Previously we showed that the vitamin A metabolite all-trans retinoic acid (atRA) induces synaptic plasticity in acute brain slices prepared from the mouse and human neocortex (Lenz et al., 2021). Depending on the brain region studied, distinct effects of atRA on excitatory and inhibitory neurotransmission have been reported. Here, we used intraperitoneal injections of atRA (10 mg/kg) in adult C57BL/6J mice to study the effects of atRA on excitatory and inhibitory neurotransmission in the mouse fascia dentate – a brain region implicated in memory acquisition. No major changes in synaptic transmission were observed in the ventral hippocampus while a significant increase in both sEPSC frequencies and synapse numbers were evident in the dorsal hippocampus 6 hours after atRA administration. The intrinsic properties of hippocampal dentate granule cells were not significantly different and hippocampal transcriptome analysis revealed no essential neuronal changes upon atRA treatment. In light of these findings, we tested for the metaplastic effects of atRA, i.e., for its ability to modulate synaptic plasticity expression in the absence of major changes in baseline synaptic strength. Indeed, in vivo long-term potentiation (LTP) experiments demonstrated that systemic atRA treatment improves the ability of dentate granule cells to express LTP. The plasticity-promoting effects of atRA were not observed in synaptopodin-deficient mice, therefore, extending our previous results regarding the relevance of synaptopodin in atRA-mediated synaptic strengthening in the mouse prefrontal cortex. Taken together, our data show that atRA mediates synaptopodin-dependent metaplasticity in mouse dentate granule cells.

Key Resource Table.

Ethics statement. All experiments were performed according to German animal welfare legislation and after positive evaluation by the local authorities (University of Freiburg, AZ G-19/152; Faculty of Medicine at the University of Frankfurt, AZ FU/1131). Animals were kept in a 12-hour light/12-hour dark cycle with access to food and water ad libitum. Every effort was made to minimize pain or distress of animals.

Pharmacological treatment. All-trans retinoic acid (atRA; Sigma-Aldrich) was dissolved in DMSO and stored at -20°C until further use. The injection solution was prepared immediately before injection by adding corn-oil to prediluted stocks to achieve a final concentration of 5% DMSO (v/v). Before use, the solution was vortexed briefly. The solution was intraperitoneally injected in adult (C57BL/6J; 6–10 weeks old) male mice at an atRA concentration of 10 mg/kg. Control animals were injected with a vehicle-only solution (5% DMSO in corn oil) but otherwise treated equally. After injection, no overt behavioral changes were observed. Experiments were performed 3–6 hours after intraperitoneal injections.

Preparation of acute mouse hippocampal slices. Adult mice were anesthetized with ketamine/xylazine (100 mg/kg ketamine and 20 mg/kg xylazine) and rapidly decapitated. Brains were removed and further dissected for the preparation of acute slices of the ventral hippocampus as previously described [30]. For the preparation of acute slices of the dorsal hippocampus, the rostral and caudal parts of the brains were removed to ensure stable coronal sectioning. Brains were immediately transferred to a cooled oxygenated extracellular solution (5°C; 5% CO₂ / 95% O₂) containing (in mM): 92 NMDG, 2.5 KCl, 1.25 NaH₂PO₄, 30 NaHCO₃, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 0.5 CaCl₂, and 10 MgSO₄; pH = 7.3 to 7.4 at ~7°C (NMDG-aCSF; [31]). 300 µm tissue sections were cut with a Leica VT1200S vibratome. Slices were transferred to cell strainers with 40 µm pore size placed in NMDG-aCSF at 34°C, and the sodium levels were gradually increased following a protocol as described before [31]. After recovery, slices were maintained for further experimental assessment at room temperature in extracellular solution containing (in mM): 92 NaCl, 2.5 KCl, 1.25 NaH₂PO₄, 30 NaHCO₃, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 2 CaCl₂, and 2 MgSO₄.

Whole-cell patch-clamp recordings. Dentate granule cells in the suprapyramidal blade of the dentate gyrus were recorded in a bath solution (35°C) containing (in mM): 92 NaCl, 2.5 KCl, 1.25 NaH₂PO₄, 30 NaHCO₃, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 2 CaCl₂, and 2 MgSO₄. Granule cell somata close to the molecular layer of the dentate gyrus were visually identified using an LN-Scope (Luigs & Neumann, Ratingen, Germany) equipped with an infrared dot-contrast and a 40x water immersion objective (Olympus, NA 0.8). Recorded signals were amplified using a Multiclamp 700B amplifier, digitized with a Digidata 1550B digitizer and visualized with the pClamp 11 software package. For recordings of spontaneous excitatory postsynaptic currents (sEPSC) and intrinsic cellular properties, patch pipettes with a tip resistance of 3-5 MΩ were used, containing (in mM): 126 K-Gluconate, 4 KCl, 10 HEPES, 4 MgATP, 0.3 Na₂GTP, 10 PO-Creatine, 0.3% (w/v) Biocytin (pH = 7.25 with KOH, 285 mOsm/kg). For sEPSC recordings, dentate granule cells were held at -80 mV in voltage-clamp mode. Intrinsic cellular properties were recorded in current-clamp mode. A pipette capacitance of 2.0 pF was corrected and series resistance was compensated using the automated bridge balance tool of the Multiclamp commander. IV-curves were generated by injecting 1 s square pulse currents starting at -100 pA and increasing in 10 pA steps until +500 pA injection was reached (sweep duration: 2 s). Spontaneous inhibitory postsynaptic currents (sIPSCs) were recorded in the same extracellular solution by adding the AMPA receptor inhibitor CNQX (10 µM, Biotrend) and the NMDA receptor inhibitor APV (10 µM, Abcam). Patch pipettes for sIPSC recordings contained (in mM): 40 CsCl, 90 K-gluconate, 1.8 NaCl, 1.7 MgCl₂, 3.5 KCl, 0.05 EGTA, 2 MgATP, 0.4 Na₂GTP, 10 PO-Creatine, 10 HEPES (pH = 7.25 with KOH, 290 mOsm) and granule cells were held at -70 mV during the recordings. Series resistance was monitored and recordings were discarded if series resistance reached > 30 MΩ.

Post-hoc labeling of patched dentate granule cells. Acute slice preparations were fixed in 4% PFA/ 4% sucrose (w/v, phosphate buffered saline, PBS) at room temperature and stored at 4°C overnight in the same solution. After fixation, slices were washed in PBS and consecutively incubated for 1 hour with 10% (v/v) normal goat serum (NGS) in 0.5% (v/v) Triton X-100 containing PBS to reduce unspecific staining. For post hoc visualization of patched dentate granule cells, sections were incubated for 3 hours with streptavidin-Alexa Fluor 488 (Invitrogen, #S32354; 1:1000 dilution in 10% (v/v) NGS, 0.1% (v/v) Triton X-100 containing PBS) at room temperature. Sections were washed in PBS and incubated with DAPI for 10 minutes (Thermo Scientific, #62248; 1:5000 dilution in PBS) to visualize the cytoarchitecture. After washing, sections were transferred onto glass slides and mounted with fluorescence anti-fading mounting medium (DAKO Fluoromount). Confocal images were acquired using a Leica SP8 laser-scanning microscope equipped with a 20x multi-immersion (NA 0.75; Leica) and a 40x oil-immersion (NA 1.30; Leica) objective. Image stacks were acquired in tile scanning mode with the automated stitching function of the LasX software package.

Electron microscopy. Adult mice of both sexes were anesthetized using intraperitoneal injection of ketamine (100 mg/kg) and xylazine (20 mg/kg). Deeply anesthetized mice were transcardially perfused using 2% (w/v; 0.1 M phosphate buffer, PB) glutaraldehyde and 4% (w/v; 0.1 M phosphate buffer, PB) paraformaldehyde. Post-hoc fixation of the brains were continued overnight in the same fixation solution. After fixation, frontal sections containing the dorsal hippocampus were generated using a Leica VT1000S vibratome. Isolated dorsal hippocampal slices were washed for 4 hours in 0.1 M PB (phosphate buffer). Subsequently, slices were incubated with 1% osmium tetroxide for 60 min, washed in graded ethanol (up to 50% (v/v)) for 5 min each, and incubated overnight with uranyl acetate (1% (w/v) in 70% (v/v) ethanol) overnight. Slices were then dehydrated in graded ethanol (80%, 90%, 98% for 5 min each, 2 times 100% for 10 min each). Subsequently, two washing steps were performed in propylene oxide for 10 min each prior to incubation with durcupan/propylene oxide (1:1 for 1 hour) and transferred to durcupan (overnight at room temperature). Slices were embedded in durcupan, and ultra-thin sectioning (55 nm) was performed using a Leica UC6 Ultracut. Sections were mounted onto copper grids (Plano), at which point an additional Pb-citrate contrasting step was performed (3 min). Electron microscopy was performed with a LEO 906E microscope (Carl Zeiss) at 4646× magnification. For each sample, 12 images from the outer two thirds of the molecular layer were acquired and further analyzed. 

RNA isolation and transcriptome analysis. Hippocampi were isolated from the brain of adult mice and immediately transferred into RNA protection buffer (New England Biolabs) and RNA was consecutively isolated using a column based RNA isolation kit according to the manufacturer’s instructions (Monarch^® Total RNA Miniprep Kit; #T2010S New England Biolabs). Strand specific cDNA library preparation from polyA enriched RNA (150 bp mean read length) and RNA sequencing was performed by Eurofins Genomics (Eurofins Genomics Europe Sequencing GmbH, Konstanz, Germany). RNA sequencing was performed using the genome sequencer Illumina HiSeq technology in NovaSeq 6000 S4 PE150 XP sequencing mode. For further analysis .fastq-files were provided. All files contained more than 45 M high-quality reads having at least a phred quality of 30 ( > 90% of total reads).

In vivo perforant path long-term potentiation. 3-month-old C57BL/6J (Synpo^+/+) or synaptopodin-deficient male animals (Synpo^-/-; with C57BL/6J genetic background) were kept in a 12-hour light/12-hour dark cycle (Scantainer) with access to food and water ad libitum. To achieve stable anesthesia, an initial dose of urethane (1.25 g/kg, in sodium chloride solution) was injected subcutaneously (s.c.); a supplemental dose of 0.1 g/kg was given as needed. After stable anesthesia was reached, atRA (10 mg/kg in 5% DMSO) or vehicle-only was intraperitoneally injected (blind to experimenter). The surgery and electrode placement were performed as previously described [32; 33]. Briefly, the mouse was placed in a stereotactic frame (David Kopf Instruments) and local anesthesia with prilocaine (Xylonest 1%, Astra Zeneca, s.c. to the scalp) was applied. Cranial access to the brain was established according to coordinates from the mouse brain atlas (Franklin and Paxinos; stimulation electrode: 2.5 mm lateral to the midline, 3.8 mm posterior to bregma; recording electrode: 1.2 mm lateral to the midline, 1.7 mm posterior to bregma). The ground electrode was placed in the neck musculature. Electrophysiological signals were amplified using a Grass P55 A.C. pre-amplifier (Astro-Med) and digitized at a 10 kHz sampling rate (Digidata 1440A, Molecular Devices). Extracellular stimulation was performed using a STG1004 stimulator (Multichannel Systems). A bipolar stimulation electrode (NE-200, 0.5 mm tip separation, Rhodes Medical Instruments) was lowered 1.5-2.2 mm below the surface of the brain to target the angular bundle of the perforant path. Then a tungsten recording electrode (TM33B01KT, World Precision Instruments) was lowered in 0.1 mm increments while monitoring the waveform of the field excitatory postsynaptic potential (fEPSP) in response to 500 µA test pulses until the granule cell layer in the dorsal part of the hippocampus was reached (1.7-2.2 mm below the surface). The correct placement of the stimulation electrode in the medial portion of the perforant path was verified electrophysiologically by the latency of the population spike (approximately 4 ms), although the activation of some lateral perforant path fibers could not be excluded. Recordings started a minimum of 3 hours after experimental treatment with atRA or vehicle-only. An input-output curve was generated by 30-800 µA current pulses, repeated three times at each intensity, 0.1 ms pulse duration, 60 pulses total at 0.1 Hz. Perforant path-dentate gyrus (PP/DG)-long-term potentiation (LTP) was recorded by applying stimuli with a current intensity set to elicit a 1-2 mV population spike (0.1 Hz, 0.1 ms pulse duration). PP/DG-LTP was induced using a weak theta-burst stimulation (TBS) protocol [34] composed of three series of six trains with six 400 Hz current pulses at double the baseline intensity and pulse duration (with 200 ms interval between trains and 20 s interval between series). Following LTP induction, evoked fEPSPs were recorded for 1 hour using the baseline stimulation parameters.

Quantification and statistics. RNA sequencing data were uploaded to the galaxy web platform (public server: usegalaxy.eu; [35; 36; 37]) and transcriptome analysis was performed using the Galaxy platform in accordance with the reference-based RNA-seq data analysis tutorial [38]. Adapter sequences, low quality and short reads were removed via the CUTADAPT tool (Galaxy version 1.16.5). Reads were mapped using RNA STAR (Galaxy version 2.7.6a) with the mm10 Full reference genome (Mus Musculus). For initial assessment of gene expression, unstranded FEATURECOUNT (Galaxy version 2.0.1) analysis was performed from RNA STAR output. Statistical evaluation was performed using DESeq2 (Galaxy version 2.11.40.6+galaxy1) with treatment as the primary factor that might affect gene expression. Genes were considered as differentially expressed if the adjusted p-value was < 0.05. Heat maps were generated based on z-scores of the normalized count table.

Single-cell recordings were analyzed offline using Clampfit 11 of the pClamp11 software package (Molecular Devices). sEPSC and sIPSC properties were analyzed using the automated template search tool for event detection [12]. Input resistance was calculated for the injection of -100 pA current at a time frame of 200 ms with a maximum distance to the initial hyperpolarization. Resting membrane potential was calculated as the mean baseline value. Action potential (AP) detection was performed using the input-output curve threshold search event detection tool, and the AP frequency was assessed by the number of APs detected during the respective current injection time. One individual cell (control group, ventral hippocampus) was excluded from the analysis of intrinsic membrane properties, since the membrane patch lost its integrity during the recordings. The action potential plots in Figures 2 and 4 depict cellular responses until 300 pA current injection. Beyond 300 pA, subsets of granule cells in both groups failed to maintain regular action potential firing. The treatment did not significantly affect action potential frequency beyond 300 pA current injection. In vivo perforant path LTP was analyzed using Clampfit 10.2 and custom MATLAB (Mathworks) scripts. In these experiments, one Synpo^-/- animal in the vehicle-only group was excluded from further analysis, since an insufficient response to increasing stimulus intensities was detected in the input-output curve.

Electron microscopy images were analyzed and cross-checked by five investigators blind to experimental conditions. Image analysis was performed using the ImageJ software package (available at http://imagej.nih.gov/ij/). For post-synaptic density (PSD) assessment, all visible PSDs in one image were counted and normalized to the image area. Subsequently, PSD length and presynaptic vesicle abundance was manually assessed. In case of perforated synapses, each PSD was analyzed individually. For statistical evaluation, each individual synapse was considered a biological replicate.

Data were statistically evaluated using GraphPad Prism 7 (GraphPad software, USA). Statistical comparisons were made using the non-parametric Mann-Whitney test. For statistical comparison of XY-plots in whole-cell patch-clamp recordings and fEPSP input-output curves, we used an RM two-way ANOVA test (repeated measurements/analysis) with Sidak’s multiple comparisons. Statistical analysis of fEPSP slope data was performed using the Mann-Whitney test for the three terminal data points. p-values smaller 0.05 were considered a significant difference. In the text and figures, values represent mean ± standard error of the mean (s.e.m.). Statistical significance in XY-plots is indicated in the figure panel. U-values were provided for significant results only. *, p < 0.05; ***, p < 0.001; ‘ns’, non-significant differences.

Digital Illustrations. Figures were prepared using Photoshop graphics software (Adobe, San Jose, CA, USA). Image brightness and contrast were adjusted.