Article Figures & Data
Figures
- Figure 1.
VIP activation vetoes SOM-mediated lateral inhibition. A, Fluorescent micrographs of tdTomato and ArchT-GFP expression in V1 of SOM-cre::LSL-TOM mouse injected with AAV-flex-ArchT-GFP. B, Representative whole-cell current-clamp trace from ArchT-expressing SOM cell during application of green light pulse. C, Quantified spontaneous firing rates from n = 5 SOM cells recorded as in B. D, Schematic of experimental paradigm where two PCs were recorded in current-clamp and one of them was forced to fire a 50–100 Hz AP train, during which the other cell received disynaptic inhibition. ArchT in SOMs was used to inhibit SOMs. E, Example recording from S1. The top traces are presynaptic APs and blue traces below them are average across trials from the postsynaptic PC (with individual trials superimposed in gray). Green bar indicates interleaved trials when the light was on to inhibit SOM firing. F, Example recording from V1, same notation as E. G, Quantified average IPSP integral across 5 experiments as in E,F (n = 2 from V1, n = 3 from S1). *p = 0.037. H, Fluorescent micrographs of tdTomato and ChETA-YFP expression in V1 of VIP-cre::TOM mouse injected with AAV-DIO-ChETA-YFP. I, Representative cell-attached recordings across three trials from a VIP cell expressing ChETA during application of 30 Hz blue light pulse trains. J, Firing fidelity across light train frequencies from four VIP cells recorded as in I. For all frequencies, pulse duration was 3 ms. K, Schematic of experimental paradigm as in D except with an excitatory opsin (ChETA) in VIPs. L, Example recordings from S1 similar to E. Blue pulses indicate interleaved trials when light was on to activate VIP firing. M, Example recording from V1, same notation as L. N, Quantified average IPSP integral across 10 experiments as in L and M (n = 5 for both V1 and S1). *p = 0.0009.
- Figure 2.
VIP axons are local, and vertical VIP→SOM connectivity is local. A, Examples of VIP cell morphology, somata denoted by white boxes, axons in blue and dendrites in red. Right, Average axon density plots for n = 19 VIPs aligned by somata (white square). B, Examples of SOM cell morphology, somata denoted by white boxes, axons in blue and dendrites in red. Right, Average axon density plots for n = 8 SOMs aligned by somata (white square). C, Left, Representative fluorescent/oblique contrast micrograph of a VIP-cre::LSL-TOM::SOM-GFP(GIN) slice with two SOMs and two VIPs being recorded with patch pipettes filled with a red dye. Right, Representative traces of presynaptic spike train in VIP (top, red star), postsynaptic response in a connected SOM (middle) and a not connected SOM (bottom). Gray traces are individual trials and green traces their average. D, Distance distribution of VIP→SOM connections, as well as not connected cells. Distributions are significantly different; connected 112 ± 25 μm, n = 15, and not connected 163 ± 14 μm, n = 46; p = 0.01 by rank sum test. E, Relative locations of connected and not connected VIP→SOM pairs recorded in coronal slices from V1.
- Figure 3.
Spontaneous correlations between VIPs and putative PCs are local. A, Schematic of experiment (left) and two-photon fluorescence micrographs of syn-GCAMP6s and TdTomato expression in V1 of a VIP-cre::TOM mouse. B, Example traces of activity in a VIP and a putative PC from the same field-of-view. C, Example field-of-view with cells colored according to their pairwise correlation coefficient with the VIP cell marked by the arrow. D, Left, Correlations between VIPs and putative PCs (all non-VIP cells were defined as PCs because they form the vast majority of non-VIP L2/3 cells) as a function of intersomatic distance quantified across 11 experiments containing a total of 68 VIPs. Right, Correlations between PCs from same experiments. E, Left, GCaMP6s fluorescence of cells in the imaged region during frames when VIP cells were active (blue) as a function of distance from the VIP cell, and the same quantification for PCs (black). Right, Comparison of data from left panel as indicated. *p = 0.005; N.S., p = 0.22 by paired t test. Data are from the same experiments as in D. F, Same data and analysis as D but only on stationary epochs. G, Data from F analyzed as in E. *p = 0.03; N.S., p = 0.24. H, Same data and analysis as D but only on locomotion epochs. I, Data from H analyzed as in E. N.S., p > 0.08.
- Figure 4.
Visual stimulus-evoked correlations between VIPs and putative PCs are local. A, Schematic of experiment (left) and two-photon fluorescence micrographs of OGB-1 and TdTomato in V1 of a VIP-cre::TOM mouse. B, Example traces of activity in a VIP and a putative PC from the same field-of-view. Visual stimuli (drifting gratings or natural images presented for 150 ms) denoted by gray bars. C, Example field-of-view with cells colored according to their pairwise correlation coefficient with the VIP cell marked by the arrow. D, Correlations between VIPs and putative PCs as a function of intersomatic distance quantified across five experiments from three mice containing a total of 18 VIPs. E, Correlations between PCs from same experiments as in D. F, OGB-1 fluorescence of cells in the imaged region during frames when VIP cells were active (blue) as a function of distance from the VIP cell, and the same quantification for PCs (black). Data are from the same experiments as in D and E. G, Comparison of data from F as indicated. *p < 0.01 by paired t test.
- Figure 5.
Individual VIP activation with two-photon optogenetics. A, Left, C1V1 was expressed in VIP cells in VIP-cre::TOM animals. Right, Confocal micrographs of DIO-C1V1-YFP expression in VIP-cre::TOM cells in fixed coronal sections. Scale bar applies for all panels. B, Spiral scan with 1040 nm two-photon laser over numbered targets corresponding to VIP-C1V1 cells. The recorded VIP #1 was only activated when the spiral was targeted to that cell. Recordings in VIP-cre::TOM slices expressing DIO-C1V1-YFP.
- Figure 6.
Individual VIP activation with spiral scan disinhibits visual responses of nearby PCs. A, Schematic of experiment paradigm. B, C, GCaMP6 imaging in vivo with simultaneous visual stimulation and spiral scan on a VIP-C1V1 cell in awake (B) and anesthetized (C) mice. From each experiment, two disinhibited example cells are shown. The visual stimulus was a 60° drifting grating in both examples. In the contour plots (top), cells are colored according to the extent of their disinhibition by the single VIP activation. D, Scatter plot of data from awake mice only (12 experiments). Small black cross in the middle denotes mean ± SEM. Red line is a linear fit to data and gray line is unity. p = 0.00003 by paired t test. E, Distribution of significantly (visual stimulus-evoked response changed by >2 × SD of ctrl data) disinhibited cells (total 33 cells) by distance from spiral target from 27 experiments from nine mice. p = 0.002 by rank sum test. There were also six cells whose responses were significantly inhibited by the spiral. Because these were a minority, we did not analyze them further. F, Average spiral-induced change in visual responses by distance from spiral target. *p < 0.05 and N.S., p > 0.05 by Z test against zero.
- Figure 7.
Individual VIP activation induced visual response changes across behavioral states. A–C, Average spiral-induced change in visual responses by distance from spiral target in anesthetized animals (A; n = 15 experiments), awake stationary (B; n = 9 experiments), and awake moving (C; n = 9 experiments). **p = 0.002, *p < 0.05, and N.S., p > 0.05 by Z test against zero.
