Mice
Female 5XFAD and C57BL/6 mice used in the studies were group housed (maximum of 5 animals per cage) in the Department of Animal Resources at Emory University under standard conditions. 5XFAD transgenic mice (The Jackson Laboratory, #034848) were purchased and maintained as hemizygotes on a C57BL/6 background. 5XFAD transgenic mice were confirmed by polymerase chain reaction and non-transgenic WT littermates were used as controls. All experiments were conducted in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Emory Institutional Animal Care and Use Committee.
Construction of miR-29a mimic
miR-29a mimic was constructed by first amplifying fragment encoding pre-miR-29a from the genomic DNA isolated from HEK-293T cells. The PCR product was then purified, digested with EcoRI and XbaI, and inserted into digested pAAV-MCS vector (Addgene, VPK-410). The construct sequenced was verified by Sanger sequencing. The primer sequences for miR-29a mimic are given in Supplementary Table 1.
miR-29a sponge design and cloning
miR-29a sponge was seven tandemly arrayed miR-29a binding sites separated by a 4-nt spacer, each of which was perfectly complementary in the seed region but with a bulge at positions 9–12 to prevent degradation by Argonaute 210. We annealed, ligated, gel purified and cloned miR-29a sponge into 3’UTR of psiCHECK-2 vector (Promega, C8021) and pAAV-GFP vector (Cell biolabs, AAV-400), respectively. All constructs were verified by Sanger sequencing. The primer sequences for miR-29a are given in Supplementary Table 1.
Luciferase assays
We plated HEK-293T cells into 24-well plates the day before transfection and transfected them in triplicate with psiCHECK-2 vector containing miR-29a sponge together with miR-29a mimic or empty vector (pAAV-MCS) at a ratio of 1:20 (300ng total DNA/well). Lipofectamine 2000 (Invitrogen, 11668019) was used as the transfection reagent. Cells were lysed in passive lysis buffer 48 h after transfection and assayed in triplicate using the Dual-Luciferase Reporter Assay System (Promega, E1910). Renilla luciferase activity was normalized to Firefly luciferase activity measured on a GloMax 96 microplate luminometer (Promega, E6521) and was then calculated relative to the negative control in each independent replicate.
Western blots
HEK-293T cells were plated into 6-well plate the day before transfection. miR-29a mimic together with pAAV-GFP vector containing miR-29a Sponge or empty vector (pAAV-GFP) were co-transfected into HEK-293T cells at a ratio of 20:1 with lipofectamine 2000 (1000ng total DNA/well). Fluorescence microscopy was used to check the percentage of GFP positive cells 24 h after transfection. 48 h after transfection, cells were lysed in RIPA buffer (Thermo Fisher Scientific, 89900) supplemented with Protease Inhibitor Cocktail (Roche, 4693159001). Protein concentrations were determined with the BCA Protein Assay kit (Thermo Scientific, 23235) according to the manufacturer’s instruction. Equal amounts of total proteins were resolved on a 10% Mini-Protean TGX precast gel (Bio-Rad, 4561033), and transferred to a PVDF membrane (Bio-Rad,1704156) with the Trans-Blot Turbo transfer system (Bio-Rad). The membranes were incubated with blocking solution (5% skim milk in Phosphate-buffered saline with 1% Tween-20) for 1 h, blotted by a blocking solution containing a primary antibody (DNMT3A, Cell Signaling Technology, #3598, 1:1000; GAPDH, Invitrogen, #39-8600, 1:10000) overnight at 4℃, and incubated with a horseradish peroxidase conjugated secondary antibody for 1 h. Proteins were visualized by ECL prime detection reagent (Cytiva, RPN2232) and ChemiDoc imaging system (Bio-Rad). Immunoreactive bands were quantified with ImageJ software. The experiment was performed in triplicate.
Adeno-associated virus (AAV) production and stereotaxic injection
The plasmid templates for AAV generation were pAAV-GFP vector containing miR-29a sponge or the empty vector. AAV-miR-29a sponge and control AAV were generated by Emory University Viral Core. Before AAV injection, female 5XFAD and WT mice were anesthetized and placed in a stereotaxic frame. After a skin incision was made, holes were drilled at x (± 1.5 mm from bregma) and y (− 2.0 mm from bregma). AAV-miR-29a sponge or control AAV were injected into the left and right hippocampi (z = − 1.9 mm from bregma) respectively, with 6.2X1010 total viral particles per side and delivered at a rate of 0.2µL/min. The syringe was left in place for 5 min and withdrawn slowly after each injection. When the injection was complete, skin was sutured and sterilized.
Morris water maze
Experiments were conducted by the Emory University Rodent Behavioral Core by trained personnel who were blinded to the mouse condition. In a circular 52-inch-diameter tank filled with opaque water kept at 23°C with a hidden circular platform (30 cm diameter) present 1 cm below the water in the northwestern quadrant of the tank, each mouse had four training trials to find the platform per day over 5 consecutive days. Each training trial lasted a maximum of 60 s. If a mouse did not find the platform in time, it was manually guided to it and placed on the platform for 10 s. Escape latency to the platform as well as swim speed were recorded by an automated tracking system (TopScan, CleverSys). A probe trial was conducted on the sixth day where the platform was removed, and the mice were released from the south start point and allowed to swim for 60 s. The tracking system recorded the percentage of search time in the quadrant where the platform was previously located.
Fear conditioning
Fear conditioning was conducted by the Emory University Rodent Behavioral Core by trained personnel who were blinded to the mouse condition. Fear conditioning occurred over 3 consecutive days in a chamber (H10-11M-TC, Coulbourn Instruments) equipped with a house light, a speaker, a ceiling-mounted camera and an electric grid shock floor that could be replaced with a non-shock wire mesh floor. Fear conditioning training on day 1 began with a 3 min acclimatization period followed by 3 tone-shock pairings during which the tone lasted 20 s and was co-terminated with a 3s, 0.5mA foot shock. Mouse behavior was recorded for 60 s after a tone-shock pairing before the next round. Contextual fear testing on day 2 was conducted in the same chamber as day 1 without any tone or shock. Cued fear testing on day 3 was conducted in a different chamber with a non-shock wire mesh floor and began with a 3 min acclimatization period followed by a 5 min tone without any shock. FreezeFrame software (Coulbourn Instruments) was used to record freezing behavior and the percentage of freezing time was determined.
RNA isolation and real-time quantitative PCR
Samples were homogenized in TRIzol reagent (Thermo Fisher Scientific, 15596018) and shaken for 15 s after addition of chloroform. The samples were transferred to pre-spun Phase Lock Gel-Heavy tubes (Quanta bio, 2302830) and incubated at RT for 5 min and then centrifuged at 12,000 g/4°C for 15 min. The upper phase aqueous solution was collected in a fresh tube and RNA was precipitated by isopropanol. Samples were gently mixed and left at -80°C overnight and then centrifuged at 14,000 rpm/4°C for 25 min. RNA pellet was washed twice in 75% ethanol and resuspended in nuclease-free water. miR-29a was reverse transcribed to cDNA using miR-29a specific primers from TaqMan (Thermo Fisher Scientific, Assay ID: 002112). Real-time polymerase chain reaction was performed using the Applied Biosystems TaqMan Gene Expression assay following the manufacturer’s instruction. Data were analyzed by the △△Ct methods using U6 as an endogenous control.
Tissue preparation
After cervical dislocation, mouse brains were removed and dissected at the midline. For biochemical analysis, hippocampi were dissociated and immediately snap-frozen in liquid nitrogen and stored at -80°C for protein and RNA sequencing. For immunofluorescence staining, mice were anesthetized and transcardially perfused with 0.9% sodium chloride and then fixed in ice-cold 4% paraformaldehyde in 1× PBS. The brains were removed and postfixed in 4% paraformaldehyde overnight at 4°C and transferred to 30% sucrose at 4°C for 48 h before being embedded for cryostat sectioning.
Immunofluorescence staining
Mouse brains were embedded with optimal cutting temperature compound (Tissue-Tek, 4583) and cut into serial 10-µm-thick coronal sections with a cryostat (Leica Biosystems). The sections were washed three times in 1XPBS for 5 min each and incubated with blocking buffer (PBS with 10% normal goat serum and 0.25% Triton X-100) for 1 h at RT. The sections were then incubated with primary antibodies overnight at 4°C, followed by incubation with Alexa-fluorophore-conjugated secondary antibodies (Thermo Fisher Scientific, A-11004, A-11011, 1:500) for 1 h at RT in the dark. Sections were rinsed and mounted onto slides using Vectashield mounting medium with DAPI (Vector Laboratories, H-1200). The following primary antibodies were used for immunofluorescence staining: anti-beta amyloid (Abcam, ab2539, 1:200); anti-Iba1 (Wako Chemicals, 019-19741, 1:200); and anti-GFAP (Abcam, ab7260, 1:200). Immunoreactivity (IR) was calculated as mean gray value (area of IR within ROI divided by total area of ROI) within ImageJ as previously described11. Images were obtained on a ZEISS 710 confocal laser-scanning microscope.
RNA seq data processing
Total RNA used in the sequencing study was isolated from the hippocampus using TRIzol reagent (Thermo Fisher Scientific, 15596018). RNA quality was measured on an Agilent 2100 Bioanalyzer system based on the 28S/18S ratio and the RNA integrity number (RIN). For each sample, 1 µg RNA was used to construct sequencing libraries using Illumina’s TruSeq RNA Sample Prep Kit. Samples were sequenced on Illumina’s HiSeq 2000 system with a sequencing depth of 40M total reads per sample (20M each direction), producing sequencing result in FastQ format. The QC on the sequence reads was done with FastQC and all samples were carried forward in the analysis. Sequenced data were aligned to mouse reference genome using STAR aligner version 2.7.3a12 and STAR produces a read count file for each sample using the algorithm of htseq-count13 with default settings.
Gene differential expression analysis
The differential expression analysis across experiment and control groups was implemented in R (version 4.1.2). Variance stabilizing transformation (VST) function offered by DESeq2 R package (version 1.32.0)14 was used to log2 transform the raw counts, normalize for library sizes, and reduce heteroskedasticity. The Surrogate Variable Analysis (SVA) method in the sva R package (version 3.40.0) 15 was used to detect potentially hidden variables from the normalized data. Given our sample size is small (n = 8), we only included the first surrogate variable (SV1) in the design matrix15. The R package DESeq2 was used to perform the differential expression analysis adjusted for SV1. The Benjamini-Hochberg method was used to control for the false discovery rate (FDR) and considered significant at FDR less than 0.1. The R package clusterProfiler16 (version 4.0.5) was used to perform gene ontology enrichment analysis for the significant genes following instructions in the package vignette. GO terms were considered significant at an FDR adjusted p-value less than 0.05.
Protein digestion and TMT labeling
Each mouse hippocampus sample was digested individually using the EasyPep™ mini sample preparation kit according to manufacturer instructions (ThermoFisher Scientific, A40006). Briefly, each sample was homogenized in 200 uL kit lysis buffer with Halt protease inhibitors and nuclease. A protein concentration assay was performed and 80ug went through digestion and desalting. Resulting peptides were desalted with a Sep-Pak C18 column (Waters, WAT054945) and dried under vacuum. Peptides were reconstituted in 100ul of 100mM triethyl ammonium bicarbonate (TEAB) and labeling performed as previously described17,18. One batch of 16-plex TMTPro isobaric tags (Thermofisher Scientific, A44520) was used to label all 16 samples. All 16 channels were then combined and dried by SpeedVac (LabConco) to approximately 100 µL and diluted with 1 mL of 0.1% (vol/vol) TFA, then acidified to a final concentration of 1% (vol/vol) FA and 0.1% (vol/vol) TFA. Peptides were desalted with a 60 mg HLB plate (Waters). The eluates were then dried to completeness.
High pH Fractionation
High pH fractionation was performed essentially as described19 with slight modification. Dried samples were re-suspended in high pH loading buffer (0.07% vol/vol NH4OH, 0.045% vol/vol FA, 2% vol/vol ACN) and loaded onto a Water’s BEH (2.1mm x 150 mm with 1.7 µm beads). An Thermo Vanquish UPLC system was used to carry out the fractionation. Solvent A consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 2% (vol/vol) ACN; solvent B consisted of 0.0175% (vol/vol) NH4OH, 0.01125% (vol/vol) FA, and 90% (vol/vol) ACN. The sample elution was performed over a 25 min gradient with a flow rate of 0.6 mL/min with a gradient from 0 to 50% B. A total of 96 individual equal volume fractions were collected across the gradient and dried to completeness using a vacuum centrifugation.
Liquid Chromatography Tandem Mass Spectrometry
All samples (~ 1ug for each fraction) were loaded and eluted using Dionex Ultimate 3000 RSLCnano (Thermofisher Scientific) an in-house packed 15 cm, 100 µm i.d. capillary column with 1.9 µm Reprosil-Pur C18 beads (Dr. Maisch, Ammerbuch, Germany) using a 23 min gradient. Mass spectrometry was performed with a high-field asymmetric waveform ion mobility spectrometry (FAIMS) Pro equipped Orbitrap Eclipse (Thermo) in positive ion mode using data-dependent acquisition with 2 second top speed cycles. Each cycle consisted of one full MS scan followed by as many MS/MS events that could fit within the given 2 second cycle time limit. MS scans were collected at a resolution of 120,000 (410–1600 m/z range, 4x10^5 AGC, 50 ms maximum ion injection time, FAIMS compensation voltage of -45). All higher energy collision-induced dissociation (HCD) MS/MS spectra were acquired at a resolution of 30,000 (0.7 m/z isolation width, 35% collision energy, 1.25×10^5 AGC target, 54 ms maximum ion time, TurboTMT on). Dynamic exclusion was set to exclude previously sequenced peaks for 20 seconds within a 10-ppm isolation window.
Protein identification and quantification
All raw files were searched using Thermo's Proteome Discoverer suite (version 2.4.1.15) with Sequest HT. The spectra were searched against a mouse uniprot database downloaded August 2020 (91414 target sequences). Search parameters included 20ppm precursor mass window, 0.05 Da product mass window, dynamic modifications methione (+ 15.995 Da), deamidated asparagine and glutamine (+ 0.984 Da), phosphorylated serine, threonine and tyrosine (+ 79.966 Da), and static modifications for carbamidomethyl cysteines (+ 57.021 Da) and N-terminal and Lysine-tagged TMT (+ 304.207 Da). Percolator was used filter PSMs to 0.1% FDR. Peptides were group using strict parsimony and only razor and unique peptides were used for protein level quantitation. Reporter ions were quantified from MS2 scans using an integration tolerance of 20 ppm with the most confident centroid setting. Only unique and razor (i.e., parsimonious) peptides were considered for quantification.
Protein differential expression analysis
The normalization and differential analysis of the proteomics data were performed in R (version 4.1.2). We included proteins with TMT abundance values in at least 50% replicates per group and display a high protein FDR confidence (FDR < 0.01). To normalize the raw data, for each sample, each protein’s abundance was first divided by the sum of abundance values of all the proteins profiled for that sample, followed by the log2 transformation. For each protein, we then constructed a linear model of normalized abundance as a function of group. The p values were adjusted for multiple comparisons using the FDR method. We next selected proteins that are predicted targets of miR-29a based on miRDB20,21 database and then performed the differential expression analysis. Proteins were declared to be significant at an FDR adjusted p-value less than 0.1 using the Benjamini-Hochberg control for the FDR.
Experimental design and statistical analysis
For AAV injection, female mice at 6–7 months of age (WT or 5×FAD) were injected with control AAV or AAV-miR-29a sponge. For simplicity, we refer the injection groups as WT-Control, WT-Sponge, FAD-Control and FAD-Sponge. The number of mice in each group was: WT-Control (n = 12), WT-Sponge (n = 11), FAD-Control (n = 9), FAD-Sponge (n = 10). The behavioral tests were performed 3 months after injection. We then randomly selected 3 mice from each group for immunofluorescence staining. For RNA-Seq and proteomics analyses, 4 mice were randomly selected from each group and one side of hippocampus was harvested for RNA-Seq analysis and the other side was for proteomics.
The expression of miR-29a and behavioral testing (i.e., Morris water maze and fear conditioning) were tested using two-way ANOVA or repeated measures (RM) ANOVA, followed by post hoc methods to control for multiple comparisons. Values were considered significant at p < 0.05 and a tendency at p < 0.1. Calculations were performed and figures created using Prism version 8.3 for Windows.