2P-NucTag: on-demand phototagging for molecular analysis of functionally identified cortical neurons

Abstract

Neural circuits are characterized by genetically and functionally diverse cell types. A mechanistic understanding of circuit function is predicated on linking the genetic and physiological properties of individual neurons. However, it remains highly challenging to map the transcriptional properties to functionally heterogeneous neuronal subtypes in mammalian cortical circuits in vivo. Here, we introduce a high-throughput two-photon nuclear phototagging (2P-NucTag) approach optimized for on-demand and indelible labeling of single neurons via a photoactivatable red fluorescent protein following in vivo functional characterization in behaving mice. We demonstrate the utility of this function-forward pipeline by selectively labeling and transcriptionally profiling previously inaccessible 'place' and 'silent' cells in the mouse hippocampus. Our results reveal unexpected differences in gene expression between these hippocampal pyramidal neurons with distinct spatial coding properties. Thus, 2P-NucTag opens a new way to uncover the molecular principles that govern the functional organization of neural circuits.

Fig. 1.. In vivo two-photon phototagging with 2P-NucTag.

(A) Schematics of the 2P-NucTag pipeline. Top: bicistronic rAAV construct, injection to the hippocampus. Middle: in vivo two-photon (2P) GCaMP- Ca²⁺ population imaging followed by 2P PAmCherry photoactivation, fluorescence-activated cell sorting (FACS), and mesoscale sequencing (Meso-seq). (B) Top: representative in vivo time-averaged (6 frames average) 2P images of individual cells before (Pre) and after (Post) in vivo two-photon PAmCherry photoactivation in the CA1 pyramidal layer of the mouse dorsal hippocampus. Individual nuclei were photoactivated with 810-nm 2P laser chessboard scanning region-of-interest (ROI, yellow boxes) over target nuclei (70 x 70 pixel for each ROI, 0.1 μm/px, 1.3 ms/px total pixel dwell time, 6,370 ms total scan time per ROI. Laser power was 40 mW measured after the objective) with a 3-dimensional acousto-optical deflector microscope (3D-AOD). Gray: GCaMP7f (940 nm excitation), magenta: PAmCherry, (1040 nm excitation). Scale bar: 50 μm. Middle and bottom: imprints of letters ‘BI’ and ‘ZI’ following patterned in vivo two-photon photoactivation in the hippocampal CA1 pyramidal layer (scale bar, 50 μm). (C) Characterization of in vivo 2P photoactivation parameters for PAmCherry: duration, wavelength, laser power (measured after the objective) (n = 11-12 cells per condition). Relative change in PAmCherry red fluorescence (ΔF) is based on normalizing the tagged nuclei fluorescence to the fluorescence of neighboring untagged nuclei measured with 1040 nm excitation. (D) In vivo stability of the PAmCherry fluorescence signal over days after a single photoactivation scan (n = 8 cells). (E) Representative time-averaged images from z-stacks of photoactivated nuclei in vivo (magenta: PAmCherry, scale bar: 100 μm). Middle: ex vivo post hoc confocal z-stack image of the same field of view (FOV, magenta: PAmCherry). Right: registered in vivo and ex vivo images following non-rigid image transformation (magenta: in vivo, yellow: ex vivo, see methods). (F) Left: 3D overlay of tagged nuclei registered between in vivo (magenta) and ex vivo (yellow) z-stacks with increasing lateral resolution (as in E). Gray box represents the segmented area for subsequent images. Right: normalized lateral (x-y, left) and axial (z, right) fluorescence profiles (mean ± s.e.m.) of tagged cells in vivo (magenta, n = 1 mouse, 200 cells). Yellow: mean ± s.e.m. of ex vivo confocal images (as in E and F, same mouse and nuclei). Inset: (x-y) top: average in vivo maximum z–projection, bottom: average ex vivo maximum z projection; (z) top: average in vivo lateral projection, bottom: average ex vivo lateral projection. Scale bar: 10 μm. Boxplots show the 25^th, 50^th (median), and 75^th quartile ranges, with the whiskers extending to 1.5 interquartile ranges below or above the 25^th or 75^th quartiles, respectively. Outliers are defined as values extending beyond the whisker ranges.

Fig. 2.. Selective phototagging of place cells in the hippocampus with 2P-NucTag.

(A) Pipeline for two-photon (2P) phototagging of functionally identified hippocampal neurons during spatial navigation. (B) Left: schematics of 2P imaging setup in virtual reality (VR). Head-fixed mice are trained to run for a water reward in a 4-m long linear VR corridor projected onto LCD screens surrounding the animal. At the end of the corridor, mice are teleported back to the start position after a 2-second delay. Right: example 2P field of view (FOV) of GcaMP in the CA1 pyramidal layer. Scale bar: 100 μm.(C) Left: Traces of relative GcaMP-Ca²⁺ fluorescence changes (ΔF/F) from five example CA1 place cells during VR spatial navigation. Right: heatmaps of normalized ΔF/F activity from three example place cells over 20 laps during VR navigation. (D) Left: heatmap of all CA1PNs detected with Suite2p/Cellpose in the FOV shown in B. Identified place cells are marked with an orange box. Right: Zoomed-in heatmap of place cell tuning curves. (E) Left: spatial mask (orange) of identified place cells from D in the FOV. Right: PAmCherry fluorescence (magenta) of tagged nuclei after 2P phototagging. Scale bar: 100 μm. (F) Left: overlay of spatial masks of identified CA1PNs and tagged nuclei for the FOVs in E. Scale bar: 100 μm. Right: tagging efficacy, defined as the fraction of successfully tagged place cell nuclei (93.3% ± 4.2%, mean ± s.e.m., n = 5 mice). (G) Left: Proportion of single, double, and triple-tagged nuclei following phototagging of a single place cell. Right: relative change in PAmCherry red fluorescence (1070 nm excitation) for non-tagged cells in the FOV after 2P imaging (green), after 2P phototagging of targeted place cell nuclei (orange) and off-target nuclei (gray, n = 5 mice). Boxplots show the 25^th, 50^th (median), and 75^th quartile ranges, with the whiskers extending to 1.5 interquartile ranges below or above the 25^th or 75^th quartiles, respectively. Outliers are defined as values extending beyond the whisker ranges.

Fig. 3.. Post hoc transcriptional profiling of phototagged place and silent cells.

(A) Schematics of in vivo photoactivated nuclei. ‘Place’ cell sample and ‘Silent’ cell samples from different mice were collected for FACS and Meso-seq. (B) Representative FACS graph. Gating for mCherry was set after the first 5000 events of DAPI+ nuclei to the border of the ‘dim’ mCherry+ population to separate out the sparse and high-intensity mCherry+ population. Bright mCherry+ NeuN+ populations were collected as the photoactivated nuclei. (C) Top: number of FACS sorted nuclei from ‘place’ and ‘silent’ samples (n = 9, 17-79 sorted nuclei, 40.3 ± 6.43, mean ± s.e.m.). Bottom: Proportion of FACS sorted nuclei compared to the number of in vivo photoactivated nuclei (n = 9, 18.45% to 66.39% FACS recovery, 39.94 ± 6.30%, mean ± s.e.m.). (D) Volcano plot of Meso-seq differential expressed gene (DEG) analysis for ‘place’ and ‘silent’ cells (significantly different genes are shown in orange and blue. Orange: enriched in place cells; blue: enriched in silent cells). (E) Meso-seq MA plot depicting DeSeq2 normalized gene counts versus log2 fold change of silent/place samples. Genes that are significantly different are labeled in orange and blue (same as above). Genes shown in panels F & G are highlighted and labeled in E. (F) Bar graph showing the normalized counts for genes that are not differentially expressed (FDR adjusted p-value. *<0.05, **<0.001, ***<0.001, PyDeSeq2. Otherwise, comparisons are not significant). (G) Bar graph showing the normalized counts for differentially expressed genes (FDR adjusted p-value. *<0.05, **<0.001,***<0.001, PyDeSeq2. Otherwise, comparisons are not significant). (H) Gene ontology analysis performed on all differentially expressed genes. Vertical line: FDR-adjusted p value of 0.05. NES = normalized enrichment score. Boxplots show the 25^th, 50^th (median), and 75^th quartile ranges, with the whiskers extending to 1.5 interquartile ranges below or above the 25^th or 75^th quartiles, respectively. Outliers are defined as values extending beyond the whisker ranges.