Animals and treatment

Eight-week-old male C57BL/6 mice with an average body weight (BW) of 20 ± 2 g were provided by the Experimental Animal Centre of Nantong University. The mice were allowed to acclimatize to the surrounding environment for 3 d before the experiments began. The mice were maintained in a temperature-controlled room at 26 ± 1 °C with a 12 h light–dark cycle and free access to a standard diet and water. During the HH treatment, the mice were exposed to a simulated altitude of 7600 m (25,000 ft, 282 mmHg) at a 300 m/min velocity in a hypobaric chamber for 24 h. Following this, the mice descended to sea level at the same speed. The mice were immediately anaesthetized and perfused with physiological saline solution (0.9%) to remove the blood. To inhibit caveolae-mediated endocytosis, the mice were injected via the tail vein with MβCD (300 mg/kg) 24 h before HH exposure. After the HH treatment, mice were reinjected with MβCD and immediately anaesthetized and perfused approximately 1 h after circulation.

Brain water content assay

Brain water content was measured by the wet‒dry weight method [40]. Brain tissue of the mice was collected after the HH treatment and weighed as wet weight. The brain tissue was fully dried in a drying oven at a constant temperature of 100 °C and weighed several times to obtain a stable dry weight (average weighing error < 0.002 g). The percentage of brain water content was calculated as follows:

$$}\;}\;}\; = \left( }\;}\; - }\;}} \right)/}\;}\; \times \;100\%$$

(1)

BBB permeability assay

Evans blue (EB) staining was performed to evaluate brain vascular leakage. After HH treatment, mice were injected via the tail vein with 4 mL/kg of 2% EB in saline. Approximately 1 h after circulation, the mice were anaesthetized and perfused with physiological saline solution (0.9%) to remove the intravascular dye. Half of the brain was immediately weighed, homogenized in 1 mL of 50% trichloroacetic acid solution, and then centrifuged at 15,000 × g for 30 min. The supernatant was then diluted 1:3 with ethanol, and its absorbance was measured at 630 nm using a microplate reader (Synergy 2™, Bio-Tek, US). The other half of the brain was fixed and sectioned for confocal imaging with a 647 nm laser by a Leica SP8. The mice were also intravenously injected with 40 kD FITC-dextran (D1845, Thermo; 4 mL/kg of 10 mg/mL in saline), which circulated for 15 min. At the end of the circulation period, the mice were promptly euthanized and perfused with physiological saline solution (0.9%) via the left ventricle to remove the blood.

Cell culture and treatments

Mouse brain microvascular endothelial cells (bEnd.3, Bioleaf Biotech) were maintained in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, US) supplemented with 10% foetal bovine serum (FBS, Sigma, US) at 37 °C in a 5% CO2 incubator. Human umbilical vein endothelial cells (HUVECs) were purchased from ScienCell Research Laboratories. The cells were cultured in ECM (1001, ScienCell, US) supplemented with FBS, endothelial cell growth supplement (ECGS), and antibiotic solution. After forming a confluent monolayer, the cells were exposed to hypoxia (94% N2, 5% CO2, and 1% O2) for 24 h. BEnd.3 cells or HUVECs were then incubated with MβCD (5 mM) for 1 h after hypoxia treatment. For the autophagy analysis, a monolayer of cells was incubated with a medium containing 10% FBS, 10 mmol/L 3-methyladenine (3-MA), and 50 nmol/L rapamycin (Rapa) for 2 h. The cells were subsequently exposed to hypoxia and cultured under consistent concentrations of 3-MA and Rapa. [41]

Transendothelial electrical resistance (TEER) and transcytosis assay

BEnd.3 cells were seeded on 12-transwell inserts (3401, Corning) and cultured to 90% confluence. For the TEER assay, cells were incubated with Hank’s balanced salt solution (HBSS) for 30 min. The total resistance of cells was measured by a Millicell ERS-2 Epithelial Volt-ohm Meter (Merck Millipore). The resistance of HBSS was recorded as the blank resistance. TEER was calculated as follows:

$$}\;\left( }^} } \right)\; = \;\left( }\;} - }\;}} \right)\left( \Omega \right) \times }\;}\left( }^} } \right)$$

(2)

Lower resistance indicates damage to the tight junction. For the transcytosis assay, 100 μg/mL 40 kD FITC-dextran (D1845, Thermo) was added to the inserts for 30 min. The medium in the lower plate was then collected and detected by a microplate reader (excitation wavelength 490 nm; emission wavelength 520 nm). The values were converted to FITC-dextran concentrations based on the standard curve.

Endocytosis assay

After the cells were cultured to a confluence over 90%, the monolayer of either bEnd.3 cells or HUVECs was incubated with 100 μg/mL 40 kD FITC-dextran after hypoxia exposure and then cultured at 37 °C in a 5% CO2 incubator for 30 min. The cells were immediately chased with fresh medium for confocal imaging at 488 nm by a Leica SP8.

Western blot

Cells were lysed in RIPA buffer, and the protein concentration was calculated by bicinchoninic acid assay (BCA). Proteins were isolated through SDS‒PAGE and transferred to PVDF membranes. The membranes were blocked with 5% nonfat dry milk and then incubated overnight at 4 °C with primary antibodies, including anti-NRF1 (CST, 46743S), anti-CAV-1 (CST, 3267), anti-VE-cadherin (Santa Cruz, sc-9989), anti-occludin (Proteintech, 13,409-1-AP), anti-claudin-5 (Thermo, 35-2500), and anti-β-actin (Sigma, A5316). The binding of primary antibodies was visualized with a goat anti-rabbit HRP conjugated secondary antibody (Jason, 115-035-033) or goat anti-mouse HRP conjugated secondary antibody (Jason, 111-035-003).

Transfection siRNA

Small interfering RNAs (siRNAs) targeting NRF1 (siNRF1) and CAV1 (siCAV1) were synthesized by Suzhou GenePharma Co., Ltd. (Suzhou, China). For transfection, bEnd.3 cells or HUVECs were incubated and cocultured in 12-well culture dishes at 50–60% confluence without antibiotics. siRNA was transfected into bEnd.3 cells or HUVECs using Lipofectamine 2000 (Invitrogen) following the manufacturer’s instructions. After 48 h of transfection, the cells were exposed to hypoxia for 24 h and harvested. Then, the knockdown efficiency was evaluated by Western blotting.

Transmission electron microscope (TEM)

At the end of the HH treatments, the mouse brains were quickly removed, placed into precooled fixative solution (4% glutaraldehyde; Sigma), and incubated at 4 °C overnight. Samples were dehydrated in a graded ethanol series and embedded in plastic. The sections were then cut at 70–90 nm and stained with 4% uranyl acetate and 0.3% lead citrate. Representative areas from the sections were viewed with HT7700 transmission electron microscope (HITACHI, Japan).

Real-time PCR

Total RNA was isolated from cells using TRIzol (Invitrogen), and cDNA was generated from 1 μg RNA using the HiScript ®III RT SuperMix kit (Vazyme, R323-01) according to the manufacturer’s instructions. Real-time PCR was performed using the SYBR qPCR Master Mix kit (Roche, USA) at 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. All primers used for real-time PCR were designed as follows: Nrf1 forward: 5′ TAT GGC GGA AGT AAT GAA AGA CG 3′, reverse: 5′ CAA CGT AAG CTC TGC CTT GTT 3′; Cav1 forward: 5′ AGC AAA AGT TGT AGC GCC AG 3′, reverse: 5′ GAC CAC GTC GTC GTT GAG AT 3′; Actb forward: 5′ CAT CCG TAA AGA CCT CTA TGC CAA C 3′, reverse: 5′ ATG GAG CCA CCG ATC CAC A 3′. Relative gene expression was calculated using ΔΔCt for normalization to the reference gene.

Immunofluorescence (IF)

Tissue sections (40 μm) and cultured cells were fixed with 4% paraformaldehyde and permeabilized with 0.3% Triton X-100. The samples were then blocked with 10% donkey serum and probed with anti-CD31 (RD, AF3628), anti-Laminin (Abcam, ab11575), anti-VE-cadherin (Santa Cruz, sc-9989), anti-occludin (Proteintech, 66,378-1-Ig), anti-claudin 5 (Thermo, 35-2500), anti-NRF1 (CST, 46743S), anti-CAV-1 (CST, 3267), anti-ZO-1 (Proteintech, 21,773-1-AP) or anti-LC3 (CST, 12741S) antibodies. The binding of primary antibodies was visualized with either Alexa Fluor 555-conjugated donkey anti-rabbit IgG (Thermo, A31572), Alexa Fluor 647-conjugated donkey anti-mouse IgG (Thermo, A32787), or Alexa Fluor 488-conjugated donkey anti-goat IgG (Abcam, ab150133). The samples were then counterstained with DAPI (Thermo) and imaged using a Leica SP8 confocal microscope.

Statistical analysis

All data were analysed using GraphPad Prism software version 8.0 (GraphPad Software, San Diego, CA). All sets of continuous data were tested for normality using the Shapiro‒Wilk test, and fewer than 5% of the tests concluded that the set was nonnormal at the 0.05 significance level, confirming that the datasets met the assumption of a normal distribution. Two variants were compared by t test (two-tailed) for independent samples for the results of the NN vs. HH comparison, and comparisons between several groups were performed using ordinary one-way ANOVA. Then, two-way ANOVA followed by Tukey’s multiple comparisons test was used when more than 1 variable was compared. The data are presented as the mean values ± SEMs for in vivo experiments or mean values ± S.D.s for in vitro experiments. Statistical significance was set at *P < 0.05, **P < 0.01, and ***P < 0.001, and n.s indicates no significance.