Resource availibility
Lead Contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by Lead Contact, Dr. David Talmage ([email protected]).
Materials Availability
Plasmids generated in this study have been deposited to Addgene and will be available upon publication under Talmage Lab.
Data and Code Availability
This study did not generate/analyze datasets. Code for fiber photometry data was previously published in (Crouse, Kim et al. 2020).
Experimental model and subject details
Adult (3-6 month) male and female Chat-IRES-Cre (B6;129S6-Chattm2(cre)Lowl/J, Jax stock number: 006410, (Rossi et al. 2011), Fos-tTA,Fos-shGFP (TetTag, Jax stock number: 018306, referred to as Fos-tTA/shGFP or Fos-shGFP), and Chat-IRES-Cre X Fos-tTA/shGFP mice were used. Mice within each cage were randomly assigned to experimental and control conditions. In all electrophysiology experiments, hemizygous Fos-tTA/shGFP mice on a C57BL/6 background were used. Mice were housed in a 12-hour light/dark cycle environment that was both temperature and humidity controlled. Mice had free access to food and water. All animal care and experimental procedures were approved by the Animal Care and Use Committees (ACUC) of the National Institute of Neurological Disorders & Stroke (NINDS) (Protocol #1531), SUNY Research Foundation at Stony Brook University (Protocol #1618), and Yale University (Protocol #2019–07895).
Method details
Viral construct
Construction of the ADCD probe
All cloning unless otherwise specified was performed using In-Fusion HD (Clontech). “mCherry-P2A” was amplified using Phusion High-Fidelity DNA Polymerase (NEB) from pV2SGE (obtained as a gift from Dr. Shaoyu Ge Stony Brook University). “oChIEF-LoxP-Lox2272” was amplified from pV2.2 (synthesized gene block from IDT). The two fragments were cloned into pAAV-WPRE linearized by BamHI. The resulting plasmid was linearized by Pml I. “7xTetO-LoxP-Lox2272-tTAH100Y.SV40” was amplified from pV2.1 (synthesized gene block from IDT) and cloned into the Pml I site. The final plasmid was packaged into AAV9 viral particles. Viral packaging was performed by the University of Pennsylvania Vector Core.
Note re: ADCD expression in BLA neurons in the presence of doxycycline:
As shown in Figure 2-Supplement 1C, we noted “leaky” expression of ADCD-mCherry in the presence of doxycycline, in the BLA of Fos-tTA mice when co-injected with a Cre expression vector expressed from a camk2a promoter. Co-injection of camk2a-Cre and ADCD-mCherry into cortex and hippocampus of wild-type (C57) mice was also found to result in “leaky” expression despite the absence of genetically encoded tTA. Injection of ADCD-mCherry in hippocampus of PV-Cre mice did not result in expression similar to injection in Chat-IRES-Cre mice (Figure 2-Supplement 1A, bottom). These findings underscore the importance of performing the appropriate controls when using these vectors in vivo.
Construction of the ADCD-DREADD probe
“BglII-hM4Di.mCherry-AscI” was amplified using CloneAmpTM HiFi PCR Premix (Takara) from pAAV-hSyn-DIO-hM4D(Gi)-mCherry (Krashes MJ, et al. 2011) (gift from Dr.Bryan Roth; Addgene plasmid # 44362; http://n2t.net/addgene:44362; RRID:Addgene_44362). A backbone with TRE and Lox sites was ligated with “BglII-hM4Di.mCherry-AscI” using T4 DNA Ligase (NEB). The final plasmid was packaged into AAV9 viral particles. Viral packaging was performed by the University of North Carolina Vector Core.
Stereotaxic surgery & viral delivery:
Three-four-month-old ChAT-IRES-Cre mice were anesthetized and stereotaxically injected bilaterally. Coordinates were calculated based on the Paxinos Mouse Brain Atlas (Franklin, K & Paxinos, G, 1997): BLA (-1.4mm A/P, ±3.5mm M/L, -4.8mm D/V), NBM (-0.7mm A/P, ±1.7mm M/L, -4mm D/V).
Tracers: 3% w/v solution of fast blue (FB) (17740-1, Polysciences Inc.) was prepared in sterile milliQ water. ~0.2µL of 3% FB was injected into the BLA bilaterally of Fos-GFP or Chat-IRES-Cre X Fos-tTA/shGFP mice. Mice were euthanized 7 days following injection.
Behavioral testing & analysis:
Threat conditioning: All training and assessments were completed with experimenter blind to condition. Both training and recall sessions were analyzed using FreezeFrame v.3 (see below).
Habituation: All mice were handled for a minimum of five minutes daily for three consecutive days before behavioral training began. For DREADD experiments, all mice were additionally habituated to restraint and injection with 100 µL saline administered i.p. daily.
Training: On training day, all chambers were cleaned with 70% ethanol. Mice were placed into the behavioral chamber for a 10 min session which consisted of 3 min of habituation, followed by 3 tone-shock pairings (30 s 80dB, 5kHz tone, co-terminated with a 2 s 0.7mA foot shock with a 1.5 min interval between each pairing), and finally 2 min of exploration. For DREADD experiments, mice were given 0.1 mg/kg Clozapine (administered i.p.) (Sigma Aldrich) 10 minutes prior to being placed in the chamber.
Recall: Recall session took place 24 - 72 hrs after completion of the training. To specifically test the response to tone-cued recall, the contextual features of the chambers were altered including texture of the floor, color of the walls, and scent of cleaner (mild lemongrass citrus-based solution). Mice were placed in the behavioral chamber for another 5 min session during which a single tone was delivered (30 s 80dB 5kHz tone) 2 min after being placed in the chamber. No shock was administered.
Analysis: Percent time spent freezing was quantified using FreezeFrame v.3 (Actimetrics). Bout duration (defined as minimum required duration when animal is frozen) was set to 1 s, and threshold was manually defined as highest motion index with no movement other than breathing. Percent time spent freezing (defined as periods of no movement) was quantified across the 10 min session in bins of 30s. The following periods were defined for statistical analysis: Baseline (average of all bins prior to tone onset), Tone response (average of all bins following tone onset).
High Responders were defined as those mice that exhibited at least a 10-percentage point increase in % time spent freezing in the 30s bin during the tone from the average of the pre-tone period (e.g. Pre-tone freezing 10% to tone-induced freezing of ≥20%). All other mice were considered Low responders. Prior to any behavioral manipulation, mice showed up to 10% (of total time in given time bin) freezing indicating this level of freezing to be non-associative (potentially related to novelty or generalized fear). This criterion was found to give statistically significant difference between pre-tone vs. tone only for high-responders and not for low-responders providing further validity to the delineation of the Low and High Responder groups.
Analysis of population composition of High and Low responders (Figure 6-Supplement 1) was performed within experiment. Cross-experiment comparisons for population composition of High and Low responders was not possible due to differences in conditions and variability within and between cohorts.
Engram labeling:
Mice were placed on doxycycline hyclate-containing chow (Cat# TD.08541 Envigo) at least 2 days prior to injection of activity-dependent viral markers. Threat conditioning was performed as mentioned above. During doxycycline withdrawal, mice were transferred to a clean cage to prevent mice from eating dox food that was dragged into the cage or buried in the bedding. To minimize stress, some bedding containing fecal pellets and urine, and nest from the old cage were transferred to the new cage.
Predator odor exposure
Habituation: All mice were habituated to restraint and injection with 100 μL saline administered i.p. daily for 3 days prior to behavioral testing for DREADD experiments. On exposure day, mice were transported to the lab several hours prior to exposure and habituated to the room and ambient sounds.
Exposure: For exposure to predator odors, a vented mouse cage (L 13in x W 7.5in x H 5.5in) with corncob bedding (EnviroDri) was placed in a designated location in a laminar flow hood with overhead fluorescent lighting. Mt. Lion Pee (Maine outdoor solutions LLC) was obtained from predatorpee.com and stored at 4°C. 200µL of urine was pipetted onto a 3in x 3in 12 ply gauze pad (Cat#6312, Dukal corp.) placed in a polystyrene petri dish (VWR) at the vented end of the cage. Mice were placed into the cage in the end away from the odor and the cage was covered using a clear plexiglass barrier. Mice were exposed for 5 min and the session was filmed using an overhead digital camcorder (Sony). Following exposure, mice were returned to their home cage or a holding cage in the case of multiple housed mice to prevent any odor transfer. Control mice were exposed to 0.9% saline. For DREADD experiments, mice were given 0.1 mg/kg clozapine (administered i.p.; Sigma Aldrich) 15 minutes prior to being placed in the chamber.
Analysis: behavior was manually scored using Jwatcher (v0.9). Defensive digging was defined as vigorous digging performed by the mice using their snout, flinging bedding up and away from the animal. Freezing was defined as immobility without any obvious motion besides breathing. Cloth contacts were defined as front paw touches to the odor pad.
Fiber Photometry
Acquisition
Fiber photometry recordings were made using a Doric Lenses 1-site Fiber Photometry System. Signal was recorded using Doric Neuroscience Studio (V 5.3.3.4) via the Lock-In demodulation mode with sampling rate of 12.0 kS/s. Data was downsampled by a factor of 10 and saved as a comma-separated file. For details on connection of the setup refer to Crouse RB., et al. 2020.
Analysis
Preprocessing of the raw data was performed using a MATLAB script provided by Doric. The baseline fluorescence (F0) was calculated using a least mean squares regression over the duration of the recording session. The change in fluorescence for a given timepoint (ΔF) was calculated as the difference between it and F0, divided by F0, and multiplied by 100 to yield % ΔF/F0. The % ΔF/F0 was calculated independently for both the signal (465 nm) and reference (405 nm) channels and a final “corrected % ΔF/F0” was obtained by subtracting the reference % ΔF/F0 from the signal % ΔF/F0 at each timepoint. The corrected % ΔF/F0 was z-scored to give the final “Z % ΔF/F0” reported. Area under the curve was calculated for 1s duration before (baseline) and 1s after tone onset. The average of all the baseline periods within each analysis was used as the baseline reading for the AUC analysis.
Electrophysiology:
Brain slice preparation
For slice physiology, mice were anesthetized and transcardially perfused with cutting solution (sucrose 248 mM, KCl 2 mM, MgSO4 3 mM, KH2PO4 1.25 mM, NaHCO3 26 mM, glucose 10 mM, sodium ascorbate 0.4 mM and sodium pyruvate 1 mM, bubbled with 95% O2 and 5% CO2) at 40ºC. The brain was then rapidly removed and sliced, coronally, at 300 µM in oxygenated cutting solution at 40ºC. Prior to recording, slices were incubated in oxygenated incubation solution (sucrose 110 mM, NaCl 60 mM, KCl 2.5 mM, MgCl2 7 mM, NaH2PO4 1.25 mM, NaHCO3 25 mM, CaCl2 0.5 mM, MgCl2 2 mM, glucose 25 mM, sodium ascorbate 1.3 mM, and sodium pyruvate 0.6 mM) at room temperature.
Electrophysiological recording:
During recording, slices were superfused with oxygenated artificial cerebral spinal fluid (Jiang et al. 2016). Fos+ neurons were identified by GFP expression. Signals were recording using patch electrodes between 4-6 MΩ, a MultiClamp 700B amplifier, and pClamp10 software. Pipette internal solution was as follows: 125 mM K‐gluconate, 3 mM KCl, 1 mM MgCl2, 10 mM HEPES, 0.2 mM CaCl2, 0.1 mM EGTA, 2 mM MgATP, and 0.2 mM NaGTP (pH = 7.3). Following recording, cytoplasm was harvested via aspiration for cell-type identification using single-cell RT-PCR. Ten-twelve basic electrical properties were determined and defined as previously described(López-Hernández, Ananth et al. 2017). Recordings were excluded if they did not meet the following criteria: 1. membrane potential less than or equal to -50 mV, 2. input resistance between 100-300 MΩ, 3. series resistance < 10 MΩ that was unchanged throughout the recording, and 4. firing a 45 mV action potential at rheobase
Single cell reverse transcription-PCR:
Single cell samples were pressure ejected into a fresh RT buffer prep (Applied biosystems). Samples were sonicated in a total volume of 20 µL at 400C for 10 min before addition of RT enzyme mix (Applied Biosystem). Tubes were incubated at 370C for 60 minutes and then 950C for 5 minutes. Two rounds of amplification (30 cycles each) were done for the detection of Chat transcripts. For the first round of amplification (reaction volume 25 µL) included 2X mastermix, sterile water, 0.2 mM of each primer, 1 mL of cDNA sample). For the second amplification, the reaction included 1 µL of the previous (first-round) PCR product, 2X mastermix, sterile water, and 0.2 mM of each primer. Whole brain cDNA was run in parallel with the single cell samples. After amplification, the PCR products (159 bp) were analyzed on 3% gels.
Immunohistochemistry:
Following perfusion, brains were fixed overnight at 4°C in 4% PFA (in 1XPBS) and were then transferred to a 30% sucrose solution (in 1XPBS). Brains were flash frozen in OCT Compound (Tissue Tek) and stored at -80°C until cryosectioning. 50 µm cryosections were mounted onto Superfrost slides (Fisher Scientific) in sets of 3 and allowed to dry overnight at room temperature. Sections were blocked overnight at 4°C in a PBS solution containing 0.3% TritonX-100 and 3% normal donkey serum and then incubated with primary antibody in a PBS-T solution (0.1% TritonX-100 and 1% normal donkey serum), overnight (24h at 4 ̊C). The next day, sections were rinsed in PBS-T and incubated in secondary antibody for 2 hr at room temperature in PBS-T along with NeuroTrace-435 (Invitrogen). Sections were treated with an autofluorescence eliminator reagent (EMD Millipore) according to the manufacturer's guidelines and mounted in Fluoromount-G (Southern Biotech). Details regarding antibodies can be found in the Key Resources Table (KRT).
Quantification and statistical analysis
Imaging and analysis:
All imaging was conducted on an Olympus wide-field slide-scanner microscope at 20X magnification (VS-120 and VS-200 systems, Z-step= 3 µm). Images were processed using the cell counter plugin on ImageJ. For Fos+ cell counts in the amygdala, only neurons (Nissl/ Neurotrace positive) with nuclear Fos stain were counted. The amygdala was identified, and a region of interest (ROI) defined using ROI manager in Image J. Total area of the ROI was measured and noted. Fluorescence threshold was set to eliminate background fluorescence in ImageJ (defined as hazy background signal detected in space between neurons and white matter). This eliminated non-specific fluorescence and out of focus signals. Fos+ nuclei were then counted using the cell counter plugin.
For ADCD cell counts, mCherry+ neurons at the NBM/SI injection site were counted. NBM was consistently identified as the cluster of cholinergic cell bodies at the base of the internal capsule in the Globus Pallidus and the SI as the area located directly ventral to the GP as denoted by the Paxinos Mouse Brain Atlas (3rd Edition). 100% of the analyzed area of every third brain section was counted (~150 µm apart). Since the NBM/SI regions lack defined boundaries, we present the data as cell counts as opposed to cell density.
For Fos analysis in the BLA, Fos+ cells were counted in the area enclosed within the external and amygdalar capsules. Since the shape of the BLA changes along the anterior-posterior axis, Fos+ cell counts were normalized to the area enclosed within the external and amygdalar capsules and presented as density of Fos+ cells.
Statistical analysis:
Statistical analyses were done using GraphPad Prism (GraphPad Software Inc., San Diego, CA, USA), Sigmaplot 12.5 (Systat Software, Inc., San Jose, CA, USA) and OriginPro 9.1 (Origin Lab Corporation, Northampton, MA, USA). Normality of the data was assessed using Shapiro-Wilk and Smirnov-Kolmogorov tests. Data that were not normally distributed according to both normality tests, were analyzed using appropriate non-parametric tests. Detailed information on statistical tests used, p-values, and sample sizes, and other descriptive statistics can be found in the text (Figure Legends) and/or in the statistical reporting table (Supplementary File 1). Sample sizes for behavior experiments were determined using a power calculation based on effect sizes in pilot experiments with power set to 0.8.
Parametric tests used: Repeated Measures (RM) One-way ANOVA, RM Two-way ANOVA, Welch’s ANOVA, paired t-test (two tailed), Welch’s t-test.
Non-parametric tests used: Mann-Whitney test, Wilcoxon matched-pairs signed rank test, Kruskal-Wallis Test, Friedman Test.
p-value criteria: * p≤0.05, ** p≤0.01, *** p≤0.001, **** p≤0.0001