Viruses, reagents and antibodies

Human TFEB was subcloned into pCDNA3.1-CMV-MCS-3flag plasmid, and single, double or triple point mutation based on the above plasmid were constructed by SyngenTech Biotechnology (Beijing, China). The TFEB luciferase plasmid was constructed by Obio Technology (Shanghai, China). The HDAC7 and P301S-tau (1N4R) plasmids, and the Lenti-GfaABC1D-Cre-NLS, LV-GfaABC1D-HDAC7-GFP, LV-GfaABC1D-GFP, LV-GfaABC1D-TFEB (WT)−3xFlag, LV-GfaABC1D-TFEB (S122D)−3xFlag, LV-GfaABC1D-TFEB (S211D)−3xFlag, LV-GfaABC1D-TFEB (K310Q)−3xFlag, LV-GfaABC1D-TFEB (K310R)−3xFlag, LV-GfaABC1D-3xFlag, LV-GfaABC1D-Tau (human P301S, 1N4R)−3xFlag and LV-GfaABC1D-mCherry-Tau (human P301S, 1N4R) were constructed by Obio Technology (Shanghai, China). The AAV-G1-shHD7-GFP and LV-GfaABC1D-GFP were constructed and packaged by Brain Case Technology (Shenzhen, China), and the target sequence of HDAC7 shRNA is TGCGCTACAAACCCAAGAAAT. The AAV11 was used in animal experiments. Lentiviruses were used in primary astrocytes with Multiplicity of infection (MOI) of 10. TMP195 (a selective class IIa HDAC inhibitor), Bafilomycin A1 (a H+-ATPase inhibitor) and Torin1 (a mTORC1 inhibitor) were bought from Selleck (Houston, TX, USA). LysoTracker Red DND-99 was from Invitrogen (Carlsbad, CA, USA). Antibodies used in this study are listed in Supplementary Table 1.

Human samples

Dr. Chao Ma from the Human Brain Bank at the Chinese Academy of Medical Sciences and Peking Union Medical College (Beijing, China) kindly provided postmortem human brain samples. The diagnosis of Alzheimer's disease (AD) was conducted in accordance with the criteria established by the consortium to Establish a Registry for AD and the National Institute on Aging. This study was performed in accordance with institutional regulatory guidelines and principles of human subject protection in the Declaration of Helsinki. Detailed information regarding the human samples can be found in Supplementary Table 2.

Animals, stereotaxic surgery and drug treatment

The PS19 mice (PS19 line, Stock No: 008169) were obtained from The Jackson Laboratory. Heterozygotes were bred to C57BL/6 wild-type mice to maintain the line. HDAC7flx/flx mice were obtained from GenPharmatech. Adult C57BL/6 mice were obtained from Changzhou Cavens Laboratory Animal Co., Ltd (Changzhou, China). The mice were housed in groups of four to five per cage with ad libitum access to food and water, and were maintained under a 12-h light/dark cycle (lights on at 7:00 p.m., off at 7:00 a.m.) at a stable temperature (22 ± 2 °C). In the present study, we have complied with all relevant ethical regulations for the animal testing and research. All procedures were approved by institutional guidelines and the Animal Care and Use Committee (Shenzhen University, Shenzhen, China) of the university’s animal core facility.

For brain stereotactic injection, mice were anesthetized with isoflurane and placed on a stereotaxic apparatus, and then sterilized with iodophor and the scalp was incised along the midline of the head. Hole was stereotaxically drilled in the skull at posterior 2.1 mm, lateral 1.4 mm, and ventral 2.0 mm relative to the bregma. Using a microinjection system (World Precision Instruments), LV-GfaABC1D-HDAC7-GFP (2.5 μL, 1.0 × 109 vg/mL), LV-GfaABC1D -GFP (2.5 μL, 1.0 × 109 vg/mL), AAV-GfaABC1D-shHDAC7-GFP (1μL, 5.0 × 1012 vg/mL) or AAV-GfaABC1D-shHDAC7-GFP (1μL, 5.0 × 1012 vg/mL) was bilaterally injected into the hippocampal DG region (posterior 2.1 mm, lateral 1.4 mm, ventral 2.0 mm) at a rate of 0.125 μL/min. The needle was kept for 5 min before withdrawal, the skin was sutured and mice were placed beside a heater for recovery.

For intraperitoneal injection, TMP195 was diluted to 7.5 mg/mL with sterile 0.9% saline containing 5% (vol/vol) Tween-80 and 20% (vol/vol) PEG-300. TauTg or WT mice were then intraperitoneally injected with the diluted TMP195.

Novel object recognition

Arenas (50cm × 50cm × 50 cm) were utilized to house object A and object B in the corner. For the training trials, the mice were given a 5-minute period to acclimate themselves on the first day. Following each familiarization period, the arenas were cleansed with 70% ethanol. After a 24-hour interval, object B was substituted with object C, and the mice were allotted 5 minutes to investigate both objects. The behavioral responses were documented via a video camera situated overhead the arenas, and the duration of exploration in the arena containing the novel object C and familiar object A was quantified respectively.

Morris water maze

The Morris Water Maze (MWM) test was employed to assess spatial learning and memory, as previously described [26]. During the acquisition training phase, mice underwent daily training in the water maze for a duration of 5 consecutive days, with each day consisting of 3 trials. These trials were conducted between 14:00 and 20:00 p.m., with a 30-second interval between each trial. In each trial, mice were allowed 60 s to find the hidden platform. In the event that the mice were unable to locate the platform within 60 s, they were guided to it and required to remain on it for 30 s. The time taken to find the platform and the swimming path within the 60 s timeframe were recorded using a fixed video camera positioned 1.5 meters above the water surface. Spatial memory was assessed 48 h after the training phase, wherein the platform was removed. A digital device connected to a computer was employed to record the latency of the initial platform crossing, the number of target platform crossings within the 60-second period, and the time spent in the target quadrant during the 60-s probe phase.

Electrophysiology recordings

Mice were anesthetized with isoflurane and brains were removed in ice-cold artificial cerebrospinal fluid (a-CSF): 119 mM NaCl, 2.5 mM KCl, 26.2 mM NaHCO3, 1 mM NaH2PO4, 11 mM glucose, 1.3 mM MgSO4, and 2.5 mM CaCl2 (pH 7.4). Coronal slices (350 μm thick) were cut in ice-cold a-CSF using a Leica VT1000S vibratome and then transferred into an oxygenated chamber for a 2 h recover at room temperature.

For LTP recordings, brain slices were transferred to a recording chamber and immersed in a-CSF. The slices were placed in a chamber containing an 8 × 8 microelectrode array (Parker Technology, Beijing, China) on the bottom surface, with each microelectrode measuring 50 × 50 um and spaced 150 μm apart. The slices remained submerged in a-CSF throughout the experiment. Electrical signals were captured using the MED64 system (alpha MED Sciences, Japan). Stimulation of the Schaffer collaterals from the CA3 region was achieved using a 0.1MΩ tungsten monopolar electrode, while field excitatory postsynaptic potentials (fEPSPs) in the CA1 region were recorded using a glass microelectrode filled with a-CSF with a resistance of 3–4 MΩ. The signals were amplified utilizing a MultiClamp 700 B amplifier (Axon), digitized through a Digidata 1440A (Axon) with a 2 kHz low pass filter and a 3 Hz high pass filter, and subsequently recorded and saved using Clampex 10.4 software (Axon) for subsequent offline data analysis. LTP of fEPSPs was induced by three times of high-frequency stimulation (HFS; 100Hz, 1-s duration) with a 200-ms interval. The LTP magnitude was assessed as the normalized percentage change in the fEPSP slope (10%−90%) taken during the 60 min interval after LTP induction.

Golgi staining and analysis

Golgi staining was performed according to the manufacturer's protocol (FD NeuroTech, cat# PK401). Briefly, mouse brains were removed after anesthetization with isoflurane and immersed in a mixture of Solutions A + B for 2 weeks at room temperature in dark. Then the brains were transferred into a new tube and added with Solution C for another 7 days at 4°C in dark. All brains were sliced into 100‐μm‐thick sections using a Vibratome (VT1200S; Leica, Germany). Images were obtained using a microscope (Ni‐E, Nikon, Japan). The dendritic complexity of hippocampal DG neurons was assessed through Sholl analysis utilizing the Simple Neurite Tracer plugin and ImageJ (Fiji, Japan) software. Imaging of dendritic spines of DG neurons was conducted using a 100× oil immersion lens. The analysis in this study focused solely on secondary dendrites.

Cell culture

Primary mouse astrocytes were isolated and purified from the mouse cortex as described previously [26]. In brief, cortices from C57BL/6 or HDAC7flx/flx mice were isolated and dissociated cells were centrifuged and resuspended in Dulbecco’s modified Eagle’s medium (DMEM) F-12 supplemented with 10% Fetal bovine serum (FBS) and 1% penicillin/streptomycin (complete medium). The cells were then plated in a T75 flask, and the culture medium was changed every 3 days. 7–10 days later, the astrocytes were sorted at shake cultivation rotating for 14 h at 200–220 rpm. The cultured cells were then treated with trypsin and the disassociated cells were re-plated in culture plate or 12-well glass chambers at a density of 5 × 105 cells per well for Western blotting and quantitative real-time PCR (RT-qPCR) analysis and 1 × 105 cells per well for immunofluorescence imaging.

HEK293T cells were cultured with DMEM-high glucose medium supplemented with 10% Fetal FBS and 1% penicillin/streptomycin at 37℃ in the presence of 5% CO2. HEK293T cell transfection was performed with lipoFactMaxTM (ABP biosciences, Beltsville, USA).

Western blotting

Western blotting was performed as previously described [25]. To analyze soluble and insoluble tau fractions in mice, mouse brain tissues were extracted following an established protocol [27]. Briefly, mouse brain tissues were isolated on ice-cold PBS, homogenized in RIPA lysis buffer (Beyotime, cat# P0013B) with 1% protease inhibitors and phosphatase inhibitors, and centrifuged at 14,000 g for 15 min at 4°C. Then, the supernatant was transferred to new tubes. The pellet was washed with ice-cold RIPA buffer and then lysed with 10% SDS buffer (10% SDS, 250 mM Tris, pH 6.8) and sonicated in a water bath sonicator until no particulate material was visible. Protein concentration was measured using the BCA method (Pierce BCA protein assay kit, 23225) for both RIPA-soluble and RIPA-insoluble extracts. After SDS/PAGE, the protein was transferred onto nitrocellulose membranes (Whatman) and then blocked with blocking buffer (Beyotime, cat# P0023B) for 1 h at room temperature. Indicated primary antibodies were incubated at 4℃ overnight. The corresponding peroxidase-conjugated secondary antibodies were incubated for 1 h at room temperature. Protein signals were detected by the ECL detection system and analyzed with ImageJ software.

Immunoprecipitation

The immunoprecipitation assay was conducted in accordance with established protocols. Briefly, cultured cells were lysed on ice for 30 minutes using IP buffer (Beyotime, cat# P0013) containing 1% protease inhibitors and phosphatase inhibitors, followed by centrifugation at 14,000 g for 10 minutes. The resulting supernatants (1 μg protein) were then incubated overnight at 4°C with specified primary antibodies, with gentle rocking, and subsequently incubated for 2 hours with protein A + G agarose. The immunoprecipitates were washed three times with PBS, resuspended in 2× loading buffer, eluted by boiling for 10 minutes and analyzed by western blotting.

Immunostaining

For animal experiments, mice were subjected to anesthesia using a 1% solution of pentobarbital sodium, followed by intracardial perfusion with saline and subsequently with a 4% solution of paraformaldehyde (PFA) in 0.1 M phosphate buffer at a pH of 7.4. The brains of the mice were then extracted and further fixed in a 4% PFA solution for a duration of 12 hours, after which they were cryoprotected using successive solutions of 20% and 30% sucrose. Brain sections with a thickness of 35 μm were obtained using a cryostat microtome (CM1900, Leica, Germany). In order to perform immunofluorescence staining, the aforementioned brain sections were first washed in PBS, then blocked in a buffer solution containing 3% bull serum albumin and 0.5% Triton X-100 for a period of 1 hour, and subsequently incubated with primary antibodies at 4℃ overnight. After washed with PBS, slices were incubated with secondary antibodies conjugated to Alexa-Fluor 488/546/647 at 37℃ for 1 h, followed by DAPI staining for 10 min.

For immunohistochemistry staining, endogenous peroxidase activity was eliminated by incubating brain slices in 0.3% H2O2 (in PBS) at 37℃ for 30 min before serum blocking. Immunoreactions were performed using a DAB-staining kit (ZSGB-BIO). For immunostaining of human brain slices, antigen-retrieval was performed in boiling sodium citrate buffer (10 mM, pH 6.0) and auto fluorescence of the brain slices was blocked with Sudan black (0.3%, room temperature, 10 min). For primary astrocytes, after washed with PBS, cells were fixed with paraformaldehyde (4%, room temperature, 20 min) and permeabilized in PBS containing triton X-100 (0.5%, 30 min) before serum blocking.

The slices or coverslips were washed with PBS and mounted onto slides. Imaging was performed using a Zeiss LSM880 confocal microscope or Leica Aperio CS2 scanner.

Mass spectrometric analysis

The mass spectrometric (MS) analysis was performed based on previous report [28]. Briefly, Flag-TFEB together with HDAC7 or vector was transfected into HEK293T cells for 24 h, and then immunoprecipitation was conducted with Flag antibody as described above. The immunoprecipitated protein samples were separated by SDS-PAGE and stained by Coomassie brilliant blue staining, and then the interested bands were carefully cut off and digested with trypsin. The peptides were analyzed with a ThermoFisher Q-Exactive mass spectrometer (ThermoFisher, USA) fitted with a Nano Flex ion source (Thermo Fisher, USA). The LC-MS/MS data were subjected to analysis for the purpose of protein identification and quantification utilizing PEAKS Studio 8.5. The local false discovery rate at PSM was 1.0% after searching against the Uniprot human database with a maximum of two missed cleavages. Precursor and fragment mass tolerance were set to 10 ppm and 0.05 Da, respectively. Shortlisted acetylated peptides were further subjected to manual examination of MS/MS spectra to identify the precise locations of acetylation.

To assess the distribution of TMP195 in the brain after intraperitoneal administration, the mouse brain was harvested at various time points (2, 4, 6, 8, 10 and 12 h post-injection) and immediately immersed in methanol. Subsequently, the brain tissue was homogenized at 4°C, followed by centrifugation at 14,000 g for 15 min and centrifugation at 30,000 g for 30 min. The supernatant along with a TMP195 standard, was subjected to LC-MS/MS analysis using AB SCIEX QTRAP6500+. The content of TMP195 in each sample was determined by referencing a standard curve.

Tau fibril preparation

The untagged full-length 2N4R human tau was cloned into the pET29b vector in the E. coli strain BL21. Briefly, BL-21 cells transfected with tau were cultured for 12–16 h in LB medium containing 100 μg/mL ampicillin at 180 rpm, at 37 °C. The cells were then collected and lysed by sonication (5 s sonication, 2 s interval, total 15 min) in bacterial lysate with 1 mM PMSF. Sequentially, the tau protein was purified by anion exchange chromatography (HiPrep CM FF 16/10, Uppsala, Sweden) and agarose chromatography (Hiload 16/600 Superdex 75 pg, Uppsala, Sweden). The buffer used for tau purification is as following: for anion exchange chromatography, start buffer: 25 mM Tris-HCl, 20 mM NaCl, pH 8; elution buffer: 25 mM Tris-HCl, 1 M NaCl, pH 8. For agarose chromatography, running buffer: 0.05 M NaPO4, 0.15 M NaCl, pH 7.2. A fast protein liquid chromatography system (AKTA, GE Healthcare, Boston, MA, USA) was used in this study. Coomassie brilliant blue staining was used to identify the purified fractions. The in vitro tau fibrillization assay was performed under the following conditions: 50 μM purified tau protein, 12.5 μM heparin, 2 mM dithiothreitol (DTT), and a protease inhibitor cocktail comprising 10 μg/ml leupeptin, 5 μg/ml chymostatin, 3 μg/ml elastatinal, and 1 μg/ml pepstatin, all in phosphate buffer saline (PBS). The mixed solution was incubated at 37℃ for a duration of two weeks, with the addition of 1 mM fresh DTT to the solution every 24 h. Fibrillization was confirmed using the thioflavin T fluorescence assay, Coomassie brilliant blue staining and transmission electron microscopy. For transmission electron microscopy, the protein samples were diluted to 5 μM and spotted onto a 230-mesh carbon-coated copper grid. At 10,000 x magnification, pffs were imaged using a Jeol JEM1230 transmission electron microscope. After incubation for 30 min, the residual solution on the surface of the grid was removed by filter paper. The grids were washed with ddH2O and mixed with 1% uranium acetate for 30 s. After confirmation, the pffs were frozen as single use aliquots (50 μM) at −80℃. The purified tau fibrils were labelled with 5-FAM label at 4℃ for 8 h, and dialysis for 72 hours (dialysate was changed three times) to remove unbound 5-FAM. The specific absorption peak at 490 nm of FAM-labeled tau fibrils was further confirmed.

Tau uptake and degradation assay

Primary astrocytes were incubated with FAM-labelled tau-pff (1 μM) for 4 h, then washed with PBS three times, followed by incubation of lysotracker red for 30 min (100 nM) and staining with Hoechst for 5 min. Cells were subjected to live imaging using a Zeiss LSM880 confocal microscopy. The colocalization of tau-pff and Lysotracker was quantified with ImageJ software using Manders' Colocalization Coefficients. the M1 Manders' coefficient is defined as: \(_=\frac__}_C1}_}\), where \(_\) represent the intensity of individual pixels for tau-pff, and \(_\) represent the colocalizing pixels of tau-pff with Lysotracker. For tau uptake assay, astrocytes were incubated with tau-pff (1 μM) for 1–12 h. At various time points, cells were washed with PBS three times and lysed in RIPA buffer. Intracellular tau levels were quantified by ELISA analysis.

For tau degradation assay, astrocytes were incubated with tau-pff (1 μM) for 4 h, followed by the removal of the tau-containing media and thorough washout. Cells were continuously cultured at 37℃ for 30–180 min. At various time points, cells were washed with PBS three times and lysed in RIPA buffer. Intracellular tau levels were quantified by ELISA analysis. Bafilomycin A1 (100 nM) was added to the cells 1 h before washout of tau-containing media and cells were cultured in its presence for 30–120 min until ELISA analysis.

ELISA

The levels of intracellular tau in primary mouse astrocytes following incubation with tau-pff were analyzed with a total tau ELISA kit (57519, Cell Signaling Technology, Danvers, MA, USA). Experiment was performed according to the manufacturer’s protocol.

Quantitative real-time PCR (RT-qPCR)

The total RNA of astrocytes was extracted using the TRIzol (Invitrogen, cat# 15596026) and reverse transcription was conducted using the PrimeScript real time Master Mix (TAKARA, cat# RR047A). Total reaction volume of RT-qPCR systems is 20 μl containing 10 μL 2◊ SYBR Green Master Mix (AG11701, AGbio), 2 μL forward primer (2 mM), 2 μL reverse primer (2 mM), 5 μL RNase/DNase-free H2O, and 1 μL pre-amplified cDNA (100 ng/μl). Samples were assayed in a BIO-RAD CFX96 RT-qPCR system. Primer sequences used for RT-qPCR can be found in Supplementary Table 3.

Luciferase reporter assay

The HDAC7 (or vector) or shHD7 (or scramble), TFEB-3xFlag, pTFEB‐Luc reporter and Renilla luciferase (pRL‐TK) constructs were transfected into HEK293T cells by LipoFectMax. 24 h after transfection, the cells were harvested and lysed in 100 μL Passive Lysis Buffer. Luciferase activity was analyzed using a Lumat LB9507 luminometer (Berthold, Germany) and the Dual-luciferase Reporter Assay Systerm (Promega, Madison, WI, USA) according to the manufacturer’s protocol. Relative light units of TFEB luciferase were normalized to Renilla luciferase light units to control for transfection efficiency.

Statistical analysis

Statistical analyses were performed with Graphpad Prism 9 (LaJolla, CA, USA). Unpaired Student’s t-test was used to determine differences between two groups. One-way or two-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was used to determine differences among three or more groups as indicated in the figure legends. The statistically significance levels were set at p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) with a confidence interval of 95%. Data were expressed as mean ± SEM. All samples or animals were included for statistical analysis unless otherwise noted in pre-established criteria.