Increased excitability of hippocampal neurons in mature synaptopodin-knockout mice

Synaptopodin (SP) is localized within the spine apparatus, an enigmatic structure located in the neck of spines of central excitatory neurons. It serves as a link between the spine head, where the synapse is located, and the endoplasmic reticulum (ER) in the parent dendrite. SP is also located in the axon initial segment, in association with the cisternal organelle, another structure related to the endoplasmic reticulum. Extensive research using SP knockout (SPKO) mice suggest that SP has a pivotal role in structural and functional plasticity. Consequently, young adult SPKO mice were shown to be deficient in cognitive functions, and in ability to undergo long-term potentiation of reactivity to afferent stimulation. However, although SP expresses differently during maturation, its role in synaptic and intrinsic neuronal mechanisms in adult SPKO mice is still unclear. To address this knowledge gap we analyzed hippocampus bulk mRNA in SPKO mice, and we recorded the activity of CA1 neurons in the mouse hippocampus slice, with both extracellular and patch recording methods. Electrophysiologically, SPKO cells in CA1 region of the dorsal hippocampus were more excitable than wild type (wt) ones. In addition, exposure of mice to a complex environment caused a higher proportion of arc-expressing cells in SPKO than in wt mice hippocampus. These experiments indicate that higher excitability and higher expression of arc staining may reflect SP deficiency in the hippocampus of adult SPKO mice.


Introduction
Synaptopodin (SP) is an actin-associated protein localized in the neck of dendritic spines of mature cortical and hippocampal neurons, in physical association with the enigmatic spine apparatus (Aloni et al. 2019;Deller et al. 2000;Mundel 1998;Segal et al. 2010). It is also found in the cisternal organelle of the axon initial segment (Orth et al. 2007;Segal 2018). These are two strategic locations, where SP is associated with calcium stores of the endoplasmic reticulum (ER) and can control the inputs and outputs of the neuron. Older mice are shown to have a higher density of mature spines, spine apparatus and SP puncta than young ones (Czarnecki et al. 2005). Recent studies have assigned a role for SP in synaptic plasticity Vlachos et al. 2009), and assumed that it links morphological changes in actin cytoskeleton with functional synaptic changes generated in response to plasticity-producing stimulation. Studies conducted with SP-knockout mice (SPKO) (Deller et al. 2003), found that these mice are deficient in cognitive tasks and that slices taken from SPKO mice are deficient in ability to generate long-term plasticity. Later studies proposed that only some LTP generating protocols are sensitive to the absence of SP (Jedlicka and Deller 2017). Concerning the role of SP in spine plasticity, it has been shown that overexpression of (transfected) SP maintains the activitydependent spine enlargement (Okubo-Suzuki et al. 2008), and that SP-positive spines are more amenable to plastic changes, which are associated with activation of calcium stores (Harris 1999). Moreover, LTP deficiency occurs only in young SPKO mice (2-8 weeks) (Aloni et al. 2019;Zhang et al. 2013), leaving the issue of what changes in the SPKO mice when they grow up, to enable plasticity, unanswered. In the present study we explored some molecular and cellular mechanisms in the adult (6-month-old) SPKO mice, and report that these mice express behavior-induced higher arc activation than wt mice and that hippocampal neurons in SPKO mice are more excitable than those of wt controls. These mechanisms may counteract the lack of SP in the adult mouse with regard to neuronal plasticity.

Animals
Experiments were conducted by the rules of the Institutional Animal Care and use Committee. Forty-four male mice, aged 6-8 months were used for all of the experiments, All mice groups originated from breeding Heterozygous SPKO mice received from Dr. Frotscher (Hamburg U.) with the 129P2/ OlaHsd genetic background (for ref. Stock No: 028822 The Jackson laboratory), and were divided into groups after genotyping for the SPKO gene. Mice were maintained on a 12 h light/dark cycle and were allowed free access to food and water. Since in earlier studies there was no apparent difference between wt and heterozygous mice, in the behavioral experiments the results of the two groups were merged (see below).

Hippocampal RNA sequencing
RNA was purified using RNeasy Mini Kit (qiagen) from hippocampal tissue. Assessment of nucleic acids (ng/uL), and their purity (A260/A280, A260/A230) was made using Nanodrop (One C, Thermo Fisher). RNA Integrity Number (RIN) was also measured using (Tapestation 2200, Agilent). Libraries were prepared using the INCPM-RNAseq. briefly, the polyA fraction (mRNA) was purified from 500 ng of total RNA following by fragmentation and the generation of double-stranded cDNA. Then, end repair, A base addition, adapter ligation and PCR amplification steps were performed. Libraries were evaluated by Qubit (Thermo fisher scientific) and TapeStation (Agilent). Sequencing libraries were constructed with barcodes to allow multiplexing of 24 samples in 2 lanes. Around 20 million single-end 60-bp reads were sequenced per sample on Illumina HiSeq 2500 V4 instrument (performed at the G-INCPM Genomics unit). The output was ~ 20 million reads per sample. Poly-A/T stretches and Illumina adapters were trimmed from the reads using cutadapt (MARTIN 2011). Resulting reads shorter than 30 bp were discarded.
The reads were mapped to the mus musculus reference genome using STAR (Dobin et al. 2013), supplied with gene annotations downloaded from Ensemble (the following options were used: EndToEnd and out Filter Mismatch NoverLmax was set to 0.04). Expression levels for each gene were quantified using htseq-count (Anders et al. 2015), using the Ensemble annotations (release 92). Differentially expressed genes were identified using DESeq2 (Love et al. 2014) with the betaPrior, cooksCutoff and independent Filtering parameters set to False. Raw p values were adjusted for multiple testing using the procedure of Benjamini and Hochberg. Pipeline was run using snakemake (Köster and Rahmann 2018). A gene was considered to be significantly differentially expressed by using the following criteria: Absolute Log2 fold was higher than 0.585, the adjusted p value was lower than 0.05 and at least one of the samples had a count higher than 30 reads.

Behavioral experiments
Animals were divided into 2 × 2 groups: (1) wt/hetero and SPKO mice that were sacrificed directly out of their home cages (control, Ctrl); (2) wt/hetero and SPKO mice that explored a new open field environment for 5 min twice, separated by 20 min. The open field was a square box (30 × 30 cm) with 15 cm high walls. A standard small (8 cm) rotating disk was placed in the middle of the box. In each exploration session, mice were lifted and randomly placed in the box. Fifty minutes after the first exposure to the open field, mice were sacrificed. All mice explored the space and ran in the rotating disk, but their behavior was apparently not different between the two groups, and was not quantified.

Immunohistochemistry
Mice were anesthetized with chloral hydrate and perfused with PBS, followed by 4% paraformaldehyde (PFA). The brains were kept in 4% PFA for 24 h followed by 48 h in 30% sucrose and 1% PFA. 25 µm thick coronal sections were then cut with cryostat and collected. Sections were blocked with 20% fetal calf serum (FCS) in 0.2% Triton X-100 containing PBS for 1 h and incubated overnight at 4 °C with the primary antibodies in 2% FCS and 0.2% Triton in PBS [Antibodies: (arc: Novous Biologicals, Cambridge, UK. gift of M. Cohen-Armon, TAU], cFos: Sigma-Aldrich, a gift from Dr R. Eilam, Weizmann Institute). Sections were then rinsed in PBS and incubated with the biotinylated antirabbit antibody in 2% FCS in PBS (1:100, JAC) for 1.5 h at room temperature followed by incubation with streptavidin (1:150, JAC) in PBS for 1 h. and with the Cy2 anti-mouse antibody in PBS for 1.5 h and DAPI for 1 h.

Imaging and analysis
Confocal image stacks were taken using a Zeiss LSM 880 laser scanning microscope equipped with EC plan-Neofluar × 5/0.16 M27, plan-apochromat 20 ×/0.8 and planapochromat 63 ×/1.40 oil DIC objectives. Detector and amplifier gain were initially set to obtain pixel densities within a linear range. Eight image stacks were recorded for each hippocampus. Arc-positive and c-Fos -positive cells were counted from each field size of 135 × 135 μm (63 ×/1.40 oil DIC objectives). Cell count and fluorescence levels were measured using Image-J software. Measurements were made in a double-blind procedure by an independent observer to assure unbiased analysis. Statistical comparisons were made using Origin software.

Extracellular electrophysiology
Mice were rapidly decapitated with a guillotine, their brain removed and the hippocampus was sliced into transverse 400 µm slices on a McIllwain tissue chopper. Slices were incubated at room temperature for 1.5 h in carbogenated (5% CO 2 /95% O 2 ) ACSF (124 mM NaCl, 4.2 mM KCl, 26 mM NaHCO 3 , 1.24 mM KH 2 PO 4 , 2.5 mM CaCl 2 , 2 mM MgSO 4 and 10 mM glucose, at pH 7.4). Recordings were made from interface slices in a standard chamber at 33.8-34.0 °C. Field excitatory postsynaptic potentials (EPSPs) were recorded through a glass pipette containing 0.75 M NaCl (4 MΩ) in the stratum radiatum of CA1 region. Synaptic responses were evoked by stimulation of the Schaffer collaterals through bipolar handmade Nickel-Chromium electrode. Two stimulating electrodes were located on both sides of the recording electrode, with both stimulating the schaffer collateral pathway. Data acquisition and offline analysis were performed using pCLAMP 9.2 (Axon Instruments) in a blind procedure.

Action potential kinetics analysis
Current-clamp recordings were imported in Matlab where the first ten action potentials (AP) that did not arrive in bursts or too close to the end of the current pulse where collected with 5 ms pre-peak and 65 ms post-peak; these spikes were aligned at peak, averaged and this average was used to calculate a phase plot. The average and standard error were calculated for these phase plots within every group. Numeric voltage derivative was calculated as the difference between the voltages recorded at neighboring sampling points; multiplied by sampling rate (per ms) when appropriate. AP threshold was calculated as a point where the voltage derivative increased significantly over a reference baseline value. Reference value was chosen as maximal derivative in 5-2.5 ms before AP peak, threshold was defined as a point where said derivative is more than 50% greater than the reference baseline value. AP shape characteristics were calculated individually for each of the 10 APs per cell, and these characteristics where averaged for every cell. AP amplitude and AP after-hyperpolarization were calculated relative to AP threshold. Half-width was calculated as the difference between time points where voltage reached (AP peak + AP threshold)/2 at rise and at decay. Rise and decay slopes were calculated as maximal and minimal numeric voltage derivative, respectively.

Statistical analysis
All experiments were analyzed using non paired Student's t tests, ANOVA or chi-square test, as the case may be. Results are expressed as mean ± SEM, Statistical significance was set at p < 0.05.

A: hippocampal RNA sequencing
To screen for the differences in gene expression in the SPKO mouse, a whole hippocampal RNAseq was analyzed in 6-7-month-old male mice and we compared differential expression (DE) genes between SPKO and wt mice. Among numerous genes that were DE (Fig. 1A), we found that the immediate early genes (IEGs) (Tyssowski et al. 2018) DE and were significantly elevated in SPKO mice in 10 out of 18 IEG's (Fig. 1B). These results indicate higher activity in the hippocampus of SPKO mice.

B: synaptic properties of SPKO: extracellular experiments
To test directly for higher excitability of hippocampal neurons in SPKO mice, experiments were conducted with acute hippocampal slices. First, population EPSPs in an extracellular field recording of CA1 pyramidal neurons were measured at three different stimulation intensities. CA1 cells produced significantly larger population EPSPs in SPKO compared to wt ( Fig. 2A). Furthermore, SPKO slices produced population spikes at lower stimulation intensities than wt slices. In addition, Paired pulse facilitation was measured at three different inter-pulse intervals (IPI). We found a trend of higher paired pulses facilitation in the SPKO group compared to wt mice, in all three IPI's tested. However, there was no statistically significant difference between the two groups (Fig. 2B). . wt n = 6, SPKO n = 4. Data presented as mean ± SEM, *** represents p < 0.005, ** represents 0.005 ≤ p < 0.01, * represents 0.01 ≤ p < 0.05. p-values measured using unpaired Student t-test (for specific statistical information see "Methods")

C: electrophysiological properties of patch-clamped CA1 neurons
Patch-clamp recording was used to examine the intrinsic properties of CA1 pyramidal neurons. No differences in passive properties were found between SPKO and wt neurons (resting membrane potential, membrane time constant, input resistance), nor did they differ in action potential (AP) threshold or the numbers of APs produced per current pulse (data not shown). However, the AP's in SPKO cells had markedly different kinetics; in particular, AP's were larger in amplitude, shorter in width (Fig. 3A,  D), had smaller after-hyperpolarization (AHP) and a steeper rise (Fig. 3B, D). In another series of experiments, CA1 neurons were voltage clamped at − 70 mV, and spontaneous synaptic currents were recorded under standard conditions. No synaptic blockers or intracellular anesthetics were used. The spontaneous events represent a mix of excitatory and inhibitory currents. The overall frequency of these events was about 50% higher in slices obtained from SPKO mice compared to wt controls (Fig. 3C, E).

D: activation of arc and cFos in active SPKO mice
As arc and cFos are activity-dependent genes, their expression in the dentate gyrus (DG) of the dorsal hippocampus was examined by IHC in wt and SPKO mice after exposure to a new environment. The active mice expressed a significantly higher number of arc-positive neurons in the SPKO than the wt mice (Fig. 4C). cFos expression was also elevated in the active mice but did not significantly differ between the two groups (Fig. 4D).

Discussion
The present study analyzed the role of SP in the adult mouse brain. SP is found at strategic locations in the neck of mature dendritic spines and in the axon initial segment. Both locations are critical in the transfer of information into and out of the host neuron. (Deller et al. 2003;Orth et al. 2007;Segal 2018), We now found that at 6 months of age, CA1 pyramidal neurons of SPKO mice are hyper-excitable compared to wt controls. It is conceivable that SP regulates excitability via regulation of [Ca 2+ ] i in both the dendritic spine and the axon initial segment, which are enriched with an expression of SP and with calcium stores (Deller et al. 2003;Segal 2018). However, since a single gene mutation can lead to diverse effects on gene expression, our results relating a single gene to neuronal properties should be interpreted with caution and further investigations are needed to analyse the types of molecular regulators of synaptic and intrinsic properties in these neurons. An earlier study (Verbich et al. 2016) showed no differences in morphological and functional attributes of CA1 pyramidal cells from SPKO compared with wt in slice cultures. Orth et al. (2007) showed that 3 month-old SPKO animals are not different in several properties of action potential discharges, expected from an organelle located in the axon initial segment. Interestingly, though not significantly different, the SPKO animals in the Bas Orth study expressed higher firing rates and other spike properties than wt mice. while our current experiments, conducted with 6-month-old mice showed significant differences in synaptic properties ( Fig. 2A), in spontaneous PSC's ( Fig. 3C, E), and in itrinsic properties with differences in AP's kinetics in the CA1 neurons (Fig. 2D,   Fig. 3 Properties of patch-clamped CA1 neurons in 6-month-old wt and SPKO mice. A Sample plots showing averages of 10 AP recorded from a WT cell (red) and a SPKO cell (blue), aligned at AP threshold. Asterisks denote AP threshold (see "Methods"), AP half-amplitude, AP peak and AP after-hyperpolarization. E). In addition, earlier studies found that SPKO mice show a maturation effect in synaptic plasticity, LTP magnitude at Schaffer collateral CA1 synapses was reduced in young mice (2-4 weeks old) but this reduction was absent in older mice (2-6 month-old) (Aloni et al. 2019;Zhang et al. 2013) suggesting that older animals are less prone to the SP knockout compared to young ones. Interestingly, Zhang et al. (2013) demonstrated that LTP induced by a tetanic stimulation was not different between SPKO and controls at 6 months of age while there were differences in NMDAmediated LTP in younger age. These earlier studies (except for Bas Orth et al.) did not explore intrinsic properties of the studied neurons, and so such differences could contribute to the age-dependent restoration of plasticity in the older neurons. Our findings showing hyper-excitability at 6 months of age might act as a compensatory mechanism for restoring normal LTP in the SPKO mice CA1 pyramidal neurons at an older age. Further analysis revealed that SPKO mice exposed to a new environment show elevation in the expression of arc in DG granular cells compared to wt mice (Fig. 4C), since the DG is essential for spatial learning processes (Hainmueller and Bartos 2018), these results indicate that spatial learning activates more cells in SPKO mice compared to wt. Importantly, the changes in IEGs of SPKO mice were found in the whole hippocampus using bulk RNA sequencing and in the DG with exposure to new invironment, while the electrophysiological analysis was conducted in CA1 region to reveal synaptic and intrinsic correlates of higher exitability. Both structures are involved in the cognitive functions of the hippocampus. and both structures demonstrate an increase in excitability attributes.
Author contributions EA conducted the RNAseq analysis, and the extracellular electrophysiological experiments, and wrote the manuscript. SV conducted and analyzed the patch recording experiments, LK conducted the behavioral and immunohistochemical studies, EK and MS wrote the manuscript. All co-authors have read and commented on the manuscript and agreed to have their names listed as authors.
Funding Supported by a grant from the Clore Center for Biological Physics of the Weizmann Institute of Science.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Conflict of interest
The authors declare no competing financial interests.
Ethical approval Experiments were conducted by the rules of the Institutional Animal Care and use Committee, approval number: 38390917-2.