes via binding to the intronic RNA (Hayakawa-Yano et al., 2017). A transcriptomic analysis of the Qki5-silenced E14.5 mouse cortex revealed increased cell differentiation, with preferential neuronal rather than glial cell generation. Qki5 and Qki6 are specifically expressed at the stemness state and decrease upon neural commitment. The authors elucidated the Qki5 RNAPB role, showing that it interferes with cell-cell interactions rather than stemness maintenance of the neural cells. Indeed, Gene Ontology (GO) analysis revealed a massive de-regulation in several biological pathways upon Qki5 silencing. Upregulation of cell adhesion, actin filament, cell projection, and microtubule suggest an important role of this RNABP in the progenitor to neuron transition (Hayakawa-Yano and Yano, 2019).

RNABPs are also responsible for translational regulation via binding untranslated regions (UTR; 5’ and 3’) during neurodevelopment (DeBoer et al., 2013; Zhang et al., 2016). Indeed, Celf1 RNAPB, an upstream regulator of 5’ UTR-driven translation, regulates the translation of the Elavl4 RNABP acting on radial progenitors to produce a balanced neocortical region (Popovitchenko et al., 2020).

Preservation of the adult SVZ-NSCs has been shown to be regulated by the RNABP Ars2, a conserved protein that acts as a transcriptional regulator of the Sox2 TF, involved in stem cell maintenance (Andreu-Agullo et al., 2011). Ars2 has been evaluated longitudinally during embryonic to adult hippocampal differentiation by Chip-seq, indicating time-dependent distinct Ars2 activities (Yu et al., 2020). Ars2 induces expression of the Sox2 TF and acts in the permissive chromatin conformation. The other enriched peaks include developmental genes along with oligodendroglial genes expressed only during the phase that precedes differentiation. Thus, Ars2 is a critical DNA regulator involved in progeny differentiation.

Environmental stimuli are no less important transcriptional regulators than TFs in shaping the NSC niche toward proliferation and fate commitment (Chiu and Dawes, 2012). For example, the maternal environment influences the proliferation and differentiation proprieties of NSCs during the embryonic stage (Wang et al., 2021). Bulk RNA-seq has shown that high levels of maternal estradiol decrease the proliferation rate of hypothalamic NSCs, with the involvement of long non-coding RNAs (lncRNAs). Single-cell transcriptomic analyses will help to clarify which cell type/s influence the rate of neuronal generation.

Notably, an RNA-seq study indicates that signaling via reactive oxygen species (ROS) regulates the fate of DG NSCs (Adusumilli et al., 2021). Neurons can adopt specific defense mechanisms against oxidative stress, however ROS levels in the DG seem to be crucial for quiescent maintenance, compared to the SVZ. In fact, DG-NSCs retain a higher level of ROS compared to SVZ stem cells, which contribute to the expression of less committed neuronal subsets, while genes involved in cell cycle inhibition are enriched. This result suggests that critical ROS levels shape proliferation in DG early-stage quiescent NSCs. On the contrary, low ROS levels observed in the SVZ niche are likely to drive cells toward cell cycle and differentiation (Adusumilli et al., 2021).

ROS levels are also associated with accelerated neuroepithelial cell differentiation (Özsoy et al., 2021). Intriguingly, a novel model of enhanced neurodevelopment has been proposed, which takes place during neurulation, involving specialized “cannibalized” neuroepithelial cells that are able to “eat” erythroblasts via endocytosis from the yolk sac, eventually acquiring the hemoglobin precursor (heme) molecule. The proposed mechanism leads to accelerated neuroepithelial cell development in a spatially restricted fashion, due to the activation of mitochondria that supply massive production of ROS by the electron transport chain. The specific heme-enriched NSCs showed a transcriptome enriched in the expression of the Notch1 repressor Numb that boosts β-catenin pro-differentiation activity, accompanied by the stabilization of adherent junction (Hirabayashi et al., 2004). Altogether, the proposed model needs further validation to demonstrate the strength of the suggested energy supply to support the overall machinery, and to test this hypothesis under hypoxic or anemic conditions.

Overall, these findings emphasize that ROS levels present in a critical time and space regulate brain development.

Stem cell niche-dependent cues affect the behavior of NSCs (Silva-Vargas et al., 2016). The NSCs in the V-SVZ, which is close to the lateral ventricle choroid plexus (LVCP), the main source of CSF, are indeed influenced by the LVCP secretome, as confirmed by proteomic analysis with antibody arrays (Silva-Vargas et al., 2016). Transcriptome analysis of the whole LVCP confirmed that it plays a central role in integrating local and systemic cues to dynamically modulate the V-SVZ stem cell niche.

Time-dependent specification and cortical layers distribution of newborn neurons derived from RG require strict orchestration of events (Gao et al., 2014). Single-cell transcriptomic analysis, enriched by direct measurement of specific cell-type transcripts in defined CNS areas, compared to bulk analysis, has brought insights into the understanding of the heterogeneity of the NSC niche during development, thus revealing the complex mechanisms required during neurogenesis to regulate neuronal specification. Notably, a large single-cell transcriptomic work has developed a time-related atlas describing the heterogeneity and specificity of NSCs starting from apical and basal embryonic stem cells towards the neuronal lineage (Yuzwa et al., 2017). By analyzing E11.5, E13.5, E15.5, and E17.5 mouse cortex, the authors describe the critical genes involved in the developmental process by trajectory analysis. Increased percentages of neuronal genes and decreased proliferative capacity during time have been reported. The authors also describe how RG progenitors enter a non-proliferative state during adult neurogenesis, recapitulating a transcriptional program similar to qNSCs. Levels of transcripts in this critical period do not always match protein levels, specifically between E13.5 (deep and superficial layer), E15.5 (superficial layers), and E17.5 (neuronal specification finished). RG progenitors exhibit high levels of several transcripts of both deep and superficial neuronal subtype specification from E13.5 and E15.5, while some of those transcripts are still detectable at later time points, indicating RG priming toward the arrangement of both layers. However, the specific repressor effector Pum2/4E-T determines which of those subtypes will be generated at each time point of murine development. In the absence of Pum2/4E-T, aberrant co-expression of deep layer transcripts in the superficial layer occurs, miss-regulating the whole critical machinery and confirming that RG is primed toward the generation of diverse neuronal types (Yang et al., 2014; Zahr et al., 2018).

As mentioned in the introduction, tight regulation characterizes neuronal and glial development, involving many players such as TFs, transcriptional coactivators/repressors, and gene expression in a time-dependent manner along with environmental cues (Ma, 2006). The transcriptional activators YAP and TAZ play a prominent role in regulating the cell proliferation rate during embryogenesis (Lavado et al., 2021). In the absence of YAP, impaired astrocyte proliferation was reported (Park et al., 2016). Double YAP and TAZ KO however demonstrates the necessary role of these two co-transcriptional activators in maintaining RG proliferative capacity, structural organization, and ependymal cell generation (Lavado et al., 2021).

Bone morphogenetic proteins (BMPs) are crucial during neurogenesis and brain development (Bond et al., 2012). Hence, their dynamic changes during differentiation direct neurogenesis as well as gliogenesis are responsible for the hippocampal stem cell niche maintenance. Combined RNA-Chip-ATAC-seq analysis revealed that BMP2 influence neuronal differentiation early (E11) in mouse development, while at a later time (E14) it impacts astrogenesis. BMP-induced differentiation involves the Smads TFs that interact with Sox11 to drive neuronal differentiation, and switch to interact with Sox8 to promote astrogenesis (Katada et al., 2021).

Among BMPs known antagonists, Gremlin1 (Grem1) has been studied to elucidate its role in cortical neurogenesis (Ichinose et al., 2021). Indeed, Grem1 can induce excitatory deep cortical layer neurons, inducing upregulation of a set of genes involved in glutamate secretion. Moreover, Grem1 is specifically expressed by neuronal committed immature precursor cells, thus, being a good candidate for specific neuronal differentiation program enhancement (Ichinose et al., 2021).

Glutamatergic neuron generation is driven also by the master transcriptional regulator Sox1, shown to be sufficient to induce neuronal generation and glial fate repression. Specifically, Sox1 overexpression induces upregulation of neural stem cell genes and growth factors leading toward glutamatergic progeny (Baumann et al., 2019).

The generation of newborn cortical neurons depends on the activity of the COMPASS family histone methyltransferase co-factor Ashl2 (Li et al., 2019). Indeed, by RNA-Chip-seq mapping of NSCs in absence of the histone-modifier Ashl2, NSCs exhibit abnormal progeny generation, indicating a crucial role in NSC number maintenance as well as NSC state. Ashl2 affects also the WNT signaling pathways, as upon Ashl2 conditional KO, WNT activators were reduced. Data confirmed by Chip-seq show that Ashl2 binds promoter regions of those activators. Notably, Ashl2-related genes were found enriched in schizophrenia, epilepsy, and autism spectrum disorders.

NSCs and neural progenitors coexist in the embryonic as well as neonatal and adult stages, hence unveiling the mechanism of differentiation since early stages might bring an enormous insight into the global characterization of neurogenesis (Codega et al., 2014). A study assessing global gene expression by RNA-seq has shown the role of the chromatin remodelers Chd5 at an early stage of development (Hwang et al., 2018). Chd5, expressed in the post-mitotic neurons, in the neural progenitors of E14.5 rat cortical region, and hippocampal progenitor cells, plays a critical role during neurogenesis (Egan et al., 2013; Vestin and Mills, 2013; Nitarska et al., 2016). Upon Chd5 depletion, NSCs differentiation was directed toward the astrocytic rather than the neuronal lineage. Boosting of astrocyte differentiation was due to translation disruption of Mash1, a main pro-neural TF that is essential during early differentiation (Hwang et al., 2018). These results suggest a strong Chd5 involvement in the cell fate decision in both cortical and hippocampal NSC niches at an early stage of differentiation.

The ganglionic eminence (GE) has been investigated to unveil positional diversification of NSCs in the lateral (LGE), medial (MGE), and caudal (CGE) ventricular regions. GE is the area where forebrain GABAergic neurons differentiate during the initial embryonic cell fate commitment (Lee et al., 2022). Many investigations focused on late/adult corticogenesis but there is a lack of knowledge regarding the heterogeneity and the transcriptional profile of embryonic forebrain including LGE, MGE, and CGE. In a scRNA-seq study, the analysis performed on E12.5 mice revealed distinct gene expression profiles in the cortex vs. ventricular regions and a great overlap between LGE and CGE. A GE atlas reports several clusters belonging to progenitor and post-mitotic cells in which Nkx2-1, Lhx6, and Sst genes were associated with MGE, while Ebf1 and Mapt were found in the LGE and CGE, which identify GABAergic projection neurons. A heterogeneous cluster expressing Npy and Sst transcripts was present in the LGE and CGE indicating migratory interneurons deriving from the MGE. The Igfbp5 gene transcript was observed to be exclusively expressed by CGE at E12.5 and could be used to direct future high-throughput studies of this specific population, unveiling the specific transcriptional profile of GE (Lee et al., 2022).

The differences between embryonic/adult GE and cortical NSCs have been analyzed based on the hypothesis that the two populations coexist and might share the same transcriptional profile (Borrett et al., 2020). The analysis covering the embryonic and the adult stage (E14; E17; P2; P6; P20; P34; P61) identified very few differences between the two populations, mainly associated with neuronal lineage specification and positioning. Overall, the environment shapes the activation, proliferation, and fate determination of future neurons also after birth. Some of the involved environmental factors have been identified, still, the switch program from excitatory toward inhibitory progeny directed by the environmental niche has not been fully elucidated.

Analysis of individual transcripts using RiboTag mouse models allows identifying the mRNA transcripts that are being actively translated in specific cell types. It has been shown that the protein product and the corresponding gene expression do not always correlate when the NSCs differentiate into neurons in vivo (Baser et al., 2019). Adult neural stem cells translate abundant transcripts with little discrimination, and post-transcriptional regulatory mechanisms mark the onset of differentiation (Baser et al., 2019). As an example, translational repression of mRNAs that encode the stem cell identity factors Sox2, Pax6, and components of the translation machinery, is a necessary condition to allow the differentiation of the neurogenic lineage.

In the last years, lncRNAs, specific transcripts that do not encode proteins, have been shown to regulate important biological functions, including neurogenesis (Briggs et al., 2015; Duran et al., 2019). The lncRNA Pnky has been described to be crucial during corticogenesis by RNA-seq (Andersen et al., 2019). The Pnky gene is localized near the pro-neural TF Pou3f2, and although its deletion does not affect Pou3f2 expression, it leads to loss of proliferation and decreased cortical neurogenesis. Another lncRNA named RUS, located upstream of the transmembrane protein Slitrk3 is involved in neurite outgrowth by regulating in cis this nearest upstream gene (Aruga et al., 2003; Schneider et al., 2022). However, upon RUS silencing, not only Slitrk3 expression was found downregulated, but also Notch- and Shh-related genes were affected. Indeed, activation of Notch signaling leads to qNSC maintenance while inactivation of the Shh pathway results in lower NSC proliferation (Engler et al., 2018; Mase et al., 2021). Such results are in line with transcriptomic data that, upon RUS depletion, show reduced expression of proliferative markers as well as neuronal differentiation genes, while pro-apoptotic and cell death genes were found to increase. Altogether, these studies highlight that lncRNAs are crucial regulators of brain cell development (Schneider et al., 2022).

LncRNAs have been also assessed by applying a novel technique for chromatin conformation capture, named NG Capture-C, which allows to simultaneously reconstruct gene networks in multiple cell types, and to reveal how chromatin conformations shape gene activity (Davies et al., 2016, 2017). This technique has been specifically applied to the lncRNAs that are co-expressed with their near protein-coding genes, specifically the Paupar lncRNA-Pax6 gene pair (Pavlaki et al., 2022). Data revealed complex cis-regulatory interactions in both Paupar and Pax6 promoter regions related to their paired co-expression in the brain.

MicroRNAs (miRNAs) are also relevant for basal progenitor amplification and neurogenesis (Cremisi, 2013). Two miRNAs are of particular interest: miR137, involved in the enhancement of ECM molecule Cd63 along with downregulation of pro-neural gene Myt1l, thus leading to proliferation maintenance, and miR-122, which acts at a post-mitotic level in late neurogenesis reducing neuronal maturation. These two effectors are crucial for cortical layer I-II identity acquisition (Tomasello et al., 2022).

Besides miRNAs and lncRNAs, also circular RNAs have emerged to play a role during neuronal differentiation and brain development (Piwecka et al., 2017). Circular RNAs include transcripts that are covalently linked together 3’- 5’ in a circular manner thus making them more resistant to exonuclease actions compared to linear RNAs (Jeck et al., 2013). A study performed in proliferative NSCs during differentiation and in newborn neurons revealed several up- and down-regulated circular transcripts that were associated with synaptogenesis/neuronal differentiation and cell cycle respectively (Dori et al., 2019). The same investigators in 2020 published a complete catalog of miRNA transcripts expression during corticogenesis in several neural cell types (Dori et al., 2020). Lateral embryonic cortices were analyzed in a double reporter mouse for characterization of proliferative, differentiating, and newly formed neurons revealing 163 putative novel miRNAs sequences, actively involved in post-transcriptional modifications during brain development.

A study based on RNA-seq analysis of postnatal neurogenesis investigated the regional specification of neuronal commitment and stem cell control systems (Tiveron et al., 2017). In the dorsal SVZ, Pax6, Tbr1/2, Emx1/2, and Ngn1/2 genes are key actors during corticogenesis, inducing predominantly GABAergic and dopaminergic neurons, while GABAergic neurons are induced in the lateral SVZ. The analysis of postnatal OB time-dependent neurogenesis unveiled the role of the zinc finger TFs Zic1 and Zic2 in the dorsal neuronal subtype commitment, which induces GABAergic interneurons differentiation by suppressing dopaminergic fate commitment. Zic1 and Zic2 were found evolutionary conserved, as also C. elegans represses the dopaminergic commitment by the same mechanism (Tiveron et al., 2017). Pax6, a gene involved in fate commitment, has also been shown to induce interneurons and dopaminergic phenotype exclusively in the presence of Meis2 TF during the pre-differentiation stage (Hau et al., 2021).

The heterogeneity of the NSCs is well known, however, whether the stem cell progenitors in the adult pool at different anatomical locations are transcriptionally similar remains under investigation, and different results lead to discordant hypotheses. Indeed, spatial localization drives neuronal specification during the early developmental stage (Delgado and Lim, 2015; Delgado et al., 2020). B cells have been profiled by scRNA-seq techniques identifying specific mRNA transcripts of the dorsal and ventral spatial organization (Cebrian-Silla et al., 2021). Urah, the gene for urate hydrolase, and Dio2 encoding the iodothyronine deiodinase define the dorsal domain, linking them to glial lineage specification, while Crym, encoding thiomorpholine-carboxylate dehydrogenase, defines the ventral territory, pushing different neuronal differentiation overlapping with the transcription factor Gsx2 found in dorsal clusters. The interesting point is the fact that all these genes are linked to the thyroid hormone signaling pathway, as other thyroid hormone-related genes were found upregulated (Rudqvist et al., 2012; Cebrian-Silla et al., 2021; Luongo et al., 2021). Altogether, these results indicate a specific molecular pathway linked to the dorsal SVZ region.

A multi-omics study investigated the generation of dopaminergic (DA) medium spiny neurons expressing DA receptors 1 and 2 (DRD1, DRD2) by the combination of bulk RNA and Chip-seq (Xu et al., 2018). In the LGE, Sp8, and Sp9 TFs are crucial for the differentiation of the DA progeny. Sp9 binds the Six3 promoter and other Six3-related enhancers, inducing DRD2 expression by directly regulating the Six3 gene, which controls cell proliferation and differentiation. Sp9 has been also associated with MGE-controlling cortical interneurons, as a 50% reduction of this progeny has been found in the absence of Sp9 (Liu et al., 2019).

An important aspect of neurogenesis is the NSC positional identity and maintenance of regional differences along the dorsoventral axis (Merkle et al., 2007). Ventral NSCs express Nkx2-1 that under Shh control induces ventral cell progeny (Delgado and Lim, 2015). Recently mixed-lineage leukemia 1 (Mll1) gene was also found to be part of the preservation of the Nkx2-1 positive cell subset (Delgado et al., 2020). Indeed, mRNA expression and chromatin analysis revealed that transient MIl1 inhibition induces NSC progeny shift to anterodorsal specification.

Single-cell transcriptomic analysis of the OB region, including cortical and V-SVZ NSCs, has allowed genetic fate mapping of neuronal progenitors toward mature OB neurons (Mizrak et al., 2020). A thorough representation of the OB newborn neuronal landscape revealed an important network of cellular crosstalk among NSCs, which is critical to maintaining homeostasis of the progeny and preventing tissue damage. In the adult V-SVZ stem cell niche that is constantly exposed to the WNT ligands from the CSF, the intermediate stem cell express Notum, a negative regulator of WNT, orchestrates normal neurogenesis by repressing the proliferative state of neighboring NSCs, thus maintaining a regular rate of proliferation and differentiation (Lehtinen et al., 2011; Mizrak et al., 2020).

Most recently, it has been shown that in the postnatal V-SVZ, two different NSCs pools co-exist (Baur et al., 2022). Specific adult basal NSCs were found, which seem to mainly contribute to the OB interneurons generation, rather than the already known apical progenitors. The two populations were transcriptionally profiled indicating that different microenvironments finely regulate the apical and basal NSCs post-natally, depending on the Notch signaling. Indeed, Notch signaling effectors Tlx/Nr2e1, along with Delta1 (Dll1) and Jagged1 (Jag1) ligands, and Hes1/Hes5 effectors were differentially expressed in the apical and basal quiescent, activated, and transitioning progenitors. Apical qNSC revealed lower expression of Notch-related genes, while basal qNSC was enriched with Jag1 expression. Basal aNSCs showed high expression of both ligands, while Jag1 was the only up-regulated in activated apical NSCs. Finally, in intermediate progenitors, Hes5 was higher in basal NSCs. Altogether, this data describes the heterogeneity of the postnatal SVZ niche showing a Notch signaling-dependent specific cell distribution (Baur et al., 2022). Notch signaling pathways inducing neurogenesis are also shaped by the expression of Foxp1 (Braccioli et al., 2017). Upon Foxp1 gene silencing, several up- and down-regulated genes involved in the Notch signaling pathway and neurogenesis, respectively, were detected. Indeed, Foxp1 absence induces cells toward oligodendrocyte (Ol) lineage specification, as higher myelin-related Mbp gene expression was found. Moreover, Foxp1 absence enriches the expression of Notch ligand Jag1, lowering neural differentiation (Braccioli et al., 2017).

Variants of the transcription factor Foxp1 have been associated with neurodevelopmental disorders by multi-omics analysis including RNA-seq and Chip-seq (Braccioli et al., 2017). Foxp1 has also been linked to differentiation and neurogenesis (Li et al., 2015). In fact, Chip-seq analysis revealed an association of the Foxp1 gene to Sox3 and Nr2f1/Tlx transcription factors suggesting possible interactions during neurogenesis. Moreover, Foxp1 acts as a Notch modulator by binding the Jag1 promoter, thus inducing Jag1 repression with a subsequent increase in differentiation. RNAseq data confirm the enrichment in neurogenesis, synaptic regulation, and Notch signaling pathways associated with Foxp1 (Braccioli et al., 2017).

Among neurodevelopmental disorders, intellectual disability, and autism spectrum disorders are the most common (Mefford et al., 2012). SET Domain Containing 5 (SETD5), a gene whose mutation has been linked to those disorders, has been functionally characterized in neural development (Sessa et al., 2019). SETD5 is a chromatin accessibility regulator that has been shown to promote transcription elongation and exhibit H3K36 lysine methyltransferase activity during neuronal development (Deliu et al., 2018).

A recent article that performed an RNA-seq analysis of the mouse SVZ demonstrated that NSCs express the insulin growth factor binding protein L1 (IGFBPL1) that can be used as a marker of adult NSCs, specifically of neuroblasts. This protein is involved in the insulin-like growth factor-1 (IGF1) signaling pathway, which regulates the proliferation, survival, and axonal growth of the striatal neurons. Indeed, mice with specific ablation of IGFBPL1-positive NSCs show morphological and functional alterations of medium spiny neurons, and cognitive impairment (Butti et al., 2022).

It is important to consider that, in the post-natal V-SVZ niche, the ependymal cells co-exist with qNSCs, RG progenitors, and NBs. Ependymal cells are ciliated structures that interact with ventricle fluid forming the vertical system, and may represent a dormant stem cell niche of endogenous NSCs (Quaresima et al., 2022). Transcriptional profiling from transgenic mice specifically targeting ependymal cells revealed a unique gene expression pattern of those cells despite they share some of the classical NSC transcriptional patterns (Shah et al., 2018). Ependymal cells display a specific set of genes related to cilia, which makes them a unique cell cluster separated from the stem cell pool, clearly identifiable by the single cell transcriptome analysis. Yet, further studies would support this notion and clarify the role of common NSCs genes in the ependymal cluster. Furthermore, clia-related genes were confirmed in a GO analysis of different published datasets at several developmental stages (MacDonald et al., 2021). This study established a solid role and evolutionarily conserved mechanism of ependymal cells in metal ion-related function, setting the basis for further investigation of the ependymal cell role in perturbed environments.

Adult neural stem cell niches, defined based on their neurogenic potential (subependymal and medial-ependymal zone- S/MEZ and OB) and non-neurogenic potential (cerebral cortex), have been profiled by proteomic analysis (Kjell et al., 2020). A first insight links SEZ and MEZ to similar proteome profiles as they well separate from cortical and OB specimens in the principal component analysis (PCA). However, many proteins (4,786) were found differentially abundant in the four regions analyzed. Specifically, SEZ was enriched for oxidative phosphorylation and gene regulatory proteins, while OB was enriched for proteoglycans key for cell migration during neurogenesis. The cortical region was, on the other hand, enriched for synapse-associated proteins. MEZ and SEZ differences reside in proteins related to neurogenesis, which were higher in SEZ, specifically Tenascin-c (Tnc), an extracellular matrix protein. Indeed, matrix-associated proteins (matrisome) provide crucial structural support to sustain signaling pathways (Naba et al., 2012). Moreover, tissue stiffness, another critical parameter in neurogenesis and neurite outgrowth, is more pronounced in the SEZ compared to the MEZ niche, suggesting a high abundance of basement membrane structures and associated ECM proteins in this specific region.

Finally, the mechanisms of long-term self-renewal and the potential for differentiation of NSCs in the adult brain are beginning to unveil thanks to multi-omics studies. A central role of topoisomerase IIa (Top2a) has been recently described by integrating RNA and Chip-seq of mouse adult brain (Qin et al., 2022). Top2a acts on the cell cycle regulation by increasing NSCs proliferation. The whole machinery includes direct binding of Top2a to the ubiquitin protein Usp37, which is shown to be its direct genomic target.

Besides gene expression, chromatin accessibility, and TFs activity, adult NSC maintenance and division are regulated also by mitochondrial dynamics and metabolic plasticity (Xie et al., 2016; Iwata et al., 2020). Combined in vitro NSC metabolomics and proteomics profiling reveals an important role of the protease Ymel1 in controlling mitochondrial NSC metabolism and proliferative rate (Wani et al., 2022). Indeed, premature neuronal differentiation accompanied by decreased proliferation occurred in Ymel1 absence, leading to dysregulated mitochondrial proteome dynamics.

During brain development, non-neural cells—differentiated astrocytes and oligodendrocytes—are generated and play a crucial role during adult life in maintaining CNS homeostasis (Arai and Lo, 2017).

In the course of cortical neurogenesis, intermediate progenitor cells enriched in the expression of the Gsx2 TF were identified by scRNA-seq, which could not only generate OB interneurons but also promote glial cell differentiation (Zhang et al., 2020). The appearance of Gsx2-expressing progenitors with multipotent features indicates that gliogenesis occurs before the end of neurogenesis, as shown by the high expression of Ols markers in the Gsx2 positive cell subset upon induction of the Shh signaling cascade, which upregulated expression of the zinc finger Gli1 and the Shh receptor Ptch1, along with expression of markers of OB interneurons and cortical Ols lineage.

Gli3 and Shh orchestrate the switch between differentiation of neuronal and glial progenitors: when Shh is upregulated, Gli3 expression decreases in a specific population, driving the differentiation of Gsx2 positive intermediate progenitors toward the cortico-oligodendroglial lineage (Zhang et al., 2020). The presence of Olig1-Gsx2-expressing NSCs was also confirmed in another study showing that these cells represent the majority of NSCs in the postnatal brain, and indicating that different newborn neurons across various brain regions derive from heterogeneous NSCs populations residing in different SVZ sub-regions (Del Águila et al., 2022). These findings should be complemented by spatial single-cell transcriptomics approaches to complete an atlas of NSC differentiation with higher spatial and temporal resolution.

Sex differences and a wall-specific neurogenic fate commitment have been detected by scRNAseq study investigating the whole SVZ stem cell niche (Mizrak et al., 2019). While both lateral and septal walls are enriched in neurogenic potential, the septal wall shows a pronounced gliogenic fate, with a higher rate of oligodendrocyte precursors cells (OPCs) in males compared to female mice. A possible mechanism contributing to this difference could be the specific activation of the myelinating system under the influence of sex hormones. However, this speculation needs to be validated on specific demyelinating diseases such as multiple sclerosis (MS), that predominantly affects females (Swamydas et al., 2009; Sellner et al., 2011).

Bulk gene expression analysis of in vitro differentiated Ols revealed that the Sox4 TF inhibits Ols differentiation via expression of the Hes5 TF (Braccioli et al., 2018). GO indicated that CNS differentiation arises from the activation of the Notch signaling pathway, suggesting that Sox4 enhances neurogenesis and downregulates oligodendrogenesis towards mature and myelinating Ols. Indeed, Hes5 silencing induced Ols differentiation in vitro.

The involvement of specific lncRNA in cell fate determination of the oligodendroglial lineage has also been investigated by integrating RNA and Chip-seq (Wei et al., 2021). As reported in subchapter 2.2.3, lncRNAs play a fundamental role during brain development by binding to the chromatin modifiers and leading to changes in the chromatin landscape, thus affecting gene expression (Roberts et al., 2014). Notably, during Ols differentiation, the Olig2 gene regulates many lncRNAs acting as activators or repressors, as they exhibit OLIG2 biding sites (Wei et al., 2021).

The characterization of neural and non-neural cell subsets by computational analysis of RNA kinetics has unveiled the dynamics of changes in gene expression under developmental transition (Zywitza et al., 2018). The single-cell fate can be evaluated by calculating the ratio between spliced and unspliced mRNA, which defines RNA velocity as the time derivative of the gene expression state. This approach was applied to the single cell analysis of the mouse SVZ, showing that astrocytes and qNSCs exhibited stable gene expression, while great changes in the transcriptome occurred during the transition toward amplifying progenitors and returned stable when late NBs are produced (La Manno et al., 2018).

The balance between neurogenesis and oligodendrogenesis is strictly related to transcription regulators such as Dlx1/Dlx2 that promote neuronal progeny, and Olig1/Olig2 that induce Ols differentiation (Petryniak et al., 2007; Silbereis et al., 2014; Lindtner et al., 2019). In this regard, chromatin conformations are crucial in determining fate commitment through accessibility to regulatory element regions such as putative cis-regulatory elements (CRE). By applying ATAC-seq, a strong interaction between Dlx1 and Rbbp4/Rbbp7 NuRD chromatin repressive complex has been shown (Price et al., 2022). Notably, during SVZ cell differentiation, chromatin accessibility is reduced and mediated by Dlx1–Rbbp4/7 contact leading toward repression of Ols differentiation. On the contrary, in the absence of Dlx1, Ols differentiation prevailed. Thus, nNuRD chromatin remodelers are crucial at this step to decide whether to initiate neuronal or oligodendroglia commitment.

Astrocyte differentiation has been hypothesized to be influenced by the activation of NF-κB, a TF known to be involved in brain development (Kovács et al., 2004; Birck et al., 2021). Indeed, an inflammatory stimulus like TNF-α induced the maintenance of SVZ glial progenitors rather than promoting their differentiation toward astrocytes, as shown by RNA-Seq analysis. Future in vivo studies applying omics techniques will provide insights into the crucial role of NF-κB during astrocytic differentiation. Astrogenesis is also supported by proteins of the high-mobility group nucleosome-binding Family (Hmgn1, Hmgn2, Hmgn3; Nagao et al., 2014). Furthermore, RNA-ATAC-Chip-seq highlight specific TFs activity during astrocyte differentiation in the mouse brain, which involves Nfia and Atf3, while Runx2 drives their maturation (Tiwari et al., 2018).

Gene expression perturbations and environments shaping different NSCs subtypes in the adult SVZ are depicted in Figure 1.

Rodent models are important for studying certain aspects of neurodevelopment but results cannot be directly translated to humans, as the human brain is indeed much more intricate than rodents in terms of its structure, organization, and connectivity (Boldog et al., 2018). Furthermore, difficulties in obtaining fresh human specimens make the research on human neurodevelopment much more problematic, with fewer studies available. A summary of primate and human studies on endogenous NSCs profiling by different omics techniques is reported in Table 1.

Two bulk transcriptomic human studies reported new insights, focusing on the human fetal brain and the spinal cord stem cell niche (Yoon et al., 2017; Ghazale et al., 2019).

Yoon et al. (2017) performed N⁶-methyladenosine (m⁶A) mRNA sequencing on mouse E13.5 forebrain, human forebrain-derived organoids, and post-conception day 11 human fetus. This study focused on mRNA modifications by RNA methyltransferases, including Mettl3 (methyltransferaselike3-)-Mettl14 complex, Wtap (Wilms tumor 1-associated protein), Kiaa429, Rbm15 (RNA-binding motif protein 15), and its paralog (Rbm15b). In the fetal brain, enrichment of m⁶A, a conserved mRNA modification set up by the Mettl3/Mettl14 complex, was detected. m⁶A controls human cortical neurogenesis by regulating the expression of genes involved in cell cycle progression and differentiation. As m⁶A was found highly enriched in genes related to mental disorders, such as schizophrenia, it has been hypothesized that its perturbation could be a causative pathological mechanism. Depletion of m⁶A by Mettl3 knockdown showed alterations in the post-natal forebrain transcriptional profile, where a significant glutamatergic subset surprisingly persisted after cortical neurogenesis, indicating that neuronal differentiation could also occur postnatally. Indeed, m⁶A depletion correlated with the differentiation of a dysregulated glutamatergic population with low neurogenic potential and downregulation of genes involved in cell differentiation and migration (Donega et al., 2018).

The murine spinal cord stem cell niche has been described thanks to single nucleus RNA-seq, however, a deep functional and molecular characterization is necessary to unveil the potential role of spinal cord NSCs during lifespan (Habib et al., 2016). Efforts have been made in this direction, by profiling transcriptome from both human and mouse, young and adult spinal cord stem cell niches (Ghazale et al., 2019). More than one hundred transcription factors were found conserved in both mouse and human specimens. Pathway analysis revealed that smoothened/Shh and Hippo/YAP genes, known to be involved in stem cell maintenance, were enriched along with 34 members of the solute carrier family. The Tmem212 gene confirmed its relevance in the NSC spinal cord niche, specifically for the ependymal cells, in three independent studies (Rosenberg et al., 2018; Zeisel et al., 2018; Ghazale et al., 2019). These investigations are particularly relevant as they compare bulk transcriptomes of human and mouse NSCs. The highlights of those studies are reported in Table 2, with a particular focus on the similarities and differences between human and mouse transcriptomic outcomes.

An interesting bulk study aimed at unraveling mechanisms of human brain evolution has been carried out in Rhesus monkey, which shows high similarity with Homo sapiens in terms of expression of genes involved in cell cycle regulation, apoptosis, DNA repair, and chromatin remodeling (Shi et al., 2019). The gene encoding for microcephalin-1, Mcph1, which is involved in the DNA damage response is believed to be a strong candidate that drove the evolution of the human brain size (Pulvers et al., 2015). Thanks to the generation of transgenic monkeys overexpressing human Mcph1, researchers analyzed pre- and postnatal stages of neuronal differentiation, identifying a possible explanation for the delayed neuronal corticogenesis. Indeed, Mcph1 contributes to this delay as, at the prenatal stage, transgenic monkeys showed elevated levels of transcripts related to translation, protein localization, and endoplasmic reticulum functions rather than synaptic functions, when compared to their wild-type counterparts. Furthermore, at the postnatal stage, downregulation of neuronal differentiation and maturation were observed, underscoring the role of Mcph1 in the neuronal maturation delay observed in transgenic monkeys. Altogether, these data set the basis for further studies aimed at characterizing other potential genes involved in the regulation of human CNS development (Shi et al., 2019).

The precise specification of excitatory neurons and inhibitory interneurons in the human cortex is key for the proper formation and function of neural circuits in the mammalian brain. Thanks to the enormous bioinformatics progress and to the development of specialized tools to reconstruct lineage tracing, such as RNA velocity, it has been possible to track brain development since early events, from gestational week 6–10, in different brain regions including the telencephalon, diencephalon, midbrain, hindbrain, cerebellum, ganglionic eminences, thalamus, hypothalamus and cortex (La Manno et al., 2018; Eze et al., 2021). At the earliest time points, no specific transcriptional programs across different brain regions have been observed. Mef2c, the gene encoding for the Myocyte Enhancer Factor 2C, was expressed in younger specimens (gestational week 6), and shown to play a role in early neurogenesis and synaptic formation. The analysis of samples at 10 gestational weeks showed several genes that were specifically expressed in different neuronal subsets: Nhlh1 expressed by intermediate progenitors-derived newborn neurons; Neurod6 and Bcl11b expressed by deep later neurons, and Calb2 expressed by ventrally migrating interneurons and excitatory neurons. Interestingly, the RNA velocity analysis for fate developmental mapping revealed a high number of genes associated with neurodevelopmental and psychiatric disorders that influence the differentiation program (Eze et al., 2021).

A combination of ATAC-seq and Hi-C techniques allowed to investigate chromatin accessibility and its physical interactions, and to build a comprehensive map of non-coding regulatory elements fundamental during human cortical development (Torre-Ubieta et al., 2018). Distal regulatory elements, enriched mainly in the enhancers, were found to impact neurogenesis. Transcriptome analysis confirmed and added further insights into the regulation of brain development. Specifically, germinal zone distant enhancers regulate glial differentiation genes, neuroblasts proliferation, and neurogenesis, while cortical plate distal enhancers were associated with neuron projection morphogenesis, axon guidance, and glutamate receptor activity. Moreover, the Fgfr2 enhancer revealed its fundamental role during human brain expansion. Thanks to the multimodal integration of different omics techniques, a list of enhancers regulating human brain development could be the basis for further studies that will analyze also non-coding transcripts whose genetic variations are involved in neurodevelopmental and psychiatric disorders (Torre-Ubieta et al., 2018).

Another important example of a multi-omics application is presented in the work of Ziffra et al. (2021) in which a combination and integration of transcriptome, chromatin state, and accessibility analysis create a comprehensive human developing brain atlas including prefrontal, motor, visual, somatosensory, parietal, temporal and insular cortex. Moreover, in this study, the authors describe regulatory programs during corticogenesis, which are related to risk factors of neurodevelopmental and neuropsychiatric disorders. Indeed, it was found that common schizophrenic variants were positively enriched in neuronal excitatory and inhibitory putative enhancers, confirming a previous investigation (Ripke et al., 2014). Finally, this work brings fundamental insights into the trajectory of the developing brain by associating chromatin state with transcriptome analysis, highlighting the significance of multi-omics analyses to better understand complex biological systems.

Interneurons originate in the GE in the embryonic mammalian brain and orderly migrate and settle in cortical layers before they establish cortical connections (Tremblay et al., 2016). Different types of interneurons have specific origins: OB interneurons originate in the LGE, striatal, hippocampal, and cortical interneurons originate in the MGE, while cortical interneurons originate in the CGE. This diversity has been investigated in the human fetal subpallium reporting a distinct transcriptional pattern among MGE, LGE, and CGE (Yu et al., 2021). The adult interneuron specification is already present at gestational week 9. Thanks to single-cell RNA in situ sequencing and analysis of the developmental trajectory, a map of each GE precursor towards a specific localized interneuron has been established. Precisely, MGE and LGE precursors induce two distinct lineages in the dorsal/ventral region respectively, thus highlighting spatial differentiation and patterning that occurs in the human fetal subpallium. A comparison of similarities and differences in the interneuron origin from initial classes of MGE, LGE, and CGE is reported in Table 3, which reports the terminal classes characterized by specific gene expression markers in mouse, primate, and human brain omics datasets recently published (Yu et al., 2021; Schmitz et al., 2022). Indeed, the work performed by Schmitz et al. (2022) dissects the initial and terminal interneurons classes across the development of mice and primates by applying RNA velocity tools and unveiling conserved initial interneurons classes of mammals.

In the aging human brain, progenitors of the SVZ remain mostly in a quiescent state leading to neurogenesis decline. A single-cell transcriptomic study of aged subjects (88–95 years old) identified potential targets that enable the exit of NSCs from quiescence and promote their proliferation (Donega et al., 2022). The endogenous stem cell gene expression profiling in the aged is compared with published fetal datasets, showing major batch effects and issues in the correct clustering of different cell types (Zhong et al., 2018; Jäkel et al., 2019). Nowadays, studies integrating a high number of different datasets should consider applying novel tools such as SCENIC to integrate datasets in an unbiased manner which helps to avoid the batch effect (Aibar et al., 2017). Interestingly, in the dataset of aged sorted CD271 positive progenitors, the authors describe two cell clusters that were enriched in Sox10 and Rgcc gene expression, the latter being involved in cell cycle dysfunction, thus indicating that progenitors of the aged human SVZ are late OPCs (Counts and Mufson, 2017). When integrating the aged with several fetal datasets, the oligodendrocyte progenitors clustered together, nevertheless, aged OPCs were largely quiescent. Thus, the absence of mature Ols during the lifespan may be due to the persistent quiescent state of their progenitors, which is maintained via inhibition of the WNT signaling pathway by its antagonist Sfrp1 (Donega et al., 2022).