Article Figures & Data
Figures
- Figure 1.
Widespread and stable co-expression of jGCaMP8s and stChrimsonR in mouse cortical neurons with a bicistronic viral vector. A, Schematic of the Cre-dependent vector. B, Viral injections were made in primary visual cortex (V1) to achieve targeted expression in excitatory neurons using an Emx1-Cre mouse line. C, Widefield fluorescence imaging through 3-mm optical windows in two example mice shows widespread expression over the cortical surface surrounding injection sites (white asterisks); see also Extended Data Figure 1-1. D, Two-photon calcium imaging (at 920 nm) in mouse V1 throughout layer 2/3 shows robust expression of the virus across many cells (FOV 414 × 414 µm); imaging produces minimal to no stimulation response (Extended Data Fig. 1-2). This figure: N = 3 mice. C, Mice 1–2. D, Mouse 3.
- Figure 2.
V1 cells expressing jGCaMP8s-P2A-stChrimsonR show expected visually-evoked and spontaneous activity. A, Two-photon imaging FOV (414 × 414 µm) in layer 2/3 of mouse V1 expressing jGCaMP8s-P2A-stChrimsonR. Imaging: 30-Hz frame rate, 15 mW, 920 nm. B, Example trial-average visual responses to a retinotopically-aligned 15° FWHM Gabor. One orientation shown. (In total, 8 orientations presented in random order, each 2-s duration, N = 20 repetitions/orientation.) C, Trial-average responses in four example cells showing direction and orientation tuning. DSI and OSI calculated using average ΔF/F0 across the 2-s visual period. Polar plots, Responses normalized to the peak response across all eight orientations for each cell. D, ΔF/F0 traces in example cells (N = 20), showing stimulus-evoked responses across three consecutive trials. E, Cumulative distribution functions showing distributions of tuning metrics and average responsivity (average response across 8 directions) for all cells from P2A mice (N = 931 cells across 3 mice, green lines) and dual-virus mice (N = 560 cells across 2 mice, orange lines). DSI: direction selectivity index, OSI: orientation selectivity index, gOSI: global orientation selectivity index. This figure: N = 5 mice. A–D, Mouse 4. E, Mice 3–5, and two dual-virus mice.
- Figure 3.
V1 cells expressing jGCaMP8s-P2A-stChrimsonR maintain visual responses for weeks following injection. A, Two-photon imaging FOV (414 × 414 µm) in layer 2/3 of mouse V1. Imaging: 30-Hz frame rate, 15-mW power, 920 nm. B, Example trial-average visual responses to a retinotopically-aligned Gabor (15° FWHM) showing evoked activity (ΔF/F0) for a single orientation. C, Cumulative distribution functions showing distributions of tuning metrics and average responsivity (average response of all 8 directions) across all cells from the mouse in A at two time points a month apart in the same FOV (N = 192 cells at 3 months postinjection, gray line, N = 167 cells at 4 months, black line). D, Same as in C, but for a second mouse (N = 149 cells at 4 months postinjection, gray line, N = 175 cells at 5 months, black line). This figure: N = 2 mice. A–C, Mouse 3. D, Mouse 6.
- Figure 4.
Simultaneous holographic stimulation and calcium imaging in cells expressing jGCaMP8s-P2A-stChrimsonR. A, Two-photon imaging FOV (340 × 210 µm) in mouse V1, depth 122 µm from the pia. Imaging: 30-Hz frame rate, 10-mW power, 920 nm. White rings, Holographic stimulation locations (N = 20). Red, gray arrows, Examples of stimulated and nearby unstimulated cells, shown in panel D. B, Responses to stimulation (ΔF/F0) in same FOV as A (N = 50 stim repetitions, 2.0 mW/target, 300-ms stim duration, 10-µm diameter holographic disk pattern, stim. delivered during imaging acquisition and gated off at edges of scan field, reported powers are adjusted for fast-gated duty cycle; Materials and Methods). C, Trial-average activity traces for all cells (N = 89 cells). Black horizontal bar separates stimulation targets (above), other cells in the FOV (below). D, Trial-average cell traces (mean ± SEM, N = 50 repetitions) for nine example cells (N = 6 stimulated, red, and N = 3 unstimulated, gray). Error bars (SEM) are small enough to be visible only as slightly thickened lines. Decay comparison to dual-virus expression (Extended Data Fig. 4-1). Distributions of responses in stimulated and unstimulated cells, Extended Figure 4-2. E, Average response map centered on targeted cells from A, B. F, Average neuropil responses centered on 50 randomly selected coordinates. G, Difference of maps from E, F. Horizontal black bar: cross-section in H. Dashed circle: 10-µm diameter holographic disk pattern. H, Horizontal cross-section of response map in G. FWHM of response = 13.5 µm. I–K, Same as A–C, in a different FOV in the same mouse one month later (414 × 414 µm, depth = 130 µm, N = 50 targets, N = 348 total cells, N = 100 stim reps, 16-mW imaging power, stimulation power 2.5 mW/target, 150-ms stim duration, 10-µm diameter disk pattern). L, Trial-average cell traces (N = 100 stim repetitions). Gray lines, 50 targeted cells in I; red line, mean ± SEM cell trace. In N = 3 animals with 10–12 stimulated target cells: 30/32 targeted cells responsive, 35/1037 unstimulated cells responsive (see Extended Data Figs. 4-3, 4-4). N = 4 photostimulation animals, N = 1 this figure, mouse 1. Extended Data Figure 4-3: N = 3 mice, mice 3, 4, 6.
Extended Data
Extended Data Figure 1-1
Widefield fluorescence imaging of jGCaMP8s-P2A-stChrimsonR expression. Widefield fluorescence imaging through 3-mm optical windows in experimental mice. White stars: injection sites for each mouse (mouse 3: N = 5 sites, mouse 4: N = 10 sites, mouse 5: N = 10 sites, mouse 6: N = 6 sites). Images were acquired 22–53 d after injection. Download Figure 1-1, TIFF file.
Extended Data Figure 1-2
Minimal signs of cross talk activation from resonant-galvo scanning at imaging onset. Imaging cross talk should lead to elevated fluorescence at onset of imaging. Cell-averaged fluorescent traces in imaging FOVs (N = 12 FOVs from N = 5 mice) across first 2 s of imaging do not show signs of widespread cross talk in activation from the imaging laser (920 nm). Only a few sessions (2/12) show small increases in fluorescence over the first few seconds after imaging onset, as would be expected if imaging cross talk were occurring (significantly higher levels of fluorescence in the 2nd second of imaging vs the 1st second p < 0.05, Mann–Whitney U test with Bonferonni correction for multiple comparisons). On the other hand, however, four of 12 imaging sessions show significantly lower levels of fluorescence, arguing against cross talk activation (p < 0.05, Mann–Whitney U test with Bonferroni correction for multiple comparisons). Together, this suggests that elevated fluorescence at the start of imaging, due to cross talk, is not a concern with this preparation. Download Figure 1-2, TIFF file.
Extended Data Figure 4-1
GCaMP decay times to photostimulation responses are similar between P2A and dual-virus preparations. A, Two-photon imaging FOV (414 × 414 µm) in layer 2/3 of mouse V1 for both a P2A mouse (top) and a dual-virus mouse (bottom). Imaging collected at 30-Hz frame rate, 10-mW power, 920 nm. B, Trial-average photostimulation responses in five example cells from a P2A (left, green lines) or dual-virus (right, orange lines) mouse. Black lines, Exponential decay functions fit to the 600-ms period following stimulation offset for all photostimulated cells to estimate decay times (Materials and Methods). Stimulation power was 2.0 mW/target (P2A) or 2.5 mW/target (dual-virus) for 300 ms using 10-µm diameter disk patterns. C, Half-decay times for photostimulated cells from both mice. Horizontal black lines, Mean half-decay time amongst cells. Vertical black lines, SEM. Half-decay times are not significantly different between P2A-expressing and dual-virus-expressing cells (p = 0.37, two-sample t test). This figure: N = 2 mice. A–C, Mouse 1 and a dual-virus mouse. Download Figure 4-1, TIFF file.
Extended Data Figure 4-2
Distribution of responses to photostimulation in stimulated and unstimulated populations. A, Cumulative distribution functions of photostimulation responses (from data in Fig. 4A–D) averaged over a 1-s period following stimulus onset in both stimulated cells (red) and unstimulated cells in the FOV (black). Using a 7.5% ΔF/F0 threshold, we found 19/20 stimulated cells and 5/69 unstimulated cells show reliable activation. B, Same as in A, but for data from Figure 4I-L. Using a 7.5% ΔF/F0 threshold, we found 33/50 stimulated cells and 45/298 unstimulated cells show reliable activation. To account for concerns of stronger neuropil contamination in response calculations when stimulating larger groups of cells, we also used a ring-based neuropil correction via Suite2p and found minimal differences in responses (35/50 stimulated cells and 41/298 unstimulated cells above a 7.5% ΔF/F0 threshold), implying neuropil signal was not contaminating cell responses. This figure: N = 1 mouse. A, B, Mouse 1. Download Figure 4-2, TIFF file.
Extended Data Figure 4-3
Holographic stimulation of cells expressing jGCaMP8s-P2A-stChrimsonR. A, Two-photon imaging FOV (414 × 414 µm) in mouse V1 in three example mice. [Depths: 130 µm (top), 140 µm (middle), and 110 µm (bottom); each in L2/3, depth differences do not produce systematically different responses.] Imaging at 30-Hz frame rate, 15-mW imaging power, 920 nm. White dashed rings, Holographic stimulation pattern (top: N = 10 targets, middle: N = 12 targets, bottom: N = 10 targets). B, Corresponding stimulation response maps showing mean optogenetically-evoked activity (ΔF/F0; N = 50 stim reps). Stimulation power 2.5 mW/target, stim duration 300 ms, 10-µm diameter disk patterns. C, Trial-average activity of all stimulated cells (light gray lines) plotted with mean ± SEM activity across all stimulated cells (red lines) and all unstimulated cells (black lines). SEM is present for unstimulated cells, but small relative to the size of the plotted line. Fraction of responsive stimulated cells: (top) 9/10, (middle) 11/12, (bottom) 10/10 (using 7.5% ΔF/F0 threshold, see Materials and Methods for details). Fraction of responsive unstimulated cells: (top) 11/436, (middle) 19/430, (bottom) 5/171. Download Figure 4-3, TIFF file.
Extended Data Figure 4-4
Spatial extent of photostimulation effects in unstimulated cells. Trial-average responses to photostimulation. Red points, Stimulated cells. Gray points, Unstimulated cells. Unstimulated cells are plotted as a function of distance to the nearest stimulated cell in the FOV. Stimulation power, duration, and the mean pairwise distance between stimulated targets is shown for all photostimulation experiments in P2A animals. This figure: N = 4 mice. Download Figure 4-4, TIFF file.
In this issue
