Laminarly orthogonal excitation of fast-spiking and low-threshold-spiking interneurons in mouse motor cortex

Abstract

In motor cortex, long-range output to subcortical motor circuits depends on excitatory and inhibitory inputs converging on projection neurons in layers 5A/B. How interneurons interconnect with these projection neurons, and whether these microcircuits are interneuron and/or projection specific, is unclear. We found that fast-spiking interneurons received strong intralaminar (horizontal) excitation from pyramidal neurons in layers 5A/B including corticostriatal and corticospinal neurons, implicating them in mediating disynaptic recurrent, feedforward, and feedback inhibition within and across the two projection classes. Low-threshold-spiking (LTS) interneurons were instead strongly excited by descending interlaminar (vertical) input from layer 2/3 pyramidal neurons, implicating them in mediating disynaptic feedforward inhibition to both projection classes. Furthermore, in a novel pattern, lower layer 2/3 preferentially excited interneurons in one layer (5A/LTS) and excitatory neurons in another (5B/corticospinal). Thus, these inhibitory microcircuits in mouse motor cortex follow an orderly arrangement that is laminarly orthogonalized by interneuron-specific, projection-nonspecific connectivity.

Figure 1.

Excitatory inputs to FS interneurons. A, Bright-field (BF) (left) and epifluorescence (right) images of a G42 transgenic mouse line motor cortex slice. B, High-resolution image of biocytin-labeled FS interneurons. C, Train of action potentials recorded in a GFP-positive cell during step current injection (0.5 s, 200 pA pulse). Inset, Single action potential. D, Example of excitatory input traces and map recorded from a layer 5A FS interneuron. E, Same, for a layer 5B FS interneuron. F, Side-view projection of FS interneuron input maps (n = 33), ordered by soma distance from the layer 5A/B border. Maps were projected onto one plane by averaging along the rows of the individual maps. Different symbols mark the absolute distances from the pia to the soma (white circles), the layer 5A/5B border (gray dashes), and layer 6/white matter (WM) border (black circles). Plots on the right show the average amount of input to layer 5A (lighter blue) and layer 5B (darker blue) FS interneurons (mean ± SEM). G, Average excitatory input from layer 2/3 (left) and 5B (right) as a function of soma location relative to the layer 5A/5B border (pia is leftward and white matter is rightward). H, Average maps of FS interneurons in layers 5A (n = 15; left) and 5B (n = 18; middle), and difference map (right), generated by subtracting the 5B map from the 5A map. The brackets indicate regions of interest (ROIs). I, Mean amplitude of layer 2/3 (left) and layer 5B (right) excitatory input to layer 5A and 5B FS interneurons, for the ROIs bracketed in the maps. *p < 0.01.

Figure 2.

Excitatory inputs to LTS interneurons. A, Bright-field (BF) (left) and epifluorescence (right) images of a GIN transgenic mouse line motor cortex slice. B, Left, Bright-field image of biocytin-stained cells. Right, Morphological reconstructions of two LTS interneurons (dendrites, red; axons, blue). C, Train of action potentials recorded in a GFP-positive cell during step current injection (0.5 s, 50 pA pulse). Inset, Single action potential. D, Example of excitatory input traces and map recorded from a layer 5A LTS interneuron. E, Example of excitatory input traces and map recorded from a layer 5B LTS interneuron. F, Side-view projection of LTS input maps (n = 31). Symbols are as defined in Figure 1F. G, Average excitatory input from layer 2/3 (left) and 5B (right) as a function of soma location relative to the layer 5A/5B border. H, Average maps of LTS interneurons in layer 5A (left; n = 17) and 5B (middle; n = 14), and difference map (right; 5B map minus 5A map). The brackets indicate ROIs. Note that here the layer 2/3 ROI is centered on the lower part of layer 2/3. I, Mean amplitude of layer 2/3 (left) and layer 5B (right) excitatory input to layer 5A and 5B LTS interneurons, for the ROIs bracketed in the maps. *p < 0.05.

Figure 3.

Comparison of FS and LTS interneurons, and of excitatory inputs to inhibitory and excitatory neurons in layers 5A and 5B. A, Difference map, calculated by subtracting the average map for layer 5A FS (n = 15) from that of LTS (n = 17) interneurons. B, ROI-averaged excitatory input from layers 2/3 (for an ROI spanning all of layer 2/3), 5A, and 5B to FS and LTS interneurons in layer 5A. ROIs are marked by brackets in the maps. *p ≤ 0.05. C, Difference map, calculated by subtracting the average map for layer 5B FS (n = 18) from that of LTS (n = 14) interneurons. D, ROI-averaged excitatory input from layers 2/3, 5A, and 5B to FS and LTS interneurons in layer 5B. ROIs are marked by brackets in the maps. *p < 0.05. E, Average maps of layer 5A corticostriatal (n = 16) and layer 5B corticospinal (n = 12) neurons. ROIs are marked by brackets in the maps. The layer 2/3 ROI was centered onto the lower layer 2/3 as indicated by the brackets. F, Average excitatory inputs (±SEM) to layer 5A corticostriatal neurons (red), layer 5A LTS interneurons (black), layer 5B corticospinal neurons (blue), and layer 5B LTS interneurons (gray) as a function of distance of the stimulus location (i.e., location of presynaptic neurons) from the pia. G, Schematic depicting the laminar pattern of excitatory output from lower layer 2/3 to LTS interneurons in layer 5A and corticospinal neurons in layer 5B.

Figure 4.

Interlaminar versus intralaminar driving of FS and LTS interneurons. A, Left, Bright-field images of motor cortex slices prepared from the G42 (left) or the GIN (right) mouse line, showing the locations of the recording interneurons (FS, circle; LTS, diamond), layer 2/3 stimulation sites (asterisks), and examples of responses recorded in cell-attached mode; the LTS but not the FS interneuron fired an action potential (arrow). B, Average excitation profiles for FS (left; n = 8) and LTS (right; n = 7) interneurons. Grid spacing, 100 μm. Plots on the right show the average number of action potentials to layer 5A/5B FS (circles) and layer 5A/5B LTS (diamonds) interneurons (mean ± SEM). C, Latency to first action potential in response to glutamate uncaging, recorded in FS and LTS interneurons, for layer 5A/B (left) and layer 2/3 (right) photostimulation locations (symbols: individual data points for each cell; bars: group means ± SEM).

Figure 5.

Latency of the layer 2/3 inhibitory inputs to corticostriatal and corticospinal neurons. A, Example image showing several biocytin-labeled corticospinal neurons in layer 5B of motor cortex. B, Examples of EPSCs (green trace) recorded at −70 mV and IPSCs (red trace) recorded at 0 mV from a layer 5B corticostriatal (top) and corticospinal (bottom) neurons. C, Example of inhibitory input traces recorded from layer 5B corticostriatal (top) and corticospinal (bottom) neurons. Dashed ellipse, Region in layer 2/3 where IPSCs were evoked. D, Example traces showing latency differences between onset of IPSCs (red trace) versus EPSCs (green trace) for corticostriatal (top) and corticospinal (bottom) neurons. E, Onset latencies recorded in corticostriatal (n = 15; left) and corticospinal (n = 10; right) neurons, for EPSCs (green circles) and IPSCs (red circles). The filled circles show group averages (±SEM). Right, Plot of IPSC–EPSC latency differences calculated for individual IPSC–EPSC pairs, for corticostriatal and corticospinal neurons, including group averages (±SEM).

Figure 6.

CRACM analysis of layer 2/3 disynaptic inhibition onto layers 5A and 5B corticostriatal neurons and layer 5B corticospinal neurons. A, Epifluorescence image (left) of a mouse motor cortex slice showing expression of ChR2-Venus in layer 2/3 (dendrites and axons of transfected layer 2/3 pyramidal neurons) and layer 5A/B (axons). Example array of traces (middle) showing excitatory (green traces) and inhibitory (red traces) responses. Traces to right show EPSC (green trace) and IPSC (red trace) before and after application of ionotropic glutamate receptor antagonists (5 μm CPP, 10 μm NBQX: black trace). B, Average maps of layer 2/3 inhibitory input to layers 5A and 5B corticostriatal neurons and layer 5B corticospinal neurons. C, Onset latencies of EPSCs (green circles) and IPSCs (red circles) of corticostriatal (n = 5; left) and corticospinal (n = 5; right) neurons. The filled circles show group averages (±SEM). Right, Plot of IPSC–EPSC latency differences calculated for individual IPSC–EPSC pairs, for corticostriatal and corticospinal neurons, including group averages (±SEM).

Figure 7.

RV-ChR2 photostimulation-evoked excitation from corticospinal or corticostriatal neurons to FS and LTS cells. A, Epifluorescence image of a mouse motor cortex slice showing retrogradely labeled corticostriatal neurons following injection of RV-ChR2 into contralateral striatum. Right, Photostimulation-evoked action potential recorded from ChR2+ corticostriatal neuron, in response to 5 ms pulse of light from a blue LED (top trace). B, Same as in A, but for corticospinal neurons. Corticospinal neurons were retrogradely transfected by injection of RV-ChR2 into spinal cord. C, Experimental paradigm for photostimulating ChR2+ projection neurons while recording responses in FS and LTS interneurons. These experiments were performed in the GIN mouse line. D, Examples showing morphology and firing patterns of FS (top left) and LTS (bottom left) interneurons, as seen in bright-field microscopy during patch recordings, and EPSPs recorded in layer 5B FS (top right) and LTS interneurons (bottom right) during ChR2-corticostriatal photostimulation. E, Example showing an action potential recorded in an FS (top left) but not an LTS (bottom left) interneuron during ChR2-corticostriatal photostimulation. Right, Plot of photoevoked corticostriatal output to pairs of layer 5B FS and LTS interneurons. Pairs are connected by lines; group averages (±SEM) are plotted to either side. *p = 0.005. F, Same as in E, for ChR2-corticospinal photostimulation. *p < 0.01. G, Normalized version of the group data in E and F. H, Schematic depicting major excitatory pathways to FS and LTS interneurons.

Figure 8.

RV-ChR2 photostimulation-evoked disynaptic inhibition between corticospinal and corticostriatal neurons. A, Experimental paradigm for stimulating ChR2+ projection neurons while recording from untransfected (ChR2−) neurons identified as corticostriatal or corticospinal by anatomical retrograde labeling (bead+), to evaluate disynaptic inhibition. B, ChR2-corticostriatal photostimulation evoked a small EPSC (green traces) and large IPSC (red traces) in bead-labeled corticostriatal (left) and corticospinal (right) neurons. The excitatory and inhibitory currents were abolished by application of ionotropic glutamate receptor antagonists (5 μm CPP, 10 μm NBQX). C, Amplitudes of IPSPs (left) and EPSPs (right) recorded in corticospinal and corticostriatal neurons while photostimulating ChR2-expressing corticostriatal neurons. D, EPSPs (green traces) and IPSCs (red traces) recorded in a layer 5B corticostriatal neuron during ChR2-corticospinal photostimulation (top), and in a layer 5B corticospinal neuron during ChR2-corticospinal photostimulation (bottom). E, Amplitudes of IPSPs (left) and EPSPs (right) recorded in corticospinal and corticostriatal neurons while photostimulating ChR2-expressing corticospinal neurons. F, IPCS/EPSC ratio of photoevoked corticostriatal output to layer 5B corticospinal (n = 6), and IPCS/EPSC ratio of photoevoked corticospinal output to layer 5B corticostriatal (n = 4) neurons. Group averages (±SEM) are plotted to either side. *p < 0.05. G, Schematic depicting disynaptic inhibitory pathways between pyramidal neurons.