Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Apr 27;186(9):1930-1949.e31.
doi: 10.1016/j.cell.2023.03.025. Epub 2023 Apr 17.

Pyramidal neurons form active, transient, multilayered circuits perturbed by autism-associated mutations at the inception of neocortex

Martin Munz  1 , Arjun Bharioke  1 , Georg Kosche  2 , Verónica Moreno-Juan  2 , Alexandra Brignall  2 , Tiago M Rodrigues  2 , Alexandra Graff-Meyer  3 , Talia Ulmer  1 , Stephanie Haeuselmann  2 , Dinko Pavlinic  2 , Nicole Ledergerber  1 , Brigitte Gross-Scherf  1 , Balázs Rózsa  4 , Jacek Krol  1 , Simone Picelli  2 , Cameron S Cowan  2 , Botond Roska  5
Affiliations

Pyramidal neurons form active, transient, multilayered circuits perturbed by autism-associated mutations at the inception of neocortex

Martin Munz et al. Cell. .

Abstract

Cortical circuits are composed predominantly of pyramidal-to-pyramidal neuron connections, yet their assembly during embryonic development is not well understood. We show that mouse embryonic Rbp4-Cre cortical neurons, transcriptomically closest to layer 5 pyramidal neurons, display two phases of circuit assembly in vivo. At E14.5, they form a multi-layered circuit motif, composed of only embryonic near-projecting-type neurons. By E17.5, this transitions to a second motif involving all three embryonic types, analogous to the three adult layer 5 types. In vivo patch clamp recordings and two-photon calcium imaging of embryonic Rbp4-Cre neurons reveal active somas and neurites, tetrodotoxin-sensitive voltage-gated conductances, and functional glutamatergic synapses, from E14.5 onwards. Embryonic Rbp4-Cre neurons strongly express autism-associated genes and perturbing these genes interferes with the switch between the two motifs. Hence, pyramidal neurons form active, transient, multi-layered pyramidal-to-pyramidal circuits at the inception of neocortex, and studying these circuits could yield insights into the etiology of autism.

Keywords: autism; embryonic development; in vivo imaging; in vivo patch clamp; layer 5; neuronal activity; pyramidal neurons; single cell sequencing; synapses; transient circuits.

PubMed Disclaimer

Conflict of interest statement

Declaration of interests The authors declare no competing interests.

Figures

None
Graphical abstract
Figure 1
Figure 1
Rbp4-Cre neurons have L5-PN identity (A) Top: Single-cell RNA sequencing workflow (Figure S1). Bottom: Expression profile of layer-specific genes,, in Rbp4-Cre neurons. (B) Top: Rbp4-Cre neurons (green), Bcl11b (magenta), Hoechst (blue). Bottom: fraction of Rbp4-Cre neurons expressing Bcl11b. n = number of Rbp4-Cre neurons. (C) Correlation of cortical layer-specific neuronal genes’ expression between Rbp4-Cre neurons and adult cortical layers for up to 150 genes., (D) UMAP embedding of Rbp4-Cre neurons’ single cell transcriptomes. Color: Leiden clusters. (E) Top: Rbp4-Cre types (colored as in D) embedded in a triangle representing the similarity between each cell’s expression profile from the three adult L5-PN types (NP, IT, and PT) (Figure S1). Bottom: For each adult type, percent of Rbp4-Cre neurons of each embryonic type associated with that adult type. (F) Percent of neurons from each type on each embryonic day (colored as in D and labels derived from E). Scale bars: 20 transcripts (A), 10 μm (B). See also Figure S1.
Figure S1
Figure S1
Rbp4-Cre neurons appear in cortex from E13.5 onwards, and divide into three embryonic types, related to Figure 1 (A) Rbp4-Cre neurons at E13.5, near surface of developing cortex (dotted line), immunostained against either tdTomato (left) or GCaMP6s (right) (white), counterstained with Hoechst (blue). (B) Single cell RNA sequencing workflow overview demonstrating two distinct stages of enrichment for tdTomato-expressing Rbp4-Cre neurons, from dissociated embryonic cortical tissue (red brackets). (C) Initial enrichment of Rbp4-Cre neurons was performed by isolating cells positive for the tdTomato marker from the dissociated cortical tissue using FACS. Gates were selected based on the size and granularity of sorted events. Gate 1 was chosen to exclude debris, while gates 2 and 3 were chosen to exclude doublets. Gate 4 was chosen to select cells with higher tdTomato fluorescence. Box: events filtered at each gate. (D) Additional enrichment for positively identified tdTomato-expressing Rbp4-Cre neurons. Relative enrichment of cells over a negatively FACS sorted population (blue) was used to define a baseline tdTomato expression level. Cells with significantly greater tdTomato expression than in the baseline were selected as positively identified Rbp4-Cre neurons (red). Green line: 6σ threshold of tdTomato expression. (E) UMAP embedding of all 25681 excitatory neuron transcriptomes (blue) demonstrating the location of the subset of 1098 positively identified Rbp4-Cre neurons (red). (F–H) Rbp4-Cre neuron divide into three embryonic types, each associated with one adult layer 5 type, stably across an increasing number of genes. Fraction of neurons within each Rbp4-Cre neuron type (1 (F), 2 (G), and 3 (H)) with a significant conditional probability of being sampled from the expression profile of one of the three adult cell types. Adult types: near-projecting neurons (NP) (beige), intratelencephalic neurons (IT) (red), and pyramidal tract neurons (PT) (pink) (as sequenced and identified in from VISp). Genes were selected to best discriminate the embryonic clusters, while still being differentially expressed across the three adult types. Dotted line: 24 differentially expressed genes used to generate Figure 1E. Scale bars: 50 μm (A).
Figure 2
Figure 2
Rbp4-Cre neurons distribute into superficial, intermediate, and deep layers (A) Top: Rbp4-Cre neurons (green), Hoechst (blue). Bottom left: Rbp4-Cre neuron depths normalized to the cortical plate and subplate thickness, (Figure S2). Bottom right: distribution of Rbp4-Cre neuron depths across days (blue: superficial layer; gray: intermediate layer; beige: deep layer). (B) Top: Fraction of all superficial layer Rbp4-Cre neurons found on each embryonic day. Bottom left: Immunostaining Rbp4-Cre neurons (green), cleaved Caspase 3 (magenta), Hoechst (blue). Bottom right: Fraction of Rbp4-Cre neurons expressing cleaved Caspase 3. Fisher’s exact test. (C) Top: Example in situ hybridizations of type-specific markers (magenta) (Figure S2), Rbp4-Cre neurons (green), Hoechst (blue). Middle: Fraction of Rbp4-Cre neurons containing embryonic-NP, IT, and PT-specific in situ hybridization markers; bold: genes shown in examples, above. Bottom: fractions of embryonic-NP, IT, and PT Rbp4-Cre neurons (based on in situ hybridizations). (D) Immunostaining Rbp4-Cre neurons (green), BrdU (red), Hoechst (blue). Arrowheads: example Rbp4-Cre neurons incorporating BrdU. (E) Normalized depth of Rbp4-Cre neurons incorporating BrdU (left), compared to Rbp4-Cre neurons without BrdU in the same embryos (center) and control embryos without BrdU injection (as in A). χ2 test. (F) Correlation of gene expression: comparing across time vs. type (on E18.5). Scale bars: 50 μm (top, A), 10% (bottom right, A), 20 μm (bottom left, B), 10 μm (top, C), 25 μm (D). See also Figure S2.
Figure S2
Figure S2
Rbp4-Cre neurons are located in the region from the subplate to the surface of cortex, are distinct from Cajal-Retzius cells and subplate neurons, and express layer 5 markers, related to Figure 2 (A) Within the UMAP embedding of all 25681 sequenced excitatory neuron transcriptomes (gray), cells expressing a number of genes previously associated with Cajal-Retzius cells,, overlap in a location (green outline) distinct from positively identified Rbp4-Cre neurons (as in Figure S1) (red outline; Rbp4-Cre neurons) (labeled in bottom right). Clusters of Rbp4-Cre neurons are additionally labeled based on the cluster identification from Figure 1E. (B) Within the UMAP embedding of all 25681 sequenced excitatory neuron transcriptomes (gray), cells expressing genes previously associated with subplate neurons overlap in a location (green outline) distinct from positively-identified Rbp4-Cre neurons (as in Figure S1) (red outline; Cre neurons) (labeled in the bottom right). Clusters of Rbp4-Cre neurons are additionally labeled based on the cluster identification from Figure 1E. (C) Rbp4-Cre neurons (green), counterstained with Hoechst (blue), show that neurons in the deep layer lie physically within the subplate (SP; dotted line), as localized by the expression of a common subplate marker, Nr4a2 (Nurr1) (magenta). (D) Only a small fraction of Rbp4-Cre neurons express Nr4a2 from E13.5 to E18.5. (E) Rbp4-Cre neurons are located in the region from the subplate to the surface of cortex. Rbp4-Cre neurons (stained using GFP antibody labeling GCaMP6s, green) in both spatial configurations (on E14.5, E16.5, and E18.5) co-labeled with antibodies labeling different zones in the developing cortical wall (Pax6 (first row), Tbr2 (second row), Tbr1 (third row), Satb2 (fourth row), red), counterstained with Hoechst (blue). Dotted line outlines the area from subplate to surface of cortex. (F) Rbp4-Cre neurons (stained using GFP antibody, green), counterstained with Hoechst (blue), show that neurons within both the superficial and deep layers at E14.5, as well as neurons within both the intermediate and deep layers at E18.5, all colocalize with Bcl11b (red), the expression of which is restricted in the adult cortex to layer 5. Arrows: example Rbp4-Cre neurons. (G) Genes selective for each embryonic layer 5 type. Transcript counts for each gene shown as colored circles in all types at all ages. Radius of circle: fraction of cells expressing the gene; color of circle: mean normalized transcripts per cell (log2). Bold text: genes used for in situ hybridization, shown in Figure 2C. (H) Spatial distribution of Rbp4-Cre neurons into layers is indistinguishable when expressing GCaMP6s compared to tdTomato on both E14.5 and E18.5, and Rbp4-Cre neurons. Quantifying the distribution of Rbp4-Cre neurons into layers at E14.5 (top) and E15.5 (bottom) in embryos generated by using the GCaMP6s-tTA2 reporter line, or a tdTomato reporter line. Probability: χ2 test comparing the fraction of Rbp4-Cre neurons in each layer for the two different reporter lines, p = 0.05. Scale bars: 20 μm (C, E, F).
Figure 3
Figure 3
Rbp4-Cre neurons show two phases of increased spontaneous activity (A) Schematic diagram of in vivo para-uterine two-photon calcium imaging (Figure S3). Top right: single embryonic neuron; arrowhead: soma activity; arrow: neurite activity; color: normalized calcium activity. (B) Mating strategy to drive GCaMP6s expression in Rbp4-Cre neurons. (C and D) Two-photon imaging of somas (red, C) and neurites (blue, D) of Rbp4-Cre neurons. Two regions of interest (ROIs) (left) and their recorded activity traces (right). (E and F) Activity of individual somas (E) and neurites (F) of Rbp4-Cre neurons. Circles: activity of each ROI; box (25–75 percentile) and whisker (5–95 percentile); white line: median; n = number of somas or neurites. Recordings from 3 (E13.5), 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos (Figures S4 and S5). (G) Activity (mean ± SEM) (data from E [soma, red] and F [neurite, blue]). Dotted line: separation of active phases and transition phase. (H) Distribution of activity (data from E [soma, red] and F [neurite, blue]) in log-scale. Horizontal lines: median (black: soma; white: neurite). (I) Activity in the two active phases across the three layers. (J) Schematic of Rbp4-Cre neuron development, highlighting the two circuit motifs and phases of activity. (K) Left: schematic of electroporations. Right: Immunostaining of electroporated Rbp4-Cre neurons (red), Bcl11b (white), Hoechst (blue). (L) Distribution of mKir2.1- or Kir2.1-positive Rbp4-Cre neurons’ normalized depths (as in Figure 2A) (10 mKir2.1-tdTomato and 10 Kir2.1-tdTomato electroporated embryos). Colored lines: medians. n = number of neurons. (E, F, H, L) Wilcoxon rank-sum test. Scale bars: 10 μm (inset, A), 40 μm (left, C, D), 25 s and 25 %ΔF/F (right, C), 25 s and 50 %ΔF/F (right, D), 2 ave. %ΔF/F (G), 20 μm (K), 10% (L). See also Figures S3, S4, and S5.
Figure S3
Figure S3
Characterizing in vivo para-uterine method for imaging cortical neurons, related to Figure 3 (A) Mean (red dot) of embryonic weights on each day (black dots) ranges from 0.17 g at E13.5 to 1.3 g at E18.5. N = number of embryos. (B) Stability of in vivo para-uterine two-photon imaging of cortical neurons in embryos is similar to stability of two-photon imaging of cortical neurons in adult mice. Movement per frame (recorded from 5 to 10 Hz), averaged per recording, computed via rigid motion correction between frames.In vivo embryonic recordings were made as schematized in Figure 3A. Adult recordings were made in head-fixed Rbp4-Cre mice, injected with AAV expressing Cre-dependent GCaMP. Probability: Wilcoxon rank-sum test; n = number of recordings from 3 (E13.5), 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos and 3 adult mice. (C–E) Embryonic blood flow does not change following 5 h of imaging, but degrades rapidly following severing of the umbilical cord. Blood flow at the surface of the brain was imaged in visible light prior to and following 5 h of imaging (C). The difference in blood flow is quantified (left), at E14.5 (D) and E18.5 (E). Blood flow at the surface of the brain was also imaged immediately prior to, and 10 min following the severing of the umbilical cord. The difference in blood flow is quantified (right), at E14.5 (D) and E18.5 (E). Box-and-whiskers: distribution of changes in blood flow across each time window, as box (25–75 percentile) and whisker (5–95 percentile); red lines: median. Probability: Wilcoxon rank-sum test. (F) Temperature of embryo stabilized para-uterine under the two-photon microscope. Infrared image is aligned with a visible light image, where the embryo can be observed within the holder (as schematized in Figure 3A). The 36.5°C marker labels the embryo,, which is visible through the opening in the holder allowing for the exit of the umbilical cord. The second 50°C marker labels the objective heater, providing a secondary source of heat during imaging. Image was taken following 5 h of imaging. (G) Fluorescence of Rbp4-Cre neurons is significantly increased over background fluorescence from E13.5 to E18.5. Mean cellular fluorescence, normalized by the pixel size of each cell, compared against the mean pixel fluorescence within three background regions selected within the imaging window, collected across all neurons, across all embryonic days (left) and on each embryonic day (right). Black circles: ratio of fluorescence within individual background regions in each imaging plane compared to each other; gray circles: ratio of fluorescence within individual neurons compared to each background region in the same imaging plane; box-and-whiskers: distributions across background fluorescence ratios (black) and cellular fluorescence ratios (gray) as box (25–75 percentile) and whisker (10–90 percentile); black and red line: median; blue line and text: mean. Probability: Wilcoxon rank-sum test comparing cellular fluorescence ratios to background fluorescence ratios. Recordings from 3 (E13.5), 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos. (H) Imaging region shown with respect to the embryonic brain at E13.5 (left), E15.5 (middle), and E18.5 (right). Imaging region was centered over the posterior dorsal pallium. Images taken from Allen Developing Brain Atlas (http://atlas.brain-map.org/).
Figure S4
Figure S4
Fraction of Rbp4-Cre ROIs with calcium events, as well as the frequency, length, and size of calcium events all vary across embryonic days, related to Figure 3 (A) Spontaneous calcium activity recorded from a single Rbp4-Cre neuron, showing three detected events, with event properties quantified for each event. (B–I) Quantification of event properties for Rbp4-Cre somas (B, D, F, and H) and neurites (C, E, G, and I). On each embryonic day from E13.5 to E18.5, these event statistics include the event rate (B and C), fraction of Rbp4-Cre ROIs showing spontaneous calcium events in each 10-min recording (D and E) and, for each detected calcium event, the size (F and G) and length (H and I) of the event. Circles: event property of each ROI; box (25–75 percentile) and whisker (5–95 percentile); white line: median. Inset: mean ± SEM. (B – E) n = number of Rbp4-Cre ROIs (somas or neurites) recorded from 3 (E13.5), 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos. (F – I) n = number of detected calcium events. Probability: Wilcoxon rank-sum test. Scale bar: 5s (A), 2⋅10−3 Hz (inset, B), 4⋅10−3 Hz (inset, C), 2 ΔF/F (inset, F), 2 ΔF/F (inset, G), 10s (inset, H), 10s (inset, I).
Figure S5
Figure S5
Characterizing calcium activity and migration in embryonic Rbp4-Cre neurons, related to Figure 3 (A and B) Activity in the somas (A) and neurites (B) of Rbp4-Cre neurons (data from Figures 3E and 3F) with neurons recorded in the same embryo colored identically. (C) Changing the anesthetic used in the dam does not change the overall calcium activity in Rbp4-Cre neurons, during both embryonic phases of increased activity. In vivo para-uterine two-photon imaging of Rbp4-Cre neurons was performed at E14.5 (left) and E18.5 (right), to characterize the activity during both active phases, with the dam anesthetized using two different anesthetics: a mixture of Fentanyl-Medetomidine-Midazolam (FMM), and 1.75% Isoflurane. Probability: Wilcoxon rank-sum test. n = number of Rbp4-Cre neurons recorded from 8 (E14.5) and 6 (E18.5) embryos. (D) Activity in Rbp4-Cre neurons’ somas (red) and neurites (blue), in each layer, on each embryonic day. Spatial layers colored as in Figure 2A (blue: superficial layer; gray: intermediate layer; beige: deep layer). Somas and neurites were not assigned to layers at E13.5 (light gray) because of the limited thickness of cortex combined with the limited depth resolution of 2p imaging. Deep layer neurons were poorly sampled >E17.5, because the increasing depth made them difficult to image from with 2p imaging. (E–I) Rbp4-Cre neurons migrate from the deep layer to the intermediate layer. (E) Fraction of all neurons within the intermediate and deep layers, from E14.5 to E18.5, found in each of the two layers. Layers colored as in Figure 2A (gray: intermediate layer; beige: deep layer). (F) In vivo para-uterine time lapse imaging of a population of Rbp4-Cre neurons labeled with tdTomato, at E16.5, over 5 h, in the cortical plate (magenta) and subplate (cyan). White arrow: migrating cell; horizontal dotted line: surface of cortex; yellow line: distance of cell to surface; vertical dotted line: distance migrated. (G) Distribution of migration velocities for all recorded migrating neurons, averaged across 5 h of time lapse imaging. n = 48 neurons from 2 E16.5 embryos. Box (25–75 percentile) and whisker (5–95 percentile); white line (median). (H) Locations of layer 5 neurons in intermediate (gray) and deep (beige) layers (colored based on position at t = 0), followed using para-uterine time lapse imaging for 5 h. Shown as a fraction of the combined cortical plate (CP) and subplate (SP) thickness. Box-and-whiskers: distribution of neuronal locations at the start of imaging (left) and after 300 min (right) as box (25–75 percentile) and whisker (5–95 percentile); black line: median; dotted line: boundary between deep and intermediate layer at t = 0; p value: Wilcoxon rank-sum test; n = 48 neurons from 2 E16.5 embryos. (I) Fraction of Rbp4-Cre neurons within each layer at the start of imaging (left) and after 300 min (right), normalized to the maximum within each layer (data from (H)). 21% of neurons in the deep layer move to the intermediate layer within the imaging period (colored shading), and no neurons move in the reverse direction. Scale bar: 10 μm (F).
Figure 4
Figure 4
Rbp4-Cre neurons display active conductances during both active phases of embryonic development (A) Left: schematic of in vivo two-photon targeted patch clamp recordings from Rbp4-Cre neurons. Right top: imaging field. Rbp4-Cre neurons (red), Alexa Fluor 488 filled pipette (green) and patched neuron (yellow). Right bottom: voltage change with peak and steady state voltage labeled, in response to current injection. (B) Rbp4-Cre neurons’ voltage responses to graded current injections (bottom). (C) Peak versus steady state voltage. n = number of neurons (Figure S6). (D) Calcium activity following TTX application. Wilcoxon signed rank test. n = number of neurons recorded in 3 E14.5 and 3 E18.5 embryos. Scale bars: 20 μm (inset, A), 10 mV (top, B), 50 ms and 40 pA (bottom, B). See also Figure S6.
Figure S6
Figure S6
All Rbp4-Cre neurons display voltage gated sodium channels dependent active conductances, at both E14.5 and E18.5, related to Figure 4 (A–D) All Rbp4-Cre neurons (A, C), but not all nearby unlabeled cells (B, D), display active conductances, at both E14.5, and E18.5. Voltage responses (top) of an Rbp4-Cre neuron and a nearby unlabeled cell at E14.5 (A and B) and E18.5 (C and D) to graded intracellular current injections (middle), recorded in current clamp mode. The voltage responses to current are quantified in both current-voltage curves and the peak voltage compared to the steady-state voltage (as in Figure 4C) for both Rbp4-Cre neurons (C) and nearby unlabeled neurons (D), to visualize any nonlinearity in the peak voltage response. The line type labels the embryonic layer in which each recorded neuron was found (E14.5: Dotted: superficial layer, solid: deep layer; E18.5: Solid: intermediate layer). Due to visibility at E18.5, it was not possible to record from deep layer neurons. n = number of neurons. (E) Left: schematic of para-uterine imaging combined with patch clamp recording to perform in vivo two-photon targeted electrophysiology from Rbp4-Cre neurons together with application of TTX. Right: Voltage responses of an Rbp4-Cre neuron (at E14.5 (top) and E18.5 (bottom)) to graded intracellular current injections, recorded in current clamp mode before (left, pre) and after (right, post) application of TTX. Scale bar: 10 mV (top, A), 50 ms and 20 pA (middle, A), 10 mV (top, B), 50 ms and 20 pA (middle, B), 10 mV (top, C), 50 ms and 20 pA (middle, C), 10 mV (top, D), 50 ms and 20 pA (middle, D), 20 mV (top, E), 10 ms and 40 pA (bottom, E).
Figure 5
Figure 5
Rbp4-Cre neurons have synapses already at E14.5 (A) Expression (circles) of selected genes (Data S1) related to neuronal communication., Radius of circles: fraction of cells expressing the gene; color of circles: mean normalized transcripts per cell (log2). (B) Immunostaining of Rbp4-Cre neurons (green), Snap25 (presynaptic; gray), PSD-95 (postsynaptic; magenta), Hoechst (blue). Right: Zoom (arrows: colocalization of Snap25 and PSD-95). (C and D) DAB staining of Rbp4-Cre neurons in tissue prepared for EM. (E) Synaptic contacts involving Rbp4-Cre neurons (DAB, darker cells), presynaptic vesicles (red arrow) and postsynaptic densities (blue arrowhead). Scale bars: 10 μm (left, B), 2 μm (right, B), 25 μm (C), 25 μm (D), 100 nm (E).
Figure 6
Figure 6
Rbp4-Cre neurons form active circuits already at E14.5 (A) Left: schematic of NMDA+AMPA injection during in vivo embryonic two-photon imaging. Right: Rbp4-Cre neurons; color: normalized calcium activity). (B) Change in fluorescence before (Pre) and after (Post) application of either cortex buffer (blue) or NMDA+AMPA (red). Wilcoxon signed rank test. n = number of Rbp4-Cre neuron ROIs from 3 E14.5 and 3 E18.5 embryos (Figure S7). (C) Pairwise correlations of Rbp4-Cre neurons’ calcium activity, that are significantly greater than random (Figure S7). Shuffled data is on the left. Filled circles: correlations; gray shading: distribution; red line: median. Bars: percent of neuron pairs with correlations significantly greater than random (red). Wilcoxon rank-sum test. n = pairs of Rbp4-Cre neurons recorded from 3 (E13.5), 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos. (D) Immunostaining of Rbp4-Cre neurons (green), Map2 (dendrites, red), NF (axons, white), Hoechst (blue). (E) Left: schematic in vivo para-uterine imaging using 3D acousto-optic two-photon microscope. Top middle: mean projections around each soma. Bottom middle: Three zoomed examples (red outline, top middle). Right: Δf/f activity from examples. Cells 1 and 3 have high correlation. (F) Pairwise correlations of E14.5 Rbp4-Cre neurons’ activity that are significantly greater than random, within and across layers. Dots: pairwise correlations; box (25–75 percentile) and whisker (5–95 percentile); line: median. Scale bars: 10 μm (A), 30 μm (top, D), 100 μm (bottom, D), 20s and 5 %ΔF/F (E). See also Figure S7.
Figure S7
Figure S7
Rbp4-Cre neurons display spontaneous excitatory synaptic potentials, and correlated activity which does not extend across the population, related to Figure 6 (A–F) Spontaneous excitatory synaptic potentials are found in both active phases of embryonic development, at E14.5 (A – C) and E18.5 (D – F). (A and D) Traces recorded in current clamp mode during in vivo two-photon targeted patch clamp recordings from Rbp4-Cre neurons, following the application of TTX to the surface of cortex. Synaptic potentials were defined as deflections greater than 5 mV (dotted line: event threshold). (B and E) Distribution of amplitudes greater than the event threshold. (C and F) Average across events, normalized to the peak voltage of each event, showing the characteristic shape of a synaptic potential. Decay time (τ) shown in ms. Black line: mean; gray shading: SEM. (G and H) Population activity across Rbp4-Cre neurons is less correlated than activity across pairs of Rbp4-Cre neurons. (G) Schematic comparing the “correlation between pairs” of neurons (black), and “correlation with the sum” of neuronal activity (blue) within all neurons recorded in an imaging field (top). Pairwise correlations of neuron 1 with each other neuron among all imaging field (middle). Correlation of the activity of neuron 1 with the sum of activity in all other neurons in the imaging window (bottom). (H) From E14.5 to E18.5, the distribution of correlations between pairs of neurons (that are significantly greater than random) (black shaded area) is compared to the distribution of correlations with the sum (for all recorded neurons) (blue shaded area). Probability: Wilcoxon rank-sum test. Correlations computed from recordings in 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos. (I–M) Pairwise correlations of spontaneous calcium activity, that are significantly greater than random, between Rbp4-Cre neurons, on each embryonic day from E14.5 to E18.5, do not decrease with distance. (N) Combining all pairwise correlations, significantly greater than random, across all embryonic days. (I–N) Red line: best fit trend of correlations across distance (red: R2 quality of fit); n = number of pairs from recordings in 9 (E14.5), 5 (E15.5), 4 (E16.5), 5 (E17.5), and 6 (E18.5) embryos. Scale bars: 5s and 2.5 mV (A), 50 ms and 25% of peak (C), 5s and 2.5 mV (D), 50 ms and 25% of peak (F).
Figure 7
Figure 7
Perturbing autism-associated genes selectively in Rbp4-Cre neurons disrupts circuit organization and activity during embryonic development (A) Expression (circles) of selected genes (Data S2) associated with autism spectrum disorder in the three Rbp4-Cre neuron types and adult L5-PN types. Radius of circles: fraction of cells expressing the gene; color of circles: mean normalized transcripts per cell (log2). (B) Fraction of genes with a mean transcript count greater than the number of transcripts shown on the x axis, for all genes (black), and genes associated with autism spectrum disorder (magenta) in Rbp4-Cre neurons (top) and adult L5-PNs (bottom). Inset: Fold change of autism-associated gene expression compared to all genes in embryos and adult. (C) Immunostaining of cortex of Rbp4-tdTomato-Chd8+/− (top) and Rbp4-tdTomato-Grin2b+/− (bottom) mice (Figure S8). Rbp4-Cre neurons (red), Bcl11b, (white), Hoechst (blue). (D) Normalized depths of Rbp4-Cre neurons (as in Figure 2A) in Rbp4-tdTomato (WT) and the two mutant (top, Chd8+/−; bottom, Grin2b+/−) embryos (Figure S9). 125 neurons from each mouse line, sampled at random. χ2 test. (E) Mating strategy to generate Rbp4-GCaMP6s-tTA2-Chd8+/− (Chd8+/−) and Rbp4-GCaMP6s-tTA2-Grin2b+/− (Grin2b+/−) embryos. (F) Example recordings from Rbp4-Cre neurons’ dendrites in Rbp4-GCaMP6s-tTA2 (WT), Grin2b+/−, and Chd8+/− embryos at E16.5 using 3D acousto-optic two-photon microscope. (G) Distribution of activity in E16.5 embryos, shown in log-scale, for WT and two mutant genotypes. Circles: activity of each neurite; red line: median; shading: distribution. Wilcoxon rank-sum test. n = number of neurites. (H) Immunostaining of local patches of cortical disorganization in Rbp4-tdTomato-Chd8+/− and Rbp4-tdTomato-Grin2b+/− mice, at E18.5. Rbp4-Cre neurons (red), Bcl11b, (white), Hoechst (blue). (I) Fraction of mutant mice, of each genotype, showing at least one patch, summed across E16.5 to E18.5. Red line: Average across all four genotypes. Data from 14 (Rbp4-tdTomato-Chd8+/−), 11 (Rbp4-tdTomato-Chd8−/−), 9 (Rbp4-tdTomato-Grin2b+/−), and 6 (Rbp4-tdTomato-Grin2b−/−) embryos. Fisher’s exact test (p = 0.05, prior to Bonferroni correction). (J) Fraction of neurons within the superficial layer that are located within patches of disorganization, in embryos with at least one patch. n = number of superficial layer neurons on each embryonic day. Scale bars: 20 μm (C), 25s and 25 %ΔF/F (F), 50 μm (H). See also Figures S8 and S9.
Figure S8
Figure S8
Crossing floxed Chd8 and Grin2b mouse lines with Rbp4-Cre results in the conditional knockout of Chd8 and Grin2b in Rbp4-Cre neurons, related to Figure 7 (A) Schematic of strategy to demonstrate Cre-mediated excision and resulting selective knockout of Chd8 and Grin2b in E18.5 Rbp4-tdTomato-Chd8−/− or Rbp4-tdTomato-Grin2b−/− embryos. (B and C) Schematic diagram of Chd8 (B) and Grin2b (C) transcripts (Chd8: Ensembl 212; Grin2b: Ensembl 202), with approximate location of inserted loxP sites, region amplified in D and E (and sequenced in F and G), translational start sites, and translation end sites. Introns not shown to scale. (D and E) Regions within Chd8 (D) and Grin2b (E) genes, including the loxP sites, were PCR amplified and electrophoretically separated. Above: genotype of embryo for Cre and either floxed Chd8 or Grin2b; ref. wt: commercial mouse genomic DNA. Inset: Zoom in (with increased contrast) showing one example of each genotype and examples of longer (red) and shorter (blue) bands, with relative intensity in agreement with the presence of a small fraction of Cre-expressing cells (i.e. weaker band of the shorter fragment) and a large fraction of cells not expressing Cre (i.e. stronger band of the longer fragment) in the cortical samples. (F and G) Aligned sequences of excised bands from D and E. Red: sequence of Chd8 (F) and Grin2b (G) bands from longer fragment (derived from cells not expressing Cre) (yellow overlay: loxP sites); blue: sequence of bands from shorter fragment (derived from cells with Cre expression) (blue box: excised region). (H and I) Left: In situ hybridization of probes against Chd8 (H) or Grin2b (I) (green) and tdTomato (red: marking Rbp4-Cre neurons) in E18.5 Rbp4-tdTomato-Chd8−/− (H) or Rbp4-tdTomato-Grin2b−/− (I) embryos. Example Cre-positive (dashed circles: tdTomato-expressing) and Cre-negative cells (solid circles: lacking tdTomato expression). Right: Number of puncta per cell in Cre-positive compared to Cre-negative cells (mean ± sem). n = 25 cells of each type. (J and K) Left: Immunostaining against Chd8 (J) or Grin2b (K) (green) and tdTomato (red: marking Rbp4-Cre neurons) in E18.5 Rbp4-tdTomato-Chd8−/− (J) or Rbp4-tdTomato-Grin2b−/− (K) embryos. Example Cre-positive (dashed circles: tdTomato-expressing) and Cre-negative cells (solid circles: lacking tdTomato expression). Right: Quantification of change in fluorescence between Cre-positive cells compared to Cre-negative cells (mean ± sem). Fluorescence was normalized by mean fluorescence within Cre-negative neurons in each slice. n = 50 cells (J) and 30 cells (K) of each type. (H–K) Probability: Wilcoxon rank-sum test. Scale bars: 10 μm (H–K).
Figure S9
Figure S9
Perturbing autism-associated genes selectively in Rbp4-Cre neurons disrupts organization of layer 5 during embryonic development, related to Figure 7 (A) Rbp4-Cre neurons (stained using tdTomato antibody, red) in Rbp4-tdTomato-Chd8−/− (top) and Rbp4-tdTomato-Grin2b−/− (blue) mice, from E14.5 to E18.5, within the cortical plate (magenta), subplate (cyan), and intermediate zone (yellow), counterstained with Hoechst (blue). (B) Distribution of Rbp4-Cre neuronal locations as a fraction of the cortical plate and subplate thickness, from E14.5 to E18.5, in control (WT, black), Rbp4-tdTomato-Chd8−/− (top, green), and Rbp4-tdTomato-Grin2b−/− (bottom, orange) mice. 85 neurons from each mouse line, sampled at random, displayed on each embryonic day. Layer boundaries derived from Figure 2A (blue: superficial layer; gray: intermediate layer; beige: deep layer). Probability: χ2 test comparing the fraction of Rbp4-Cre neurons in each layer between the conditional knockout mouse and control mice, on each embryonic day; p = 0.05. (C) Local patches of disorganization in Rbp4-tdTomato-Chd8 and Rbp4-tdTomato-Grin2b conditional knockout (cKO) mice (examples from each embryonic day from E16.5 to E18.5) show Rbp4-Cre neurons (stained using tdTomato antibody, red) at the surface and disrupted intermediate layer, including both neurons expressing Cre (red) and not expressing Cre (stained using Bcl11b antibody, white), counterstained with Hoechst (blue). (D) Superficial layer Rbp4-Cre neurons in E18.5 control mouse, without disorganization of the underlying intermediate layer. Scale bar: Scale bar: 20 μm (A, C, D).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Harris K.D., Shepherd G.M.G. The neocortical circuit: themes and variations. Nat. Neurosci. 2015;18:170–181. doi: 10.1038/nn.3917. - DOI - PMC - PubMed
    1. Kim E.J., Juavinett A.L., Kyubwa E.M., Jacobs M.W., Callaway E.M. Three types of cortical layer 5 neurons that differ in brain-wide connectivity and function. Neuron. 2015;88:1253–1267. doi: 10.1016/j.neuron.2015.11.002. - DOI - PMC - PubMed
    1. Brown A.P.Y., Cossell L., Strom M., Tyson A.L., Vélez-Fort M., Margrie T.W. Analysis of segmentation ontology reveals the similarities and differences in connectivity onto L2/3 neurons in mouse V1. Sci. Rep. 2021;11:4983. doi: 10.1038/s41598-021-82353-7. - DOI - PMC - PubMed
    1. Wertz A., Trenholm S., Yonehara K., Hillier D., Raics Z., Leinweber M., Szalay G., Ghanem A., Keller G., Rózsa B., et al. Single-cell–initiated monosynaptic tracing reveals layer-specific cortical network modules. Science. 2015;349:70–74. doi: 10.1126/science.aab1687. - DOI - PubMed
    1. Vélez-Fort M., Rousseau C.V., Niedworok C.J., Wickersham I.R., Rancz E.A., Brown A.P.Y., Strom M., Margrie T.W. The stimulus selectivity and connectivity of layer six principal cells reveals cortical microcircuits underlying visual processing. Neuron. 2014;83:1431–1443. doi: 10.1016/j.neuron.2014.08.001. - DOI - PMC - PubMed

Publication types

LinkOut - more resources