Cortical Neurogenesis

Cortical Neurogenesis is the embryonic and prenatal developmental process through which neural stem and progenitor cells lining the lateral ventricles of the dorsal telencephalon (pallium) proliferate, diversify, and guide newborn post-mitotic neuroblasts radially outward to construct the six horizontal laminae (Layers I–VI) of the mammalian neocortex (isocortex). Whereas excitatory glutamatergic projection neurons (~80% of adult neocortical neurons) arise locally within dorsal pallial germinal zones and migrate radially along glial fibers, cortical inhibitory GABAergic interneurons (~20%) originate ventrally in the medial and caudal ganglionic eminences and invade the cortex tangentially—assembling canonical microcircuit columns underlying perception, motor control, language, and cognition.

From Symmetric Neuroepithelial Expansion to Asymmetric Radial Glia

Under Pasko Rakic's Radial Unit Hypothesis (1988), neocortical surface area is determined early by the number of founder proliferative units lining the ventricular zone (VZ), whereas cortical thickness reflects neuronal output per column. Before neurogenesis begins, pseudostratified neuroepithelial cells coupled apically by tight and adherens junctions divide symmetrically, doubling the founder pool each cycle. At neurogenesis onset (mouse E10.5; human gestational week GW7), neuroepithelial cells downregulate tight junctions, express astroglial markers (VIM, FABP7, SLC1A3) alongside PAX6 and SOX2, and convert into bipolar apical radial glia (aRG) spanning the cerebral wall from ventricular endfeet to the pial basal lamina.

Combining retroviral lineage tracing with live slice microscopy, Stephen C. Noctor, Arnold R. Kriegstein, and colleagues proved in 2001 that radial glia act as the primary mitotic neural stem cells of the mammalian cortex. Dividing asymmetrically at the ventricular surface, an aRG self-renews one radial glial daughter while budding either a neuron (direct neurogenesis) or a multipolar transit-amplifying intermediate progenitor cell (IPC) expressing EOMES (TBR2) (indirect neurogenesis). IPCs delaminate into the subventricular zone (SVZ) and divide symmetrically into pairs of neurons, amplifying output per radial unit.

Primate Outer Subventricular Zone (oSVZ) and Evolutionary Gyrification

Whereas rodent SVZ compartments remain thin, David V. Hansen, Jan H. Lui, Arnold R. Kriegstein, and colleagues demonstrated in 2010 that gyrencephalic primates and humans markedly expand a specialized germinal layer—the outer subventricular zone (oSVZ)—above an inner fiber layer. Between human GW11 and GW24, the oSVZ generates the bulk of upper-layer neurons via self-renewing outer (basal) radial glia (oRG) co-expressing PAX6, SOX2, HOPX, PTPRZ1, and LIFR. Shedding apical ventricular contact while keeping a long basal fiber toward the pia, oRG cells undergo repeated divisions with mitotic somal translocation (leaping 50–100 μm basally via ROCK–myosin II contractility prior to cytokinesis), fanning radial scaffolds tangentially to drive the ~1,000-fold cortical surface expansion and sulcal–gyral folding between mouse and human—amplified by hominin-specific gene duplications including mitochondrial ARHGAP11B and Notch-potentiating NOTCH2NL.

Excitatory cortical projection neurons (Layers II–VI) populate the cortical plate via an inside-out chronological rule established by Richard Sidman and Pasko Rakic: early-born neurons settle in deep Layer VI, whereas successively later-born cohorts migrate past older somata to settle progressively closer to the pial surface (Layer V → Layer IV → Layers II/III):

1. Multipolar-to-Bipolar Transition: Newborn neuroblasts in the SVZ and intermediate zone adopt a transient multipolar morphology, downregulating N-cadherin, extending exploratory neurites, and specifying a trailing axon via LKB1–SAD kinases before repolarizing their centrosome apically into a bipolar locomoting neuron.
2. Glia-Guided Saltatory Locomotion: Adhering to radial glial fibers via astrotactin-1 (ASTN1) and connexin-26/43 adhesive hemichannels, bipolar neuroblasts climb at 15–35 μm/hour as cytoplasmic dynein complexed with LIS1 (PAFAH1B1), NDEL1, and Doublecortin (DCX) pulls the centrosome into the leading process ahead of myosin-II somal translocation; mutations in PAFAH1B1 or X-linked DCX arrest migration, causing human four-layered lissencephaly or subcortical band heterotopia.
3. Marginal-Zone Reelin Stop Signal: Superficial Layer I Cajal–Retzius cells (TP73⁺/RELN⁺) secrete the 388 kDa glycoprotein Reelin (RELN). Reelin binding to VLDLR and ApoER2 (LRP8) receptors on arriving neuroblasts activates Src/Fyn phosphorylation of Disabled-1 (DAB1), driving Rap1–integrin α₅β₁ and N-cadherin terminal somal translocation past older neurons and radial-fiber detachment; RELN loss causes an outside-in inverted cortex.

Over time, cortical progenitors progressively restrict fate competence and activate cross-repressive post-mitotic transcription factors mapped by Jeffrey D. Macklis, Nenad Sestan, and Susan K. McConnell:

After Layer II/III neurogenesis concludes (~E17.5 in mice; ~GW24–27 in humans), newborn neurons secrete cardiotrophin-1 and Notch ligands activating JAK–STAT3 and NFIA/NFIB in radial glia alongside epigenetics DNA demethylation of astrocytic promoters (GFAP, AQP4) to trigger the neurogenic-to-gliogenic switch, generating cortical astrocytes, ependymal cells, and OLIG2⁺/PDGFRA⁺ oligodendrocyte precursors.

Human cortical development exhibits pronounced neoteny: whereas mouse neurogenesis finishes in 7 days and dendritic spines stabilize within a month, human cortical neurogenesis spans ~18 weeks in utero, followed by 5 to 25 years of postnatal synaptogenesis, elevated childhood glucose metabolism, and adolescent pruning of glutamatergic synapses sustaining prolonged experience-dependent synaptic plasticity.

Human stem-cell-derived cortical neurons xenografted into neonatal mouse cortex—or cultivated in human cerebral organoids and assembloids—still require 9 to 18 months to reach mature spine densities and firing properties rather than adopting the 3-week mouse schedule. In 2023, Ryohei Iwata, Pierre Vanderhaeghen, and colleagues showed that mitochondrial tricarboxylic acid (TCA) cycle and NAD⁺/NADH oxidative rates pace this intrinsic clock: human newborn neurons maintain lower mitochondrial respiration than mouse neurons, and pharmacologically boosting human pyruvate/lactate oxidative flux accelerates neuronal arborization and synaptic maturation, coupling bioenergetic pacing with human-specific synaptic modifiers (SRGAP2C, MEF2C).

Why are excitatory and inhibitory cortical neurons born in separate telencephalic domains?

Opposing dorsal BMP/Wnt and ventral Sonic hedgehog morphogen gradients segregate neurotransmitter identity early in forebrain patterning: dorsal PAX6⁺/NEUROG1/2⁺ progenitors produce glutamatergic pyramidal neurons radially, whereas ventral NKX2-1⁺/DLX1/2⁺ medial and caudal ganglionic eminence progenitors generate GABAergic interneurons that migrate tangentially into the cortex.

How do human-specific genes ARHGAP11B and NOTCH2NL expand the neocortex?

Hominin segmental duplications ~3 to 5 million years ago produced ARHGAP11B (carrying a C→G splice mutation targeting the protein to mitochondria to stimulate basal-progenitor glutaminolysis) and three active NOTCH2NL paralogs (potentiating Notch2 signaling to delay differentiation); expressing either human gene in mouse, ferret, or marmoset embryonic cortex expands outer radial glia and induces gyral folding.

What becomes of radial glia once cortical neurogenesis ends?

In late gestation, radial glia retract their ventricular endfeet, translocate somata into the parenchyma to differentiate into protoplasmic and fibrous astrocytes, or remain along the ventricular wall to form multiciliated ependymal cells and postnatal subventricular B1 neural stem cells.

1. Noctor, S. C., Flint, A. C., Weissman, T. A., Dammerman, R. S., & Kriegstein, A. R. (2001). Neurons derived from radial glial cells establish radial units in neocortex. Nature, 409(6821), 714–718. https://doi.org/10.1038/35055553
2. Hansen, D. V., Lui, J. H., Parker, P. R. L., & Kriegstein, A. R. (2010). Neurogenic radial glia in the outer subventricular zone of human neocortex. Nature, 464(7288), 554–561. https://doi.org/10.1038/nature08845
3. Lui, J. H., Hansen, D. V., & Kriegstein, A. R. (2011). Development and evolution of the human neocortex. Cell, 146(1), 18–36. https://doi.org/10.1016/j.cell.2011.06.030
4. Greig, L. C., Woodworth, M. B., Galazo, M. J., Padmanabhan, H., & Macklis, J. D. (2013). Molecular logic of neocortical projection neuron specification, development and diversity. Nature Reviews Neuroscience, 14(11), 755–769. https://doi.org/10.1038/nrn3586
5. Iwata, R., Casimir, P., Erkol, E., Boubakar, L., Planque, M., Gallego López, I. M., ... & Vanderhaeghen, P. (2023). Mitochondria metabolism sets the species-specific tempo of neuronal development. Science, 379(6632), eabn4705. https://doi.org/10.1126/science.abn4705