Regulation of PV interneuron plasticity by neuropeptide-encoding genes

Abstract

Neuronal activity must be regulated in a narrow permissive band for the proper operation of neural networks. Changes in synaptic connectivity and network activity-for example, during learning-might disturb this balance, eliciting compensatory mechanisms to maintain network function^1-3. In the neocortex, excitatory pyramidal cells and inhibitory interneurons exhibit robust forms of stabilizing plasticity. However, although neuronal plasticity has been thoroughly studied in pyramidal cells^4-8, little is known about how interneurons adapt to persistent changes in their activity. Here we describe a critical cellular process through which cortical parvalbumin-expressing (PV⁺) interneurons adapt to changes in their activity levels. We found that changes in the activity of individual PV⁺ interneurons drive bidirectional compensatory adjustments of the number and strength of inhibitory synapses received by these cells, specifically from other PV⁺ interneurons. High-throughput profiling of ribosome-associated mRNA revealed that increasing the activity of a PV⁺ interneuron leads to upregulation of two genes encoding multiple secreted neuropeptides: Vgf and Scg2. Functional experiments demonstrated that VGF is critically required for the activity-dependent scaling of inhibitory PV⁺ synapses onto PV⁺ interneurons. Our findings reveal an instructive role for neuropeptide-encoding genes in regulating synaptic connections among PV⁺ interneurons in the adult mouse neocortex.

Fig. 1. Inhibitory synapses mediate the homeostatic response of PV⁺ interneurons to increased activity.

a, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, FOS and PV expression in coronal sections through S1. c, Quantification of FOS staining intensity in hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice (vehicle, n = 3 mice; CNO, n = 3 mice; two-tailed Student’s t-test, P = 0.02). a.u., arbitrary units. d, Traces of mEPSCs recorded from hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice. e, Amplitude (vehicle, n = 13 cells, 10 slices, 5 mice; CNO, n = 18 cells, 15 slices, 8 mice; two-tailed Student’s t-test, P = 0.71) and frequency (vehicle, n = 13 cells, 10 slices, 5 mice; CNO, n = 18 cells, 15 slices, 8 mice; two-tailed Student’s t-test, P = 0.48) of mEPSCs recorded from hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice. f, Traces of mIPSCs recorded from hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice. g, Amplitude (vehicle, n = 13 cells, 10 slices, 5 mice; CNO, n = 18 cells, 15 slices, 8 mice; two-tailed Student’s t-test, P = 0.003) and frequency (vehicle, n = 13 cells, 10 slices, 5 mice; CNO, n = 18 cells, 15 slices, 8 mice; Mann–Whitney U test, P = 0.001) of mIPSCs recorded from hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice. h, E/I ratio in hM3Dq-mCherry-infected PV⁺ interneurons in vehicle- and CNO-treated mice (vehicle, n = 13 cells, 10 slices, 5 mice; CNO, n = 18 cells, 15 slices, 8 mice; two-tailed Student’s t-test, P = 0.005). i, Presynaptic VGaT⁺ puncta and postsynaptic gephyrin⁺ clusters in infected and uninfected PV⁺ interneurons in vehicle- and CNO-treated mice. Insets show PV and mCherry immunoreactivity in cell somata. j, Quantification of the change in synaptic density between infected and uninfected PV⁺ interneurons in vehicle- and CNO-treated mice (vehicle, n = 8 mice; CNO, n = 9 mice; two-tailed Student’s t-test, P = 0.008). Data are mean ± s.e.m. Scale bar, 50 µm (b) and 1 µm (i). Source data

Fig. 2. Homeostatic inhibition originates from PV⁺ interneurons.

a, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, Recording and stimulation configuration in Pvalb^cre/Flp;RCL^Chr2/+ mice. c, Traces of evoked inhibitory postsynaptic currents (eIPSCs) recorded from hM3Dq⁺ PV⁺ interneurons following full-field stimulation of PV⁺ interneurons in vehicle- and CNO-treated mice. d, Peak amplitude (vehicle, n = 13 cells, 13 slices, 7 mice; CNO, n = 11 cells, 11 slices, 7 mice; two-tailed Student’s t-test, P = 0.02) and charge (vehicle, n = 13 cells, 13 slices, 7 mice; CNO, n = 11 cells, 11 slices, 7 mice; two-tailed Student’s t-test, P = 0.04) of LED-evoked eIPSCs. e, Recording and stimulation configuration in Sst^cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. f, Traces of eIPSCs recorded from hM3Dq⁺ PV⁺ interneurons following full-field stimulation of SST⁺ interneurons in vehicle- and CNO-treated mice. g, Peak amplitude (vehicle, n = 10 cells, 10 slices, 6 mice; CNO, n = 11 cells, 11 slices, 5 mice; two-tailed Student’s t-test, P = 0.64) and charge (vehicle, n = 10 cells, 10 slices, 6 mice; CNO, n = 11 cells, 11 slices, 5 mice; two-tailed Student’s t-test, P = 0.48) LED-evoked eIPSCs. h, Recording and stimulation configuration in Vip^cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. i, Traces of eIPSCs recorded from hM3Dq⁺ PV⁺ interneurons following full-field stimulation of VIP⁺ interneurons in vehicle- and CNO-treated mice. j, Peak amplitude (vehicle, n = 13 cells, 12 slices, 6 mice; CNO, n = 12 cells, 10 slices, 5 mice; two-tailed Student’s t-test, P = 0.67) and charge (vehicle, n = 13 cells, 12 slices, 6 mice; CNO, n = 12 cells, 10 slices, 5 mice; two-tailed Student’s t-test, P = 0.25) LED-evoked eIPSCs. Data are mean ± s.e.m. Source data

Fig. 3. Increased activity in PV⁺ interneurons leads to upregulation of Scg2 and Vgf.

a, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, Volcano plot showing ribosome-associated mRNAs identified through RNA-seq. Differentially expressed RNAs (fold change > 1.5, P < 0.05, two-sided Wald’s test with adjustment for multiple testing) are labelled in teal. c, Ranking of the top DEGs on the basis of four selection criteria (Methods). Darker shades indicate higher scores (values 0 to 1). d, Partial STRING network of the most prominent node of DEGs, highlighting IEGs, and genes encoding secreted and membrane-bound proteins. e, Scg2 and Vgf expression in neighbouring hM3Dq⁺ and uninfected PV⁺ interneurons in layer 2/3 of S1 from CNO-treated mice. Scale bar, 10 µm. f, Expression of Scg2 (n = 6 mice; two-tailed paired Student’s t-test, P = 0.01) and Vgf (n = 5 mice; two-tailed paired Student’s t-test, P = 0.01) in neighbouring hM3Dq⁺ and uninfected PV⁺ interneurons from CNO-treated mice. Data are mean ± s.e.m. CoV, coefficient of variation. Source data

Fig. 4. VGF is necessary and sufficient for scaling PV–PV connectivity.

a, Experimental strategy. b, Presynaptic SYT2⁺ puncta and postsynaptic gephyrin⁺ clusters in shLacZ, shScg2 and shVgf expressing PV⁺ interneurons in vehicle- and CNO-treated mice. Insets show PV and mCherry immunoreactivity in cell somata. Scale bar, 1 µm. c, Quantification of the change in synaptic density between infected and uninfected PV⁺ interneurons in CNO-treated mice injected with hM3Dq-shLacZ (n = 10 mice), hM3Dq-shScg2 (n = 9 mice, two-tailed one-sample t-test, P = 0.049) and hM3Dq-shVgf (n = 8 mice, two-tailed one-sample t-test, P = 0.003). d, Experimental strategy. e, Presynaptic SYT2⁺ puncta and postsynaptic gephyrin⁺ clusters in mCherry- and Vgf-mCherry-expressing PV⁺ interneurons. Insets show PV and mCherry immunoreactivity in cell somata. Scale bar, 1 µm. f, Quantification of change in synaptic density between neighbouring uninfected and infected PV⁺ interneurons in mice injected with Vgf-mCherry (uninfected, n = 5 mice; CNO, n = 5 mice; two-tailed Student’s t-test, P = 0.003). Data are mean ± s.e.m. Schematics in a,d adapted from ref. (reprinted with permission from AAAS) and ref. . Source data

Fig. 5. Contextual fear conditioning increases Vgf expression and PV–PV connectivity in activated PV⁺ interneurons.

a, Experimental strategy. b, Presynaptic SYT2⁺ puncta and postsynaptic gephyrin⁺ clusters on FOS⁻ and FOS⁺ PV⁺ cells 2 h after cFC. Insets show PV and FOS immunoreactivity in cell somata. c, Quantification of PV⁺ synaptic density onto FOS⁻ and FOS⁺ PV⁺ interneurons 2 h after cFC (FOS⁻, n = 5 mice; FOS⁺, n = 5 mice; two-tailed Student’s t-test, P = 0.044). d, Experimental strategy. e, Vgf expression 1 h after cFC in activated (tdTomato⁺) and neighbouring, non-activated (tdTomato⁻) PV⁺ interneurons. f, Quantification of Vgf intensity in tdTomato⁻ and tdTomato⁺ PV interneurons (tdTomato⁻, n = 5 mice; tdTomato⁺, n = 5 mice; two-tailed paired Student’s t-test, P = 8 × 10^-4). g, Presynaptic SYT2⁺ puncta and postsynaptic gephyrin⁺ clusters in tdTomato⁻ and tdTomato⁺ PV⁺ interneurons 24 and 72 h after cFC. h, Quantification of the change in synaptic density between activated (tdTomato⁺) and non-activated (tdTomato⁻) PV⁺ interneurons 24 and 72 h after cFC. 24 h: n = 5 mice; 72 h: n = 5. tdTomato⁻ versus tdTomato⁺ at 24 h: two-tailed paired Student’s t-test, P = 0.001; tdTomato⁺ at 24 h versus tdTomato⁺ at 72 h: two-tailed Student’s t-test, P = 0.033. Data are presented as mean ± s.e.m. In box plots, the centre line indicates the median, box limits delineate middle quartiles, and whiskers indicate maximum and minimum. Scale bars: 10 µm (e) and 1 µm (b,g). Schematics in a,d adapted from ref. . Source data

Extended Data Fig. 1. hM3Dq infection density and CNO control.

a, mCherry and PV expression in coronal sections through S1. b, quantification of the percentage of mCherry⁺ PV⁺ interneurons in layer 2/3 (n = 6 mice). c, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . d, Traces of mEPSCs recorded from mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. e, Quantification of amplitude (vehicle, n = 9 cells, 9 slices, 3 mice; CNO, n = 10 cells, 6 slices, 3 mice; two-tailed Student’s t-test, p = 0.51) and frequency (vehicle, n = 9 cells, 9 slices, 3 mice; CNO, n = 10 cells, 6 slices, 3 mice; two-tailed Student’s t-test, p = 0.45) of mEPSCs recorded from mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. f, Traces of mIPSCs recorded from mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. g, Quantification of amplitude (vehicle, n = 9 cells, 9 slices, 3 mice; CNO, n = 10 cells, 6 slices, 3 mice; two-tailed Student’s t-test, p = 0.64) and frequency (vehicle, n = 9 cells, 9 slices, 3 mice; CNO, n = 10 cells, 6 slices, 3 mice; two-tailed Student’s t-test, p = 0.84) of mIPSCs recorded from mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. h, Excitation/Inhibition (E/I) ratio in mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice (vehicle, n = 9 cells; CNO, n = 10 cells; two-tailed Student’s t-test, p = 0.24). Data are presented as mean ± s.e.m. Scale bar, 50 µm. Source data

Extended Data Fig. 2. Bidirectional regulation of PV+ interneuron activity has opposite effects on inhibition.

a, Presynaptic VGaT⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and hM3Dq⁺ PV⁺ interneurons in vehicle-treated mice. Inserts show PV and mCherry immunoreactivity in cell somata. b, Quantification of the density of VGaT + /Gephyrin+ puncta on neighbouring uninfected and hM3D⁺ PV⁺ interneurons in vehicle-treated mice (n = 8; two-tailed paired Student’s t-test, p = 0.86). c, Presynaptic VGaT+ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and hM3Dq-mCherry+ PV+ interneurons in CNO-treated mice. Inserts show PV and mCherry immunoreactivity in the cell somata. d, Quantification of the density of VGaT + /Gephyrin+ puncta on neighbouring uninfected and hM3Dq-mCherry+ PV+ interneurons in CNO-treated mice (n = 9; two-tailed paired Student’s t-test, p = 8×10⁻⁴). e, Quantification of the density of VGaT⁺/Gephyrin⁺ puncta on neighbouring uninfected PV+ interneurons in vehicle-treated and CNO-treated mice (Vehicle: n = 8; CNO: n = 9; two-tailed Student’s t-test, p = 0.44). f, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . g, Traces of mEPSCs recorded from hM4Di-mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. h, Quantification of amplitude (vehicle, n = 15 cells, 14 slices, 7 mice; CNO, n = 16 cells, 12 slices, 7 mice; two-tailed Student’s t-test, p = 0.20) and frequency (vehicle, n = 15 cells, 14 slices, 7 mice; CNO, n = 16 cells, 12 slices, 7 mice; two-tailed Student’s t-test, p = 0.46) of mEPSCs recorded from hM3Dq-mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. i, Traces of mIPSCs recorded from hM4Di-mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. j, Quantification of amplitude (vehicle, n = 15 cells, 14 slices, 7 mice; CNO, n = 16 cells, 12 slices, 7 mice; two-tailed Student’s t-test, p = 0.14) and frequency (vehicle, n = 15 cells, 14 slices, 7 mice; CNO, n = 16 cells, 12 slices, 7 mice; two-tailed Student’s t-test, p = 0.03) of mIPSCs recorded from hM3Dq-mCherry-infected PV+ interneurons in vehicle- and CNO-treated mice. k, Excitation/Inhibition (E/I) ratio in hM4Di-mCherry infected PV interneurons in vehicle- and CNO-treated mice (vehicle, n = 15 cells, 14 slices, 7 mice; CNO, n = 16 cells, 12 slices, 7 mice; two-tailed Student’s t-test, p = 0.009). Data are presented as mean ± s.e.m. Scale bars, 1 µm. Source data

Extended Data Fig. 3. Validation of Flp-dependent hM3Dq-mCherry AAV and Channelrhodopsin control experiments.

a, PV and mCherry expression in coronal sections through S1. b, Quantification of the percentage of mCherry+ PV+ interneurons in layer 2/3 (n = 8 mice). c, Fos and mCherry expression in coronal sections through S1 from vehicle- and CNO-treated mice. d, quantification of the intensity of Fos staining in hM3Dq+ PV+ interneurons in vehicle- and CNO-treated mice (vehicle, n = 4 mice; CNO, n = 4 mice; two-tailed Student’s t-test, p = 0.03). e, Traces of Channelrhodopsin (ChR2) evoked currents recorded from hM3Dq-mCherry expressing PV+ interneurons at a holding voltage of −60 mV and +10 mV in the presence of picrotoxin (PTX) at 100% LED power. f, Quantification of ChR2-evoked charge in the presence of PTX (−60 mV, n = 4 cells; +10 mV, n = 4 cells; two-tailed paired Student’s t-test, p = 0.01). g, Traces of eIPSCs recorded from an hM3Dq+ PV+ interneuron at a holding voltage of +10 mV following stimulation of ChR2 expressing PV+ interneurons at 100% LED power in the absence or presence of picrotoxin (PTX). The PTX condition reflects the contribution of direct ChR2-evoked current (compare to panel a). h, The resultant trace of the subtraction of the traces is shown in panel (g). i, Overlay of panels (g) and (h). j, Quantification of eIPSC charges recorded from hM3Dq+ PV+ interneurons following stimulation of ChR2+ PV+ interneurons at 100% LED power in the absence or presence of PTX (control, n = 4 cells; PTX, n = 4 cells; two-tailed Student’s t-test, p = 0.03). Blue boxes indicate LED stimulation. Data are presented as mean ± s.e.m. Scale bars, 50 µm. Source data

Extended Data Fig. 4. Excitability of channelrhodopsin-expressing interneurons following LED stimulation at different intensities.

a, Experimental strategy to record the excitability of uninfected PV⁺ interneurons in a cell-attached configuration in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. b, Quantification of the number of elicited action potentials (APs) following LED stimulation at different intensities in uninfected PV⁺ interneurons from vehicle- and CNO-treated mice (5% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.97; 10% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.59; 25% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.76; 50% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.32; 100% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.29). c, Quantification of the selected stimulation intensity from panel (b). d, Experimental strategy to record the excitability of infected PV+ interneurons in a cell-attached configuration in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. e, Quantification of the number of elicited APs following LED stimulation at different intensities in infected PV⁺ interneurons from vehicle- and CNO-treated mice (5% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.23; 10% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 1.00; 25% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.21; 50% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.04; 100% LED-intensity, vehicle, n = 10 cells, 9 slices, 5 mice; CNO, n = 11 cells, 10 slices, 7 mice; two-tailed Student’s t-test, p = 0.04). f, Quantification of the selected stimulation intensity from panel (e). g, Experimental strategy to record the excitability of SST⁺ interneurons in a cell-attached configuration in vehicle- and CNO-treated Sst^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. h, Quantification of the number of elicited APs following LED stimulation at different intensities in SOM⁺ interneurons in vehicle- and CNO-treated mice (2% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.65; 3% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.61; 5% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.59; 10% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.67; 25% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.48; 50% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.55; 100% LED-intensity, vehicle, n = 13 cells, 8 slices, 5 mice; CNO, n = 8 cells, 5 slices, 2 mice; two-tailed Student’s t-test, p = 0.42). i, Quantification of the selected stimulation intensity from panel (h). j, Experimental strategy to record the excitability of VIP⁺ interneurons in a cell-attached configuration in vehicle- and CNO-treated Vip^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. k, Quantification of the number of elicited APs following LED stimulation at different intensities in VIP⁺ interneurons in vehicle- and CNO-treated mice (2% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.56; 3% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.77; 5% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.94; 10% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.38; 25% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.98; 50% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.85; 100% LED-intensity, vehicle, n = 15 cells, 10 slices, 3 mice; CNO, n = 17 cells, 10 slices, 3 mice; two-tailed Student’s t-test, p = 0.69). l, Quantification of the selected stimulation intensity from panel (k). m, Experimental strategy to record the inhibitory inputs from infected PV+ interneurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice from pyramidal neurons. n, Traces of eIPSCs recording from layer 2/3 pyramidal cells following full-field stimulation of PV+ interneurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. o, Quantification of the peak amplitude (vehicle, n = 13 cells, 13 slices, 5 mice; CNO, n = 12 cells, 12 slices, 5 mice; two-tailed Student’s t-test, p = 0.76) and charge (vehicle, n = 13 cells, 13 slices, 5 mice; CNO, n = 12 cells, 12 slices, 5 mice; two-tailed Student’s t-test, p = 0.66) of eIPSCs evoked by LED stimulation in layer 2/3 pyramidal cells following full-field stimulation of PV⁺ interneurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. Data are presented as mean ± s.e.m. Dotted boxes indicate the selected stimulation strength. Source data

Extended Data Fig. 5. Spontaneous IPSC recordings confirm the increase in PV⁺ interneuron activity.

a, Traces of spontaneous IPSCs (sIPSCs) recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. b, Quantification of amplitude (vehicle, n = 13 cells, 13 slices, 7 mice; CNO, n = 11 cells, 11 slices, 7 mice; two-tailed Student’s t-test, p = 0.003) and frequency vehicle, n = 13 cells, 13 slices, 7 mice; CNO, n = 11 cells, 11 slices, 7 mice; two-tailed Student’s t-test, p = 1.1×10⁻⁷) of sIPSCs recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. c, Traces of sIPSCs recorded from layer 2/3 pyramidal neurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. d, Quantification of amplitude (vehicle, n = 13 cells, 13 slices, 5 mice; CNO, n = 12 cells, 12 slices, 5 mice; two-tailed Student’s t-test, p = 0.28) and frequency (vehicle, n = 13 cells, 13 slices, 5 mice; CNO, n = 12 cells, 12 slices, 5 mice; two-tailed Student’s t-test, p = 0.43) of sIPSCs recorded from layer 2/3 pyramidal neurons in vehicle- and CNO-treated Pvalb^Cre/Flp;RCL^Chr2/+ mice. e, Traces of sIPSCs recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Sst^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. f, Quantification of amplitude (vehicle, n = 10 cells, 10 slices, 5 mice; CNO, n = 11 cells, 11 slices, 5 mice; two-tailed Student’s t-test, p = 0.009) and frequency (vehicle, n = 10 cells, 10 slices, 5 mice; CNO, n = 11 cells, 11 slices, 5 mice; two-tailed Student’s t-test, p = 0.005) of sIPSCs recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Sst^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. g, Traces of sIPSCs recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Vip^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. h, Quantification of amplitude (vehicle, n = 13 cells, 12 slices, 6 mice; CNO, n = 12 cells, 10 slices, 5 mice; two-tailed Student’s t-test, p = 0.03) and frequency (vehicle, n = 13 cells, 12 slices, 6 mice; CNO, n = 12 cells, 10 slices, 5 mice; two-tailed Student’s t-test, p = 0.006) of sIPSCs recorded from hM3Dq⁺ PV⁺ interneurons in vehicle- and CNO-treated Vip^Cre/+;Pvalb^Flp/+;RCL^Chr2/+ mice. Data are presented as mean ± s.e.m. Source data

Extended Data Fig. 6. Increasing the activity of PV⁺ interneurons leads to an increase in inhibitory synaptic density.

a, Schematic outline of the experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, Quantification of change in synaptic density between infected and uninfected PV⁺ interneurons in vehicle- and CNO-treated mice (vehicle, n = 6 mice; CNO, n = 6 mice; two-tailed Student’s t-test, p = 0.01). c, Confocal images illustrating presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and hM3Dq-mCherry expressing PV⁺ interneurons in vehicle-treated mice. Inserts show PV and mCherry immunoreactivity of the cell’s soma. d, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and hM3Dq expressing PV⁺ interneurons in vehicle-treated mice (n = 6; two-tailed paired Student’s t-test, p = 0.79). e, Confocal images illustrating presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and hM3Dq-mCherry expressing PV⁺ interneurons in CNO-treated mice. Inserts show PV and mCherry immunoreactivity of the cell’s soma. f, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and hM3Dq expressing PV⁺ interneurons in CNO-treated mice (n = 6; two-tailed paired Student’s t-test, p = 0.006). g, Quantification of the density of Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected PV⁺ interneurons in vehicle-treated and CNO-treated mice (vehicle: n = 6; CNO: n = 6; two-tailed Student’s t-test, p = 0.87). h, Validation of HA-vTRAP-hM3Dq-myc AAV showing PV and mCherry expression in coronal sections through S1. i, Quantification of the percentage of mCherry⁺ PV⁺ interneurons in layer 2/3 (n = 7 mice). j, Fos and mCherry expression in coronal sections through S1 from vehicle- and CNO-treated mice. k, Quantification of the intensity of Fos staining in HA-vTRAP-hM3Dq-myc infected PV⁺ interneurons in vehicle- and CNO-treated mice (vehicle, n = 4 mice; CNO, n = 4 mice; two-tailed Student’s t-test, p = 0.004). Scale bar, 1 µm (c and e) and 50 µm (h and j). Data are presented as mean ± s.e.m. Source data

Extended Data Fig. 7. Scg2 and Vgf expression following manipulations.

a, Experimental strategy. b, Confocal images showing Scg2 expression in in neighbouring infected and uninfected PV+ interneurons in vehicle-treated mice. c, Quantification of showing Scg2 expression in in neighbouring infected and uninfected PV+ interneurons in vehicle-treated mice (n = 6 mice; two-tailed paired Student’s t-test, p = 0.54). d, Confocal images showing Vgf expression in in neighbouring infected and uninfected PV+ interneurons in vehicle-treated mice. e, Quantification of showing Vgf expression in in neighbouring infected and uninfected PV+ interneurons in vehicle-treated mice (n = 5 mice; two-tailed paired Student’s t-test, p = 0.58). f, Experimental strategy. g, Scg2 and Vgf mRNA expression in neighbouring infected and uninfected PV⁺ interneurons in CNO-treated mice. h, Quantification of Scg2 and Vgf mRNAs (shScg2-mCherry: n = 6 mice; two-tailed paired Student’s t-test, p = 0.007; shVgf-mCherry: n = 6 mice; two-tailed paired Student’s t-test, p = 0.02). i, Fos, mCherry and PV expression in coronal sections through S1 from vehicle- and CNO-treated mice infected with LacZ-mCherry, shScg2-mCherry, and Vgf-mCherry. j, Quantification of the percentage of mCherry⁺ PV⁺ interneurons in layer 2/3 of mice infected with LacZ-mCherry (n = 3 mice), shScg2-mCherry (n = 5 mice), and Vgf-mCherry (n = 5 mice). k, Quantification of the intensity of Fos staining in PV⁺ interneurons in vehicle- and CNO-treated mice infected with LacZ-mCherry, shScg2-mCherry, and Vgf-mCherry (shLacZ-mCherry: vehicle, n = 3 mice; CNO, n = 3 mice; two-tailed Student’s t-test, p = 0.02; shScg2-mCherry: vehicle, n = 3 mice; CNO, n = 3 mice; two-tailed Student’s t-test, p = 0.001; Vgf-mCherry: vehicle, n = 3 mice; CNO, n = 3 mice; two-tailed Student’s t-test, p = 0.04). Data are presented as mean ± s.e.m. Scale bars, 10 µm (b,c and g) and 50 µm (i). Schematics in a,f adapted from ref. (reprinted with permission from AAAS) and ref. . Source data

Extended Data Fig. 8. Downregulation of Scg2 and Vgf prevents the increase in synaptic density induced by increased activity.

a, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, Quantification of the change in synaptic density between infected and uninfected PV+ interneurons in vehicle-treated mice injected with hM3Dq-shLacZ (n = 8 mice), hM3Dq-shScg2 (n = 9 mice, two-tailed one-sample t-test, p = 0.27) and hM3Dq-shVgf (n = 7 mice, two-tailed one-sample t-test, p = 0.94). c, Presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shLacZ-mCherry. Inserts show PV and mCherry immunoreactivity in the cell somata. d, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shLacZ-mCherry (vehicle: n = 8; two-tailed paired Student’s t-test, p = 0.67; CNO: n = 10; two-tailed paired Student’s t-test, p = 0.003). e, Presynaptic Syt2⁺puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shScg2-mCherry. Inserts show PV and mCherry immunoreactivity in the cell somata. f, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shScg2-mCherry (vehicle: n = 9; two-tailed paired Student’s t-test, p = 0.40; CNO: n = 9; two-tailed paired Student’s t-test, p = 0.07). g, Presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shVgf-mCherry. Inserts show PV and mCherry immunoreactivity in the cell somata. h, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and infected PV⁺ interneurons in vehicle- and CNO-treated mice injected with shVgf -mCherry (vehicle: n = 7; two-tailed paired Student’s t-test, p = 0.70; CNO: n = 8; two-tailed paired Student’s t-test, p = 0.72). Data are presented as mean ± s.e.m. Scale bar, 1 µm. Source data

Extended Data Fig. 9. Overexpression of Vgf increases PV-PV interconnectivity but does not affect synaptic input or inhibitory release probability of PV⁺ interneurons.

a, Experimental strategy. Adapted from ref. (reprinted with permission from AAAS) and ref. . b, Vgf mRNA expression in neighbouring uninfected and Vgf-mCherry⁺ PV⁺ interneurons. c, Quantification of Vgf mRNA expression (n = 3 mice; two-tailed paired Student’s t-test, p = 0.03). d, Presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and infected mCherry⁺ PV⁺ interneurons. Inserts show PV and mCherry immunoreactivity in the cell somata. e, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and mCherry⁺ PV⁺ interneurons (n = 5; two-tailed paired Student’s t-test, p = 0.23). f, Presynaptic Syt2⁺ puncta and postsynaptic Gephyrin⁺ clusters in neighbouring uninfected and Vgf-mCherry⁺ PV⁺ interneurons. Inserts show PV and mCherry immunoreactivity in the cell somata. g, Quantification of the density of colocalising Syt2⁺/Gephyrin⁺ puncta on neighbouring uninfected and Vgf-mCherry⁺ PV⁺ interneurons (n = 5; two-tailed paired Student’s t-test, p = 0.01). h, Traces of mEPSCs recorded from mCherry expressing PV⁺ interneurons and VGF-mCherry expressing PV⁺ interneurons. i, Quantification of amplitude (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 7 cells, 7 slices, 4 mice; two-tailed Student’s t-test, p = 0.59) and frequency (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 7 cells, 7 slices, 4 mice; two-tailed Student’s t-test, p = 0.12) of mEPSCs recorded from mCherry expressing PV+ interneurons and VGF-mCherry expressing PV⁺ interneurons. j, Traces of mIPSCs recorded from mCherry expressing PV⁺ interneurons and VGF-mCherry expressing PV⁺ interneurons. k, Quantification of amplitude (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 7 cells, 7 slices, 4 mice; two-tailed Student’s t-test, p = 0.26) and frequency (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 7 cells, 7 slices, 4 mice; two-tailed Student’s t-test, p = 0.42) of mIPSCs recorded from mCherry expressing PV⁺ interneurons and VGF-mCherry expressing PV⁺ interneurons. l, Excitation/Inhibition (E/I) ratio in mCherry expressing PV⁺ interneurons and VGF-mCherry expressing PV⁺ interneurons (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 7 cells, 7 slices, 4 mice; two-tailed Student’s t-test, p = 0.38). m, Traces of inhibitory pair pulses recorded from mCherry expressing PV⁺ interneurons and VGF-mCherry expressing PV⁺ interneurons with a 50 ms inter-stimulus interval. n, Quantification of paired-pulse ratio (mCherry, n = 9 cells, 9 slices, 4 mice; VGF-mCherry, n = 10 cells, 10 slices, 4 mice; two-tailed Student’s t-test, p = 0.46). Bar graphs are presented as mean ± s.e.m. Box plot centre line indicates the median, box limits indicate upper and lower quartiles, whiskers indicate maximum and minimum. Source data

Extended Data Fig. 10. Contextual fear conditioning activates a sparse population of PV⁺ interneurons.

a, Experimental strategy. b, Quantification of the percentage of time spent freezing for control (unshocked) and previously shocked mice during re-exposure to the cFC environment after 24 h (unshocked, n = 4 mice; shocked, n = 5 mice; two-tailed Student’s t-test, p = 8 × 10⁻⁵). c, Experimental strategy. d, Quantification of the number of Fos⁺ PV⁺ interneurons 2 h after cFC (n = 117 cells). e, Experimental strategy. f, Example images of tdTomato+ cells 1 (6 mice) and 7 (7 mice) days after cFC. Data are presented as mean ± s.e.m. Scale bar, 25 µm (f). Schematics in a,c,e adapted from ref. . Source data