(A,B) In situ hybridization for Gad67 in wild type (A) and Fezf2^−/− (B) cortex at P28 shows reduced numbers of interneurons in layer V and increased numbers in superficial layers of the Fezf2^−/− cortex. (C) Total number of Gad67-interneurons is unchanged between wild type and Fezf2^−/− cortex. (D–R) Quantification of the layer distribution of Gad67-interneurons across motor (D–H), somatosensory (I–M) and visual (N–R) cortex. (D,E,I,J,N,O) Representative sections of wild type and Fezf2^−/− cortex. (F,K,P) Unbiased binned distribution of Gad67-interneuron percentages shows decreased percentages in deep bins and increased percentages in superficial bins of the mutant. (G,L,Q) β-galactosidase immunocytochemistry in Fezf2^+/− heterozygote mice demonstrates that layer V corresponds to bins 4–5 (in motor and somatosensory cortex) and bins 3–4 (in visual cortex). (H,M,R) Quantification of interneurons within layers demonstrates a specific reduction in interneuron percentages in mutant layer V across all cortical areas, and an increase in interneuron percentages in layers II/III–IV. All results are expressed as the mean ± s.e.m.. LV, lateral ventricle. Scale bars, 500 μm (A,B); 200 μm (D,E,G,I,J,L,N,O,Q). (See also Figures S2 and S3).

(A,B,G,H) In situ hybridization for SST (A,B) and immunocytochemistry for PV (G,H) in wild type and Fezf2^−/− somatosensory cortex at P28 show reduction of both interneuron populations in layer V and increase in upper layers. (C,I) Total number of SST- (C) and PV- (I) interneurons is unchanged between wild type and Fezf2^−/− cortex. (D,J) Unbiased binned distribution of SST- and PV-interneuron percentages shows decreased numbers in deep bins and increased numbers in superficial bins of the mutant. (E,K) β-galactosidase immunocytochemistry in Fezf2^+/− heterozygote mice highlights layer V in bins 4–5. (F,L) Quantification of SST- (F) and PV- (L) interneurons within layers demonstrates a reduction in the percentages of both interneuronal subpopulations in mutant layer V and an increase in layers II/III–IV. All results are expressed as the mean ± s.e.m.. LV, lateral ventricle; str, striatum; cc, corpus callosum. Scale bars, 500 μm (A,B,G,H); 100 μm (E,K). (See also Figures S4 and S5).

(A–F′) In situ hybridization for Lhx6 (A–B′), SST (C–D′) and VIP (E–F′) combined with immunocytochemistry for TBR1 on electroporated brains demonstrates that aggregates of corticofugal neurons (B′,D′,F′, arrows) contain deep layer Lhx6-positive (B,B′, arrows) and SST-positive (D,D′, arrows) interneurons and very small numbers of VIP-positive (F,F′, arrows) interneurons. Aggregates of interneurons are absent from contralateral locations (A′,C′,E′). (GO) Immunocytochemistry for NPY (G–J), GABA, Reelin and GFP (K–O) show that interneuron subpopulations that are mostly found in cortical superficial layers or layer Va are absent from the Fezf2-positive aggregates (H′,J,L′,O). (K″) Coexpression of GABA and Reelin in layer Va. Scale bars, 500 μm (A,B,C,D,E,F,G,H,K,L); 100 μm (A′B′,C′,D′,E′,F′,G′H′,I,J,M–O).

(A) Schematic of experimental design and quantification of first-generation, BrdU-positive and GABA-positive interneurons within aggregates expressing either Fezf2^GFP or APP shRNA. GABAergic interneurons in the corticofugal aggregates (Fezf2^GFP) are largely early-born, while those in the aggregates of layer II/III CPN (APP shRNA and CAG-GFP) are largely late-born. (B) Quantification of first-generation, BrdU-positive and GABA-positive interneurons within the deep and superficial layers of the cortex overlying the aggregates. Late-generated corticofugal neuron aggregates and synchronically generated CPN show distribution of E12.5- and E14.5- interneurons as observed in the deep and superficial cortical layers, respectively. (C–N) GABA-positive interneurons within Fezf2^GFP aggregates are born at E12.5 (C–E, arrows) and not at E14.5 (F–H, arrowheads), while GABA-positive interneurons within the APP shRNA aggregates are born at E14.5 (L–N, arrows) and not at E12.5 (I–K, arrowheads). Scale bars, 100 μm (C–N).

(A–E) Retrograde labeling of CPN in contralateral Fezf2^−/− cortex shows colocalization of FG with the expanded population of TBR1- and SATB2 -positive neurons in layer V. (F–K) In situ hybridizations at P6 for Lpl, Inhba and Limch1 (CPN markers), showing that the expanded population of CPN in Fezf2^−/− layer V are of deep layer identity. Scale bars, 100 μm (F–K), 20 μm (B–E). (See also Figure S1)

(A) Schematic of recording area (red box) indicating the position of the stimulating electrode in the white matter (WM). (B,C) Pseudocolor peak response frames from VSDI movies of wild type (B) and Fezf2^−/− (C) slices 15 ms after the stimulus (arrowhead), showing that wild type mice exhibit a strong stimulus response that propagates rapidly to the upper layers (B); whereas, the response in Fezf2^−/− slices rarely reaches the upper layers, remaining largely confined to the lower layers (C). Black and white squares indicate quantified regions of interest in the upper (UL) and lower layers (LL), respectively. Scale bars, 250 μm. (D,F) Stimulus response curve in the regions of interest for upper (D) and lower (F) layers. (E, G) Response at half max. *** p<0.001, t-test. (See also Movies S1, S2, S3 and S4).

(A) Schematic of experimental approach. (B–C′) In utero overexpression of Fezf2 in cortical progenitors at E14.5 causes a switch of fate to corticofugal projection neurons, which develop as ectopic aggregates below the corpus callosum. (C′–F) GFP- and CTIP2-positive aggregates of corticofugal neurons contain GABA-positive interneurons (arrows). (G–H′) In situ hybridization for Gad67 and immunocytochemistry for TBR1 demonstrates that TBR1-positive aggregates of corticofugal neurons contain Gad67-interneurons (arrows). Scale bars, 500 μm (B,C,G,H); 100 μm (B′,C′,D–F,G′,H′). (See also Figures S6, S7 and S8).

(A) Schematic of experimental approach. (B–C′) In utero overexpression of APP shRNA combined with CAG-GFP in cortical progenitors at E14.5 blocks the migration of superficial layer projection neurons, which differentiate as ectopic aggregates below the corpus callosum. (C′, D–F) GFP- and SATB2-positive aggregates of upper layer CPN contain GABA-positive interneurons (arrows). Scale bars, 500 μm (B,C); 100 μm (B′,C′,D–F).