Interplay between hippocampal TACR3 and systemic testosterone in regulating anxiety-associated synaptic plasticity … – Nature.com

All the experiments carried out here were previously approved by the committees for ethical care and use of animals for experimentation at the Ben-Gurion University (b14764_30) and the University of the Basque Country (M20/2016/001; M20/2018/296; M20/2016/019). The experiments were all carried out in accordance with the guidelines of the European Community Council Directives (2010/63/EU).

Experiments were conducted using adult Wistar rats weighing between 280 and 450 grams and aged 34 months, as well as rats ranging from embryonic day 18 (E18) to 30 days old for developmental studies. With the exception of a biochemical investigation focusing on TACR3 expression during the estrous cycle, all other experiments exclusively utilized male rats. Prior to the experimental phase, animals were acclimated for a minimum of 15 days in Plexiglas cages, housing five rats per cage unless specified otherwise. These rats were maintained under controlled environmental conditions, with a temperature of 212C and a 12:12-h light-dark cycle (lighting from 07:00 to 19:00h). Throughout the acclimation period, the rats had ad libitum access to food and water, which was restricted only during active experimentation.

All behavioral, electrophysiological, and morphological experiments were conducted under blinded conditions to minimize bias. The investigator performing the assessments was unaware of the treatment groups or experimental conditions until the completion of data analysis.

Animals were randomized into different experimental groups based on their weight to ensure an equitable distribution across all conditions. Random assignment was then carried out within each stratum to ensure that each experimental group had a comparable range and mean weight.

Gene expression was analyzed using RNA extracted from the ventral hippocampus of male rats displaying Severe (SA) or Moderate Anxiety (MA: Supplementary Fig.1). Differentially expressed genes (DEGs) were defined as those with a absolute log-fold change between SA and MA anxiety above 1, and P value<0.05. Gene ontology (GO) and KEGG enrichment analysis was performed using DAVID [26].

We assessed anxiety-related behavior with the elevated plus-maze (EPM) test. The maze has two open arms (measuring 4510cm) and two enclosed arms (measuring 451050cm) located opposite each other, and it is connected to a central platform (measuring 1010cm) elevated 65cm above the floor. For juvenile rats, we used a smaller maze (arms dimensions: 35540cm). Each rat was placed in the central compartment facing one of the enclosed arms and allowed to freely explore the maze for 5min. The rats movements were recorded with a video camera and analyzed with a computerized tracking system (AnyMaze, Stoelting) that registered their entry into an arm as soon as all four paws were in it. The time spent in the open and closed arms and in the central compartment was recorded. To quantify the anxiety-like behavior of the rats, we divided the time spent in the open arm by the total time in both arms. This value was used to classify rats as having moderate (MA), intermediate (IA) or severe (SA) anxiety based on their scores relative to the overall distribution of the scores from four experiments involving 186 rats (Fig.1a, b). Specifically, MA rats had scores above the 90th percentile, SA rats had scores below the 10th percentile, and IA rats had scores between the 40th and 60th percentiles.

a Experimental design. Rats were categorized based on their performance in the elevated plus maze (EPM), and two weeks later, their hippocampus was extracted for gene expression analysis. b Classification of rats in the EPM. Left: Representative traces from the EPM showing the path (left) and color-coded time spent in each location of the maze (right) by rats categorized with moderate (MA) or severe anxiety (SA). Right: Frequency distribution of the EPM scores for all rats: the rats with extreme scores, indicated in color, were selected for gene expression analysis. c Volcano plot of the differential gene expression in MA and SA rats. Upregulated genes are shown in red, downregulated genes in blue, and non-significantly changed genes in gray, based on their statistical significance (-log10 p-value) and fold change (log2 fold change) values. d Hierarchical clustering was performed on eight samples (four with SA and four with MA) using Euclidean distances calculated from the expression of 172 differentially expressed genes (DEGs). The clustering analysis resulted in the formation of distinct clusters, and the colors in the heat map represent row-scaled expression values, with blue indicating weak expression and red indicating strong expression. Dot plots illustrating the enriched (e) KEGG pathways, (f) GO biological processes, and (g) GO cellular components associated with the DEGs. Each dots position on the x-axis represents the number of genes out of the 172 DEGs enriched for the corresponding term displayed on the y-axis. The dots size and color indicate the GeneRatio (proportion of DEGs within the pathway/process/component out of the 143 DEGs found in the DAVID database) and the level of significance, respectively. The terms are ordered based on the number of DEGs on the x-axis. Terms with an FDR (False Discovery Rate)<0.1 or containing TACR3 or CAMK2B genes are marked [26] (For a comprehensive list of genes, see 10.5281/zenodo.8305270).

Adult male Wistar rats (3 months old) were injected subcutaneously on five consecutive days with the vehicle alone (mineral oil) or with testosterone propionate (5mg/kg/day: #86541-5G Sigma-Aldrich, Fig.3d). The hippocampus was then extracted and lysed by sonication, and TACR3 was analyzed in western blots.

Animals were divided into two groups, control or osanetant treated (Sigma, SML0798), and the latter were administered osanetant (25mM) prepared from a stock solution diluted in DMSO while the control animals were injected with the same volume of saline (vehicle). Before treatment, the stock solution was diluted in 0.9% sterile saline up to 1ml for each animal according to their body mass. Osanetant was administered intraperitoneally (ip) at a dose of 5mg/kg.

Blood samples (approximately 200l) were collected twice from the tail vein: before treatment and one day after the last treatment with osanetant or testosterone. Blood samples were centrifuged at 10,000g for 5min at 20C, and the serum retrieved was stored at 20C for further analyses.

Following Testosterone treatment, serum testosterone was measured with the testosterone Parameter Assay Kit (R&D Systems, #KGE010) as indicated by the manufacturer. For osanetant treatment, serum testosterone was measured at the Endocrinology Lab of the Soroka Medical Center by competitive Immunoassay using direct chemiluminescent Technology on an ADVIA Centaur XPT machine (SIEMENS). The threshold for detection was 0.07ng/mL.

Anesthetization of male Wistar rats (age: 3 months) was achieved using 2.5% isoflurane. Intracerebroventricular (i.c.v.) delivery cannulas from Alzets brain infusion kit II were surgically implanted using a stereotaxic frame (KOPF Instruments). The implantation was carried out at specific coordinates relative to the bregma: AP, 0.8mm; ML, +1.6mm; and DV, 4.0mm. Osmotic minipumps (Alzet; model #2004) were loaded with either 100nM of Osanetant (Sigma-Aldrich; SML0798) or a control vehicle (sterile 0.9% NaCl of medical grade). These pumps were pre-equilibrated in 0.9% NaCl solution at 37C for 48h. Subsequently, the osmotic minipumps were connected to the i.c.v. cannula tubing and were subcutaneously implanted on the rats back. For post-surgery pain management, subcutaneous injections of long-acting Buprenorphine at a dosage of 0.65mg/kg were administered, with a second injection of the same dosage given 72h later. Following a 10-day recovery period, behavioral testing was performed.

Rat Tacr3-mcherry (NM_017053.1) was sub-cloned from a synthesized template (VectorBuilder), and the whole construct was amplified by PCR using the primers: GCTCTAGAGCCACCATGGCCTCAGTCC, AACATGCATGCTTACTTGTACAGCTCGTCC. The resulting construct was cloned into the Sindbis vector pSinRep5 between XbaI and PaeI restriction sites. The rat Tacr3-IRES-EGFP construct was first sub-cloned into pHA-IRES-EGFP between BcuI and PstI, and TacR3 was PCR amplified from a template using the primers: GACTAGTGCCACCATGGCCTCAGTCC, GCACTGCAGTTAGGAATATTCATCCACAGAGGTA. The whole TacR3-IRES-EGFP construct was then amplified using specific primers (GCTCTAGAGCCACCATGGCCTCAGTCC, AACATGCATGCTTACTTGTACAGCTCGTCC) and cloned into the pSinRep5 Sindbis vector between at the XbaI and PaeI restriction sites. After successful ligation, the plasmids were linearized for transcription, and recombinant RNA transcripts were then synthesized using the SP6 promoter and transfected into BHK cells (Supplementary Fig.2).

Sindbis virus was prepared as described previously [27,28,29]. Briefly, plasmids containing the protein of interest (pSinRep5) and the helper plasmid (pDHtRNA) were linearized and purified using phenol-chloroform extraction, followed by ethanol precipitation. In vitro RNA transcription was performed using the mMESSAGE mMACHINE SP6 Transcription Kit (Thermo Scientific, AM1340), and the RNA obtained was then purified using phenol-chloroform extraction, followed by isopropanol precipitation.

For each nucleofection, a total of 10106 BHK-21 cells were electroporated and resuspended in 100l of Cell Line Nucleofector solution (Lonza, VCA-1005), along with 10g of the transcript of interest and 10g of the helper RNA. Electroporation was carried out using the protocol for the BHK-21 cell line with the Amaxa Nucleofector II system. Immediately after electroporation, the cells were plated onto a 150mm dish and maintained at 37C in 5% CO2. BHK 21 (Clone 13) from Hamster Syrian kidney was from Sigma-Aldrich (#85011433). This commercial cell line is tested by the ECACC for mycoplasma.

At 4872h post nucleofection, the medium containing viral particles was recovered and concentrated by ultracentrifugation for 2h on a 20% sucrose cushion at 25,000rpm using a SW28 rotor. The supernatant was discarded, and the pellet was resuspended in 5% Fetal Bovine Serum (FBS) in a neurobasal medium (NBM: Life Technologies, 21103049). The virus was then stored at 80C for further use.

Neuronal cultures in 96-well plates maintained at 37C in 5% CO2 and at a controlled humidity were transferred to a SPARK Multimode Microplate reader (Tecan), and after obtaining a baseline recording of 560min, the cultures were treated as desired. For osanetant treatment, primary neuronal cultures (15 DIV) were infected over 24h with Sindbis virus expressing SEP-GluA1 (a pH-sensitive fluorescence protein). The next day, the media was replaced with an equilibrated bathing solution, and the cells were incubated for 10min before reading in the SPARK reader for 30min. A portion of the plate was then treated with osanetant (final concentration 100nM), and any fluorescence changes were recorded every 30min for 4h. For the induction of chemical LTP (cLTP), glycine was added for 5min at a final concentration of 200M, while control samples were treated with the bathing solution without glycine. After a 5-min incubation, the solution was replaced with the bathing solution (glycine-free), and the plate was read every 30min over 4h. Fluorescence readings were obtained using 475Ex/535Em nm filters.

The electrophysiological activity was recorded using an Axion Maestero Edge recording system with 16 extracellular recording electrodes and a ground electrode per well on a 24-well multielectrode array (MEA: Axion Biosystems, M384-tMEA-24W). Neurons were plated at a density of 30,000 per well in NBM (5L) with 10% FBS (Atlanta Biologicals, S11550), and they were allowed to attach to the plate for 2h, after which 300L of serum-free NBM was added. At 9 DIV, 50% of the medium was changed and supplemented with BrainPhys medium, and at 12 DIV, 50% of the medium was changed with supplemented NBM. From 13 DIV, neurons were recorded using Axion AxIS Navigator software over 10120-min intervals. Electrical activity was measured with an interface board at 12.5kHz, digitized, and transmitted to an external computer for data acquisition and analysis.

All voltage data were filtered using dual 200Hz (high pass) and 3000Hz (low pass) filters, and action potential thresholds were set automatically using an adaptive threshold for each electrode (>6 standard deviations from the electrodes mean signal). Waveforms collected with the Axion AxIS Navigator were exported to a Plexon Offline Sorter (v4) for automatic completion, and the Principal Components Analysis (PCA) was plotted on the waveforms for each electrode. The K-means clustering algorithm was used to split the waveforms by source units, allowing per-neuron analyses using NeuroExplorer (v5), Axion Neural Metric Tool, and ad-hoc Python scripts.

A cross-correlation analysis was used to assess the connectivity between pairs of neurons [30], with correlograms presenting the conditional probability of a spike from one neuron, given that a spike occurred in a reference neuron. Only sorted neurons with firing rates above 20 spikes per recording (1/30Hz) were filtered for cross-correlation calculations. Cross-correlograms were calculated between each possible pair of neurons in the same well using NeuroExplorer software, and the average cross-correlogram for each treatment was calculated considering all the neuron pairs under the same treatment for each record separately. The average cross-correlogram peak for each treatment and recording was also calculated. All data analysis was performed using ad-hoc in-house Python scripts.

RNA was extracted from the neuronal cell lysate using the NucleoSpin RNA mini kit (Macherey-Nagel), and this RNA was reverse transcribed with All-In-One 5X RT MasterMix (ABM, #G592), diluting the resulting cDNA to 100ng/l. Gene-specific primers were designed with the Primer-BLAST NCBI tool, and their sequences are listed. Real-time PCR (RT-PCR) was carried out on a LightCycler 480 (Roche) and using SYBR Green PCR Master Mix (Applied Biosystems), with an initial denaturation at 95C for 20s, followed by 40 cycles at 95C for 3s and 60C for 30s. Each sample was run in triplicate, and the 2-Ct method [31] was used for the relative quantification of gene expression, with changes in gene expression normalized to an internal control gene (Actin). The following primers were used: TACR3 Forward CACAAGCGCATGAGAACTGT; TACR3 Reverse: AAGTTCTGGAAGCGGCAGTA; Actin Forward: CCCTACAGTGCTGTGGGTTT; Actin Reverse: GCAAGGAGTGCAAGAACACA.

A protocol described previously was followed for cLTP induction with minor changes [29, 32]. Specifically, neuronal cultures at 37C and in 5% CO2 were incubated in 200M glycine-containing extracellular solution at pH 7.4 (in mM): 129 NaCl, 4 KCl, 4 CaCl2, 10 HEPES, 10 Glucose. The controls were incubated in a glycine-free extracellular solution alone (vehicle).

Primary neurons: To assess the overall morphology of dissociated neurons, cells (2024 DIV) were infected with the EGFP Sindbis virus for 24h to visualize the dendrites and dendritic spines. The cells were fixed in fresh 4% PFA in PBS for 10min at room temperature and washed three times with PBS. The cells were then covered with Prolong Gold Antifade Reagent (Thermo Fisher Scientific, P36934), and after 24h, they were visualized on a Zeiss LSM880 Airyscan confocal microscope equipped with an Argon 488nm laser line. A tile-scan application was used to obtain images of whole neurons (10x) or dendrites (63x), quantifying spines using Imaris 9.7.2 software (Bitplane Inc.) and dividing the number of spines by the corresponding dendritic length to calculate the spine density on each dendrite.

To test the effect of testosterone and osanetant on dendritic spines, primary hippocampal cultures from rats were infected with a Sindbis virus expressing EGFP at 16 DIV. At 20h post-infection (hpi), the neurons were exposed to either osanetant (100nM) or testosterone (10nM) for 2h in a growth medium. In rescue experiments, neurons were initially exposed to testosterone (10nM) for 2h, and then osanetant (100nM) was added for a further 2-h incubation. The neurons were then fixed in 4% PFA and washed in PBS with 4% sucrose for 10min and then three times with PBS prior to mounting the coverslips with ProLong Gold Antifade Mounting medium (Invitrogen) and visualizing them on an Olympus IXplore SpinSR10 microscope.

Brain slices. Intracellular injections of Lucifer Yellow: Rats were anesthetized with pentobarbital (0.04mg/kg) and transcardially perfused with 300ml of 4% PFA (pH 7.4) prior to removing their brain. Each brain was coded (codes were not broken until after the quantitative analysis), post-fixed in 4% PFA (pH 7.4) for 24h, and coronal microtome sections (150m: Leica VT1000 S Vibrating blade) were labeled with 4,6-diamidino-2-phenylindole (DAPI: Sigma D9542) for 12min. Cells in the ventral dentate gyrus (DG) and the lateral nucleus of the amygdala were injected individually with 4% Lucifer Yellow (CH: Sigma-Aldrich) in 1M LiCl (pH 7.4) by passing a steady hyperpolarizing current through the electrode (0.5 to 1.0nA).

Following injection, the sections were probed overnight with an antibody specifically targeting Lucifer Yellow (1:200 Rabbit: Thermo Fisher Scientific #A-5750) and then for 4h with an Alexa Fluor Plus 488 conjugated Goat anti-Rabbit IgG (H+L) secondary antibody (1:1000: Invitrogen, # A32731). The sections were preserved and then mounted in fresh ProLong Gold antifade reagent (Invitrogen, Eugene, OR), the slides were left in the dark at room temperature for 24h to cure the mounting medium, and finally, the coverslips were sealed using nail polish.

Sections were visualized on a Zeiss laser scanning multispectral confocal microscope equipped with an argon laser. The acquired image stacks had a physical size of 76.976.9m and a logical size of 10241024 pixels. These stacks consisted of 100350 image planes captured through a 63 glycerol immersion lens (NA 1.3, working distance 280m, and refraction index 1.45). To optimize the imaging, a zoom factor of 3.2 was calculated, resulting in a voxel size of 75.175.1136.4nm with a z-step of 0.14m. For each rat (5 neurons per rat), 15 randomly selected dendrites were scanned from the soma to the tip. Subsequently, the acquired stacks were processed using a 3D blind deconvolution algorithm (ClearView GPU Accelerated Deconvolution), applying 10 iterations to reduce the impact of out-of-focus light.

Dendrites were traced using the Neurolucida 360 software (MicroBrightField Inc., Williston, VT). Dendritic spine densities were determined on granular neurons in the DG or on pyramidal-like neurons in the lateral nucleus of the amygdala, traced from their proximal to distal tips and marking the presence of spines during the tracing process. This analysis was performed on five neurons per rat for each area, and all protrusions observed were considered spines and included in the analysis without applying any factors to correct the spine counts. The reconstructed data were then exported to Neurolucida Explorer (MicroBrightField Inc., Williston, VT) for further quantification. The spine density was automatically calculated, as indicated previously.

Spine head volume was measured using Imaris 9.7.2 software (Bitplane AG, Zurich, Switzerland) [33, 34] and for each dendritic segment, various intensity thresholds were applied to generate a data model that was visualized as a solid surface using the IsoSurface module. Subsequently, the solid surface corresponding to the contour of each spine head was selected. The three-dimensional image of each dendrite was rotated and carefully examined to verify the accuracy of the solid surface selected for each spine head. Spines with no visible head were extremely rare and were not included in the analysis.

For the in vivo studies, rats were anesthetized with urethane (1.6g/kg i.p.) to assess LTP induction in the DG. Surgical procedures and recordings were performed while the animals were situated in a Kopf stereotaxic device. Field potentials were obtained using Nichrome microelectrodes (<1M, 120m thick), and the perforant pathway was stimulated using a bipolar electrode (World Precision Instruments) at double the threshold intensity to elicit a response (1050A). The experimental protocol consisted of a 10-min baseline period to establish stable activity, with the stimulation pathway activated at 0.5Hz. Subsequently, three stimulation trains of 100Hz were delivered for 500ms each, with a 2s interval between trains to induce LTP. Following LTP induction, the pathway was stimulated again at 0.5Hz for 30min, and the average evoked field potential was calculated every minute (30 stimuli). The slope of the evoked field potential was measured and plotted, with the mean slope during the control period considered as 100%.

Rats were anesthetized with sodium pentothal (20mg/kg of body weight, intraperitoneal), decapitated, and their brain was rapidly removed and placed in oxygenated, ice-cold dissection solution: 10mM D-glucose, 4mM KCl, 26mM NaHCO3, 233.7mM sucrose, 5mM MgCl2, 1:1000 Phenol Red. Coronal microtome slices (300m, Leica VT1000 S Vibrating blade) were placed in a recovery chamber containing Artificial CSF (aCSF) at 2426C for at least 1.5h before recording. The aCSF, with an osmolarity adjusted to 290mOsm, was used for recovery and recording: 119mM NaCl, 2.5mM KCl, 1mM NaH2PO4, 11mM glucose, 1.2mM MgCl2, 2.5mM CaCl2. A concentric bipolar platinum-iridium stimulation electrode and a low-resistance glass recording microelectrode filled with aCSF (34M resistance) were placed in the middle molecular layer to record extracellular field excitatory postsynaptic potentials (fEPSPs). In each slice, an input-output (I/O) curve was recorded to compare the basal synaptic transmission in different animals.

The primary antibodies used here were raised against: NK3R (TACR3, Assay Biotech R12-3093), NeuN (polyclonal: Synaptic Systems, 266 006), GluA1 (Cell signaling 13185S or Abcam #ab31232), phospho-CaMKII, T286 (Millipore #05-533), CaMKII (Sigma Aldrich #C6974), PSD95 (NeuroMab 75-028), synaptophysin, (Millipore #MAB329), -actin (Cell signaling 4970S or Cell Signaling Technology 4970S), Phospho-(Ser) PKC Substrate Antibody (Cell signaling #2261) GAPDH (Santa Cruz Biotechnology sc-47724). The secondary antibodies used were: anti-mouse and anti-rabbit IgG HRP-linked secondary antibodies (Cell Signaling 7076S and 7074S), goat anti-Chicken Alexa Fluor 633 (Invitrogen 10444562), cross-adsorbed Alexa Fluor 594 and 488 goat anti-Mouse IgG (H+L) antibodies (Thermo Scientific A-11005 and A-21121), and cross-adsorbed Alexa Fluor 488 and 594 goat anti-Rabbit IgG (H+L) secondary antibodies (Thermo Fisher Scientific A-11008 and A-11012).

The drugs used in this study were: glycine (Bio Lab Ltd UN #071323), strychnine (Sigma-Aldrich #S0532-5G), senktide (Tocris #1068), osanetant (Sigma-Aldrich SML0798-25MG), and testosterone propionate (Sigma-Aldrich # 86541-5G).

Analyses were carried out using GraphPad Prism software (version 8.00, GraphPad Software, La Jolla, CA, USA). A Kolmogorov-Smirnov normality test was used to assess the distribution of the datasets, applying parametric or non-parametric analysis as appropriate. The data was presented as the meanstandard error of the mean (SEM), and the number of animals, cells, spines, or cultures are indicated in each figure. All experiments were carried out at least three times and the data presented are the combined results of all these repetitions. The statistical tests used and the P values are indicated in the figures or their corresponding legends.

We did not assume equal variances among groups, and as a precautionary measure, non-parametric tests were used for statistical comparisons unless explicitly stated otherwise. All tests are two-sided. Adjustments for multiple comparisons were made using the Bonferroni method. Throughout the study, center values are defined as the mean, and error bars represent the standard error of the mean (SEM).

No a priori sample size calculation was performed to detect a pre-specified effect size for this study. The sample size was determined based on previous similar experiments and the feasibility of the study within the given time frame. Additionally, no animals were excluded from the analysis; all animals that were subjected to experimental conditions were included in the final data set. The inclusion of all subjects and the absence of a pre-calculated sample size should be considered when interpreting the study results.

The other methods used in the study are described in theSupplemental Materialsand Methods.

Continue reading here:
Interplay between hippocampal TACR3 and systemic testosterone in regulating anxiety-associated synaptic plasticity ... - Nature.com