General Anesthetic Conditions Induce Network Synchrony and Disrupt Sensory Processing in the Cortex

Abstract

General anesthetics are commonly used in animal models to study how sensory signals are represented in the brain. Here, we used two-photon (2P) calcium activity imaging with cellular resolution to investigate how neuronal activity in layer 2/3 of the mouse barrel cortex is modified under the influence of different concentrations of chemically distinct general anesthetics. Our results show that a high isoflurane dose induces synchrony in local neuronal networks and these cortical activity patterns closely resemble those observed in EEG recordings under deep anesthesia. Moreover, ketamine and urethane also induced similar activity patterns. While investigating the effects of deep isoflurane anesthesia on whisker and auditory evoked responses in the barrel cortex, we found that dedicated spatial regions for sensory signal processing become disrupted. We propose that our isoflurane-2P imaging paradigm can serve as an attractive model system to dissect cellular and molecular mechanisms that induce the anesthetic state, and it might also provide important insight into sleep-like brain states and consciousness.

Keywords: anesthetics; cerebral cortex; genetically encoded calcium indicators; isoflurane; two-photon imaging.

Figure 1

Chronic in vivo two-photon activity imaging. (A, Top) Titanium head-plate in rectangular design is shown with a top and a front view. Scale bar: 5 mm. (Bottom) Headplate in the corresponding fixation device. (B) Schematics for experimental setup for two-photon Ca²⁺ imaging of neural activity evoked by sensory stimulation. Sensory stimulation was achieved through an air-puff to the mystacial whiskers on the right side. (C) YC2.60 is expressed under control of the CAG promoter (CAGpro) and delivered into the barrel cortex by rAAV injection. WPRE, woodchuck-hepatitis posttranscriptional regulatory element; BGHpA, bovine growth hormone polyadenylation signal. (D) Wide-field images showing superficial blood vessels used for orientation and virus injected region within the barrel cortex. Scale bar: 1 mm. (E) Fluorescence images of brain slices from animals that were used for 2P-imaging show YC2.60 fluorescence in cortical layers. Note the absence of gliosis in the virus-injected region. Scale bar: 1 mm. (F, Left) Example images for immunofluorescence analysis for co-expression of YC2.60 and NeuN or GFAP in layer 2/3. Arrow indicates YC2.60-expressing glial cell. Scale bar: 30 μm. (Right) Quantification of co-expression of YC2.60 and NeuN or GFAP reveals strong expression tropism for neurons (90 ± 6%) over glial cells (6 ± 3%). Error bars = SEM, ^****p < 10⁻⁶; unpaired Student's t-test, n = 6 mice (600 cells in total). (G) Chronic in vivo Ca²⁺ imaging in layer 2/3 of the barrel cortex. (Left) Images depict example brain region for functional imaging on day 1 and day 4 (depth = ~120 μm from the dura mater); scale bar: 25 μm. (Right) Example activity traces from all cells for a single trial on two different imaging days are shown.

Figure 2

Deep anesthesia enhances neural synchrony. (A) Random example fluorescence traces of the same cell under different anesthetic conditions. Vertical bars mark detected peaks, inset: 5% ΔR/R (y-axis), 5 s (x-axis). (B) Extracted peak time points of all cells shown as probabilities over adjacent frames (Top panel, scale bar: 5 s), color-coded correlation matrices for Pearson's r of 3 example trials (Bottom panel, small pictures) and example correlation maps (Bottom panel, big pictures, scale bar: 30 μm). Inset: color-code for peak probability (white = 0 to black = 1). Dotted lines indicate time points of whisker stimulation. Correlation is expressed as Pearson's correlation coefficient (r) and color-coded from r = 0 (blue, no correlation) to r = 1 (red, perfect positive correlation). (C) The number of detected peaks per cell (mean over all trials and cells ± SD; ANOVA, p = 0.71). (D) Correlation r's filtered for significance (p < 0.05) shift toward higher values with high isoflurane: cumulative frequency distribution for all pairwise correlation r's (p < 0.01 High Iso vs. Low Iso). (E) High isoflurane concentrations induce a marked increase in overall synchrony that is independent of sensory stimulation. (Left) Mean pairwise correlation of all cell pairs for the indicated isoflurane concentration over all trials per mouse (mean ± SEM, ^*p < 0.05, ^**p < 0.01 ^***p < 0.001). (Right) Summary of correlation r-values under a high isoflurane concentration (mean ± SEM). The same analysis as above was additionally performed for trials where only auditory stimulation was present (All peaks vs. subtracted p = 0.58, All peaks vs. Sound only; p = 0.43). (F) High isoflurane concentrations preferentially increase synchrony between nearby neurons. Shown is the mean pairwise correlation (Pearson's r) plotted against pairwise distance (bin size: 25 μm) (mean ± SEM).

Figure 3

Anesthesia-dependent synchrony is similarly induced by chemically distinct anesthetics. (A) Example traces of three cells for one trial of each anesthetic condition are shown. (Top) Ca²⁺ signal traces; (Bottom) peak probability matrix. Dotted lines: time of air-puff stimulus (B) Cumulative frequency distributions of Pearson's r over all cell pairs and trials. Values were pre-filtered for significance level of p < 0.05 as in Figure 2D (see Section Materials and Methods for details); the low isoflurane dataset is shown in black and is taken as baseline condition and as a reference for statistical testing. A substantial rise in correlation between cell pairs was observed for all anesthetic conditions except Ketamine + Xylazine (K + X) (^***p < 0.001 for all datasets vs. Low Iso except K + X). (C) Mean number of events per trial, mean amplitude of all events per trial and mean amplitude of all stimulus evoked responses per trial over all trials are plotted (mean ± SEM, ^*p < 0.05).

Figure 4

Deep isoflurane anesthesia affects cortical information processing. (A) Example field of view showing the raw image (Left) and an overlay of analyzed ROIs (Right), scale bar 25 μm. (B) Example traces of imaged cells shown in (A) for whisker puff and sound only trials for each isoflurane concentration, inset: 10% ΔR/R (y-axis), 2 s (x-axis), thick lines indicate group average. (C) Fraction of neurons responding to whisker stimulation by an air-puff and sound-only stimulation under low (0.5%), medium (1.0%), and high (1.5–2.0%) isoflurane doses, and after recovery (0.5%) (Mean ± SEM, ^*p < 0.05, only significant differences between stimulation groups and isoflurane treatments are marked in the figure). The differences in sensory responsiveness at low isoflurane vanished when the stimulus detection windows were randomized within datasets (RND) (see Section Materials and Methods) (mean ± SEM, ^**p < 0.01, ^***p < 0.001).