Fig 1.
Viral transduction of LC-NE neurons in different model systems.
(A) Either CAG-DIO-eGFP (in Dbhcre, Netcre, and Thcre) or PRS×8-eGFP (in WT) was bilaterally injected into the LC of mice (n = 7 animals in each group). Coronal brain slice image from [37]. (B) Example LC hemi-brain slice (vertically mirrored) from a Dbhcre animal. Noradrenergic neurons were identified by immunostaining against TH (left, magenta), while transgene expression was visualized with an anti-GFP staining (right, green). (C) Example LC cells in each of the four model systems (columns) immuno-stained against TH (first row, magenta outlines) and GFP (second row, green outlines). Cells were identified with CellPose and boundaries are denoted with colored outlines. Resulting cell masks were then overlaid (third row), and each TH+ cell was labeled as GFP+ (successful transduction, black) when cell masks overlapped 50% or more the size of the TH+ cell, or GFP− otherwise (missed transduction, magenta). GFP+ and TH− (erroneous transduction, green) was counted as unspecific transduction. (D) Efficacy (i.e., the true positive rate), computed as N(GFP+|TH+)/N(TH+), across model systems. (E) Specificity (i.e., the inverse of the false positive rate), computed as N(TH+|GFP+)/N(GFP+), across model systems. (F) Difference in normalized eGFP fluorescence of neurons that co-expressed eGFP and TH (i.e., true positive neurons) and neurons that expressed eGFP only (i.e., false positive neurons) across model systems. Negative values indicate that eGFP fluorescence of TH-negative neurons was brighter than eGFP fluorescence of TH-expressing neurons. In (D–F) error bars denote the standard error of the mean (with n = 7), while statistical significance is denoted by */**/*** for p < 0.05/0.01/0.001, respectively. Only significant differences are indicated. Numerical data underlying panels D–F can be found in S1 Data.
Fig 2.
Ectopic transgene expression across model systems.
(A–D) On the left panels: example brain slices showing ectopic expression in Dbhcre/Netcre/Thcre/PRS×8 model systems. Immuno-staining against GFP and TH is displayed in green and magenta, respectively. On the right panels: summary of ectopic expression across all mice in Dbhcre/Netcre/Thcre/PRS×8 model systems. Brain regions are color-coded to reflect the number of mice which showed ectopic expression (at least 1 neuron) in the corresponding brain region. Brain slice images from [37].
Fig 3.
Effects of promoter, serotype, and injection volume on viral transduction of LC-NE neurons.
(A) To investigate effects of the promoter on viral transduction, rAAV2/9-hSyn-DIO-eGFP (a construct analog to rAAV2/9-CAG-DIO-eGFP used in Figs 1 and 2) was bilaterally injected into the LC of Dbhcre mice (n = 7 animals). Coronal brain slice image from [37]. (B) Example LC section from an rAAV2/9-hSyn-DIO-eGFP-injected Dbhcre mouse immuno-stained against TH (top left) and GFP (top right), as well as color-coded overlay (bottom left). Cells were identified with CellPose. (C) Efficacy (N(GFP+|TH+)/N(TH+)) and (D) Specificity (N(TH+|GFP+)/N(GFP+)) were comparable upon injection of hSyn-DIO-eGFP (yellow) or CAG-DIO-eGFP (red; re-plotted from Figs 1D and 2) in Dbhcre mice (ns = not significant; two-sample t test, n = 7 each). (E) To investigate effects of the viral serotype on viral transduction, 300 nl of either rAAV2/2-hSyn-eGFP or rAAV2/9-hSyn-eGFP were bilaterally injected into the LC of wild-type mice (n = 6 LC from n = 3 mice each). (F) Example images of hemi-brain slices of an rAAV2/2-injected (orange, left) and an rAAV2/9-injected mouse (blue, center). Images were smoothed and fluorescence was normalized, before thresholding images with an increment of 0.1 and calculating the area above threshold (see example with an increment of 0.2 on the right). (G) Viral spread was quantified at various thresholds, revealing a more restricted spread upon injection of rAAV2/2 as compared to rAAV2/9 (*/**/*** for p < 0.05/0.01/0.001, two-sample t-tests, n = 6 LC per group). (H) To investigate effects of the injection volume on viral transduction, either 300, 100, or 50 nl of rAAV2/9-hSyn-eGFP were bilaterally injected into the LC of wild-type mice (n = 6 LC from n = 3 mice each). (I) Example images of hemi-brain slices of mice injected with 300 nl (dark blue, left), 100 nl (light blue, center) or 50 nl of rAAV2/9-hSyn-eGFP (cyan, right). (J) Viral spread was quantified at various thresholds, revealing a more restricted spread upon injecting 100 and 50 nl of viral suspension as compared to 300 nl of viral suspension (*/**/*** for p < 0.05/0.01/0.001, one-way ANOVA and Tukey’s test, N = 6 LC per group). Asterisks denote significance between the 300 nl group and the 100 nl/50 nl group, respectively, as no significant differences were found between the 100 nl and the 50 nl group. Numerical data underlying panels C, D, G, and J can be found in S1 Data.
Fig 4.
(A) Bilateral transduction efficacy (n = 5 animals in each group) using a single micropipette to sequentially infect both LC (2 × 300 nl). (B) Same as A, except that one micropipette was used for each LC (n = 2 animals in each group). (C) Fraction of animals with bilateral (solid) or unilateral (shaded) LC transduction upon LC injections with a single pipette vs. two pipettes. (D) Assessement of viral spread by injections of a ‘cre-switch’ construct leading to eGFP expression in a cre-dependent manner and tdTomato expression in cre-negative cells [45]. (E) Transgene expression in Dbhcre, Netcre and Thcre animals (n = 3 animals each) shows conditional eGFP expression in cre-positive cells (green) and tdTomato expression in cre-negative cells (magenta). LC-NE cells are visualized by immunostaining against TH (dark blue). Importantly, in animals without eGFP expression, cre-negative cells around the LC also lacked the expression of tdTomato, likely resulting from loss of the virus suspension to the ventricle. (F) Change of mediolateral coordinates for virus injections from ±0.9 mm to ±1.1 mm. (G) Bilateral transduction efficacy after using the new coordinates in transgenic or wild-type animals. (H) Fraction of conditionally eGFP-expressing animals with bilateral (solid) or unilateral (shaded) LC transduction upon LC injections at mediolateral coordinates of from ± 0.9 mm and ± 1.1 mm (e.g., pooled data from panel A/B vs. panel G, uppermost row). (I) Fraction of unconditionally fluorophore-expressing animals with bilateral (solid) or unilateral (shaded) LC transduction upon LC injections at mediolateral coordinates of ±0.9 mm and ±1.1 mm (e.g., pooled data from panel D vs. panel G, rows 2–5). ns = not significant, * for p < 0.05, chi-squared test.
Fig 5.
Behavioral screening of cre-expressing mice.
(A) Left: To assess anxiety-like behavior, mice were placed in one corner of the arena and left to explore the open field arena for 20 min. Right: The percentage of time spent in the center of the arena indicates no difference in anxiety-like behavior of cre-expressing mice (filled bars) as compared to wild-type littermates (empty bars) of different driver lines. (B) Left: To further test anxiety-like behavior, mice were placed on the central platform of the elevated plus maze and left to explore the maze for 5 min. Right: The percentage of time spent in open arms indicates no difference in anxiety-like behavior of cre-expressing mice as compared to wild-type littermates in any driver line. (C) Left: To assess working memory, mice were placed in the center of the Y-maze and left to explore the maze until 26 transitions were made, for a maximum of 15 min. Upon the first transition between arms (black arrow), the next transition is expected to occur towards the previously unexplored arm (green arrow), rather than to the arm that was just visited before (red arrow). Right: The percentage of alterations (i.e., consecutive visits of all three arms) over the total number of transitions is above chance level (dashed line) in all groups, while no genotype-dependent differences in working memory were detected between cre-expressing mice and wild-type littermates in any mouse line. (D) Left: To test spatial memory, mice are trained to find a hidden platform in a circular pool filled with opaque water on two consecutive days. On day three, the platform is removed, and the time spent in the area where the platform was previously located serves as a proxy for spatial memory. Right: the percentage of time spent in the platform quadrant was above chance level (dashed line) in all groups, while no genotype-dependent differences in spatial memory were detected between cre-expressing mice and wild-type littermates in any mouse line. All data is depicted as mean ± standard error of the mean. ns = not significant. **/*** for p < 0.01/0.001, respectively, tested against chance level. n = 8 females (circles) and 8 males (triangles). Numerical data underlying this figure can be found in S1 Data.
Fig 6.
PRS×8-mediated constructs to monitor and manipulate LC activity.
(A) PRS×8-jGCaMP8m was injected into one LC of a Dbhcre mouse and hSyn-DIO-jRGECO1a was injected in the contralateral LC. (B) Epifluorescence images of hemi-brain slices expressing PRS×8-driven jGCaMP8m (green, left) and cre-dependent jRGECO1a (magenta, right). Immunostaining against TH is shown in black, while dashed lines indicate the optical fiber tracts. Confocal images of anti-TH and anti-GFP staining, as well as a merged image with quantification of efficiency and specificity are shown next to the overview images. (C) Simultaneous fiber photometry recordings of jGCaMP8m fluorescence (top) and jRGECO1a fluorescence (center), along with the pupil diameter of the mouse (bottom). (D) ΔF (mean ± standard error of the mean) of jGCaMP8m (green) and jRGECO1a (magenta) as well as pupil diameter (gray) locked to local peaks in the derivative of the pupil diameter (indicated by dashed line, n = 171 peaks from 20 min of recording). Only events in the absence of locomotion (gray, bottom) were analyzed. (E) Pupil-aligned functional and isosbestic fluorescence of jGCaMP8m (top) and jRGECO1a (bottom). Functional excitation was done at 470 nm for jGCaMP8m and 560 nm for jRGECO1a, while isosbestic excitation for both indicators was done at 405 nm. (F) ΔF of green-light-filtered GCaMP8m (green) and jRGECO1a (magenta), locked to peaks in the derivative of pupil diameter, exclude major cross-talk from jGCaMP8m to the jRGECO1a channel (n = 52 peaks). (G) ΔF of red-light-filtered jGCaMP8m (green) and jRGECO1a (magenta), locked to peaks in the derivative of pupil diameter, exclude major cross-talk from jRGECO1a to the jGCaMP8m channel (= 84 peaks). (H) Scheme of bilateral injections of PRS×8-ChrimsonR-tdTomato into the LC of a wild-type mouse. (I) Epifluorescence image of a hemi-brain slice expressing PRS×8-driven ChrimsonR-tdTomato (left). tdTomato was amplified using a cross-reacting antibody against mCherry (magenta), while the LC was visualized with an anti-TH staining (black). Dashed lines indicate the fiber position (only the lesion caused by the fiber tip is visible in this slice). Transgene expression was quantified on confocal images (right). (J) Pupil diameter of the mouse during bilateral optogenetic activation of the LC (red bars; 633 nm, ~5 mW, 20 ms pulses at 20 Hz for 4 s). Insets show DeepLabCut-traced videographic images at indicated time points before and after an optogenetic stimulus. (K) Relative pupil diameter (color coded) during a recording session consisting of 30 trials with an inter-trial-interval of 30s. (L) Pupil size in response to optogenetic stimulation (mean ± standard error of the mean across trials shown in K). Absence of locomotion is shown in gray (bottom).