Precise Long-Range Microcircuit-to-Microcircuit Communication Connects the Frontal and Sensory Cortices in the Mammalian Brain

Abstract

The frontal area of the cerebral cortex provides long-range inputs to sensory areas to modulate neuronal activity and information processing. These long-range circuits are crucial for accurate sensory perception and complex behavioral control; however, little is known about their precise circuit organization. Here we specifically identified the presynaptic input neurons to individual excitatory neuron clones as a unit that constitutes functional microcircuits in the mouse sensory cortex. Interestingly, the long-range input neurons in the frontal but not contralateral sensory area are spatially organized into discrete vertical clusters and preferentially form synapses with each other over nearby non-input neurons. Moreover, the assembly of distant presynaptic microcircuits in the frontal area depends on the selective synaptic communication of excitatory neuron clones in the sensory area that provide inputs to the frontal area. These findings suggest that highly precise long-range reciprocal microcircuit-to-microcircuit communication mediates frontal-sensory area interactions in the mammalian cortex.

Keywords: columnar microcircuit; cortical circuit; excitatory neuron clone; in utero retroviral labeling; long-range circuit; quadruple whole-cell recording; rabies virus tracing; top-down modulation.

Figure 1:. Identification of the presynaptic input neurons to individual excitatory neuron clones in S1 as a unit.

(A) Schematic of the experimental procedure. (B) 3D whole brain reconstruction of a representative starter excitatory neuron clone in S1 barrel field (bfd) area (violet) expressing both DsRed and EGFP (arrow). (C) Confocal images of the radially distributed starter excitatory neuron clone in B. The numbers (1-10) and arrows indicate individual starter excitatory neurons in the starter clone. Colored lines indicate the pia and laminar boundaries. Similar display and layout are used in subsequent figures. Scale bars: 100 μm and 10 μm. (D) 3D reconstruction of the starter excitatory neuron clone in C. Similar display and layout are used in subsequent figures. (E) Confocal images of the EGFP-expressing (green) long-range presynaptic input neurons stained with DAPI (blue). White arrowheads indicate presynaptic cells in M2. ACC, anterior cingulate cortex. Scale bars: 1 mm. (F) High magnification images of the long-range presynaptic neurons. Similar display and layout are used in subsequent figures. Scale bars: 100 μm. (G) 3D whole-brain reconstruction of the presynaptic neurons (green) to a starter excitatory neuron clone in S1 bfd area (yellow). RSP, retrosplenial cortex; VIS, visual cortex; HY, hypothalamus; PAL, pallidum. (H) Quantification of the numbers of neurons in individual excitatory neuron clones that were infected by RVdG (yellow) and thereby served as the starter neurons (n = 11). (I) Quantification of the laminar distribution of excitatory neurons in individual starter excitatory neuron clones in S1 (superficial layer, 51.8 ± 5.5%; deep layer, 48.2 ± 5.5%; n = 11; p = 0.6; unpaired two-tailed t test). (J) Quantification of the percentage of neurons in individual excitatory neuron clones that were yellow starter neurons. (K) Quantification of numbers of long-range input neurons per starter neuron in different brain regions. PTLp, posterior parietal association areas; TEA, temporal association area; CLA, claustrum; MB, midbrain. See also Figure S1.

Figure 2:. Spatial organization of frontal input neurons in M2.

(A) Confocal images of serial sections of M2 harboring EGFP-expressing input neurons (green) to an excitatory neuron clone in S1 stained with DAPI (blue). Scale bar: 100 μm. (B) 3D rendered images of the input neurons in A (green, top) and the randomly simulated cells in the same space (gray, bottom). (C) Quantification of the percentage of input neurons in superficial and deep layers in M2 in the experimental (n = 8) and randomly simulated datasets (unpaired two-tailed t test). (D) Quantification of the angular orientation of individual input neuron pairs in M2 relative to the pia in the experimental (n = 8) and randomly simulated datasets (unpaired two-tailed t test). See also Figure S2.

Figure 3:. Preferential synaptic connectivity between frontal input neurons in M2.

(A) Representative reconstruction images of a starter excitatory neuron clone in S1 barrel field (bfd) area (yellow) and its presynaptic neurons (green). Input neurons and adjacent non-input neurons in M2 subjected to quadruple whole-cell recording are marked as dark green and black circles, respectively. HPF, hippocampal formation. Similar display and layout are used in subsequent figures. (B) Representative confocal images of a pair of recorded input excitatory neurons (1 and 3) and two adjacent non-input excitatory neurons (2 and 4) in M2. Note the dendritic spines on the dendrites. Similar display and layout are used in subsequent figures. Scale bars: 100 μm and 20 μm. (C) Morphological reconstruction of the four recorded neurons in B. Yellow and black arrows indicate the chemical synaptic connections. Similar display and layout are used in subsequent figures. (D) Sample traces of the four recorded neurons. Average presynaptic neuron traces are shown in red and average postsynaptic neuron traces are shown in black. Individual traces are shown in gray. Black arrows and broken line boxes indicate the reliable postsynaptic responses. Similar display and layout are used in subsequent figures. Scale bars: 100 mV and 200 msec (red); 10 pA and 200 msec (black). (E) High-magnification average sample traces showing the reliable postsynaptic responses. Scale bars, 50 mV and 50 msec (red); 5 pA and 50 msec (black). (F) Quantification of the chemical connection rate of top-down input neuron pairs, input/non-input neuron pairs and non-input/non-input neuron pairs in M2 (chi-square test). Numbers of recording neuronal pairs are shown on the graph based on 61 quadruple, 44 triple, and 24 dual recordings (from 16 different brains). (G) Quantification of the chemical synaptic connection rate at different inter-soma distances (chi-square test or chi-square test for trend). (H) Quantification of the chemical synaptic connection rate in M2 of individual excitatory neuron clones located in various S1 regions (bfd, barrel field; ll/tr, low limb and trunk; ul, up limb). See also Figures S3, S4 and S5.

Figure 4:. Spatial organization of contralateral input neurons in c-S1.

(A) Confocal images of serial sections of c-S1 harboring EGFP-expressing transcallosal input neurons to an excitatory neuron clone in S1 bfd area. Scale bar: 200 μm. (B) 3D rendered images of the input neurons (green, top) in A and the randomly simulated cells (gray, bottom) in the same space. (C) Quantification of the percentage of input neurons in the superficial and deep layers in the experimental (n = 8) and randomly simulated datasets (unpaired two-tailed t test). (D) Quantification of the angular orientation of individual contralateral input neuron pairs relative to the pia in the experimental (n = 8) and randomly simulated datasets (unpaired two-tailed t test).

Figure 5:. No preferential synaptic connectivity between contralateral input neurons in c-S1.

(A) Representative reconstruction images of a starter excitatory neuron clone in S1 ul area (yellow) and its presynaptic neurons (green). (B) Confocal images of a pair of recorded input excitatory neurons (1 and 3) and two adjacent non-input excitatory neurons (2 and 4) in c-S1. Scale bars: 100 μm and 20 μm. (C) Morphological reconstruction of the four recorded neurons in B. (D) Sample traces of the four recorded neurons. Scale bars: 100 mV and 200 msec (red); 20 pA and 200 msec (black). (E) Quantification of the chemical synaptic connection rate in c-S1 (chi-square test). See also Figure S5.

Figure 6:. Preferential M2 input neuron connectivity depends on the clonality of starter excitatory neurons.

(A) Representative reconstruction images of a randomly labeled starter excitatory neuron vertical cluster in S1 bfd area (yellow) and its presynaptic input neurons (green). (B) Confocal images of a pair of recorded input excitatory neurons (1 and 3) and two adjacent non-input excitatory neurons (2 and 4) in M2. Scale bars: 100 μm and 20 μm. (C) Morphological reconstruction of the four recorded neurons in B. Black arrow indicates the chemical synaptic connection. (D) Sample traces of the four recorded neurons. Black arrows and broken line boxes indicate the reliable postsynaptic responses elicited by presynaptic action potentials (APs). Scale bars: 100 mV and 200 msec (red); 20 pA and 200 msec (black). (E) High-magnification average sample traces showing the reliable postsynaptic responses. Scale bars, 50 mV and 50 msec (red); 5 pA and 50 msec (black). (F) Quantification of the chemical synaptic connection rate in M2 (chi-square test). See also Figure S6.

Figure 7:. Preferential M2 input neuron connectivity depends on synaptic communication in starter excitatory neuron clones.

(A) Schematic of the experimental procedure. (B) Representative reconstruction images of the control starter excitatory neuron clone in S1 bfd areas (yellow) and their input neurons in M2 (green). (C) Morphological reconstruction of the four recorded neurons in D. Yellow arrow indicates the chemical synaptic connection. (D) Representative confocal images of a pair of recorded input neurons (1 and 3) and two adjacent non-input excitatory neurons (2 and 4) in M to the control starter excitatory neuron clone. Scale bars: 100 μm and 20 μm. (E) Sample traces of the four recorded neurons in D. Black arrows and broken line boxes indicate the reliable postsynaptic responses. Scale bars: 50 mV and 200 msec (red); 20 pA and 200 msec (black). (F) High-magnification average sample traces showing the reliable postsynaptic responses between neurons 1 and 3. Scale bars: 25 mV and 25 msec (red); 5 pA and 25 msec (black). (G) Representative reconstruction images of a TeLC-expressing starter excitatory neuron clone in S1 bfd area (yellow) and its presynaptic neurons (green). (H) Morphological reconstruction of the four recorded excitatory neurons in I. (I) Confocal images of a pair of recorded input neurons (1 and 3) and two adjacent non-input neurons (2 and 4) in M2 to the TeLC-expressing starter excitatory neuron clone in S1. Scale bars: 100 μm and 20 μm. (J) Sample traces of the four recorded neurons in I. Scale bars: 100 mV and 200 msec (red); 20 pA and 200 msec (black). (K) Quantification of the chemical synaptic connection rate in M2 of the control and TeLC-expressing starter excitatory neuron clone groups (chi-square test). See also Figures S7 and S8.

Figure 8:. Preferential synaptic connectivity between bottom-up input neurons in S1 to individual M2 clones.

(A) Representative reconstruction images of a starter excitatory neuron clone in M2 (yellow) and its presynaptic input neurons (green). (B) Representative confocal images of a pair of recorded input excitatory neurons (green, 1 and 3) and two adjacent non-input excitatory neurons (2 and 4) in S1. Scale bars: 100 μm and 10 μm. (C) Morphological reconstruction of the four recorded excitatory neurons in B. Yellow arrow indicates the chemical synaptic connection. (D) Sample traces of the four recorded excitatory neurons. Black arrows and broken line squares indicate the reliable postsynaptic responses. Scale bars: 100 mV and 200 msec (red); 10 pA and 200 msec (black). (E) High-magnification average sample traces showing the reliable postsynaptic responses. Scale bars, 50 mV and 50 msec (red); 5 pA and 50 msec (black). (F) Quantification of the chemical connection rate in S1 (chi-square test). (G) Precise microcircuit to microcircuit communication between the frontal and sensory areas of the mammalian cortex. Our data suggest that highly specific microcircuit to microcircuit communication exists between the frontal and sensory areas, but not between the transcallosal or ipsilateral cross-modal sensory areas, and that precise microcircuit formation in the frontal area depends on the preferential synaptic communication of the microcircuit in the sensory area.