Animals, Model Establishment, and Treatment

C57/BL6 male mice (18–22 g; Liaoning Changsheng Biotechnology Ltd., Liaoning, People’s Republic of China) were used in this experiment. The mice received normal chow and tap water, and were raised in accordance with the guidelines of the Care and Use of Laboratory Animals published by the China National Institute of Health and the ARRIVE guidelines. All experiments were approved by the Ethics Committee of the Harbin Medical University.

The mice were observed and weighed daily, and the model was established when the weight reached 20–25 g. Cerebral ischemia was induced by permanent middle cerebral artery ligation (pMCAL) as previously described [42]. The mice were first anesthetized with 2% nembutal (5 mL/kg). An incision was made from the left temporal muscle with a spring scissor, and then bone rongeurs were used to remove a piece of the skull to slightly expose the middle cerebral artery. Finally, the distal part of the artery was ligated with a vessel cauterizer to establish the cerebral ischemia model. In the sham-operated group, the vessels were bluntly dissected but without ligation.

After modeling, the sham-operated and cerebral ischemia model mice were divided into two groups: one group was administered epalrestat by gavage twice a day at 50 mg/kg (SML0527, Sigma Chemical Company, St. Louis, Missouri), which dissolved in control saline, and the other was administered equal amounts of saline (NaCl) as the control group. In addition, in order to judge the effect of epalrestat dose on the experimental results, we set three concentrations of 25 mg/kg, 50 mg/kg and 100 mg/kg for experimental verification.

Cranial Ischemia Model Confirmation

The mice were euthanized with nembutal anesthesia 12 h, 1 d, 3 d, and 5 d after model establishment. After intracardiac perfusion with phosphate-buffered saline (PBS), the mouse brain tissue was carefully removed. Brain slices (2 mm) were cut along the coronal suture with a brain slicer and incubated in 2% 2,3,5-triphenyltetrazolium chloride (TTC) solution (in PBS) for 30 min at room temperature. Areas without red staining by the TTC reagent were assumed to be injured. All sections were photographed with a digital camera and the infarct areas (white) were measured blindly using Image J software.

Assessment of BBB Damage

Evans blue is widely used to assess the integrity of the BBB. After modeling, the mice were intravenously injected with 4% Evans blue dye (E2129, Sigma-Aldrich) at 100 μL/20 g via the tail vein, and the body quickly turned blue. The vital signs were carefully monitored for 30 min, and then the mice were sacrificed and transcardially perfused with cold PBS to flush away the blood and Evans Blue from the blood vessels. The ischemic lateral brain tissue was quickly removed and placed in a solution of 50% trichloroacetic acid in PBS (1.5 mL). The brains were homogenized, sonicated in trichloroacetic acid, and centrifuged at 12,000 rpm for 20 min. The supernatant was collected, and the fluorescence emission was measured at 680 nm (with an excitation wavelength of 620 nm) using an ultraviolet spectrophotometer.

Neurological Score

After model establishment, the mice were placed in a cage alone to observe their overall state and behavior. The neurological score [43] was calculated as follows: 0 points, normal behavior; 1 point, cannot fully extend the right front legs; 2 points, turning around in a circle; 3 points, falling onto the right side; 4 points, cannot move on own and loss of consciousness; 5 points, death.

The effects of cerebral ischemia and epalrestat treatment on sensorimotor abilities were evaluated based on forelimb use, since animals with unilateral ischemic brain injury show forelimb preference. The forelimb movements of each mouse were analyzed in a transparent plexiglass rearing cylinder 9 cm in diameter and 15 cm in height. The size of the cylinder is sufficient to allow for free movement, and its weight prevents it from moving during support [44]. The mice were trained three times a day for approximately 5 min each for the 3 d prior to the trial. After modeling, each mouse was individually placed in the cylinder and observed for 5 min. To evaluate the sensorimotor deficit, the percentage of weight-bearing episodes (braces) on the side of the cylinder that were initiated with the non-impaired (ipsilateral), impaired (contralateral), and both forepaws were calculated in each animal after pMCAL. The initial forepaw placement of each weight-bearing contact with the wall was recorded as right or left. In this experiment, the damage was induced on the left side, and therefore we compared the times of using the right paw to the total number of times of weight-bearing contact with a forelimb.

Immunofluorescence and TUNEL Staining

The brains were collected from the mice in the different groups and fixed to obtain 10-μm-thick cryosections. The cryosections were washed with 0.01 mol/L PBS (pH = 7.4) and blocked with 1% bovine serum albumin (BSA) for 1 h. The cryosections were then incubated with anti-CD31 (ab7388, Abcam, 1:500), anti-cleaved-caspase3 (ab214430, Abcam, 1:500), anti-occludin (ab216327, Abcam, 1:500), anti-F4/80 (30325, Cell Signaling Technology, 1:800), anti-LC3B (L7543, Sigma-Aldrich, 1:500), anti-P62 (ab109012, Abcam, 1:500), anti-LAMP1 (ab225762, Abcam, 1:200), anti-GFAP (ab53554, Abcam, 1:500), anti-AR (sc-166918, Santa Cruz, 1:200), and anti-ZO-1 (ZO1-1A12, Thermo Scientific, 1:200) antibodies overnight at 4 °C. The sections were incubated with secondary antibodies, including fluorescein isothiocyanate (FITC)-conjugated donkey-anti-goat/rat IgG and TRITC-conjugated donkey anti-rabbit IgG (Jackson, 1:500) at room temperature for 1 h. The sections were washed with flowing water three times for 5 min and incubated for 5 min with 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI; 28718–90-3, Sigma-Aldrich) to stain the nuclei. Apoptotic cells were detected using an in situ cell death (TUNEL) detection kit (Roche). After staining with DAPI, the samples were treated with 0.3% H2O2 for 10 min to remove endogenous peroxides at room temperature and then rinsed four times with 0.01 mol/L PBS (5 min each time). The sections were then incubated with the TUNEL reaction mixture (solution A/B, 1:9) in a humidified chamber at 37 °C for 1 h under dark conditions. The sections were washed three times with 0.01 mol/L PBS and placed in a dark chamber at 37 °C for 30 min. Subsequently, the specimens were washed four times with 0.01 mol/L PBS (5 min each time). Finally, the sections were mounted and observed under a confocal microscope (Zeiss, Germany). A negative control was established for each group to which no primary antibody was added. The rest of the steps were consistent; however, no positive fluorescent areas were observed.

Flow Cytometry

Brain hemispheres were collected from the mice in different groups after intracranial perfusion with D-Hanks’ balanced salt solution after modeling for flow cytometric analysis. A 70–30% Percoll gradient centrifugation protocol was used for flow cytometry analysis of neutrophils marked with CD45-Percp (557235, BD Biosciences, 1:100), CD11b-APC (553312, BD Biosciences, 1:100), and Ly6G-FITC (551460, BD Biosciences, 1:100) infiltrating the brain with a FACS Calibur flow cytometer (BD Biosciences). All antibodies and isotype-matched controls were purchased from BD Biosciences (San Diego, CA, USA), and the analysis was conducted using FlowJo software.

The flow cytometry procedure for detecting macrophages and neutrophils in peripheral blood circulation was different from that described above. Blood was collected from the eyeballs of the mice after modeling, and the blood was placed in an anticoagulant tube. After adding red blood cell lysate, the blood was gently vortexed and shaken. Then the cell mixture was filtered, and the cells were transferred to a 96-well plate. After breaking the membrane, antibodies were added for incubation. Related flow cytometric antibodies were: CD45-Percp (557235, BD Biosciences, 1:100), CD11b-APC (553312, BD Biosciences, 1:100), and Ly6G-PE (127608, Biolegend, 1:100)、F4/80-FITC (123107, Biolegend, 1:100), Fixable Viability Dye eFluor™ 780 (65–0865-18, eBiosciences, 1:2000).

Cell culture and Treatment

Mouse brain-derived endothelial cells (bEnd.3) were obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 mg/mL), and maintained at 37 °C in a 5% CO2 incubator. When the cells state was better, epalrestat was added at 30 μm/L, untreated cells served as the control group. Additionally, bicalutamide (an AKT inhibitor) and rapamycin (an mTOR inhibitor) were added to the culture medium along with epalrestat to analyze the effect of epalrestat on the AR/AKT/mTOR signaling pathway. The cells were passaged about ten times.

In Vitro Ischemia Model

OGD is widely used as an in vitro model of ischemic stroke. The cells were placed in a modular incubator after the medium was replaced with glucose-free DMEM. The chamber was then filled with a mixture of 95% N2 and 5% CO2, and the temperature of the chamber was set to 37 °C. The cells were exposed to the OGD condition for 2 h, 4 h, 6 h, and 8 h.

Cell Viability Assay

To test the effects of epalrestat on endothelial cell activity, the bEnd.3 cell viability was determined using a Cell Counting Kit-8 (CCK-8) assay (Dojindo, Japan). The cells(1 × 104/well) were seeded in 96-well culture plates and incubated overnight in DMEM containing 10% FBS. Epalrestat was added into cells at 0, 10, 30, 50 μmol/L, followed by exposure to OGD after 24 h. The CCK-8 solution (10 μL) was added to each well of the plate and the plates were incubated for 4 h. The absorbance at 450 nm was measured using an ultraviolet spectrophotometer. Furthermore, to test the effects of bicalutamide and rapamycin on endothelial cell activity, bicalutamide was added to cells at 0,0.5,1.0,1.5 μmol/L, and rapamycin at 0, 0.1, 0.3, 0.5, 1.0 μmol/L. The rest of the steps were the same as above, and the results was in Supplementary Materials.

Permeability Assays

Permeability assays were conducted as previously described [45]. The bEnd.3 cells (1 × 104) were plated on top of a 3-μm pore-size upper Transwell chamber (Corning, USA) with epalrestat for 1 day and then the medium was replaced with glucose-free DMEM containing 50 μg/mL of BSA-FITC (A9771, Sigma) to the Transwell upper chambers with exposure to the OGD condition for 4 h. Media (both 100 μL) were collected from the upper and lower chambers of each well, and the fluorescence intensity was measured using an FL600 microplate fluorescent reader (Biotek). Cells not exposed to OGD were cultured in a cell incubator (5% CO2) in glucose-free DMEM as a control group for comparison.

Protein Preparation and Western Blot

Protein samples were extracted from the cultured bEnd.3 cells in vitro and from the ischemic brains of the mice. RIPA lysis buffer (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added to the cell/tissue homogenate, and the lysed proteins were centrifuged at 12,000 × g for 15 min at 4 °C. The supernatants were collected for protein concentration measurements using a BCA protein assay kit (Pierce, Rockford, IL, USA). The protein samples were loaded on 10% Tris–HCl sodium dodecyl sulfate–polyacrylamide gels (Bio-Rad Laboratories, Hercules, CA, USA) for electrophoresis (120 V, 60 min) and then transferred onto a polyvinylidene fluoride membrane for blocking (block solution, 5% non-fat milk solution dissolved in Tris-buffered saline with Tween). The membranes were incubated with the following primary antibodies overnight at 4 °C: anti-β-actin (TA-09, ZSGB-BIO, 1:1000), anti-AR (sc-166918, Santa Cruz Biotechnology, 1:500), anti-occludin (ab216327, Abcam, 1:1000), anti-cleaved-caspase3 (ab214430, Abcam, 1:1000), anti-Bcl2 (sc-783, ZSGB-BIO, 1:1000), anti-LC3B (L7543, Sigma-Aldrich, 1:1000), anti-Beclin1 (ab207612, Abcam, 1:1000), anti-ZO-1 (ZO1-1A12, Thermo Scientific, 1:1000), anti-Bax (sc-7480, Proteintech, 1:1000), anti-AKT (ab38449, Abcam, 1:1000), anti-p-AKT (4060, Cell Signaling Technology, 1:1000), anti-mTOR (ab109268, Abcam, 1:1000), and anti-p-mTOR (5536, Cell Signaling Technology, 1:1000). The membranes were then incubated with goat anti-rabbit or anti-mouse secondary antibodies (Santa Cruz Biotechnology). The protein levels were normalized to the level of β-actin, and immunoreactivity signals were detected relative to the corresponding control.

Cell Apoptosis Assay

Cell apoptosis was analyzed using the Annexin V-FITC/PI Apoptosis Detection Kit (Vazyme Biotech Co., Ltd., Nanjing, China) with a FACS Calibur flow cytometer (BD). When the cells reached the wall of the bottle, epalrestat was added according to the experimental conditions, followed by OGD treatment. After OGD, the cells were counted and seeded in 96-well plates at a density of 1 × 106 cells per well. The supernatants were then centrifuged, discarded, and the antibodies were configured according to kit standards using buffer buffer, 50 μl of antibodies were added to each well, the cells were resuspended, and incubated at room temperature for 20 min, finally, buffer was added to stop antibody incubation.

Statistical Analysis

All quantitative data are expressed as the mean ± standard deviation. Statistical analysis was performed using GraphPad Prism software (version 6.0). Unpaired Student’s t-tests were used to compare data between two groups. One-way or two-way analysis of variance with a Tukey post-hoc test was used for comparisons among three or more groups. Differences were considered statistically significant at P < 0.05.