Human brain samples

Postmortem human brains were banked through a willed body donation program [29], with the donor’s clinical records prior to and/or during the last hospitalization obtained whenever available. All the brains were histopathologically processed according to the Standard Brain Banking Protocol set by China Brain Bank Consortium [30]. The brains were dissected and preserved fresh-frozen at -70 °C. Three AD cases (aged over 65 years) and three age-matched control cases (aged over 65 years) were included in the analyses. The stages of Aβ and tau pathology were assessed according to the established standards [31, 32]. The age-matched control cases had no known clinical history of dementia. The patient information is provided in Table 1.

Table 1 Demographic information of the brain donors. Braak neurofibrillary tangle (NFT) stage was determined by immunostaining with the AT8 antibody. Thal Aβ phase was determined by immunostaining with the 6E10 antibodyAnimals

Wild-type C57BL6J mice were purchased from Guangdong Medical Laboratory Animal Center (Guangzhou, China). MANF transgenic mice were described in our previous study [28] and generated by Cyagen Biosciences (Guangzhou, China). 5xFAD mice were from the Jackson Laboratory (hemizygous, strain #034848-JAX). APP/PS1 mice (C57BL6J background) were from Cavens-Biogle (Suzhou, China). All the mice were maintained in a 12-h light/dark cycle in the Division of Animal Resources of Jinan University. Both male and female mice were used in every experiment of this study. The primers used for genotyping are listed as follows. For MANF transgenic mice, forward: 5’-ATT GAC CTG AGC ACA GTG GAC CTG-3’; reverse: 5’-GTC ACT GTC ACC TTG TAC TCT GG-3’. For 5xFAD mice, common: 5’-ACC CCC ATG TCA GAG TTC CT-3’; WT: 5’-TAT ACA ACC TTG GGG GAT GG-3’; mutant: 5’-CGG GCC TCT TCG CTA TTA C-3’. For APP/PS1 mice, common: 5’-GGA TCT CTG AGG GGT CCA GT-3’; WT: 5’-GTG TGA TCC ATT CCA TCA GC-3’; mutant: 5’-ATG GTA GAG TAA GCG AGA ACA CG-3’.

Antibodies

The primary antibodies used in the study are listed as follows: MANF (LSBio, B2688; Abcam, ab67271; EM572, self-made), APP (Biolegend, 803001), amyloid beta (Abcam, ab126649), NeuN (Abcam, ab177487; Millipore, MAB377), β-tubulin (Proteintech, 10094–1-AP), PSD95 (Abcam, ab238135; GeneTex, GTX133091), synaptophysin (Santa Cruz Biotechnology, sc-12737), HA tag (Abcam, ab9110), GFAP (Sigma, G3893; Abcam, ab7260), IBA1 (Abcam, ab178846), GRP78 (Zen-bioscience, 200310-4F11), GST tag (Invitrogen, MA4-004), His tag (Cell signaling, 9715), Histone H3 (Cell signaling, 2365), Vinculin (Millipore, MAB3574), Cas9 (Sigma, MAC133), GFP (Emarbio, EM33012), ELAVL2 (Proteintech, 14008–1-AP, 67097–1-Ig), GRIA1 (Proteintech, 67642–1-Ig), GRIN2A (Proteintech, 28525–1-AP).

Mouse behavioral tests

For all the behavioral tests, the mice were randomly assigned to the experimental groups, and the tests were performed by a person blinded to the experimental groups. Individual data points that deviated by more than three standard deviations from the mean were classified as outliers and were excluded from the data analysis.

Novel object recognition was performed according to a published protocol [33]. Briefly, on Day 1, the mice were placed in a white plastic box measuring 33 × 33 × 20 cm for habituation. On Day 2, two identical objects (wooden circular cones) were placed in the box and the mice were allowed to explore the objects for 10 min. On Day 3, one old object was replaced with one new object (wooden cube), and the time the mice spent exploring the old and new objects was recorded using two stop watches.

For fear conditioning, the mice were placed in the apparatus (Coulbourn) composed of Plexiglas with a metal shock grid floor. Three conditioned stimulus (CS)—unconditioned stimulus (US) pairings were presented with a 1-min inter-trial interval. The CS consists of a 20-s 85-db tone, and the US consists of a 2-s 0.5-mA foot shock. The shock was delivered via a Precision Animal Shocker (Colbourn) connected to each fear conditioning chamber. On Day 2, the mice were placed in the same chamber used on Day 1, and the amount of freezing will be recorded via a camera and the software provided by Colbourn. On Day 3, the mice were exposed to the CS in a novel compartment. Following a 2-min habituation period, the 85-db tone was presented for 6 min, and the amount of freezing behavior was recorded.

For elevated plus maze, the mice were transferred to the testing room 30 min prior to the test for habituation. The mice were individually placed in the crossover of the open and closed arm, facing the open arm. The movement of the mice was recorded by a video camera mounted above the maze for 5 min. The recorded videos were analyzed using the TopScan behavior analyzing system.

For open field, an automated system was used (Photobeam activity system, San Diego Instruments). One day before the test, the mice were transferred to the test room in their home cages to acclimate. On the test day, each mouse was placed in a separate test chamber and allowed to explore the chamber freely for 30 min. The locomotion of the mice in the chamber was automatically recorded as infrared beam breaks.

Rewarded alternation T-maze was performed according to a published protocol [34]. Prior to the test, the mice were subjected to food restriction and were placed in the testing room for 30 min for habituation. In the training phase, individual mouse was placed in the starting arm. One biscuit was placed at the end of one goal arm, and the other goal arm was closed. After the mouse entered the goal arm and consumed the biscuit, these two goal arms were swapped alternately. Three rounds of training were conducted for three days. In the testing phase, one biscuit was placed at the end of one goal arm with the other one closed. Individual mouse was placed in the starting arm and allowed to enter the goal arm to consume the biscuit. For the first trial, one biscuit was placed in the other goal arm while keeping both arms open. If the mouse entered the arm with the biscuit, it was recorded as a correct trial and the next trial ensued with the goal arms swapped. If the mouse entered the arm without the biscuit, it was recorded as an incorrect and the biscuit remained in the same arm in the next trial. The trial was repeated ten times, and the correct rate of each mouse was calculated.

Western blotting

The brain tissues were lysed in ice-cold RIPA buffer containing a protease inhibitor cocktail (Mei5bio, MF182-plus-10) and phosphatase inhibitors (Sigma, S7920 and S6508). The lysates were sonicated for 5 s × 6 times and centrifuged at 12,000 g for 15 min at 4 °C. The supernatants were collected and subjected to SDS–PAGE. The proteins in the gel were transferred to a nitrocellulose membrane. The blots were blocked with 5% milk/TBST for 1 h at room temperature and incubated with selected primary antibodies in 3% BSA/TBST overnight at 4 °C. After three washes in PBS, the blots were incubated with HRP-conjugated secondary antibodies in 5% milk/TBST for 1 h at room temperature. The blots were then washed three times in TBST and developed using ECL Prime (Millipore, WBKLS0500). The images were acquired digitally using Clinx ChemiScope 6300.

Immunofluorescent staining

The mice were intracardially perfused with warm 0.9% saline solution, followed by 4% ice-cold paraformaldehyde (PFA) in 0.1 M PB solution. The brains of the mice were separated and fixed overnight in 4% PFA solution and transferred to 15% sucrose for 24 h, and to 30% sucrose for another 24 h. The brains were embedded in the OCT solution (Sakura, 4583) and sectioned at 30 μm in a cryostat (Thermo Fisher). The slices were blocked with 3% bovine serum albumin in PBS with 0.2% Triton X-100 (Sigma, X100) for 30 min at room temperature. The slices were then incubated with selected primary antibodies (buffered in PBS with 0.2% Triton X-100) at 4 °C overnight. On the following day, the slices were washed with PBS three times and incubated with secondary antibodies and nuclear dye DAPI (buffered in PBS with 0.2% Triton X-100) for 1 h at room temperature. Images were acquired using an Olympus FV3000 confocal laser scanning microscope.

The ImageJ software (Ver 1.54d) was used for the quantification of staining intensity or area. The image type was converted to 8-bit grayscale. The “Threshold” function was used to subtract background, and the “Measure” function was used to quantify the staining intensity or area. For the quantification of PSD95 and synaptophysin co-localized puncta, high-resolution images were captured using an Olympus FV3000 confocal laser scanning microscope equipped with a 100 × objective lens. The ImageJ plugin “ComDet” (v.0.5.5) was used with advanced scaling settings. The images were processed by navigating to “Image-Color-Make composite”, followed by appropriate zooming. A 10 μm × 10 μm rectangle was framed using the "Rectangle" tool. Subsequently, the "Detect particles" function was used with parameters set as follows: a maximum distance of 6 pixels between colocalized spots, an approximate particle size of 3 pixels, and an intensity threshold (in standard deviations) ranging from 3 to 5 to encompass all staining signals. The colocalized signals were automatically counted. To the quantify the co-staining area of IBA1 and synaptophysin, both images were imported into the "JaCoP" plugin of ImageJ. The “Threshold” function was used to subtract background and the Manders' coefficients were calculated. The M2 value, which represents the overlapping area of IBA1 and synaptophysin, was used for statistical analysis.

Immunohistochemistry

The brain slices were blocked in blocking buffer (3% BSA/2% goat serum/0.1% Triton X-100 in 1 X PBS) for 1 h and incubated with selected primary antibodies in the blocking buffer at 4 °C overnight. After washing with PBS three times, the brain slices were incubated with secondary antibodies for 10 min and developed with the Mouse and Rabbit Specific HRP/DAB (ABC) Detection IHC kit (Abcam, ab64264) following the manufacturer’s protocol. Images were acquired using a Zeiss AX10 Axio microscope. The ImageJ software was used for the quantification of staining intensity. The image type was converted to 8-bit grayscale. The “Threshold” function was used to subtract background, and the “Measure” function was used to quantify the staining intensity.

Enzyme-linked immunosorbent assay (ELISA)

The ELISA kit (Invitrogen, KHB3441) specific to Aβ42 was used according to the manufacturer’s instructions. Mouse brain tissues were homogenized and sonicated in PBS containing a protease inhibitor cocktail (Mei5bio, MF182-plus-10) and phosphatase inhibitors (Sigma, S7920 and S6508), followed by centrifugation at 12,000 g for 15 min at 4 °C. The concentration of the samples was adjusted to 1 μg/μl and diluted 1:500 before use. The standard curve was calculated using the ELISACalc software. Aβ42 concentrations in the samples were determined by fitting the standard curve.

Electrophysiology

The brain slices of 250 μm were prepared in pre-cooled NMDG solution containing 93 mM NMDG, 2.5 mM KCl,1.2 mM NaH2PO4, 25 mM D-glucose, 30 mM NaHCO3, 20 mM HEPES, 5 mM sodium ascorbate, 2 mM Thiourea, 3 mM sodium pyruvate, 10 mM MgSO4 and 0.5 mM CaCl2. The solution was titrated with HCl to pH 7.25 at an osmotic pressure of 300 ~ 310 mOsm. The prepared brain slices were incubated in the NMDG solution with oxygen (95% O2 and 5% CO2) and kept at 32 ± 1 °C for 1 h. The brain slices were placed in the recording chamber with artificial cerebrospinal fluid (aCSF) containing 126 mM NaCl, 2.5 mM KCl, 1.25 mM NaH2PO4, 26 mM NaHCO3, 2 mM MgCl2, 2 mM CaCl2, and 10 mM glucose, and recorded with patch pipettes (6–8 MΩ) filled with internal solutions. Gluconate internal electrode fluid was used for recording (126 mM potassium gluconate, 4 mM KCl, 10 mM HEPES, 4 mM Mg-ATP, 0.5 mM Na-GTP, and 10 mM creatine phosphate). Whole-cell recordings of hippocampal CA1 pyramidal cells were performed via IR-DIC visualization under a Nikon Eclipse FN-1 microscope. The cells were recorded with a holding potential of − 70 mV. The data were collected using a Multiclamp 700B amplifier (Molecular Devices), with low pass filtered at 3 kHz and sampled at 10 kHz. The data were analyzed with Clampfit 10.4.

Golgi staining

The Golgi–Cox Impregnation & Staining System was used according to the manufacturer’s instruction (FD Rapid GolgiStainTM kit, FD NeuroTechnologies). After impregnation, brain sections (100 μm) were obtained using a cryostat and mounted to gelatin-coated glass slides. Sequential images were taken with an Olympus FV3000 microscope under a 63 × oil objective using the z-stack function at 10 μm intervals. The density of apical densities from the CA1 region of the hippocampus was measured by manually counting the number of spines along 100 μm of the dendrites using the “Multi-point” tool in the ImageJ software. Each group contained three mice and eight segments from each mouse were randomly selected for quantification.

Virus packaging and stereotaxic injection

The AAV-MANF, AAV-GFP, AAV-Ctrl-gRNA, AAV-Manf-gRNA, and AAV-Cas9 viral vectors were generated and described in our previous study [27]. These viral vectors were sent to PackGene Biotech Inc for packaging (AAV9 serotype, 1 × 1013 vg/ml). The method for stereotaxic injection was adopted from our previous study [27]. The mice were anesthetized with 1.5% isoflurane inhalation and stabilized in a stereotaxic instrument (RWD, 69100). The hair around the surgical site was removed using a disinfected scissor and the skin was sterilized with 70% alcohol. The injection site was determined according to the distance from the bregma: anterior–posterior =  − 2.5 mm, medial–lateral =  ± 2.0 mm, dorsal–ventral =—2.0 mm. A small hole was drilled on the skull, and a 30-gauge Hamilton microsyringe was used to deliver the virus at a speed of 200 nl per minutes (bilateral injection, 1 μl in each side). After surgery, the mice were placed on a heated blanket to recover from the anesthetic.

RNA sequencing

The mouse hippocampal tissues were dissected and sent to Novogene Co, Ltd for RNA extraction, library construction and sequencing. The raw data were quantified using Salmon software (Ver. 1.9.0) with the mapping-based mode. The differentially expressed genes (DEGs) were analyzed using the edgeR package (Ver 3.1.6) with P value < 0.05 and |log2Fold Change|> 0.5. Correlation map was plotted using corrplot (Ver 0.92) package. To explore the related GO and KEGG pathways, the clusterProfiler (Ver 4.0.2) package was used for the up-regulated or down-regulated genes with P value < 0.05. The ComplexHeamap package (Ver 2.16.0) was used to draw heatmap plot. All analyses were performed using the R (Ver 4.3.0) and the Rstudio (Ver 2023.09.1 + 494) software.

Recombinant protein production

The pET-28a vector expressing His tagged MANF protein and the method for purification of the recombinant protein was described in our previous study [27]. Full-length ELAVL2 and ELAVL2 fragments were amplified from the mouse cDNA library and cloned into the pGEX-4T-1 vector. Primers used are listed as follows. Full length ELAVL2, forward: 5’-TAG GAT CCA TGG CAG TCA GAC TGT GTG ATG-3’; reverse: 5’-TAG AAT TCT TAG GCT TTG TGC GTT TTG TTT G-3’. Fragment RRM1, forward: 5’-TAG GAT CCA TGG CAG TCA GAC TGT GTG ATG-3’; reverse: 5’-TAG AAT TCT GAG GCT GAG CTT GGG CGA G-3’. Fragment RRM2, forward: 5’-TAG GAT CCA TCA GAG ATG CAA ACT TAT ACG TC-3’; reverse: 5’-TAG AAT TCG GCC TGA TTG GTT TTT TGG CTT G-3’. Fragment RRM3, forward: 5’-TAG GAT CCT CCC AGC TGT ACC AGT CTC CA-3’; reverse: 5’-TAG AAT TCT TAG GCT TTG TGC GTT TTG TTT G-3’.

For protein production, the vectors were transformed into BL21(DE3) competent cells, and the cells were induced by IPTG for 1 h at 37 °C. The cells were lysed in lysis buffer (1% Triton X-100 and 1:500 PMSF in 1 × PBS) by sonication, and then mixed with glutathione beads (Sigma) at 4 °C overnight. The beads were washed (0.5% Triton X-100 and 1:500 PMSF in 1 × PBS) three times, re-suspended in 1 × PBS, and stored at 4 °C.

Immunoprecipitation and in vitro binding assay

The cell or brain lysates were homogenized in NP-40 buffer (50 mM NaCl, 50 mM Tris–HCl pH8.0, 0.1% Triton X-100, 0.5% NP-40), and 300 μg of protein was used for one experiment. The protein mixture was first pre-cleared with protein A-agarose (Sigma) for 1 h and then incubated with the primary antibody overnight. The next day, protein A-agarose was added to the mixture and incubated for 1 h. The precipitated antibody-protein complexes were collected for western blotting.

For in vitro binding assay, His-MANF recombinant protein was mixed with glutathione beads conjugated with full-length ELAVL2 or ELAVL2 fragments in 500 μl of NP40 buffer and kept at 4 °C overnight. On the next day, the beads were washed with NP40 buffer three times, and the precipitated beads–protein complexes were used for western blotting.

Quantitative real-time PCR

The RNA was extracted from the brain tissues using the TRIzol reagent (Invitrogen). Equal amount of RNA was used for cDNA synthesis, using PrimeScript RT Reagent Kit with gDNA Eraser (Takara, RR047A). For quantitative real-time PCR, QuantiNova™ SYBR Green PCR kit (QIAGEN, 208,052) was used. The reaction was performed in CFX Connect Real-Time System (Bio-Rad). Primers used are listed as follows. Gapdh, forward: 5’-ATC ACT GCC ACC CAG AAG ACT GT-3’; reverse: 5’-ATG ACC TTG CCC ACA GCC TTG G-3’. Adcy8, forward: 5’-GGC TTC CTA CAC CTT GAC TGT-3’; reverse: 5’-ATG ACC CCT CGG TAG CTG TAT-3’. Casp3, forward: 5’-CTC GCT CTG GTA CGG ATG TG-3’; reverse: 5’-TCC CAT AAA TGA CCC CTT CAT CA-3’. Htrb1, forward: 5’-CGC CGA CGG CTA CAT TTA C-3’; reverse: 5’-AGC GAT TAC AAA GGC GTT GGA-3’. Cyp2j9, forward: 5’-ATG CGC CTT CCT TTC GTG G-3’; reverse: 5’-CCA GGC TTA GAA CAT TCC CGT A-3’. Plk2, forward: 5’-GAC TAC TGC ACC ATA AGC ATG T-3’; reverse: 5’-CTT CTG GCT CTG TCA ACA CCT-3’. Hrh1, forward: 5’-CAG ACC TGA TTG TAG GGG CAG-3’; reverse: 5’-CAT AGA GAG CCA AAA GAG GCA G-3’. Bdnf, forward: 5’-TCA TAC TTC GGT TGC ATG AAG G-3’; reverse: 5’-ACA CCT GGG TAG GCC AAG TT-3’. Map3k1, forward: 5’-TAA ATA CCG GGT GTT TAT TGG GC-3’; reverse: 5’-TTT TCT CCA TAA CAT GGG GTC AG-3’. Nup93, forward: 5’-CGT TCC CGT ACC CTC ACA C-3’; reverse: 5’-AAG TCC CCT TGA CCC GAG AA-3’. Cyp2j9, forward: 5’-ATG CGC CTT CCT TTC GTG G-3’; reverse: 5’-CCA GGC TTA GAA CAT TCC CGT A-3’. Gng13, forward: 5’-AGA GCC TCA AGT ACC AAC TGG-3’; reverse: 5’-GGG TCC TTG GGG ATT CCA T-3’.

RNA Immunoprecipitation

The RiboCluster Profiler RIP-Assay kit (MBL, RN1001) was used according to the manufacturer’s instructions. The mouse brains were homogenized with a glass grinder in 1 mL of lysis buffer and incubated on rotation for 30 min at 4 °C. The lysate was centrifuged at 12,000 × g for 5 min at 4 °C. The supernatant was transferred to a tube containing the primary antibody-conjugated protein A magnetic beads (Invitrogen, 88846) and incubated with rotation for 3 h at 4 °C. The beads-RNP complexes were then washed and subjected to RNA isolation. All the operations were conducted on an RNA-free workbench.

Subcellular fractionation

The mouse brains were homogenized using a glass grinder in the buffer containing 0.32 M sucrose, 4 mM HEPES, and adjusted to pH 7.5. The homogenate was filtered through a 40 μm cell strainer (Biologix, 15–1040) and centrifuged at 800 × g for 10 min. The resulting pellet (P1) and supernatant (S1) were carefully separated. The P1 fraction was resuspended in 0.32 M sucrose and centrifuged again at 800 × g for 10 min to obtain the nuclear fraction, whereas the supernatant was discarded. The S1 fraction was subjected to centrifugation at 800 × g and the supernatant was collected as the cytoplasmic fraction.

Statistical analysis

The data were analyzed using the Prism 9 software (GraphPad). For comparisons between two groups, two-tailed student t-test was used. For three or more groups, one-way ANOVA with Tukey's multiple comparisons test was used. All experiments were repeated at least three times, and the quantification was presented as mean ± SEM. A P value less than 0.05 was considered statistically significant.