Animals

Sprague–Dawley (SD) rats (aged 16-18 weeks and weighed 250–280 g, male) were procured from Experimental Animal Center of Fujian Medical University. The rats had unrestricted access to both water and food and were maintained in a room (temperature: 22 ± 1 ℃; humidity: 50–60%) on a 12-h light/dark cycle.

Surgical procedure

As described previously, the CI/R model was constructed via transient middle cerebral artery occlusion (tMCAO) (Sun et al. 2018a, b). In brief, the animals underwent an intraperitoneal injection of pentobarbital sodium (40 mg/kg). Afterwards, the neck was disinfected with iodophor before making a 1.5 cm incision with surgical scissors to expose the left common carotid artery (CCA), internal carotid artery (ICA) and external carotid artery (ECA). Next, CCA and ECA were occluded using temporary atraumatic clips. A silicon-coated nylon monofilament (0.2 mm in diameter) (Doccol, Corporation, USA) was carefully inserted into the ICA (Insertion length:18-20 mm). The silicon-coated nylon monofilament was gently withdrawn 90 min after the tMCAO procedure to ascertain reperfusion.

Adeno-associated virus injection

As previously described, AAV–Empty, AAV-SIRT1 or AAV-sh_SIRT1 (at a dose of 1.04 × 1010gc; Hanbio, Shanghai, China) was respectively injected into the cortex and ipsilateral striatum of rats 21 days before tMCAO surgery (Zheng et al. 2018). In brief, a 10 ul volume micro-syringe (Hamilton, Reno, NV) was injected into the bregma origin of the coordinate system: mediolateral [−]2.5 mm, anterior-posterior 0.2 mm. The depth of insertion was 2.5 mm and 4.5 mm. The syringe remained in situ for 3 min after each injection and 15 min after the final injection to prevent any reflux after the injection.

Cultured primary rat cortical neurons and oxygen–glucose deprivation/reoxygenation (OGD/R)

As described previously, with minor modifications, the OGD/R model was established (Yi et al. 2019). Briefly, the cerebral cortex from SD rat embryos (aged 16–18 days) was dissected. The cerebral cortex was then placed in PBS on an ice plate after the removal of pia mater and blood vessels with ophthalmic forceps. Subsequently, the cerebral cortex was sliced into particles (1mm3 in size), gently blown, and resuspended with 2 mL of PBS. Next, the resuspended particles were transferred into a 15 mL centrifuge tube with 0.25% trypsin (Cat# 25200–072, Gibco, NY, USA) and placed in a cell incubator (37 °C) for digestion for 30 min. Then, the cortical neurons were isolated and centrifuged before the removal of the supernatant. The cortical neurons subsequently received a 10 mL neuronal medium (containing 96% Neurobasal, 1% 50 U/ml penicillin, 2% B27, 1% 0.5 mM glutamine) (Ca# 21103–049, Cat# 17504–044, Cat# 15140–122, Cat# 35050–061, respectively; all from Gibco, NY, USA). Scattered cells were cultured in coverslips (24 × 24 mm2) and plates. After a PBS wash, the cells were refreshed with the same medium and further cultured in an incubator with 5% CO2.

After 7 days of plating, the cortical neurons underwent 3 PBS washes and were subsequently exposed to a DMEM that did not contain glucose (Cat# 11966–025; Gibco, NY, USA) before being cultivated in an anaerobic chamber for 120 min. Afterwards, the cortical neurons were returned to their initial culture environment for 12 h.

Lentivirus transfection

As per the instructions from the manufacturer, the cortical neurons were transfected with 3 MOI of puromycin-resistant or GFP-contained LV-vector, LV-SIRT1 or LV-sh_SIRT1, respectively (Hanbio, Shanghai, China). Three days after the transfection, the cortical neurons underwent diverse treatments and were then subjected to distinct experimental procedures. Neurons transfected with puromycin-resistant lentiviruses were utilized to determine ROS levels and underwent flow cytometric analysis, whereas those transfected with GFP-containing lentiviruses were subjected to the other experimental assessments.

Chemical administration

3-(1H-1,2,3-triazol-4-yl) pyridine (3-TYP, HY-1083; MedChemExpress, USA) was dissolved in dimethyl sulfoxide (DMSO, less than 2%). The animals received an intraperitoneal injection of 3-TYP at a dosage of 50 mg/kg every other day for six consecutive days prior to the tMCAO surgery (Zhai et al. 2017). Neurons were treated with 3-TYP (5uM) for 4 h prior to the OGD/R exposure (C Wang et al. 2019; Wu et al. 2020).

Neurobehavioral tests

The modified neurological severity score (mNSS) was performed to evaluate the sensory, reflex, balance, and motor behaviors of the animals. A scale of 0 to 18 was employed to score the neurological function, where a score of 0 denoted no neurological impairment and a score of 18 represented the severest neurological deficit (N Liu et al. 2011).

As previously described, a rat rotarod 47700 tester was employed to assess the motor coordination of rats (Ugo Basile, Milan, Italy) (Shiotsuki et al. 2010). Before the experiment, the rats were trained to stay on the rotating rod as it increased progressively from 4 to 40 rpm within a span of 5 min. The trial test was repeated 10 times and the durations of rats staying on the rod were averaged and recorded.

As previously described, a YLS-13A grip strength tester was adopted to evaluate the rats’ forelimb grip strength (Yiyan Technology, Shang Dong, China) (Xiao et al. 2020). The rats were placed on a flat plate and allowed to grasp the sensing rod. They were pulled by the tail until their bodies were parallelly lifted off the plate. The experimenter pulled the rat slowly backwards until the animal released the sensing rod. The experiment was replicated 10 times and the grip strength was averaged and recorded.

Magnetic resonance imaging (MRI)

As previously described, rats underwent MRI at 72 h after CI/R injury (7-T, Bruker Medizintechnik, Germany) (Harada et al. 2007). Briefly, the animals underwent inhalational anesthesia via an exposure to a blend of 1.5–2% isoflurane and oxygen for 5 min. The respiratory status of the rats was assessed using a respiratory detection system, with the respiratory rate maintained at 25–30 breaths/min. T2-weighted images were acquired with the following sequence (TubroRARE): Slice thickness (ST) = 0.56 mm, Slices = 48, Field of view (FOV) = 35 × 35mm2, Time of repetition (TR) = 5200 ms, Matrix = 256 × 256, Time of echo (TE) = 32 ms, Total scan time = 11 min 5 s 600 ms. The images were analyzed using Image J software. The infarct volume was determined by the equation: The percentage of cerebral infarct volume = infarct area volume/brain volume × 100%.

Real-time quantitative PCR (RT-qPCR)

The RNA extraction was carried out with a Trizol Up Plus RNA Kit (Cat# ER501-01; TransGen, Beijing, China) to obtain total RNA from peri-ischemic cortex tissues. Then, mRNA was converted to cDNA with a 1st Strand cDNA Synthesis Kit (Cat# 11121ES60; YEASEN, Shanghai, China). Finally, RT-qPCR was conducted utilizing a Prism 7500 thermal cycler (Applied Biosystems, USA) and an RT-qPCR SYBR Green Kit (Cat# 11143ES50, YEASEN, Shanghai, China). The primer details are summarized in Table S1. The levels of mRNA were adjusted against GAPDH mRNA levels and determined using the 2−ΔΔCT formula.

Western blotting

As described previously, Western Blot was conducted (Karda et al. 2017; Xie et al. 2022). In brief, proteins (20–40 ug) were subjected to electrophoresis, electroporation, blocking, and incubation with the primary antibodies and the corresponding secondary antibodies (Table S2). The chemiluminescence system (Bio-Rad, CA, USA) was adopted to visualize the target proteins. Finally, the protein levels were analyzed with the ImageJ software and then standardized against Lamin B1, β-actin, or Tomm20 on an individual basis.

The isolation of mitochondria, cytoplasm, and nucleus

Mitochondria, cytoplasm, and nucleus were isolated using a Mitochondrial Fractionation Kit (Cat# 40015; Active Motif, Shanghai, China) following the protocol from the manufacturer. Briefly, the neurons were spun at 600 g for 5 min at 4 ℃. After the supernatant removal, the cell pellet was resuspended in 5 ml PBS and spun at 600 g for 5 min at 4 ℃ before another supernatant removal. Afterwards, the cell pellet was homogenized using 30–50 strokes with a homogenizer and spun at 800 g for 20 min at 4 ℃. The resulting pellet was the nucleus. Finally, the supernatant was spun at 10,000 g for 20 min at 4 ℃, producing the mitochondria and the cytosolic fraction. The protein concentration of each fraction was determined against a BSA standard curve by a Bio-Rad assay.

Measurement of mitochondrial bioenergetics

As per the instructions from the manufacturer, mitochondrial oxygen consumption rate (OCR) was determined with a Seahorse XF Cell Mito Stress Test Kit (Cat# 103,015–100, Seahorse Bioscience, Copenhagen, Denmark) (Paliwal et al. 2018). The cartridge was calibrated with purified water before OCR determination. Then, the mitochondria-containing plate was inserted into the analyzer and successively received oligomycin at 5 µmol/L, FCCP at 5 µmol/L, and rotenone/antimycin A at 0.5 µm to identify key assessment parameters.

Measurement of MDA, ATP content, SOD and GSH-Px

The level of malondialdehyde (MDA), ATP content, the activities of superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) were respectively detected with a Lipid Peroxidation MDA Assay Kit, an enhanced ATP assay kit, SOD Assay Kit (Cat# S0131S; Cat# S0027; Cat# S0103; all from Beyotime, Shanghai, China), GSH-Px Assay Kit (Cat# A005, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) as per the protocol from the manufacturer.

Immunofluorescent staining

As previously described, the immunofluorescence was performed with minor modification (Q Y Zhang et al. 2019). Briefly, the tissue sections were blocked before an overnight incubation with primary antibodies (Table S4). Next, they were subjected to a 2-h incubation with the appropriate secondary antibody (Table S4). DAPI (Cat# C3606, Beyotime, Shanghai, China) was employed to stain the cell nuclei for 15 min. Lastly, the tissue sections were visualized by confocal microscopy (LSM 750, Zeiss, Gottingen, Germany).

Detection of mitochondrial respiratory chain

As per the instructions from the manufacturer, the levels of mitochondrial complexes I − V in the cerebral mitochondria were detected with the biochemical analysis kits (Cat# BC0515; Cat# BC3235; Cat# BC3245; Cat# BC0945; Cat# BC1445; all from Solarbio, Beijing, China). The levels of mitochondrial complexes I–V were expressed in nmol/min of the cerebral mitochondrial protein.

Molecular dynamic (MD) simulation procedures

The protein structures of SIRT3 (https://www.uniprot.org/uniprotkb/ B2RZ31/entry) and SIRT1 (https://www.uniprot.org/Uniprotkb/A0A0G2JZ79/entry) were downloaded from Uniprot. The protein structures were modeled with AlphaFold2 software (Jumper et al. 2021), visualized, and depicted with Pymol software (Schrödinge, Inc.). The results were analyzed with LigPlot + software package (Laskowski and Swindells 2011).

MD simulations were performed with Gromacs 2019.6 using the amber14sb force field (Van Der Spoel et al. 2005). Subsequently, MD simulation was performed at normal temperature and pressure for 500 ns. The value of root-mean-square deviation (RMSD) represented the stability of backbone structures in the complex. Radius of gyration (Rg) was adopted to estimate the proximity of the architecture. The solvent accessible surface area (SASA) indicated the solvent-accessible surface areas of the compounds. The hydrogen bond number (HBnum) was employed to describe the hydrogen bond contact between two proteins. The algorithm was realized by Gmx_MMPBSA (Valdes-Tresanco et al. 2021).

Statistical analysis

Numeric statistics from animal and cell experiments were presented as the mean ± standard error of the mean (SEM) and processed with SPSS 20.0 software. Data normality was examined by the Shapiro–Wilk normality test and variance homogeneity was assessed by Levene's test. Data with equal variance were analyzed by one-way ANOVA with Bonferroni's post hoc test; those with unequal variance by one-way ANOVA with Dunnett's T3 post hoc test; and those collected from the same subjects at different time points by two-way repeated-measures ANOVA with Bonferroni post hoc test. The statistical significance was defined as follows: *p < 0.05, **p < 0.01, ***p < 0.001 as compared with the Sham or Control group; #p < 0.05, ##p < 0.01, ###p < 0.001 as compared with the tMCAO + AAV-Empty or OGD/R + LV-Vector group; $p < 0.05, $$p < 0.01, $$$p < 0.001 as compared with the tMCAO or OGD/R group; &p < 0.05, &&p < 0.01, &&&p < 0.001 as compared with the tMCAO + AAV-SIRT1 or OGD/R + LV-SIRT1 group.