Transgenic mouse generation

Camk2a-tTA;TRE-ATase1 (referred to as ATase1 nTg thereafter) and Camk2a-tTA;TRE-ATase2 (referred to as ATase2 nTg thereafter) mice were generated as previously described [12, 20, 21]. Human cDNA was isolated from ATase1- and ATase2-pCMV6 plasmids (Origene; RC215647 for ATase1/NAT8B and RC202157 for ATase2/NAT8) using PCR and subsequently subcloned into pTRE-Tight plasmid (Takara Bio, Inc.). The resulting pTRE-Tight-ATase1 and pTRE-Tight-ATase 2 plasmids were linearized with XhoI and then injected into C57BL/6 J mice (The Jackson Laboratory; Stock No. 000664). These mice were subsequently crossed with B6.Cg-Tg(Camk2atTA)1Mmay/DboJ (Camk2a-tTA) mice (The Jackson Laboratory; Stock No. 007004) to create ATase1 and ATase2 neuron transgenic (nTg) mice. Genotyping from tail DNA was conducted using these specified primers: ATase1 forward (5′-GCTCGTTTAGTGAACCGTCAGAT-3’), ATase1 reverse (5′-CTCCTGGTATTTGCGGATGTGAT-3’); ATase2 forward (5′-GCTCGTTTAGTGAACCGTCAGAT-3’), ATase2 reverse (5′-CTCCTGGTATTTGCGGATGTGA-3’); Camk2a-tTA forward (5′-CGCTGTGGGGCATTTTACTTTAG-3′), and Camk2a-tTA reverse (5′-CATGTCCAGATCGAAATCGTC-3’).

Animals

The University Laboratory Animal Resources provided standardized cages for mice housing. Each cage had one to five littermates and was supplied with regular chow and water ad libitum. All the experiments were done according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the UW-Madison committee for animal care (protocol #M005120). The TRE-ATase1, TRE-ATase2, and Camk2a-tTA transgenes were all bred as heterozygous. Throughout the investigation, wild type (WT), non-transgenic littermates were employed as controls. Animals were used from multiple litters at random. Both males and females were studied. The age and sex of animals at time of experimentation are specified in the figure legends. Investigators were not blind to the genotype of the animals. However, behavioral assessment, mass spectrometry analysis, Imaris-assisted morphology analysis, and computer-assisted multi-electrode array analysis were conducted in a blinded fashion.

Behavior testing

The behavioral assays were all performed at the Behavioral Testing Service of the Waisman Center (University of Wisconsin-Madison, Madison, WI, USA). Prior to each behavior assay, mice were placed in the testing room for a 30-min acclimation period. The performed behavioral assays consisted of the marble burying assay, social interaction, novel object recognition, and fear conditioning paradigm, all of which have been previously described [12, 20, 21]. Additionally, the following tests were conducted:

Open field exploration

The mice were individually removed from their cages and placed in the center of an arena for a single 30-min open field session. The Omnitech Fusion system used photobeams to monitor and record the animal’s placement during the session. Measured variables consisted of the total distance covered (in centimeters), the number of vertical activity episodes, the total time spent in ambulation (in seconds), and the distance traveled within the centroid (in centimeters). Data were recorded using the Omnitech Fusion system with a center ratio zone map.

Light/dark exploration

Mice were placed in a divided arena for a period of 10 min. Each mouse was allocated one session, with the time spent in the arena (in seconds) and the number of entrances into the light and darkened areas of the arena being recorded.

Jumping, grooming, and digging

For jumping, video recordings of the mice in their cages were taken; a red light was used to visualize the dark cycle. The number of jumps performed by each mouse was quantified within a 20-min time frame. For spontaneous grooming and digging, the mice were placed in a new empty cage with standard bedding. Grooming and digging behaviors were quantified within a 20-min time frame using Stopwatch+ [23, 24].

Primary neuron cultures

Primary neurons were obtained from P0 pups using a papain dissociation kit (Worthington Biochemical Corporation; LK003150) and cultured for a maximum of 28 days. The micro-dissected cortex and hippocampus were stained. To analyze synapse formation, Syn-1 (MACS; 130-119-358; 1:200), Psd-95 (Thermo Fisher; MAI-045; 1:500) primary antibodies and phalloidin stain (Abcam; ab176753) were used for immunofluorescence analysis. Images were collected using a Nikon A1 inverted confocal microscope with NIS-Elements AR version 5.11.01 software using 405 nm blue channel, 488 nm green channel, 561 nm red channel, and 640 nm far red laser wavelengths. The imaging process involved capturing multi-z-stack images with dimensions of 1024 × 1024 pixels and 15 z-steps every 0.2 µm. This was achieved using a 60x oil objective with a numerical aperture of 1.4 and a pixel size of 0.21 µm. The pinhole size used was 72.8 µm. Imaris (Version 9.5) was used to import the .nd2 images and convert them to the .ims format. The Filament Tracer module with the Autopath method was used to trace dendrites from the soma, with the thinnest diameter being 1.5 μm, and dendrite seed points within 30 µm of the soma were removed. Dendrite spines were then identified, with the seed point diameter being 0.7 μm and the maximum length being 7 µm. To measure dendritic branching, a Sholl analysis was conducted with 1 µm spaced spheres. To assess synapse formation, spots with a diameter of 2 µm were fitted to the pre-and post-synaptic markers; co-localized spots were analyzed using Imaris software.

Microelectrode array activity

For multi-electrode array (MEA), 48-well MEA plates (Axion Biosystems; M768-tMEA-48B-5) were pre-coated with PDL (Thermo Fisher; A3890401) for 1 h, rinsed three times with sterile water, and then dried overnight. On the following day, postnatal hippocampal and cortical cell suspensions were prepared with mouse laminin (Thermo Fisher; 23017015; 1 µg/mL) supplementation. 50,000 cells per well in a volume of 5 to 10 µL were placed in the middle of the well and incubated for one hour at 37 °C and 5% CO2; after 1 h plating, 200 µL of neuron culture media were added to each well. Cells were maintained at the same temperature and CO2 level with a half media change every 3-4 days. Spontaneous activity recordings were performed every 7 days in vitro (DIV) with the use of Axion Biosystems Maestro pro multifunctional system. Neural Real Time configured for continuous spontaneous activity data were collected for a period of 10 min using Axion Navigator software. A band-pass filter (3,000 Hz to 200 Hz) was applied utilizing variable threshold spikes detection. This was employed at ±6 standard deviation of the root mean squared of the background noise. An active electrode was determined based on a minimum spike rate of 5 spikes per minute, while a mature network was defined as having at least 8 out of 16 active electrodes. The inter-spike interval threshold was utilized to detect bursts in mature networks, with a maximum inter-spike interval of 100 ms and a minimum of 5 spikes. Additionally, network bursts were detected in mature networks with a minimum number of 50 spikes and a minimum of 35% participating electrodes, while also applying a synchronicity window of 20 ms. The neural excitability was represented by mean firing rate (Hz) and burst rate (Hz), whereas network synchronization was represented by network burst rate (Hz) and network synchronicity (index value between 0 and 1).

Electrophysiology

Extracellular recordings of field Excitatory Postsynaptic Potentials (fEPSPs), long-term potentiation (LTP), and long-term depression (LTD) were collected as per previously described methods [20, 21] with specific modifications. The slice preparation solution consisted of N-methyl-D-glucamine (NMDG, 93 mM), NaH2PO4 (1.2 mM), KCl (2.5 mM), NaHCO3 (30 mM), glucose (25 mM), sodium ascorbate (5 mM), sodium pyruvate (3 mM), thiourea (2 mM), HEPES (20 mM), CaCl2 (0.5 mM), and MgSO4 (10 mM). Also, the following composition was used for artificial cerebrospinal fluid (aCSF) recording: NaCl (124 mM), NaH2PO4 (1.25 mM), KCl (3 mM), NaHCO3 (26 mM), glucose (15 mM), sodium ascorbate (0.8 mM), CaCl2 (2.5 mM) and MgSO4 (1.3 mM). When saturated with carbogen, all the solutions were buffered to a pH of 7.3 and the osmolality was confirmed to be between 294 and 297 mOsm. Coronal slices were subject to recordings using fire polished borosilicate glass recording pipettes, which were filled with aCSF (3–5 MΩ) and a Pt/Ir concentric bipolar stimulating electrode. The potentiation and depression of the recordings were determined by dividing the average fEPSP slope over the last 10 min of the recording by the average 10 min of baseline immediately prior to induction of either LTP or LTD.

Western blotting

Western blotting was performed as previously described [12, 20, 21]. The primary antibodies used in the study were as follows: NAT8/NAT8B (Abcepta; AD4957c) and β-actin (CST; 3700; 1:1,000). The following secondary antibodies were used for infrared imaging on a LICOR Odyssey Infrared Imaging System (LI-COR Biosciences): Goat anti-mouse (926-32210) and donkey anti-rabbit (926-68070). Supplementary Figure 6 in the manuscript shows the original uncropped images from Western blot experiments.

Histology and immunostaining

Histology, immunostaining techniques, and Golgi staining were conducted according to the instructions described earlier [12, 15, 20, 21, 25, 26]. Klüver-Barrera staining was conducted on slices embedded in 10 µm of paraffin using the EMS kit (Electron Microscopy Sciences; 26681). Primary antibodies used for immunostaining included Iba1 (Abcam; ab178846), Gfap (Abcam;ab4674), Myelin Basic Protein (Abcam; ab40390; 1:200), MOG (Abcam; ab233549), Olig2 (Millipore; AB9610; 1:300), NAT8/NAT8b (Abcepta; AD4957c), and NeuN (Millipore; ABN91MI; 1:1,000).

The upright Leica DM4000 B microscope, equipped with a 10x air objective and 100x oil objective, was used to acquire bright-field images, and Image-Pro version 6.3 was utilized for image processing. All slides with fluorescent labeling were imaged on a Nikon A1 inverted confocal microscope, utilizing NIS-Elements AR version 5.11.01 software and Galvano scan head technology, with laser wavelengths of blue channel (405 nm), green channel (488 nm), red channel (561 nm), and far red (640 nm). For the slides stained with NAT8/NAT8B/NeuN, single z-slice images were obtained at a resolution of 1024 × 1024 pixels using a 10x air objective (NA = 0.3; 1.24 µm/pixel) and 60x oil objective (NA = 1.4; 0.21 µm/pixel) with pinhole size of 197.95 µm. For MBP/NeuN-stained slides, single z-slice images (1024 × 1024 pixels; 1.24 µm/pixel) were acquired usage of 10x air objective (NA = 0.3) at a pinhole size of 72.8 µm. The 10x objective (NA = 0.3) and pinhole size of 72.8 µm for MBP/MOG/Olig2/NeuN-stained slides produced single z-slice images with (1024 × 1024 pixels; 1.24 µm/pixel) resolution.

Microglial Iba1-positive cells were quantified using ImageJ (Version 1.52). This was achieved by creating binary images through an intensity threshold and then counting the objects with a Particle Analyzer tool. For Golgi staining images, pre-processing was done in ImageJ (Version 1.52) by inverting and then subtracting background three consecutive times with pixel sizes of 100, 50, and 25. The resulting images were stored in .tiff files, imported into Imaris (Bitplane; Version 9.5) and converted into the native .ims format. To analyze the secondary dendritic branches, the Autopath method of the Filament Tracer module was used. Branches were semi-manually traced with 0.25 µm and automatic dendrite volume detection was allowed. A mask was created to remove signals from reconstructed dendrites, allowing for the detection of branched spines with a minimum dendrite diameter of 0.25 μm and a maximum dendrite length of 5 μm. For further analysis, the dendrite spine density (measured in spines per 10 µm of dendrite length) and spine volume (measured in µm3) data were used.

ManAz (sialylation) assay

Primary cortical neuron cells were cultured onto coverslips in a 24-well plate at a density of 10,000 cells per well and allowed to grow for 2 weeks in a neurobasal A medium. Following this, Click-iT ManNAz was introduced at a concentration of 50 μM and the cells were incubated at 37 °C for 48 h. After the incubation period, the cells were carefully washed in PBS, fixed with 4% PFA and subsequently blocked using BSA solution. Cells were stained using a combination of Click-iT reaction buffer, cell buffer additive and alkyne dye for 30 min. Coverslips were mounted on microscope slides and images captured in a Nikon A1. The presence of ManAz at the cell membrane was measured through synapse volume changes with the aid of Imaris software and through quantifying fluorescent intensity using ImageJ.

ProteomicsProtein extraction and digestion

Cortical and hippocampal tissues were resuspended in 8 M urea lysis buffer (50 mM Tris-HCl, PH = 8), containing protease and phosphatase inhibitors (Thermo Fisher). Protein concentration was determined using a commercial bicinchoninic acid (BCA) protein assay. Proteins from all samples were reduced with 5 mM dithiothreitol (DTT) for 30 min at 37 °C and then alkylated with 15 mM iodoacetamide (IAA) for 45 min at room temperature (RT) in the dark. The samples were quenched by 5 mM DTT for 10 min at RT. The protein mixture was diluted with 50 mM Tris-HCl (pH=8) to a final urea concentration of ≤ 1 M before the addition of trypsin (Promega). Complete tryptic digestion was achieved using a trypsin-to-protein ratio of 1:50 (w/w) at 37 °C overnight. The digestion was quenched by acidification with 10% trifluoroacetic acid to a final pH below 3. Digested samples were desalted using Sep-Pak C18 cartridges (Waters) and eluted first with 0.1% formic acid (FA) in 50% acetonitrile (ACN) and then with 0.1% FA in 80% ACN. A Pierce Quantitative Colorimetric Peptide Assay was used to determine peptide concentration, with 200 μg peptides aliquoted for each sample and used for DiLeu isobaric labeling.

DiLeu labeling

Two sets of 12-plex N, N-dimethyl leucine (DiLeu) tags were used to label 24 samples, which included 12 cortex tissue samples and 12 hippocampus tissue samples. The DiLeu activation solution was prepared with anhydrous N,N-dimethylformamide, 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium tetrafluoroborate and N-methylmorpholine at 0.6x molar ratio to DiLeu tags. The tags were activated by vortexing the mixture of tags and activation solution for 45 min at RT. The supernatant of activation solution was then mixed with peptide samples reconstituted in 0.5 M triethylammonium bicarbonate buffer, followed by vortexing at RT for 2 h. The reaction was quenched with 5% hydroxylamine to a final concentration of 0.25%, and the mixtures were vortexed at RT for 10 min. Labeled peptide samples were combined at a ratio of 1:1:1:1:1:1:1:1:1:1:1:1 and dried in vacuo. Pooled samples underwent cleaning with strong cation exchange (SCX) according to the manufacturer’s protocols. To enhance proteome coverage and improve detection of low-abundance proteins, off-line high pH (HpH) fractionation was performed prior to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. Each DiLeu labeled set obtained 8 combined fractions after HpH fractionation.

LC-MS/MS

The samples were analyzed using a Q Exactive HF orbitrap mass spectrometer (Thermo Fisher Scientific) coupled with a Dionex UltiMate 3000 UPLC system. Each sample was dissolved in 0.1% FA before undergoing chromatographic separation on a 15 cm C18 column (1.7 μm, 130 Å, Waters). Mobile phase A consisted of water with 0.1% FA, while mobile phase B was composed of ACN with 0.1% FA. Separation was achieved with gradient elution of 4% to 40% mobile phase B over 120 min at a flow rate of 0.3 µL/min. Full MS data were acquired within a mass scan range of m/z 300–1500 at a resolution of 60 K, with an automatic gain control (AGC) target of 1 ×106 and a maximum injection time (IT) of 100 ms. For data-dependent acquisition tandem mass spectrometry (DDA-MS2), the top 20 precursor ions were selected for MS2 fragmentation at a resolution of 60 K, AGC target of 1 ×105, maximum IT of 200 ms, isolation width of 1.0 Da, fixed first mass at m/z 110, normalized collision energy (NCE) of 30, and dynamic exclusion was 45 seconds.

Data analysis

Protein identification and quantification were performed using MaxQuant (version 1.5.2.8). The collected mass spectral data were searched against the UniProt mouse reviewed database (November 2023). Trypsin was set as the digesting protease with two missed cleavages allowed. DiLeu labeling on peptide N termini and lysine residues (+145.12801 Da), and carbamidomethylation of cysteine residues (+57.02146 Da) were selected as static modifications. Oxidation of methionine residues (+15.99492 Da), and deamidation of asparagine and glutamine residues (+0.98402 Da) were defined as variable modifications. The first search peptide tolerance and main search peptide tolerance were set at 20 ppm and 4.5 ppm, respectively. Match between runs was enabled, and other parameters were set as default. Reporter ion intensities were normalized following correction of isotopic impurities, and subsequent analysis was performed by Perseus (1.6.15.0) and DAVID bioinformatics resources.

Statistical analysis

The data were analyzed using GraphPad Prism version 9.5.0, and all results are reported as mean ± standard deviation unless otherwise noted. For comparisons between two groups, an unpaired t-test was utilized. Ordinary one-way or two-way ANOVA was used for comparisons between three or more groups, with subsequent multiple comparison tests of either Tukey-Kramer (for all group comparisons) or Dunnett’s (for comparisons to a single control group). If ANOVA testing did not reveal any differences in sex, the data were merged to improve visual representation. The specifics of each statistical test are outlined in the figure legends. Grubb’s test was employed to eliminate outlier values identified at a significance level of P < 0.05. Statistical significance was determined as P < 0.05. All experiments were conducted at least two independent times with successful replication.