Adaptive evolution of gene regulatory networks in mammalian neocortex
Summary
This Nature paper identifies a transcription factor-centred regulatory module — focused on ZBTB18 and its targets — that helped shape mammalian neocortical excitatory neurons (ExNs). Using cell-type selective RNA-seq and H3K27ac ChIP–seq from mouse upper-layer (IT) and deep-layer (ET) neurons, plus comparative H3K27ac data from chicken and multispecies alignments, the authors show that ZBTB18 binds enhancers that are mammal- or eutherian-specific for key genes (Cux2, Satb2, Robo1, Bcl11b). Functional assays (reporter transgenics, luciferase, conditional knockouts and electroporation) demonstrate ZBTB18 activates mammalian IT enhancers, that loss of ZBTB18 shifts gene expression toward ET programmes, and that callosal and subcortical projections are disrupted in Zbtb18 mutants. Motif conservation analyses indicate ZBTB18 binding sites in IT enhancers are highly conserved across placental mammals and marsupials, implicating regulatory evolution of CREs in mammalian neocortical specialisation.
Key Points
- ZBTB18 is highly expressed in postmitotic IT neurons and its binding sites mark enhancers for IT-specific genes including Cux2 and Satb2.
- Mouse ZBTB18 ChIP–seq identifies enhancers bound by ZBTB18 that are absent or inactive in chicken dorsal pallium, indicating mammalian-specific regulatory changes.
- Cux2-E1 and Satb2-E1 enhancers are motif-dependent ZBTB18 targets; ZBTB18 and POU3F2 activate Cux2-E1, while HDAC2/SIN3A repress it.
- Luciferase assays across species show ZBTB18 activation of Cux2-E1 is a placental mammal adaptation; opossum and chicken orthologues lack the same response.
- Zbtb18 knockout or postmitotic conditional deletion reduces IT markers (Cux1/2, Satb2, Rorb), increases ET marker Bcl11b, and produces simplified laminar organisation.
- ZBTB18 loss disrupts long-range wiring: agenesis or reduction of corpus callosum and subcerebral tracts, altered intrahemispheric inputs to prefrontal cortex — effects partly rescued by Robo1 overexpression.
- Comparative motif analysis across 60 vertebrates shows enriched conservation of ZBTB18 core motifs in IT-biased enhancers among placental mammals and marsupials, suggesting a regulatory evolutionary step in Theria.
Content summary
The study combined FACS of Arpp21–Gfp (upper-layer IT) and Fezf2–Gfp (deep-layer ET) mouse neurons at neonatal stages with RNA-seq and H3K27ac ChIP–seq to map subtype-biased CREs and candidate TFs. ZBTB18 emerged as a top IT-associated TF; the authors mapped its binding by HA-tagged ZBTB18 ChIP–seq and identified mammal-specific enhancer peaks near Cux2, Satb2, Robo1 and Bcl11b. Transgenic reporter assays showed Cux2-E1 drives upper-layer, commissural IT expression and co-localises with ZBTB18. Luciferase and deletion assays established motif-dependent activation by ZBTB18; cross-species tests revealed placental-mammal specificity. Zbtb18 KO and conditional postmitotic cKO reduced IT gene programmes and increased ET-like gene expression. Anatomical tracings and axon assays showed loss of callosal and many subcerebral projections and a reorganisation of cortico-cortical inputs. Rescue experiments with Robo1 partially restored callosal projections, linking ZBTB18 regulation of axon-guidance genes to connectivity. Finally, phylogenetic conservation analysis found ZBTB18 motifs selectively conserved in IT enhancers across placental mammals and marsupials, indicating adaptive evolution of CREs rather than changes to the ZBTB18 protein itself.
Context and relevance
This paper sits at the intersection of developmental neurobiology and evolutionary genomics. It provides a concrete molecular mechanism — evolution of cis-regulatory elements bound by an otherwise ancient TF — that likely contributed to the diversification of excitatory cortical neuron subtypes and mammal-specific long-range connectivity. For researchers studying cortical development, comparative neurobiology, or the genetic bases of callosal disorders and autism, the work links non-coding regulatory evolution to both circuit architecture and disease susceptibility.
Author style
Punchy: the authors back a tight mechanistic story with cell-type genomics, cross-species comparisons and functional perturbations. If you want to understand how small regulatory changes can rewire circuits during evolution — and why that makes some human cortical programmes vulnerable — read the details.
Why should I read this?
Want the short version — forget trawling dozens of papers. This one shows how tweaking enhancers (not the TF protein) helped build mammal-style cortical neurons and connections. It explains a likely molecular step that made callosal circuits and upper-layer identity what they are in mammals, and connects those changes to genes already implicated in neurodevelopmental disorders. If you’re into brain evolution, cortical wiring or regulatory genomics, it’s worth your 10-minute skim — we did the heavy reading for you.