Treatment with β-sitosterol ameliorates the effects of cerebral ischemia/reperfusion injury by suppressing cholesterol overload, endoplasmic reticulum stress, and apoptosis

Abstract

β-Sitosterol is a type of phytosterol that occurs naturally in plants. Previous studies have shown that it has anti-oxidant, anti-hyperlipidemic, anti-inflammatory, immunomodulatory, and anti-tumor effects, but it is unknown whether β-sitosterol treatment reduces the effects of ischemic stroke. Here we found that, in a mouse model of ischemic stroke induced by middle cerebral artery occlusion, β-sitosterol reduced the volume of cerebral infarction and brain edema, reduced neuronal apoptosis in brain tissue, and alleviated neurological dysfunction; moreover, β-sitosterol increased the activity of oxygen- and glucose-deprived cerebral cortex neurons and reduced apoptosis. Further investigation showed that the neuroprotective effects of β-sitosterol may be related to inhibition of endoplasmic reticulum stress caused by intracellular cholesterol accumulation after ischemic stroke. In addition, β-sitosterol showed high affinity for NPC1L1, a key transporter of cholesterol, and antagonized its activity. In conclusion, β-sitosterol may help treat ischemic stroke by inhibiting neuronal intracellular cholesterol overload/endoplasmic reticulum stress/apoptosis signaling pathways.

Introduction

Thrombolytic drugs such as alteplase and urokinase are currently considered to be the standard strategy for treating ischemic stroke, but their clinical application is significantly limited owing to the short treatment time window and the risk of bleeding (Hankey, 2014; Wardlaw et al., 2014; Tao et al., 2020). Therefore, in-depth exploration of the molecular mechanism of cerebral ischemia/reperfusion injury and the search for potential therapeutic targets are key research focuses. The pathophysiology of ischemic stroke is a dynamic/complex process that is influenced by many intracellular and extracellular physicochemical factors (Paul and Candelario-Jalil, 2021). Studies have shown that neuronal apoptosis is an important form of cell death that occurs after ischemia/reperfusion and plays an important role in ischemic brain injury (Xia et al., 2021; Tuo et al., 2022; Yan et al., 2022; Qiu et al., 2023; Tian et al., 2023). Many genes and proteins regulating neuronal cell viability are involved in both promoting and inhibiting this apoptotic process, including Immediate early genes (IEGs), the B-cell lymphoma-2 (Bcl-2) gene family, and the cysteine protein kinase family (Uzdensky, 2019; Chen et al., 2020; Liu et al., 2021). Therefore, combatting neuronal apoptosis after reperfusion is an important therapeutic goal for patients undergoing ischemic stroke recanalization (Datta et al., 2020; Lai et al., 2020).

We previously found that β-sitosterol from Pinellia ternata (the main ingredient of Erchen decoction) is highly absorbed by the brain, where it may have a therapeutic effect (Tang et al., 2018). β-Sitosterol is a type of phytosterol that is similar in structure to cholesterol and is widely produced by plants. Many studies have shown that β-sitosterol has strong antioxidant, anti-hyperlipidemia, anti-inflammation, and anti-tumor effects in the central nervous system (Yang et al., 2013; Sharma et al., 2021). Another study showed that β-sitosterol can enhance the activity of endothelial cells subjected to hypoxic injury by reducing antioxidant stress (Zhang et al., 2000). Therefore, we hypothesized that β-sitosterol has neuroprotective effects. To verify this hypothesis, we used a mouse model of ischemic stroke induced by middle cerebral artery occlusion (MCAO), an oxygen-glucose deprivation cell model, and molecular docking analysis to explore the mechanisms by which β-sitosterol reduces the effects of stroke.

Methods

Animals

C57BL/6J mice were purchased from the Experimental Animal Center of the Fourth Military Medical University (license No. SCXK (Shaan) 2021-001). All animals were housed with free access to a regular pellet diet and appropriate water in a temperature-maintained room (21–25°C) under a 12/12-hour light/dark cycle. This study was approved by the Laboratory Animal Welfare and Ethics Committee of the Fourth Military Medical University (approval No. 20201782) on October 10, 2020). All experiments were designed and reported according to the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines (Percie du Sert et al., 2020).

After 1 week of acclimation to the housing conditions, 142 mice (23–26 g, 6–8-weeks old) were randomly allocated to six groups: sham operation (sham group, n = 38), sham + β-sitosterol high-dose group (50 mg/kg, n = 5), MCAO (ischemia/reperfusion (I/R) group, n = 31), MCAO + β-sitosterol low-dose group (2 mg/kg, n = 16), MCAO + β-sitosterol medium-dose group (10 mg/kg, n = 16), and MCAO + β-sitosterol high-dose group (50 mg/kg, n = 36). The animals in the MCAO + β-sitosterol groups were injected intraperitoneally with β-sitosterol (Cat# Y18D8C50778, Shanghai Yuanye Bio-Technology Co., Ltd, Shanghai, China) 30 minutes after reperfusion. To show that β-sitosterol can cross the blood-brain barrier in mice, we injected the sham group of mice with β-sitosterol.

MCAO model

To test the neuroprotective effects of β-sitosterol, we established the in vivo MCAO mouse model (Wang et al., 2020). Briefly, the mice were anesthetized with 2% sodium pentobarbital (0.2 mL/25 mg, Sigma, St. Louis, MO, USA) and fixed on the operating table in a supine position. A V-shaped incision was made at the base of the left external carotid artery using ophthalmic scissors. A 15-mm-long 8-0 nylon cord that had been pre-treated by soaking in heparin was carefully inserted through the external carotid artery to cause ischemia by blocking the opening of the middle cerebral artery. After 1.5 hours of ischemia, the suture was withdrawn from the external carotid artery to the common carotid artery to induce reperfusion, which was allowed to continue for 22.5 hours. During the operation, a laser Doppler scanner (Moor Instruments, Devon, UK) was used to monitor middle cerebral artery blood flow, and a decrease in blood flow by 70–80% compared with before the operation was considered a sign of successful establishment of the MCAO model (Chiang et al., 2011; Wu et al., 2019). For the sham operation group, the same procedure was performed, but no cord was inserted into the external carotid artery.

Laser Doppler analysis of blood flow

After mice were anesthetized by intraperitoneal injection of 2% sodium pentobarbital (0.2 mL/25 mg), a 1-cm incision was made in the skin on the top of the head at the sagittal position, the subcutaneous fascia was separated with cotton swabs, and the exposed fascia was gently rubbed until the fontanelle and skull were exposed. Then, the whole head was scanned by laser Doppler scanner with the following parameters: bandwidth, 250 Hz to 15 kHz; scan speed, 4 ms/pixel.

Neurological deficit assessment

The degree of neurological deficit was evaluated at 24 hours after MCAO using a modified neurological deficit severity test (Li et al., 2000). Each animal was graded on a scale of 0 to 14, where a score of 0 reflected normal function; 1 to 4 mild injury; 5 to 9 moderate injury and 10 to 14 severe injury.

Gas chromatography-mass spectrometry

The content of β-sitosterol in the cerebral cortex of experimental mice was detected by a gas chromatography-mass spectrometer (GC/MS) equipped with an autosampler system (Shimadzu, Kyoto, Japan) at 24 hours after MCAO. The carrier gas was 99.99% high-purity helium with a flow rate of 1.2 mL/min. The injection and interface temperatures were set at 300°C and 280°C, respectively. The selective ion monitoring function was set for quantitation with a dwell time of 300 ms/ion.

Cerebral infarction volume analysis

After 24 hours of reperfusion, the brains of mice sacrificed with 10% sodium pentobarbital (0.2 mL/25 mg) were removed and frozen at –20°C for 20 minutes. Five coronal sections were placed in 2% triphenyltetrazolium chloride dye solution, incubated at 37°C in water in the dark for 15 minutes, fixed in 4% paraformaldehyde overnight, and then photographed (Zuo et al., 2019). The volume of cerebral infarction was analyzed using ImageJ 1.x software. The percentage of cerebral infarction volume (%) was calculated as follows: (sum of the ischemic zone areas in each section/sum of each slice area in each brain) × 100.

Measurement of cerebral water content

After MCAO and reperfusion, the mice were anesthetized with an excessive dose of 10% sodium pentobarbital (0.2 mL/25 mg) and decapitated, and the brains were removed. The olfactory bulb cerebellum and the lower brainstem were removed, and the left and right cerebral hemispheres were separated. The wet weight was determined immediately, and then the brains were dried in a 110°C oven (Taisite, Tianjin, China) for 24 hours until they reached a constant weight, which was defined as the dry weight of the brain tissue. The cerebral water content (%) was calculated as follows: (wet weight – dry weight)/wet weight × 100.

Nissl staining

After 24 hours of reperfusion, the mouse brains were removed for Nissl staining. Hippocampal sections exhibiting intact morphology were selected and immersed in chloroform for 1 minute, and then washed with phosphate-buffered saline (PBS) three times, for 5 minutes each time. Next, the slices were defatted with 75% alcohol for 4 hours and then rinsed with distilled water for 2 minutes. Subsequently, Nissl staining solution (Beyotime, Shanghai, China) was added to the slices, and the staining time was monitored under a microscope (Nikon, Tokyo, Japan). When the Nissl bodies could be clearly seen, the staining reaction was stopped by rinsing with distilled water twice. After being allowed to dry completely, the slices were treated with alcohol for 1 minute, then with xylene to make the rest of the tissue transparent, and finally sealed with neutral gum and observed under a high-powered bright-field microscope (Nikon). The number of Nissl bodies in the hippocampal CA3 area was determined for five samples from each group.

TdT-mediated dUTP nick-end labeling

After MCAO and reperfusion, whole brains were removed from over-anesthetized mice. The brains were embedded in paraffin and cut into sections, which were then dewaxed, hydrated, and stained using a TdT-mediated dUTP nick-end labeling (TUNEL) kit (Beyotime) according to the manufacturer’s instructions. Six fields were randomly selected for counting apoptotic cells and photographing under a fluorescent microscope (Nikon).

Determination of cholesterol levels

Mouse brains were obtained as described above, and brain cholesterol content was determined using a cholesterol assay kit (ab65359, Abcam, Cambridge, UK) according to the manufacturer’s instructions.

Primary cortical neuron culture

Cortical neurons were prepared from C57BL/6J mice (n = 30) on days 14–15 of gestation (the pregnant mouse sacrificed with 10% sodium pentobarbital). The meninges and vascular membranes of the fetal mice were stripped under an anatomical microscope (Zhongxian, Chongqing, China), and the cerebral cortex tissues were cut into 1- to 2-mm³ pieces, transferred to 0.25% trypsin solution (Sangon Biotech, Shanghai, China), and digested at 37°C for 20 minutes to generate a single-cell suspension. Digestion was terminated by adding Dulbecco’s modified eagle medium (Gibco, Beijing, China) containing 20% fetal bovine serum (Gibco, Gaithersburg, MD, USA). The cells were quantified using a cell counting plate and a microscope (Thermo Fisher Scientific, Waltham, MA, USA). The cells were then inoculated in a cell culture plate precoated with polylysine (25 μg/mL) and cultured in a CO₂ incubator for 24 hours with Neurobasal medium (Gibco, Gaithersburg, MD, USA). Every 72 hours the cell growth medium was changed. Experiment were carried out after 7–14 days of culture.

Oxygen-glucose deprivation cell model

To create an in vitro oxygen and glucose deprivation model, the cell growth medium was removed, and the cells were washed three times with PBS. Glucose-free Dulbecco’s modified eagle medium was added, and the cells were incubated in a 5% CO₂/95% N₂ atmosphere at 37°C for an appropriate amount of time. The glucose-free medium was then removed, and the cells were washed three times with PBS. Next, to mimic reperfusion, normal growth medium was added, and the cells were placed in a 37°C incubator (Thermo Fisher Scientific) with saturated humidity and 5% CO₂ for 24 hours before use in subsequent experiments (Deng et al., 2020).

Lactate dehydrogenase release detection

After the oxygen-glucose deprivation (OGD) cell model was treated with β-sitosterol, lactate dehydrogenase (LDH) release was measured using a kit according to the manufacturer’s instructions. The optical density (OD) value of each well was measured at 450 nm using a microplate reader (TECAN, Shanghai, China), and the value was used to calculate LDH release.

Cell counting kit-8 assay

Primary neurons (100 μL/well) were inoculated into 96-well plates at a concentration of approximately 5000 cells per well. After a suitable pre-culture period, the OGD model was established, and then the cells were incubated in normal medium containing different concentrations of β-sitosterol (0.1, 1.0, and 10 μM) for 24 hours. Then, 10 μL of reagent from a Cell counting kit-8 (CCK-8; Beyotime) was added to each well. After incubation for 4 hours, the absorbance at 490 nm was measured using a microplate reader. Cell viability (%) was calculated as follows: [OD (experiment) – OD (blank)]/[OD (control) – OD (blank)] × 100.

Flow cytometry assay

The neurons were rinsed with pre-cooled PBS three times, then gently scraped off the cell culture plate, added to a centrifuge tube, and centrifuged at 1000 × g for 5 minutes. Annexin V-FITC (5 μL) and propidium iodide (PI) (10 μL) (Beyotime) were added to the cells after re-suspension. The solution was mixed thoroughly, and the cells were incubated in the dark for 15 minutes. The rate of cell apoptosis was detected by flow cytometry (ACEA, San Diego, CA, USA).

Assessment of the effects of cholesterol on neurons

Normal primary neurons growing in multi-well plates were washed three times with PBS. The neurons were then treated with different doses (3, 9, and 27 μM) of cholesterol (Shanghai Yuanye Bio-Technology Co., Ltd.), along with 1 μM β-sitosterol, followed by flow cytometry analysis.

Prediction of β-sitosterol targets

The Swiss Target Prediction system (http://swisstargetprediction.ch/) is an online database capable of predicting possible target proteins of small molecule compounds. It was used to predict the most probable protein targets of β-sitosterol.

Western blot assay

The protein expression levels of glucose-regulated protein 78 (GRP78)/Bip, Caspase 12, Caspase 9, Caspase 3, Bax, Bcl-2, c-Jun N-terminal kinase (JNK2), phosphorylated-c-Jun N-terminal kinase (p-JNK2), signal transducer and activator of transcription 3 (STAT3), phosphorylation-signal transducer and activator of transcription 3 (p-STAT3), and NPC1L1 were detected by Western blotting. After 24 hours of reperfusion, mouse brains were removed, placed in ice-cold radioimmunoprecipitation assay buffer (150 mM sodium chloride, Triton X-100, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 50 mM Tris, pH 8.0) containing protease and phosphatase inhibitors (Thermo Fisher Scientific), and broken down by sonication. The isolated proteins were subjected to gel electrophoresis and transferred to polyvinylidene fluoride membranes, which were then blocked with skim milk. After blocking, the membranes were incubated with rabbit polyclonal anti-GRP78/Bip (1:1000, Abcam, Cat# ab21685, RRID: AB_2119834), rabbit polyclonal anti-Caspase 12 (1:1000, Abcam, Cat# ab62484, RRID: AB_955729), mouse monoclonal anti-Caspase 9 (1:1000, Abcam, Cat# ab115792, RRID: AB_10938323), rabbit polyclonal anti-Caspase 3 (1:1000, Abcam, Cat# ab13847, RRID: AB_443014), mouse polyclonal anti-Bax (1:1000, Abcam, Cat# ab216494), rabbit polyclonal anti-Bcl-2 (1:1000, Abcam, Cat# ab196495, RRID: AB_2924862), goat polyclonal anti-JNK2 (1:1000, Abcam, Cat# ab134567), rabbit polyclonal anti-p-JNK2 (1:1000, Abcam, Cat# ab21685), mouse monoclonal anti-STAT3 (1:1000, Abcam, Cat# ab119352, RRID: AB_10901752), rabbit polyclonal anti-p-STAT3 (1:1000, Abcam, Cat# ab278669, RRID: AB_779085), rabbit monoclonal anti-Niemann-Pick C1 like 1 (NPC1L1) (1:1000, Abcam, Cat# ab124801, RRID: AB_10976493), and rabbit polyclonal anti-β-actin (1:5000, Abcam, Cat# ab8227, RRID: AB_2305186) overnight at 4°C. Next, the membranes were washed three times with Tris-buffered saline and Tween 20, then incubated with goat anti-rabbit IgG H&L (HRP) (1:10,000, Abcam, Cat# ab6721, RRID: AB_955447) and goat anti-mouse IgG H&L (HRP) (1:10,000, Abcam, Cat# ab6789, RRID: AB 955439) at room temperature for 1 hour.

Molecular docking studies

Molecular docking analysis was performed using the crystal structure of the N-terminal domain of Niemann-Pick C1 (NPC1) (Protein Data Bank file 3GKI). The co-crystal structure of NPC1 with its substrate, cholesterol, was obtained from the RCSB PDB data bank (http://www.rcsb.org). Maestro 12.8 (Schrodinger Inc., New York, NY, USA) was employed to perform the molecular docking analysis. Before docking, the protein was prepared using Protein Preparation Wizard (Schrodinger Inc.), with the removal of water molecules and the addition of H atoms. The H-bonds were optimized using OPLS2005 (Schrodinger Inc.), and energy minimization was applied to avoid any interactions. Then, the grid was prepared using the receptor grid generation wizard set to default parameters by selecting the internal ligand to specify the binding site. The β-sitosterol was prepared by LigPrep (Schrodinger Inc.) to generate conformers, tautomers, and rotamers. Energy minimization was applied, and the prepared ligands were used for ligand docking. Post-docking minimization was performed, and five positions per ligand interaction were recorded. Maestro pose viewer and Glide XP visualizer (Schrodinger Inc.), together with PyMol 3.7 (DeLano Scientific LLC, San Carlos, CA, USA), were used to generate the 2-dimensional/3-dimensional docking position.

Then, molecular dynamics simulation was performed with the above docking result as the initial conformation using the standard dynamics cascade protocol in Discovery Studio 2019 (Dassault Systèmes, Vélizy-Villacoublay, France). The simulation was carried out at natural pressure as well as at nearly 300 K in temperature. The CHARMM36 force field (Dassault Systèmes) was applied, and then the protein-ligand complex was solvated in an orthorhombic cell shape and an explicit periodic boundary model. During the calculation progress, the equilibration simulation time lasted for 20 ps, and the production simulation time was 200 ps. The molecular dynamics data were analyzed using the trajectory analysis tool in Discovery Studio 2019 (Dassault Systèmes) to obtain root-mean squared deviation, root-mean squared fluctuation, and a non-bond heatmap.

Statistical analysis

No statistical methods were used to predetermine sample sizes. No animals or data points were excluded. The experimental results are expressed as the mean ± standard error of mean (SEM), and all statistical analyses were performed using GraphPad Prism software (version 8.4.0 for Windows, GraphPad Software, San Diego, CA, USA, www.graphpad.com). The statistical significance was determined by one-way analysis of variance (ANOVA) with Bonferroni test or Welch ANOVA with Tamhane T2 test to compare multiple groups. Differences were considered to be statistically significant if P < 0.05.

Results

β-Sitosterol can reach mouse brain tissue after intraperitoneal injection

We assessed β-sitosterol concentrations in plasma and brain tissue from mice in the sham and MCAO groups to determine whether β-sitosterol mitigates the neuronal damage caused by MCAO. Figure 1A shows the chemical structure of β-sitosterol. Mice in the sham and MCAO groups were intraperitoneally injected with β-sitosterol (50 mg/kg) 30 minutes after reperfusion, and the left and right cerebral cortex tissues were collected 24 hours later (Figure 1B). Then, GC/MS was used to detect the β-sitosterol content of the brain tissues (Figure 1C–E). The results showed that the concentration of β-sitosterol in the left cerebral cortex of mice in the sham operation group was 18 ng/g, and there was no significant difference between β-sitosterol concentrations in the left cerebral cortex in the sham and MCAO groups. Within each group, there was no significant difference in β-sitosterol concentration between the left brain (infarct side) and the right brain (non-infarct side) (one-way ANOVA, F_{(3, 16)} = 0.5302, P = 0.6680; Figure 1F). Therefore, β-sitosterol can reach the mouse brain after being injected intraperitoneally.

Figure 1.

β-Sitosterol reaches mouse brain tissue after intraperitoneal injection.

(A) Chemical structure of β-sitosterol. (B) A schematic diagram illustrating the experimental design. (C–E) GC/MS analysis of normal brain tissue (C), brain tissue from animals treated with β-sitosterol (D), and brain tissue from a mouse subjected to middle cerebral artery occlusion (E). (F) The β-sitosterol concentration in the left and right brain 24 hours after intraperitoneal injection. Data are expressed as mean ± SEM (n = 5 per group) and were analyzed by one-way analysis of variance followed by Bonferroni test. GC/MS: Gas chromatography-mass spectrometer; MCAO: middle cerebral artery occlusion.

β-Sitosterol protects the brain against MCAO/R-induced neurological deficits

To confirm successful establishment of the MCAO/R model, cerebral blood flow in the experimental mice was assessed using a laser speckle Doppler imaging system. The results showed that blood flow decreased approximately 38% after MCAO/R compared with before the model was established. After reperfusion, the blood flow returned to approximately 82% of baseline flow levels (Welch ANOVA, F_{(2, 7)} = 18.71, P = 0.0013; Figure 2A), indicating that the MCAO/R model was successfully established. The cerebral infarct area and nervous system score are often used as important indicators to evaluate brain injury (Han et al., 2021; Zhao et al., 2021). The cerebral infarct volume was calculated by triphenyltetrazolium chloride staining 24 hours after MCAO/R (Figure 2B and C). Infarct volumes were smaller in the β-sitosterol (2, 10, 50 mg/kg) groups than in the MCAO group (P < 0.01), and the volumes decreased in a dose-dependent manner (one-way ANOVA, F_{(4, 20)} = 76.88, P < 0.0001; Figure 2C). Neurological scores were assigned 22 hours after MCAO/R. The results showed that there were marked neurological deficits in the model group, and that β-sitosterol treatment decreased these deficits in a dose-dependent manner (one-way ANOVA, F_{(4, 25)} = 90.24, P < 0.0001; Figure 2D). In addition, we found that the water content of the brain in the MCAO mice was increased significantly compared with the con mice. In the groups treated with β-sitosterol, the brain water content was slightly lower than that seen in the model group (one-way ANOVA, F_{(4, 25)} = 90.24, P < 0.0001; Figure 2E).

Figure 2.

Effects of β-sitosterol on neurobehavioral function and cerebral infarction volume in mice subjected to MCAO.

(A) Laser speckle Doppler results showed that blood flow to the brain was blocked after middle cerebral artery occlusion. The left side is the infarct side, and the right side is the non-operative side. (B) TTC staining of mouse brain tissues showing that the cerebral infarct volume induced by ischemia was significantly reduced by β-sitosterol treatment. The non-infarcted area is pink/red, and the infarcted area is white/cream. (C) Quantitative analysis of cerebral infarction volumes. (D) Quantitative analysis of modified neurological severity scores. (E) Quantitative analysis of cerebral water contents. Data are expressed as mean ± SEM (n = 6 per group) *P < 0.05, **P < 0.01, vs. sham group; #P < 0.05, ##P < 0.01, vs. model group (Welch analysis of variance followed by Tamhane T2 test in A, or one-way analysis of variance followed by Bonferroni test in C–E). MCAO: Middle cerebral artery occlusion; TTC: triphenyltetrazolium chloride.

β-Sitosterol decreases neuronal apoptosis in mouse brains after MCAO/R

To determine whether β-sitosterol decreases neuronal apoptosis, we performed Nissl and TUNEL staining of post-infarct brain sections (Figure 3). Nissl staining showed significant morphological changes in the peri-infarct zone in MCAO mice, including neuronal cell loss, neuronal nucleus shrinkage, cell swelling, and interstitial edema (Figure 3A). The degree of these morphological changes was significantly lower in the β-sitosterol-treated mice than in the model group mice (one-way ANOVA, F_{(2, 15)} = 93.02, P < 0.0001; Figure 3A). Similar results were obtained by TUNEL staining. As shown in Figure 3B, no apoptotic cells were observed in the brain tissue of mice in the sham group. In contrast, many apoptotic cells were observed in the brain tissue of mice in the model group. In comparison with the model group, fewer apoptotic neurons were observed in the groups treated with β-sitosterol (one-way ANOVA, F_{(2, 15)} = 43.09, P < 0.0001; Figure 3B). These findings indicated that β-sitosterol has an anti-apoptotic effect after MCAO.

Figure 3.

β-Sitosterol reduces the neuronal apoptosis caused by MCAO in mice.

(A) Nissl staining of cortical neurons and quantification of intact neurons. The Nissl bodies appear dark blue, whereas the nuclei and glia appear light blue. Arrows indicate Nissl bodies. Nissl staining showed that β-sitosterol treatment alleviates the decrease in the number of Nissl bodies caused by cerebral ischemia-reperfusion. Scale bars: 1000 μm (A upper) and 50 μm (A lower and B). (B) Neuronal apoptosis in the cortex was detected by TUNEL staining 24 hours after MCAO. The ratio of TUNEL-positive cells to total (DAPI-positive) cells was calculated. Arrows indicate double-stained cells. β-Sitosterol treatment reduced the neuronal apoptosis caused by cerebral ischemia/reperfusion. Scale bars: 1000 μm (low-magnification images) and 20 μm (enlarged images). Data are expressed as mean ± SEM (n = 5 per group). **P < 0.01, vs. sham group; ##P < 0.01, vs. model group (one-way analysis of variance followed by Bonferroni test). DAPI: 4′,6-Diamidino-2-phenylindole; MCAO: middle cerebral artery occlusion; TUNEL: TdT-mediated dUTP nick-end labelling.

β-Sitosterol alleviates OGD/R-induced cell injury in primary cortical neurons

To determine the optimal hypoxia/reoxygenation experimental conditions, we used a CCK-8 assay and LDH tests to detect neuronal activity after different durations of hypoxia. The results showed that neuronal activity was significantly decreased compared with the control group after 1 hour of hypoxia, so 1 hour of hypoxia was used for all subsequent in vitro experiments (one-way ANOVA, F_{(4, 25)} = 6.590, P < 0.0001, Figure 4A; one-way ANOVA, F_{(4, 25)} = 6.967, P < 0.0001, Figure 4B). Additionally, the results from the CCK-8 assay showed that β-sitosterol (0.1, 1.0, 10 μM) had no effect on the activity of normal neurons (one-way ANOVA, F_{(3, 20)} = 1.764, P = 0.1864, Figure 4C). We then subjected neurons to OGD/R and treated them with different concentrations of β-sitosterol and found that β-sitosterol treatment significantly improved neuronal activity and significantly reduced LDH release after OGD/R (Welch ANOVA, F_{(4, 18.69)} = 11.48, P < 0.0001, Figure 4D; one-way ANOVA, F_{(4, 25)} = 20.25, P < 0.0001, Figure 4E). To further assess the protective effect of β-sitosterol in the cellular model of stroke, primary cortical neurons were stained with an Annexin-V-FITC/PI dual staining kit and analyzed by flow cytometry. The rate of apoptosis in the OGD/R group was significantly higher than that in the control group (one-way ANOVA, F_{(3, 20)} = 60.24, P < 0.0001, Figure 4F). Furthermore, the proportion of apoptotic cells in the β-sitosterol treatment groups was lower than that in the model group (one-way ANOVA, P < 0.0001, Figure 4F), suggesting that β-sitosterol reduces neuronal apoptosis induced by OGD/R.

Figure 4.

β-Sitosterol protects primary cortical neurons from OGD/R.

(A) The survival rate of primary cortical neurons was detected by CCK-8 assay after different durations of OGD/R. (B) LDH release was measured 24 hours after different durations of OGD/R. (C) β-Sitosterol at different concentrations had no effect on the viability of primary neurons, as detected by CCK-8. (D) β-Sitosterol restored the level of viability of primary cortical neurons exposed to OGD/R, as detected by CCK-8. (E) LDH release ratio after β-sitosterol treatment. (F) The effects of β-sitosterol on apoptosis of primary cortical neurons subjected to OGD/R were assessed by flow cytometry. Data are expressed as mean ± SEM (n = 6 per group). *P < 0.05, **P < 0.01, vs. control group; #P < 0.05, ##P < 0.01, vs. OGD group (one-way analysis of variance followed by Bonferroni test in A–C and E, Welch analysis of variance followed by Tamhane T2 test in D, or unpaired t-test in F). CCK-8: Cell Counting Kit-8; CTRL: control; LDH: lactate dehydrogenase; ns: not significant; OGD: oxygen-glucose deprivation; OGD/R: oxygen-glucose deprivation/reperfusion.

β-Sitosterol inhibits the endoplasmic reticulum stress pathway in MCAO/R mice

The endoplasmic reticulum is critical for protecting the brain from apoptosis-related injuries (Pan et al., 2021; Ren et al., 2021). Therefore, we next used western blotting to evaluate the effect of β-sitosterol on components of the endoplasmic reticulum stress pathway in the ischemic cortex after MCAO/R. Compared with the sham group, the protein expression levels of endoplasmic reticulum stress markers GRP78/Bip (Welch ANOVA, F_{(4, 9.282)} = 5.955, P = 0.01) and Caspase 12 (Welch ANOVA, F_{(4, 14.25)} = 3.907, P = 0.002) were significantly increased in the model group (Figure 5A–C), while treatment with β-sitosterol significantly reversed this effect. Additionally, β-sitosterol treatment partially restored the MCAO-induced reduction in Caspase 3 expression (one-way ANOVA, F_{(4, 20)} = 10.41, P = 0.001) (Figure 5D), but did not reverse the effects of MCAO on Caspase 9 expression (Figure 5E). Furthermore, β-sitosterol treatment effectively reversed the effect of MCAO on Bax and Bcl-2 expression (Bax: Welch ANOVA, F_{(4, 8)} = 31.19, P < 0.0001; Bcl-2: one-way ANOVA, F_{(4, 20)} = 1.360, P = 0.02831; Figure 5F–H). In addition, β-sitosterol treatment significantly reversed the decrease in p-JNK2/JNK2 ratio (Welch ANOVA, F_{(4, 8)} = 7.221, P = 0.0091) and p-STAT3/STAT3 ratio (one-way ANOVA, F_{(4, 20)} = 35.42, P < 0.0001) induced by MCAO (Figure 5I and J). These results confirm that β-sitosterol protects the brain after MCAO/R by inhibiting the endoplasmic reticulum stress pathway.

Figure 5.

β-Sitosterol exerts a neuroprotective effect in MCAO/R mice via the endoplasmic reticulum stress pathway.

(A) Representative western blot images. (B–E, G–J) GRP78/Bip (B), Caspase 12 (C), Caspase 3 (D), Caspase 9 (E), Bax (G), Bcl-2 (H), p-JNK2/JNK2 (I), and p-STAT3/STAT3 (J) expression in the left cortex. (F) Representative western blot images. Data are expressed as mean ± SEM (n = 5 per group). *P < 0.05, **P < 0.01, vs. sham group; #P < 0.05, ##P < 0.01, vs. model group (one-way analysis of variance followed by Bonferroni test in A–C, G, H, K and L, Welch analysis of variance followed by Tamhane T2 test in D, E, I and J). ACTB: β-Avtin; GRP78: glucose regulated protein 78; JNK2: c-Jun N-terminal kinase; MCAO: middle cerebral artery occlusion; ns: not significant; p-JNK2: phosphorylation-c-Jun N-terminal kinase; p-STAT3: phosphorylation-signal transducer and activator of transcription 3; STAT3: signal transducer and activator of transcription 3.

β-Sitosterol directly acts on the NPC1L1 receptor

To study the mechanism by which β-sitosterol inhibits cholesterol, we used the Swiss Target Prediction system to predict the most probable protein targets of β-sitosterol. NPC1L1, a key cholesterol transporter (Hu et al., 2021), exhibited the strongest binding affinity with β-sitosterol out of all predicted protein targets. Molecular docking analysis using the crystal structure of NPC1 (PDB file 3GKI), which shares 51% similarity with NPC1L1, showed that β-sitosterol can occupy the cholesterol binding cavity in the N-terminal domain of NPC1L1 with a binding energy of –11.19 kcal/mol (Figure 6A). The 3D and 2D binding modes are shown in Figure 6A and B, respectively. β-Sitosterol formed a hydrogen bond with residue GLN79 (2.90 Å), and also displayed hydrophobic interactions with residues TRP27, LEU83, PHE108, PRO202, PHE203, and ILE205, indicating that NPC1L1 might be a direct protein target of β-sitosterol.

Figure 6.

β-Sitosterol binds closely to NPC1L1.

(A) β-sitosterol interaction with the NPC1L1 binding site. (B) Two-dimensional schematic diagram of β-sitosterol interacting with key amino acids of NPC1L1. (C) Non-bond heatmap showing that residue GLN79 of NPC1L1 forms a steady hydrogen bond with most conformations of β-sitosterol. (D) RMSD plot of the stable complex versus the various structural conformations during a 250-ps period. (E) RMSF plot of the stable complex versus the various structural conformations during a 250-ps period. NPC1L1: Niemann-Pick C1 like 1; RMSD: root mean square deviation; RMSF: root mean square fluctuation.

To assess the flexibility of NPC1L1–β-sitosterol binding, molecular dynamics simulation was performed using the docking result described above as the initial conformation. The simulation showed that the β-sitosterol–bound configuration remained generally stationary. The non-bond heatmap presented in Figure 6C showed that residue GLN79 of NPC1L1 formed a steady hydrogen bond with most conformations of β-sitosterol. Moreover, the calculated root-mean squared deviation indicated that the simulated β-sitosterol complex attained stability after 250 ps of run time (Figure 6D). The mean root-mean squared deviation value of NPC1L1-β-sitosterol (1.3 Å) was close to that of the NPC1L1-cholesterol complex (1.0 Å), implying that the above docking result was reliable (Figure 6D). In addition, the highly similar root-mean squared fluctuation values of the simulated NPC1L1–β-sitosterol and NPC1L1-cholesterol complexes (Figure 6E) suggested that both complexes are highly stable. Taken together, these in silico results suggest that β-sitosterol has high binding affinity for and a stable interaction site within NPC1L1. The highly similar structures and binding patterns between β-sitosterol and cholesterol, which is the natural substrate of NPC1L1, suggested that β-sitosterol might compete with cholesterol for the same binding site, thus acting as an NPC1L1 antagonist.

β-Sitosterol reduces apoptosis caused by excessive cholesterol accumulation in neurons

As a key cholesterol transporter, NPC1L1 plays an important role in cholesterol entry into cells (Hu et al., 2021). One study has shown that cerebral ischemia/reperfusion increases NPC1L1 expression in mouse brains (Yu et al., 2018). Our western blot results confirmed that NPC1L1 expression was significantly lower in the groups treated with β-sitosterol than in the model group (Welch ANOVA, F_{(4, 14.94)} = 6.049, P = 0.0042, Figure 7A). To determine whether β-sitosterol inhibits cellular cholesterol uptake after MCAO/R, we performed a cholesterol assay. The assay results showed that brain total cholesterol (P = 0.0002), free cholesterol (P = 0.0013), and cholesteryl ester (P = 0.03515) levels were significantly increased in the MCAO group compared with the sham group. Treatment with β-sitosterol reduced the cholesterol levels at 24 hours after MCAO (total cholesterol: one-way ANOVA, F_{(2, 12)} = 17.38, P = 0.0003; free cholesterol: one-way ANOVA, F_{(2, 12)} = 11.44, P = 0.0017; cholesteryl esters: one-way ANOVA, F_{(2, 12)} = 1.042, P = 0.03825; Figure 7B). To further confirm that β-sitosterol inhibits cholesterol accumulation, we treated neurons with 1 μM β-sitosterol and different doses (3, 9, and 27 μM) of cholesterol. Annexin V-FITC/PI double staining confirmed that direct administration of cholesterol to neurons causes apoptosis. The proportion of apoptotic cells in the 1 μM β-sitosterol + 27 μM cholesterol group was lower than that in the cholesterol-only group, suggesting that β-sitosterol treatment reduced the neuronal apoptosis caused by cholesterol (one-way ANOVA, F_{(4, 25)} = 31.98, P < 0.0001, Figure 7C). Western blotting showed that β-sitosterol significantly reversed the substantial increase in expression of the endoplasmic reticulum stress–related proteins GRP78/Bip (Welch ANOVA, F_{(4, 12.82)} = 11.87, P = 0.0003) and Caspase 12 (Welch ANOVA, F_{(4, 14.54)} = 4.953, P = 0.0100) in primary neurons induced by cholesterol (Figure 7D).

Figure 7.

β-Sitosterol inhibits cholesterol-induced neuronal apoptosis.

(A) Treatment with β-sitosterol decreased NPC1L1 expression in the cortical tissue of mice subjected to middle cerebral artery occlusion. (B) Treatment with β-sitosterol decreased the MCAO/R-induced cholesterol increase in brain tissue, as determined by a cholesterol kit assay. (C) Treatment with β-sitosterol decreased reduced the primary neuronal apoptosis induced by cholesterol. (D) Treatment with β-sitosterol decreased GRP78/Bip and caspase 12 expression in primary neurons treated with cholesterol. Data are expressed as mean ± SEM (n = 5 per group). *P < 0.05, **P < 0.01, vs. sham/control group; #P < 0.05, ##P < 0.01, vs. model/3, 9, 27 μM cholesterol group (Welch analysis of variance followed by Tamhane T2 test in A, D, one-way analysis of variance followed by Bonferroni test in B, and C). ACTB: β-Actin; GRP78: glucose regulated protein 78; MCAO: middle cerebral artery occlusion; NPC1L1: Niemann-Pick C1 like 1.

Discussion

In this study, mice with ischemic stroke induced by MCAO were treated by intraperitoneal injection of β-sitosterol 30 minutes after occlusion. Our results suggest that β-sitosterol can significantly reduce the volume of cerebral infarction, improve nerve function, increase neuronal cell viability, and reduce intracellular cholesterol accumulation after OGD/R. β-sitosterol may exert the observed neuroprotective effects by decreasing the endoplasmic reticulum stress induced by intracellular cholesterol accumulation after ischemic stroke.

Cerebral penumbra is affected by cerebral ischemia/reperfusion injury (CIRI) after vascular recanalization in the acute stage of ischemic stroke, and neuronal cell apoptosis is an important mechanism of CIRI (Sun et al., 2020; Xu et al., 2021). CIRI treatment aims to reduce apoptosis, alleviate the effects of tissue hypoperfusion, preserve the ischemic penumbra, and promote the recovery of nerve function (Jin et al., 2019). In this study, MCAO/R resulted in a low neurological function score in the model group 24 hours after injury, while mice treated with β-sitosterol exhibited significantly higher scores at the same time point. In addition, treatment with β-sitosterol significantly lowered the cerebral water content and decreased the infarct area compared with the untreated group. These results indicated that β-sitosterol improves the neurological function of mice post-CIRI. Nissl and TUNEL staining showed that neuron loss, nuclear pyknosis, and morphological and structural changes were decreased after β-sitosterol treatment compared with the model group. Additionally, CCK-8, LDH, and Annexin V-FITC/PI double staining of primary neurons after OGD/R showed that the apoptosis rate of the β-sitosterol–treated cells was significantly reduced compared with the model group. These findings show that β-sitosterol can effectively inhibit neuronal apoptosis in the ischemic area after CIRI in vivo and in vitro, and that β-sitosterol has a neuroprotective effect.

Apoptosis is a dynamic process that depends on the ratio of pro-apoptotic molecules to anti-apoptotic molecules (Fricker et al., 2018). Our western blot results showed that the expression of apoptosis-related protein Bax was significantly increased, expression of the anti-apoptotic molecule Bcl-2 was significantly decreased, and phosphorylation levels of the proliferation-related protein JNK2 and STAT3 significantly decreased in the MCAO/R model compared with the sham group. These results suggest that neuronal apoptosis occurred in the model mice. The Bax/Bcl-2 ratio is an important predictor of apoptosis (Liu et al., 2020). A reduction in the Bcl-2/Bax ratio initiates the Caspase 3 cascade, leading to cleavage of polyADP ribose polymerase, which ultimately leads to apoptosis (Li et al., 2021). In this study, we found that the expression levels not only of Caspase 3, but also of Caspase 9 and Caspase 12, were significantly increased in the MCAO/R model group. Cysteine aspartic acid-specific proteases (caspases) are a group of proteolytic enzymes known for their role in regulating cell apoptosis. Many studies have suggested that Caspase 3 is a key protease that plays a central role in mammalian cell apoptosis (D’Amelio et al., 2010; Zhao et al., 2013; Gong et al., 2022). Caspase 3 is also known as the “death protease” (Larsen et al., 2010). Caspase 9 activates the mitochondria-mediated endogenous apoptosis pathway (An et al., 2020). Caspase 12 is an endoplasmic reticulum–specific apoptosis effector protein (Nakagawa et al., 2000). Our western blot results showed that Caspase 3, Caspase 9, and Caspase 12 expression levels were significantly increased in the ischemic brain tissue of mice in the model group, and that this effect was significantly rescued by treatment with β-sitosterol. There was no significant difference in Caspase 9 expression level between the model group and the β-sitosterol-treated group; these results indicated that β-sitosterol may play a protective role in the context of MCAO/R by modulating the endoplasmic reticulum stress pathway. GRP78 is also known as the immunoglobulin heavy chain binding protein, and its level can directly reflect the level of endoplasmic reticulum stress in cells (Friedlander et al., 2000). We found that GRP78/Bip expression was significantly increased in the brain tissue of mice in the model group, suggesting that MCAO may induce neuronal apoptosis by over-activating endoplasmic reticulum stress, thereby triggering the GRP78/Caspase 12 signaling pathway. Our western blot results suggested that β-sitosterol plays a protective role in post-CIRI brain injury through this signaling pathway.

Next, we screened the Swiss Target Prediction database and found that the protein with the highest predicted affinity for β-sitosterol was NPC1L1. As a key protein involved in cholesterol transport, NPC1L1 regulates cholesterol metabolism in vivo, and is also the target of ezetimibe, the only lipid-lowering drug on the market that inhibits cholesterol absorption (Yu et al., 2018). We found that β-sitosterol effectively inhibited the increase in NPC1L1 expression induced by MCAO/R, and the molecular docking results showed compatible binding between β-sitosterol and NPC1L1.

Cholesterol is an indispensable component of cell membranes and plays a vital role in key physiological processes. Brain cholesterol accounts for a large proportion of the total cholesterol in the human body and must be strictly controlled to ensure normal brain function. Previous studies have confirmed that increased cholesterol in the brain is an important cause of stroke, and it has been reported that the total cholesterol content of the brain sharply increases in mice after MCAO/R (Yu et al., 2018). However, one study also showed that brain cholesterol synthesis is reduced in an animal model of Huntington’s disease (Valenza et al., 2010). In addition, disorders of cholesterol metabolism in the brain have been linked to Alzheimer’s disease (Di Paolo and Kim, 2011) and Parkinson’s disease (García-Sanz et al., 2021). Therefore, the specific role of disturbed cholesterol metabolism in the pathogenesis of neurodegenerative diseases has not been fully elucidated. In our study, we found that the brain cholesterol level increased significantly in response to MCAO/R compared with that in control group; this acute effect is unlike the long-term cumulative effect seen in neurodegenerative diseases, but also affected neuron viability. Cholesterol homeostasis in the brain, especially in neurons, plays an important role in the occurrence and development of ischemic stroke. Normally, inactive precursors of sterol regulatory element binding proteins bind to sterol regulatory element cleavage activation protein in the ER membrane. An increase or decrease in intracellular cholesterol disrupts the lipid and Ca²⁺ balance, inducing endoplasmic reticulum stress. Using Annexin V-FITC/PI double staining, we found that treatment with β-sitosterol significantly ameliorated neuronal cell damage in a cholesterol concentration-dependent manner. Western blotting showed that GRP78/Bip and Caspase 12 expression increased when cells were treated with cholesterol, suggesting that cholesterol indeed induced endoplasmic reticulum stress in neurons, and that β-sitosterol reversed this effect.

There were some limitations to our study. Although we demonstrated in vitro and in vivo that β-sitosterol ameliorates the effects of cerebral ischemia/reperfusion injury by modulating the cholesterol overload/endoplasmic reticulum stress/neuronal apoptosis pathways, other forms of cell death, like necroptosis, autophagy, ferroptosis, and pyroptosis, were not explored in this study. In addition, while we identified NPC1L1 as a potential target of β-sitosterol in silico, we did not validate this prediction in transgenic animals; in future studies we plan to extend this line of investigation by modifying the structure of NPC1L1 and performing an NPC1L1-targeting molecular screen.

In summary, our findings revealed that β-sitosterol mitigates brain damage after ischemic stroke, and that this effect may be attributable to anti-apoptotic activity exerted by inhibiting the cholesterol overload/endoplasmic reticulum stress pathway (Figure 8). Thus, β-sitosterol is a potential candidate for the treatment of brain diseases associated with apoptosis, such as neurodegenerative diseases. Furthermore, we presented a scientific basis for β-sitosterol as a neuroprotective phytomedicine.

Figure 8.

Schematic illustration of the potential mechanisms by which β-sitosterol exerts its neuroprotective effects.

Schematic illustrating the underlying mechanism by which β-sitosterol exerts neuroprotective effects in the context of cerebral ischemia/reperfusion injury. β-sitosterol binding to NPC1L1 modulates the cholesterol overload/endoplasmic reticulum stress-apoptosis signaling pathway. ERS: Endoplasmic reticulum stress; GRP78: glucose regulated protein 78; MCAO: middle cerebral artery occlusion; NPC1L1: Niemann-Pick C1 like 1.