Loss of Activity-Induced Phosphorylation of MeCP2 Enhances Synaptogenesis, LTP, and Spatial Memory

Loss of NAIP of MeCP2 in Mecp2^{S421A;S424A/y} mice

To compare the level of S421 phosphorylation in the silent hippocampus with that in the active hippocampus, we treated freshly prepared hippocampal slices with either tetradotoxin (TTX, to suppress neuronal activity) or potassium chloride (KCl, to depolarize neurons), and performed double immunostaining with a phospho-S421 (p-S421) specific antibody and a NeuN antibody (a neuronal marker). The intensity of p-S421 signal was dramatically higher in the KCl-treated slices than in the TTX-treated slices (Fig. 1a,e), indicating the induction of S421 phosphorylation by depolarization. As a control, the intensity of NeuN signal was the same in KCl- and TTX-treated slices (Fig. 1a). We next tested whether phosphorylation of S421 may be induced by more physiologically relevant stimuli. High frequency electrical stimulation (HFS) is commonly used to induce long-term potentiation (LTP) in hippocampal slices. Using the same p-S421 specific antibody, we detected more than 100% increase in the intensity of p-S421 signal after HFS in acute hippocampal slices, as compared to its level in slices under baseline electrical stimulation (Fig. 1b,e). Again NeuN immunoreactivity was similar in slices receiving either baseline stimulation or HFS (Fig. 1b). Finally, we tested whether phosphorylation of S421 may be induced by behavioral trainings, such as contextual fear conditioning and Morris water maze. In the contextual fear training experiment, we compared p-S421 immunoreactivity in hippocampal sections from mice that received shock immediately after being put into the box (immediate shock control) and mice that received shock two minutes after being put into the box (context trained), and found a 59% increase of p-S421 signal in the context trained mice over the immediate shock control (Fig. 1c,e). In the Morris water maze training experiment, we compared p-S421 immunoreactivity in hippocampal sections from mice that received the hidden platform version of Morris water maze (WM) training (WM trained) and mice that swam the same amount of time in the same tank with no platform (yoked/swimming control), and found a 88% increase of p-S421 signal in the hippocampus of WM trained mice over the yoked/swimming control (Fig. 1d,e). In addition to co-staining with NeuN (Fig. 1c,d), we performed immunostaining against total MeCP2 on the adjacent sections, and detected no difference in total MeCP2 expression between the context trained group and the immediate shock control, and between the WM trained group and the yoked/swimming control (Supplementary Fig. 1). Taken together, our data suggest that increased neuronal activity level, induced by either artificial depolarization, physiologically relevant electrical stimulation, or behavioral training, can lead to phosphorylation of S421 on the MeCP2 protein in the mouse hippocampus.

Recently, a detailed mass spectrometry analysis¹⁸ identifies two serines in MeCP2 (S421 and S424), whose phosphorylation is induced by neuronal activity in the mouse brain. Yet the same study finds NAIP of MeCP2 only at S421, but not S424, in the rat brain. To completely abolish NAIP of MeCP2, we generated mice that encode serine to alanine (S to A) mutations at both the 421^th and the 424^th amino acids in the endogenous Mecp2 gene¹⁸. The Mecp2^{S421A;S424A/y} mice showed none of the RTT phenotypes observed in the Mecp2 null mice (i.e. small brain or hindlimb clasping), bred normally, and had a normal lifespan (¹⁸ and data not shown). On brain sections prepared from the Mecp2^{S421A;S424A/y} mice, the temporal and spatial expression patterns of the MeCP2^S421A;S424A protein were indistinguishable from those of the wild type MeCP2 protein in the wild type mice (data not shown). As an example, MeCP2^S421A;S424A immunoreactivity (Fig. 2a) in the pyramidal neurons (co-stained by NeuN, Fig. 2b) of the CA1 region (marked by arrowheads in Fig. 2a–h) in the adult hippocampus is shown to be indistinguishable from MeCP2 immunoreactivity in the wild type (Fig. 2c,d). Moreover, the sub-cellular localization of the MeCP2^S421A;S424A protein in cortical neurons (Supplementary Fig. 2a–d) was also identical to that of the wild type MeCP2 protein (Supplementary Fig. 2e–h). In brain lysates prepared from the Mecp2^{S421A;S424A/y} mice, the MeCP2^S421A;S424A protein was expressed at the same level as that of the MeCP2 protein in wild type mice both under normal condition and after seizure (Fig. 2i). We next examined the phosphorylation status of S421 in the Mecp2^{S421A;S424A/y} and control mice. Consistent with our expectation, while phosphorylation at S421 was present at baseline (Fig. 2g,i) and greatly induced by seizure in the wild type brain (Fig. 2i), it was completely abolished in the Mecp2^{S421A;S424A/y} mice both at baseline (Fig. 2e,i) and after seizure (Fig. 2i). By RT-PCR based sequencing of the expressed Mecp2^S421A;S424A RNA transcript, we have previously shown that S424 has been changed to A424¹⁸. Since the mutation to alanine is a proven way to abolish serine phosphorylation, A424 cannot be phosphorylated under any conditions in the Mecp2^{S421A;S424A/y} mice. Taken together, these results suggest that the S421A;S424A double mutation specifically abolished NAIP of MeCP2, without changing the expression level, expression pattern and intracellular localization of the mutant protein.

Enhanced hippocampal memory in Mecp2^{S421A;S424A/y} mice

To reveal the physiological relevance of neuronal activity-induced MeCP2 phosphorylation, we performed a series of behavioral analyses of the Mecp2^{S421A;S424A/y} mice. To eliminate potential influence of different genetic backgrounds on animal behavior, we backcrossed the Mecp2^S421A;S424A mice with the C57BL/6 mice for a minimum of 10 generations. As an initial screen for phenotypes covering anxiety, exploratory activity, motor coordination, and learning and memory, we compared the Mecp2^{S421A;S424A/y} mice to their wild type littermates in the open field test, the rotarod test, and the fear conditioning test. In the open field test, the Mecp2^{S421A;S424A/y} mice showed a similar level of anxiety, as measured by the percentages of distance traveled in the middle and the corners of the field, as their wild type littermates (Supplementary Fig. 4a,b); and the same amount of exploratory activity, as measured by the total distance traveled, as their wild type littermates (Supplementary Fig. 4c). To confirm these results, we performed another widely used anxiety test--the elevated plus maze--with another cohort of Mecp2^{S421A;S424A/y} mice and their wild type littermates, and observed similar levels of exploratory activity and anxiety in the Mecp2^{S421A;S424A/y} mice and their wild type littermates (Supplementary Fig. 4d–f). In a test for motor coordination, the Mecp2^{S421A;S424A/y} mice fell off the rotating rod slightly quicker than their wild type littermate (Supplementary Fig. 4g). However, the difference was not statistically significant (p = 0.3).

After the first two screening tests yielded no difference, we observed a strong phenotype in the last test of the screening--the fear conditioning test. In this classic Pavlovian conditioning preparation aimed at studying fear learning and memory (illustrated in Fig. 3a), the Mecp2^{S421A;S424A/y} mice froze more than their wild type littermates (Fig. 3c) in the context test, while they froze the same as the wild type littermates (Fig. 3c) in the cue test. It is widely believed that performance in the context test is dependent on the function of both the hippocampus and the amygdala, while performance in the cue test is only dependent on the function of the amygdala. Thus the combination of results from the context test and the cue test strongly suggests a difference in hippocampus function between the Mecp2^{S421A;S424A/y} and the wild type mice. To confirm the context test results, we performed context-only training (no tone played during training, as illustrated in Fig. 3b) and testing with a fresh cohort of Mecp2^{S421A;S424A/y} mice and their wild type littermates. The Mecp2^{S421A;S424A/y} mice again froze more than the wild type (Fig. 3d) in the context test. In both training paradigms, no significant difference in baseline freezing (pre-shock during training sessions, and pre-tone in the cue test sessions) was observed between the Mecp2^{S421A;S424A/y} mice and their wild type littermates (Fig. 3c–d). To further confirm the involvement of the hippocampus in an independent assay, we performed the Morris water maze test (as illustrated in Fig. 4a), a well-established hippocampus-dependent spatial learning and memory test, on yet another cohort of Mecp2^{S421A;S424A/y} mice and their wild type littermates. The hidden platform version of the water maze test was used in this experiment. Both the Mecp2^{S421A;S424A/y} mice and their wild type littermates swam at the same speed during training and testing (Fig. 4b). Throughout the training period, both genotypes of mice learned the location of the hidden platform equally well, as indicated by the persistent decrease in time needed to find and climb onto the platform (Fig. 4c). In the probe trials, while both the Mecp2^{S421A;S424A/y} and the wild type mice remembered where the hidden platform was placed during training by spending significantly more time in the target quadrant than the other quadrants (indicating that both genotypes remembered the location of the platform), the Mecp2^S421A;S424A/y mice spent significantly more time than the wild type littermates in the target quadrant (Fig. 4d), which clearly demonstrated that the Mecp2^{S421A;S424A/y} mice had better spatial memory than their wild type littermates. Taken together, these results strongly suggest the Mecp2^{S421A;S424A/y} mice have altered hippocampal function.

Enhanced LTP and synaptogenesis in Mecp2^{S421A;S424A/y} mice

Since our behavioral analysis identified the hippocampus as one brain region whose function was altered by abolishing activity-induced MeCP2 phosphorylation, we decided to study synaptic physiology in the hippocampus of the Mecp2^{S421A;S424A/y} mice. In this series of experiments, we focused our analysis on two well-characterized synapses, the Schaffer collateral-CA1 synapse and the mossy fiber-CA3 synapse. In both cases, we first examined basal synaptic transmission in the Mecp2^{S421A;S424A/y} mice and their wild type littermates. By using extracellular field potential recording, we examined the input-output curve and paired-pulse facilitation (PPF, a form of presynaptic short term plasticity). The input-output curves showed no significant difference (Supplementary Fig. 5) between the Mecp2^{S421A;S424A/y} mice and their wild type littermates. Similarly, no difference was detected in paired-pulse facilitation (administered at six interstimulus intervals of 25, 50, 100, 200, 300, 400 and 500ms) between the Mecp2^{S421A;S424A/y} mice and their wild type littermates (Supplementary Figs. 6 and 7). In addition, we examined frequency facilitation at the mossy fiber-CA3 synapse, and detected no difference between the Mecp2^{S421A;S424A/y} mice and their wild type littermates (Supplementary Fig. 8). Taken together, these results suggest abolishing NAIP at S421 and S424 of the MeCP2 protein does not alter basal synaptic transmission or presynaptic short term plasticity at either the Schaffer collateral-CA1 synapses or the mossy fiber-CA3 synapses in the hippocampus.

We next examined the long term potentiation (LTP) at both the Schaffer collateral-CA1 synapse and the mossy fiber-CA3 synapses in the Mecp2^{S421A;S424A/y} mice. In hippocampal slices prepared from both the Mecp2^{S421A;S424A/y} mice and their wild type littermates, LTP was readily induced by two trains of 100 Hz stimulation for 1s with an intertetanus interval of 20 s in the Schaffer collateral-CA1 synapse and lasted for 3 hours (Fig. 5a,b). At the mossy fiber-CA3 synapse, LTP was induced in both genotypes by two trains of 100 Hz stimulation for 1s with an intertetanus interval of 20 s in the presence of 50 μM D-AP5 and lasted for more than 60 minutes (Fig. 6a,b). Analysis by two-way ANOVA with repeated measures revealed significantly stronger LTP in Mecp2^{S421A;S424A/y} mice than in their wild type littermates at both the Schaffer collateral-CA1 synapse (Fig. 5c) and the mossy fiber-CA3 synapse (Fig. 6c). At the Schaffer collateral-CA1 synapse, the late phase of the LTP was affected, since the average LTP magnitudes during the last 20 minutes of recordings were significantly different between the two genotypes (Fig. 5d). These results were consistent with the enhanced hippocampus-dependent learning and memory in these mice revealed by the fear conditioning test and the Morris water maze.

We next examined whether there were any changes at the cellular level that may underlie the altered synaptic physiology and behavioral output of the hippocampus in the Mecp2^{S421A;S424A/y} mice. Primary hippocampal and cortical neurons were isolated from postnatal day 0–1 Mecp2^{S421A;S424A/y} and wild type littermate pups, and cultured separately for 21 days in vitro (DIV). Triple immunostaining of VGLUT1 (excitatory presynaptic marker), PSD95 (postsynaptic marker), and MAP2 (dendritic marker) was performed to examine the development of excitatory synapses (Fig. 7a). Quantification of the densities of VGLUT1 puncta, PSD95 puncta, and co-localized VGLUT1/PSD95 puncta was performed along MAP2 labeled neurites (Fig. 7b,d). In the hippocampal culture, our analysis revealed a 45% increase in the number of VGLUT1 puncta (p = 0.00003), a 28% increase in the number of PSD95 puncta (p = 0.005), and a 152% increase in the number of co-localized VGLUT1/PSD95 puncta (p = 0.000002) in Mecp2^{S421A;S424A/y} neurons, compared to those in the wild type neurons (Fig. 7c). In the cortical culture, similar analysis revealed a 99% increase in the number of VGLUT1 puncta (p = 0.00002), a 37% increase in the number of PSD95 puncta (p = 0.002), and a 84% increase in the number of co-localized VGLUT1/PSD95 puncta (p = 0.04) in Mecp2^{S421A;S424A/y} neurons, compared to those in the wild type neurons (Fig. 7e). Since co-localized VGLUT1/PSD95 puncta likely represent functional excitatory synapses, our results strongly suggest an increase in excitatory synaptogenesis in both the hippocampal and cortical neurons from the Mecp2^{S421A;S424A/y} mice. These results are consistent with the enhanced hippocampal LTP and the enhanced learning and memory phenotypes observed in these mice.

Altered gene expression in Mecp2^{S421A;S424A/y} hippocampus

Given the well-established role of MeCP2 in modulating gene transcription^5,8, we suspected potential gene transcription changes may underline the cellular, synaptic, and behavioral phenotypes observed in the Mecp2^{S421A;S424A/y} mice. To reveal what gene transcription changes are caused by the loss of NAIP of MeCP2 in these mice, we first examined known MeCP2 target genes with established neuronal functions. Bdnf is such a gene^16,20. We microdissected the hippocampus from the Mecp2^{S421A;S424A/y} mice (n = 14) and their wild type littermates (n = 14), extracted RNA, performed realtime PCR, and detected an increase in Bdnf transcript covering its protein coding region (Bdnf CDS) in the Mecp2^{S421A;S424A/y} hippocampus (Fig. 8a). In addition, we found an increase in Bdnf transcript covering its neuronal specific promoter (Exon IV, Bdnf E4) in the Mecp2^{S421A;S424A/y} hippocampus (Fig. 8a). The increase in Bdnf transcription is consistent with the observed phenotypes of enhanced hippocampal LTP and hippocampus-dependent learning and memory in the Mecp2^{S421A;S424A/y} mice.

Recognizing that the phenotypes of the better contextual fear memory, the enhanced hippocampal LTP, the increased excitatory synaptogenesis, and the increase in Bdnf transcription in the Mecp2^{S421A;S424A/y} mice are similar to those observed in transgenic mice overexpressing Mecp2 (Mecp2^Tg)^8,21,22, we examined the expression of three additional genes whose levels were changed in the Mecp2^Tg mice⁸. These included myocyte enhancer factor 2 (Mef2c), bone morphogenetic protein 4 (Bmp4), and metabotropic glutamate receptor 1 (Grm1), all of which are important for the development and function of the nervous system. Moreover, MeCP2 has been shown to directly bind to the promoter of Mef2c and repress its transcription⁸. We found increased expression of Bmp4 and decreased expression of Mef2c and Grm1 (Fig. 8a) in the Mecp2^{S421A;S424A/y} hippocampus. The expression change of each of the three genes in the Mecp2^{S421A;S424A/y} mice was very similar to that in the Mecp2^Tg mice. However, it is not immediately clear how expression changes in Bmp4, Mef2c and Grm1 may help explain the observed phenotypes in the Mecp2^{S421A;S424A/y} mice.

Altered promoter association by the phospho-mutant MeCP2

As a first step to understand how the loss of phosphorylation may alter gene transcription, we examined the ability of the phospho-dead form of MeCP2 to bind to gene promoters. To better quantify the promoter occupancy of the phospho-dead MeCP2 protein, we generated two novel Mecp2 knockin alleles: the Mecp2^{S421A;S424A-flag/y} mice expressing the MeCP2^S421A;S424A-flag protein from the endogenous Mecp2 locus and the Mecp2^WT-flag/y mice expressing the MeCP2-flag protein from the endogenous Mecp2 locus. These unique mouse lines allowed us to take advantage of a very reliable anti-flag antibody to conduct chromatin immunoprecipitation (ChIP) experiments. Western blot analysis confirmed that both the MeCP2^S421A;S424A-flag protein and the MeCP2-flag protein were expressed at the same level as the MeCP2 protein (data not shown). Behavioral testing confirmed that the Mecp2^{S421A;S424A-flag/y} mice had enhanced hippocampus-dependent learning and memory (Supplementary Fig. 3i) similar to that observed in the Mecp2^{S421A;S424A/y} mice, while the Mecp2^WT-flag/y mice performed similar to the Mecp2^+/y mice (data not shown). These data suggest that the Mecp2^{S421A;S424A-flag/y} mice are functionally indistinguishable from the Mecp2^{S421A;S424A/y} mice, and that the Mecp2^WT-flag/y mice are functionally indistinguishable the Mecp2^+/y mice.

Because the transcriptional changes detected in Fig. 8a were from the in vivo hippocampus, we reasoned it would be the most appropriate to use hippocampi from the Mecp2^{S421A;S424A-flag/y} and the Mecp2^WT-flag/y mice for ChIP analysis to examine promoter occupancy of the phospho-mutant form and the wild type form of MeCP2. Results from this experiment revealed increased promoter occupancy by the MeCP2^S421A;S424A-flag protein, as compared with that by the MeCP2-flag protein, on the promoters of Bdnf, Bmp4, Mef2c, and Grm1 (Fig. 8b). Thus, the loss of phosphorylation appears to enhance the binding of MeCP2 to its target gene promoters. Because MeCP2 has been shown to be capable of activating the transcription of some genes and repressing the transcription of other genes at the same time^8,9, it is possible that the increased promoter occupancy of the phospho-mutant form of MeCP2 may lead to opposite transcriptional outcomes in the phospho-mutant hippocampus. More detailed studies in the future will help reveal the mechanisms of such regulations.

Mouse genotyping

Two PCR genotyping assays were developed. Primers for the first assay were 5′-TCACAAGTTAACAGTGCCAGCTGC-3′(forward) and 5′-ACCCCCTTGGGACTGAAGTTACAGA-3′ (reverse). A ~100 bp band in this assay represents the wild type Mecp2 allele. A ~200 bp band in this assay represents the Mecp2^S421A;S424A allele. Primers for the second assay were 5′-CGAGGAGGCGCACTGGAAGCA-3′ (forward) and 5′-ACCTGACTGTGCTTGTCGGTAAG (reverse). A 413 bp band in this assay represents the Mecp2^S421A;S424A allele.

Immunostaining and Western blot analyses of S421 phosphorylation

Anti-MeCP2 (Abcam), anti-phospho-S421 (custom made by Covance), anti-tubulin (Sigma), anti-NeuN (Santa Cruz), and anti-actin (Sigma) were used. For immunostaining, 10–12 weeks old mice were transcardially perfused with normal saline solution followed by 4% paraformaldehyde (PFA). Brains were dissected, post-fixed in 4% PFA, cryoprotected in 30% sucrose, embedded and frozen in Tissue Tek. Serial coronal sections were cut at 40 μm throughout the brain, and stained with appropriate primary and secondary antibodies. For Western blot, brains were quickly dissected out from 10–12 weeks old mice and homogenized in RIPA buffer containing both the protease inhibitor cocktail (Roche) and the protein phosphatase inhibitor cocktail (Roche). Lysate was run on 10% NuPAGE Bis-Tris gel (Invitrogen), and transferred to Protran BA 85 nitrocellulose membranes (Whatman). The membrane was incubated with appropriate primary and IR dye-conjugated secondary antibodies (Thermo Scientific), and scanned by the Odyssey infrared imaging system.

Mouse behavioral tests

10–12 weeks old mice were used in all tests. All protocols were approved by the Institutional Animal Care and Use Committee at University of Wisconsin-Madison.

Electrophysiology

8–12 weeks old mice were anesthetized and killed by decapitation. 400 μM transverse hippocampal slices were prepared using a Vibratome Tissue Chopper. Slices were recovered in oxygenated ACSF at room temperature for at least 1 hr. The ACSF contained (in mM): 124 NaCl, 2.5 KCl, 1.25 NaH₂PO₄, 26 NaHCO₃, 1.2 MgCl₂, 2.5 CaCl₂, and 15 glucose. All recordings were performed at 30 ± 1 °C in a submerged chamber (Warner instruments, Hamden, CT) continuously superfused with oxygenated ACSF. Schaffer collateral-CA1 synaptic responses and mossy fiber-CA3 synaptic responses were recorded as extracellular field potentials (fEPSPs) from the CA1 stratum radiatum and the CA3 stratum lucidum area, respectively. Unless otherwise-mentioned, test stimuli were delivered at a frequency of 0.05 Hz. The stimulus intensity was set to produce 25–30% of the maximum response in all the experiments. Paired-pulse facilitation (PPF) was performed at 25, 50, 100, 200, 300, 400, and 500 ms interstimulus intervals. Frequency facilitation was investigated by increasing the stimulation rate from 0.1 Hz to 0.5 Hz. LTP was induced by two trains of 100 Hz stimulation lasting 1 s with a 20 s intertetanus interval. No bicuculline was applied during any recording. For CA3 LTP induction, tetanus was delivered in the presence of 50 mM D-(2)-2-amino-5-phosphonopentanoic acid (D-APV) to prevent contamination of the NMDA-dependent pathway converging on CA3 neurons. Mossy fiber synaptic responses were identified by large PPF and frequency facilitation. Experiments were included for data analysis only if DCG-IV caused a greater than 75% reduction in the synaptic response. For Schaffer collateral-CA1 recordings, fEPSP rise slope was calculated as the slope of the rising phase 20–80% of the peak response. For CA3 recordings, the amplitude of individual synaptic response was measured after subtraction of the response that remained in the presence of DCG-IV. Data were acquired and analyzed with pClamp10.2 software (Axon Instruments, Sunnyvale, CA). Signals were filtered at 2 Hz and sampled at 10 kHz by Digidata 1440A (Axon Instruments, Sunnyvale, CA). Traces were averaged for every 2 min interval.

Analysis of excitatory synaptogenesis in cultured hippocampal and cortical neurons

Dissociated hippocampi and cortical neurons were isolated from postnatal day 0 or day 1 Mecp2^{S421A;S424A/y} and Mecp2^+/y pups, and plated at a density of 1 x 10⁴/cm² on poly-L-lysine (Sigma# P2636) coated coverslips. Culture medium was Neurobasal (Invitrogen #21103-049) containing 2% B27 (Invitrogen #12587-010), 1% N2 (invitrogen#17502-048), and 0.5 mM GlutaMax (Invitrogen#35050-061). Cultured hippocampal and cortical neurons were fixed at DIV (days in vitro) 21 with 4% PFA and 4% sucrose in PBS, washed with PBS, permeabilized with Tritonx-100 in PBS, blocked with 10% Normal Donkey Serum in PBS (all at room temperature), and incubated at 4 °C overnight with mouse anti-MAP2 (Millipore#05-346, 1:300), rabbit anti-PSD95 (Cell Signaling#2507, 1:200) and goat anti-VGLUT1 (Santa Cruz#sc-13320, 1:200) antibodies in PBS containing 10% normal donkey serum. The next day, after being washed in PBS, the coverslips were incubated with secondary antibodies, including AffiniPure F(ab′)₂ Fragment donkey IgG(H+L) conjugated with DyLight-488 (anti-goat),-549 (anti-rabbit), -649 (anti-mouse) respectively (1:200, Jackson immunoResearch Laboratories). Coverslips were then washed in PBS and stained with DAPI before being mounted on glass slides with FLUOROMOUNT G (SouthernBiotech#0100-01).

Image analysis

Images in Figures. 1 and 7 were captured on a Nikon A1RSI confocal microscope. For each experimental series, all the images were acquired with identical settings for laser power, photomultiplier gain and offset. Z-stacks were collapsed in a maxmum projection and analyzed with NIH ImageJ software. For quantification of MeCP2 pS421 level in CA1, all the images were quantified using a single threshold value, which is determined by the background level of the control sections in each experiment. MeCP2 pS421 level was measured as the average MeCP2 p421 intensity in the CA1 neuronal nucleus, which is identified by NeuN immunoreactivity. For pre- and post-synaptic puncta quantification of the primary neuronal cell culture, only primary dendritic branches were included in the analysis. VGLUT1 and PSD95 images were thresholded using constant settings for each experiment, and converted to binary images. Thresholds were chosen such that all recognizable punctate structures were included into the analysis. VGLUT1 and PSD95 puncta were identified by the “partial analysis” function in ImageJ, with the minimal size 0.2 μm² for VGLUT1 and 0.1 μm² for PSD95. VGLUT1 puncta with overlap or directly adjacent PSD95 puncta were scored as colocalized VGLUT1/PSD-95 puncta. Puncta density was measured as the puncta number per μm of dendritic branch.

RNA extraction, reverse transcription, and real time PCR

Total RNA from mouse hippocampus was isolated using SV Total RNA Isolation System (Promega). cDNA was made using qScript^TM cDNA SuperMix (Quanta Bioscience). Real-time PCR was performed on an iQ5 Real-Time PCR machine using iQ SYBR Green Supermix (Bio-Rad). Fold change was calculated with the 2^{−delta Ct} method after normalization to Gapdh. Primers were: 5′-AATGGGAAGCTTGTCATCAACG-3′ (Gapdh-F), 5′-GAAGACACCAGTAGACTCCACGACATA-3′ (Gapdh-R), 5′-CGCCATGCAATTTCCACTATCAATAATTTAAC-3′(Bdnf E4-F), 5′-CTTTTTCAGTCACTACTTGTCAAAGTAAAC-3′ (Bdnf E4-R), 5′-GACCATCCTTTTCCTTACTATGG-3′ (Bndf CDS-F), 5′-CCATTCACGCTCTCCAGAGTC-3′ (Bdnf CDS-R), 5′-CATCACGAAGAACATCTGGAG-3′ (Bmp4-F), 5′-CGAGGAGATCACCTCATTCTC-3′ (Bmp4-R), 5′-CAACAACATATGGTACTGAGTAC-3′ (Mef2c-F), 5′-CATGTTATGTAGGTGCTGCTGC-3′ (Mef2c-R), 5′-ACAGAAGGGAATTACGGCGAGA-3′ (Grm1-F), 5′-ATCAAAGCTCTTCTCGCCAGCA-3′ (Grm1-R).

Chromatin immunoprecipitation and gene promoter specific real time PCR

Hippocampi were dissected from 9–12 weeks old mice, diced into small pieces and crosslinked in 0.5% formaldehyde, washed with ice-cold PBS, sonicated with Misonix Sonicator 3000 in lysis buffer (1% SDS, 10 mM EDTA, 20 mM Tris-HCl, pH 8.1, Roche protease inhibitor cocktail), and cleared by centrifugation to generate sheared chromatin. Dynabeads (Invitrogen) were pre-washed with PBS/BSA (0.5% BSA in PBS) and incubated with anti-FLAG antibody (Sigma) to form the beads/antibody complex. For immunoprecipitation, sheared chromatin was diluted in dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl) and incubated with beads/antibody complex overnight on a Nutator at 4 oC. The next day, the beads/antibody/chromatin complex were washed with washing buffer (50 mM Hepes, pH 7.6, 1 mM EDTA, 0.7% DOC, 1% NP-40, 0.5 M LiCl) and TBS. The immunoprecipitated chromatin was then eluted from the beads in elution buffer (1% SDS, 10 mM EDTA, 20 mM Tris-HCl, pH 8.1 ), and reverse-crosslinked. Sheared chromatin not incubated with beads/antibody complex was processed the same way to generate input DNA. Both ChIP DNA and input DNA were treated with RNase A (Thermo Scientific) and proteinase K (Promega), purified by ethanol precipitation, and dissolved in water (Promega). 20 ng of ChIP DNA and input DNA was used for each quantitative real-time PCR. Real-time PCR was performed on an iQ5 Real-Time PCR machine using iQ SYBR Green Supermix (Bio-Rad). The ChIP DNA level was normalized to the input DNA level using the 2^{−delta Ct} method. Relative promoter occupancy was calculated by setting the MeCP2-flag binding level as 100%. Primers were: 5′-TTCTGTGTGCGTGAATTTGC-3′ (Bdnf E4-F), 5′-TGGGTGGGAGTCCACGAG-3′ (Bdnf E4-R), 5′-TACAGTAGCATTGTGGGTTC-3′ (Mef2c-F), 5′-AGGAGGGATTAATCATGCTC-3′ (Mef2c-R), 5′-CTCTAGTGTTTTCATGTGAC-3′ (Grm1-F), 5′-GGAGTAGCCAAATGTAGAGT-3′ (Grm1-R), 5′-TGAAGCTGATAGTTCCTTCC-3′ (Bmp4-F), 5′-GCTAGCTGGGAGGTGGAATG-3′ (Bmp4-R).

Statistical analysis

For Morris water maze data, two way ANOVA with repeated measures was performed to reveal any significant interaction between genotype and quadrant, which was followed by a Tukey-HSD post hoc test to complete the pairwise comparison of the percentage of time spent by each genotype in each quadrant. For LTP and basic synaptic transmission data, two-way ANOVA with repeated measures was performed. For the rest of the data, the Student’s t-test was performed. Statistical significance was determined at p < 0.05.