5-HMF plays an intervention role in ischemic stroke by regulating GluR2 to improve mitochondrial function and promote angiogenesis

Abstract

Introduction

5-Hydroxymethyl furfural (5-HMF) is a furan compound with a molecular formula of C₆H₆O₃. Studies have found that 5-HMF has many pharmacological effects, such as improving hemorheology, anti-inflammatory, antioxidant activity and anti-myocardial ischemia. Identifying the preventive effect of 5-HMF against ischemic stroke and its possible mechanism was the aim of this investigation.

Methods

The MCAO animal model was used to detect the related indexes of behavior, neurological function and mitochondrial function, and to clarify the effect of 5-HMF on ischemic stroke. The HBMECs model induced by OGD-R was intervened by 5-HMF, and the GluR2 silencing sequence was added to detect mitochondrial function and angiogenesis-related indicators, so as to further explore the mechanism of 5-HMF intervention in ischemic stroke.

Results

In the MCAO experiment, the behavioral results showed that 5-HMF increased the quantity of autonomic activities, increased the exercise time and distance, reduced the balance beam score, and prolonged the residence time of MCAO rats on the rotating rod (P < 0.01). The results of neurological function indicated that 5-HMF reduced the neurological function score, increased the blood flow velocity of middle cerebral artery (P < 0.01), reduced the area of cerebral infarction, and alleviated the damage of cerebral cortex and hippocampal neurons. The results of mitochondrial function indicated that 5-HMF could significantly reduce the apoptosis of primary brain cells, ROS, Ca²⁺ levels, increase mitochondrial membrane potential and GluR2 expression in MCAO rats (P < 0.01). In the OGD-R-induced HBMECs experiment, 5-HMF increased the expression level of GluR2, increased cell viability, and decreased LDH activity in cell supernatant (P < 0.01). The results of mitochondrial function showed that 5-HMF reduced the levels of Ca²⁺ and glutamate in the cell supernatant, reduced the levels of apoptosis and ROS, increased the mitochondrial membrane potential, and improved the expression of mitochondrial dynamics-related proteins (P < 0.05 or P < 0.01). These improvements were markedly reversed after the addition of GluR2 silencing sequence (P < 0.01).

Conclusions

5-HMF has the effect of ameliorating brain damage in MCAO rats, and by regulating the expression of GluR2, reducing intracellular Ca²⁺ concentration, improving mitochondrial function, promoting angiogenesis, thereby reducing OGD-R-induced HBMECs damage.

Highlights

• •5-HMF can play a role in improving ischemic stroke.
• •5-HMF improves neurological dysfunction in MCAO rats by regulating mitochondrial function.
• •5-HMF improves the mitochondrial function of HBMECs by regulating the expression of GluR2.
• •5-HMF improves the angiogenesis of HBMECs by regulating the expression of GluR2.

1. Introduction

5-Hydroxymethylfurfural (5-HMF) is an antioxidant that increases the binding affinity of hemoglobin to oxygen, and 5-HMF has a good protective effect against oxidative damage and can increase the expression level of the mitochondrial fusion 2 protein (MFN2) [1]; 5-HMF can reduce oxidative damage to the striatum through the Nrf2/ARE signaling pathway, prevent permanent cerebral ischemia, and improve neurological damage [2]; 5-HMF can effectively promote the proliferation and migration of human skin fibroblasts (HSF) and the production of collagen, enhance angiogenesis, increase collagen production and promote epithelial regeneration to accelerate wound healing in vivo [3]; however, the specific mechanism of its anti-ischemic stroke is still unclear [4,5]. AMPAR is an ion channel glutamate-specific receptor. AMPAR is mostly a tetrameric complex composed of GluR2 dimer and GluR1, GluR3 or GluR4, in which GluR2 is an important subunit that determines the function of AMPAR. Ischemia and hypoxia can lead to changes in the expression and distribution of AMPAR subunits in the brain, and the expression of GluR2 is reduced, which is considered to be the main cause of nerve damage in vulnerable sites caused by short-term ischemia of the central nervous system [6]. Studies have indicated that GluR2, as an ionic glutamate receptor, can cause changes in the permeability of the cell membrane to Ca²⁺. In the normal physiological environment, GluR2 is in a high expression state, resulting in Ca²⁺ can't flow into the cell through the cell membrane. When there is hypoxia and ischemia in the brain, the expression of GluR2 is down-regulated, and a large amount of Ca²⁺ flows in, which in turn causes excitatory amino acid toxicity and aggravates neuronal damage [7]. Intracellular Ca²⁺ homeostasis is very important to maintain the normal physiological function of mitochondria. Mitochondrial oxidative phosphorylation produces ATP to provide energy for the survival of neurons, and mitochondrial function plays a crucial part in preserving vascular homeostasis and regulating angiogenesis [8]. The generation of ROS in the mitochondria, the regulation of intracellular Ca²⁺ and the proper balance of apoptosis via the mitochondrial pathway is necessary. The homeostasis of these mitochondria ensures the transformation of the proliferative phenotype required for endothelial cells to angiogenesis [9].

In order to study the mechanism of 5-HMF intervention in ischemic stroke, we performed the following experiments. The MCAO model was ‌established to investigate the effects of 5-HMF on behavioral and neurological function-related indicators in rats, evaluate the efficacy of 5-HMF, and detect the levels of Ca²⁺, as well as apoptosis, ROS, and mitochondrial membrane potential related to mitochondrial function, and preliminarily investigate the possible mechanism of 5-HMF intervention in ischemic stroke. Secondly, cell experiments were performed to detect indicators related to mitochondrial function, as well as endothelial cell migration, adhesion, and angiogenesis ability, in order to explore5-HMF has a good antioxidant damage effect 5-HMF's mechanism in improving ischemic stroke.

2. Materials and methods

2.1. Chemicals and drugs

5-HMF was bought from Shanghai yuanye Bio-Technology Co., Ltd (Shanghai, China).

2.2. Animals and ethical statement

Fifty SPF SD male rats, weighing 240–280 g, were bought from Zhejiang Vital River Laboratory Animal Technology Co., Ltd. Production license number: SCXK (Zhejiang) 2019-0001. SD rats were brought up in a clean animal laboratory, 18–22 °C, free food and water. The experiment on animals was authorized by the Experimental Animal Ethics Committee of Henan University of Chinese Medicine. Animal experimental ethics approval number: DWLL201908112. Environmental use license number: SYXK (Yu) 2021-0015. Every animal test complies with the ARRIVE guidelines12 and is carried out in accordance with the U.K. Animals (Scientific Procedures) Act 1986 and associated guidelines13, or EU Directive 2010/63/EU on the protection of animals used for scientific purposes14, or the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals and, as applicable, the Animal Welfare Act15.

2.3. Animal experiments

According to the principle of uniform weight, male SD rats were randomly divided into sham operation group (Sham), model group (MCAO), 5-HMF low dose group (5-HMF-L, 10 mg/kg), 5-HMF high dose group (5-HMF-H, 20 mg/kg), positive drug nimodipine group (Nimo, 10 mg/kg). The rats in the treatment group were given the corresponding dose of drugs by gavage every day, and the rats in the sham operation group and the model group were given the same volume of normal saline by gavage. The rats were modeled after 7 days of administration. Prior to modeling, the rats were fasted for 12 h, and their body weights were measured. Except for the sham operation control group, the other rats were established MCAO/R model by suture method [10]. The rats were anesthetized by intraperitoneal injection of 1 % pentobarbital sodium (0.04 g/kg). The rats were fixed on the rat plate with their abdominal surfaces facing upwards and placed near a radiator to maintain constant body temperature during the operation, and the skin of the anterior median neck was prepared for skin disinfection. An incision of about 3–5 cm was made along the middle of the neck, and then the muscles were separated directly. A ′ V ′-shaped incision was made at a distance of about 6 mm from the bifurcation of the common carotid artery (CCA). The silicone suture was inserted from the CCA into the internal carotid artery (ICA) at a depth of about 18–22 mm. After fixing the suture for 2 h, the suture was gently removed, and the extraction time was recorded. The sham operation control group was operated as described above, but no suture was inserted.

2.4. Open field experiment and autonomous activity

After modeling, the open-field activity test system of OFT-100 rats and mice was employed in order to identify autonomous behavior, exploratory behavior and tension of rats in the new environment. Three days before the experiment, the rats were placed in the reaction box at a fixed time daily to reduce their fear. The formal experiment was conducted in a silent setting. The rats were placed in the middle of the box's bottom and photographed and timed at the same time. After 5 min of observation, the camera was stopped and the motion path and data were saved. Alcohol was used to clean the inner wall and bottom of the box and blow dry with a hairdryer to avoid the influence of the last animal 's urine and odor on the next animal 's behavior.

The number of autonomous actions of rats was determined using the YLS-1B multipurpose rat autonomous activity recorder after modeling. The rats were placed in a covered black box, and the surrounding environment was kept quiet. After 5 min of adaptation, the recording was started. The recording time was 5 min, and the number of autonomic activities of each rat was recorded and statistically analyzed.

2.5. Balance beam experiment

The balance beam experimental device was used to evaluate the motor coordination ability of the animals. Three days before the formal experiment, the rats were trained to ensure they could pass through the balance beam smoothly and quickly. The evaluation criteria were displayed in Table 1.

Table 1.

Balance beam score table.

Evaluation criterion	Score
Smoothly and smoothly through the balance beam	0
Through the balance beam, less than 50 % of the way, slip foot	1
Through the balance beam, more than 50 % of the distance, slippery feet	2
Through the balance beam, but the hind limbs cannot move	3
You can 't cross the balance beam, you can sit on it	4
Can not stay	5

2.6. Rotating rod experiment

The spinning rod was used to hold the animals, and the residence time of the animals on the rotating rod was observed to evaluate the motor coordination ability and balance ability of the animals. Three days before the formal experiment, the rats were trained to ensure that they could not fall at the rotation speed of 10–30 rpm within 3 min. The rotating rod experiment was divided into an acceleration phase and a uniform phase. The acceleration phase is divided into the first phase speed of 8–12 rpm for 20 s, the second phase speed of 12–20 rpm for 20 s, the third phase speed of 20–30 rpm for 25 s, and the uniform phase speed of 30 rpm for 90 s, the time that rats stayed on the rotating rod was recorded the duration of rats' stay on the rotating rod was recorded, and each rat was evaluated three times at an interval of 10 min.

2.7. Neurologic function score

Rats' neurological function was scored using the Longa 5 scoring system following a 24 h modeling period, and single-blind method was used to record. The specific evaluation criteria are shown in Table 2.

Table 2.

Longa 5 neurological function score.

Evaluation standard	Neurological function score	Neurological dysfunction
normal	0	nil
It is not possible to completely stretch the left front paw	1	slight
When walking, the rats circled to the left	2	moderate
The rat's body rotates to the left and excretes when it walks	3	gravity
Can not walk spontaneously, loss of consciousness	4	Extremely heavy

2.8. Middle cerebral artery blood speed

The middle cerebral artery blood flow of rats was detected by VisualSonics Vevo®2100. Three rats were randomly selected from each group, and their heads were depilated and cleaned with depilatory cream. Firstly, the rats were placed in the glass cover of the anesthesia machine. The isoflurane concentration was adjuste to around 3 %, and flow rate of isoflurane was about 3 L/min. The rats entered the anesthesia state for about 3 min. The rats were fixed on the platform for monitoring physiological data. The nose was placed in the mask of the anesthesia machine. The flow rate was set at 1 L/min and the isoflurane concentration was adjusted to around 1 %. The heart rate of the rats was maintained between 300 and 550 BPM. Once the rat's heart rate stabilized, the coupling agent was applied to its limbs and head, and the MS250 (with a center of frequency 21 MHz) probe was used to collect images. In the color Doppler (Color Mode) mode, the middle cerebral artery's blood flow position was identified, and the pulse Doppler (Power Doppler Mode) mode was used to measure the middle cerebral artery's blood flow velocity. The experimental data were exported, and the maximum velocity and average velocity of blood flow were calculated by using Version 3.2.0 to measure the velocity time integral (VTI).

2.9. Brain coefficient

After 24 h of modeling, the rats were weighed and anesthetized by intraperitoneal injection of 1 % pentobarbital sodium intraperitoneally (0.04 g/kg). The rats were then dissected to collect blood and the water on the surface of the brain tissue was wiped off using filter paper. The brain weight was measured using a ten-thousandth balance, and the brain coefficient was calculated.

2.10. Triphenyltetrazolium chloride (TTC)

The brain was taken out and quickly placed in a refrigerator at −20 °C for 2 h. At the same time, the mouse brain mold was pre-cooled in a refrigerator at −80 °C. After 15 min, the brain became hard and could be sliced. The rat brain mold was placed on pre-prepared ice pre-prepared, and then the brain tissue was put into the mold. The distance between the blade and the frontal pole was 3 mm, and the coronal continuous section was made from front to back, each with a thickness of 2 mm. The brain slices were submerged in 2 % TTC phosphate buffer that had been heated beforehand, incubated at 37 °C in the dark, and the tissue was gently shaken every 5 min. After 20 min, the tissue was taken out for observation. The region of the cerebral infarction was white, whereas the normal brain tissue was stained red. The brain sections were submerged in 4 % paraformaldehyde and stored in the dark for 48 h. The brain sections were arranged in the order of the anterior to posterior brain regions, and with a ruler placed next to them.

2.11. Hematoxylin-Eosin staining (HE)

The fresh brain tissue was immediately immersed in 4 % formaldehyde fixative and fixed for at least 24 h. After fixation, the tissue underwent dehydration, paraffin embedding, and continuous coronal sectioning at 4 μm thickness using a microtome, followed by HE staining. Hematoxylin staining: Sections were immersed in hematoxylin staining solution for 8 min, washed with tap water, differentiated with 1 % acid alcohol for a few seconds, and rinsed with ammonia water to blue the nuclei. Eosin staining: The sections were immersed in eosin staining solution for 2–4 min and rinsed with tap water. For dehydration and clearing, the sections were sequentially immersed in anhydrous ethanol and xylene, then removed, air-dried briefly, and mounted with neutral gum. The sections were observed under an Olympus microscope, and photographs were taken and saved.

2.12. Nissl staining

The paraffin sections were dehydrated, then incubated in toluidine blue staining solution for 5 min. They were differentiated with 1 % glacial acetic acid, rinsed with running water to stop the reaction, and dried in an oven. After being cleared in xylene for 5 min, the sections were mounted with neutral gum. Observations were made under a microscope, and photographs were taken.

2.13. Apoptosis, ROS, JC-1 and Ca²⁺ levels in primary brain cells were detected by flow cytometry

Primary brain cell isolation: An appropriate amount of right brain tissue was removed and placed in an EP tube containing 2 mL of PBS, then minced thoroughly with scissors. The supernatant was discarded after centrifugation, and the cells were washed with 2 mL PBS. To lyse red blood cells in the brain tissue, the supernatant was removed after centrifugation, 1 mL of red blood cell lysate was added, and the mixture was left to stand for 5 min. The supernatant was discarded by centrifugation, and 10 mL of PBS was added to resuspend the cells. The tissue fragments were filtered through a 70 μm filter, and primary brain cells were obtained by centrifugation.

Apoptosis detection: Cells were resuspended in 1 × Binding Buffer, and 100 μL of the cell suspension was taken to ensure the cell density was approximately 1 × 10⁶ cells/mL. After adding 5 μL 7-AAD and 5 μL PE sequentially, the mixture was incubated for 15 min in the dark at room temperature. The apoptosis rate of rat primary brain cells was determined using a Flowsight multidimensional panoramic flow cytometer.

ROS detection: Primary rat brain cells were collected and counted to ensure the cell density was maintained between 1 × 10⁶-2 × 10⁷ cells/mL. First, a positive tube was prepared by adding a reactive oxygen species (ROS) positive control, followed by incubation at 37 °C for 30 min. After treating the positive group, all cell samples were loaded with probes. DCFH-DA was diluted in DMEM at a ratio of 1:1000, and 100 μL of the diluted DCFH-DA was added to the collected cell samples to resuspend them. The mixture was incubated in the dark at 37 °C for 20 min, with gentle mixing every 5 min to ensure full contact between the DCFH-DA fluorescent probe and the cells. The cells were washed with 1 mL of DMEM, centrifuged at 1000 rpm for 5 min, and the supernatant was discarded; this step was repeated 3 times to fully remove unbound DCFH-DA. A Flowsight multidimensional panoramic flow cytometry was used to determine the ROS level in rat primary brain cells.

JC-1: After resuspending the cells in 0.5 mL of DMEM, 0.5 mL of 1 × JC-1 staining buffer was added to the negative tube. The remaining cells were mixed with 0.5 mL of JC-1 staining working solution, incubated in the dark at 37 °C for 20 min, and then centrifuged at 600 g for 4 min. After discarding the supernatant, the cells were washed with 1 mL of JC-1 staining buffer, centrifuged twice at 600 g for 4 min, and resuspended in 0.5 mL of JC-1 staining buffer for detection.

Ca²⁺: Fluo-3AM was diluted to 5 μmol/L, and 100 μL of the diluted Fluo-3AM was added to each tube. The cells were incubated in the dark at 37 °C for 1 h, then centrifuged at 1000 rpm for 5 min, rinsed with PBS three times, and resuspended in 500 μL of PBS for detection.

2.14. Confirmation of optimal transfection conditions

Screening of transfection conditions: HBMECs were seeded in 6-well plates at a density of 1 × 10⁵ cells/mL. After seeding, the cells were divided into the following groups: NC group (blank control), Si-NC group (negative sequence control), Lipo 2000 group (reagent control), Si-GluR2-1 group (sequence 1), Si-GluR2-2 group (sequence 2), Si-GluR2-3 group (sequence 3), and FAM group (FAM-labeled negative sequence). Transfection was performed according to the instructions for Lipo 2000. After 6 h of culture in a 37 °C incubator, the medium was replaced with normal medium to observe the transfection status of the FAM group. The sequences of SiRNA are shown in Table 3.

Table 3.

SiRNA primary structure nucleotide sequence.

Name	Justice chain sequence (5′–3′)	Antonymous chain sequence (5′–3′)
Si-NC	UUCUCCGAACGUGUCACGUTT	ACGUGACACGUUCGGAGAATT
Si-RNA-1	GGACUGUGAAAGGGAUAAATT	UUUAUCCCUUUCACAGUCCTT
Si-RNA-2	GCUCACACAACAACAAUUATT	UAAUUGUUGUUGUGUGACCTT
Si-RNA-3	GGGACAAGGUGUAGAAAUATT	UAUUUCUACACCUUGUCCCTT
FAM	UUCUCCGAACGUGUCACGUTT	ACGUGACACGUUCGGAGAATT

RT-qPCR: After 48 h, total RNA was extracted from the cells using a complete RNA extraction kit. Agarose gel electrophoresis was performed to detect RNA degradation. The concentration of RNA was measured with a NanoDrop ™ UV spectrophotometer. RNA was reverse-transcribed using a thermal cycler under the following conditions: 42 °C for 30 min, 80 °C for 10 min, and finally held at 4 °C. Tthe mRNA level of GluR2 was detected by the SYBR Grenn method using a fluorescence quantitative PCR instrument. The primer sequences are shown in Table 4.

Table 4.

Primer sequence reading table.

Gene	upstream sequence (5′–3′)	downstream sequence (5′–3′)
GRIA2	CATTCAGATGAGACCCGACCT	GGTATGCAAACTTGTCCCATTGA
GAPDH	CCAGGTGGTCTCCTCTGA	GCTGTAGCCAAATCGTTGT

Western blot: Protein extraction: Proteins in the brain tissue were extracted using a cell membrane and cytoplasmic protein extraction kit, then quantified with a BCA protein quantification kit. 50 μg of protein was subjected to SDS-PAGE electrophoresis, transferred to a PVDF membrane, degreased, and blocked. Specific primary antibodies (GluR2, 11994-1-AP, Proteintech; β-actin, AC004, ABclonal) were added, and the membrane was incubated overnight at 4 °C. The next day, after equilibration at room temperature for 30 min, unbound primary antibodies were washed off with 0.2 % PBST (5 min per wash, repeated 5 times), followed by the addition of secondary antibodies. Unbound secondary antibodies were washed off with 0.2 % PBST (5 min per wash, repeated 3 times), and the membrane was then collected. After drying, Odyssey two-color infrared fluorescence imaging technology was used for detection.

2.15. Cell grouping and administration

Based on the results of PCR and WB, the sequence with the best silencing effect was selected for the experiment. The cells were divided into five groups: NC, Si-NC, M, M − Si, 5-HMF, and 5-HMF-Si. Transfection was performed according to the instructions of Lipo 2000. 24 h after transfection, OGD-R modeling and drug administration were carried out, with the concentration of 5-HMF set at 5 μmol/L.2.16. CCK8 detection of cell viability and LDH activity detection.

100 μL of cell supernatant was aspirated from each well, and 10 μL of CCK-8 was added. The cells were incubated in the incubator for another 2 h, after which the OD value at 450 nm was measured using a microplate reader. HBMECs were seeded into 24-well plates at a density of 8 × 10⁴ cells/mL, with 500 μL per well. The modeling and administration methods were the same as described above. The LDH activity in the cell supernatant was detected, and the specific steps were performed according to the kit instructions. The protein content in the cells was measured using a BCA protein quantification kit. Results: The LDH activity in the cell supernatant was calculated using the following fomula. $\text{LDH}\text{activity}\left(\mathrm{U}/\text{gprot}\right)=\frac{\text{Determination}\text{of}\text{OD}\text{value}‐\text{control}\text{OD}\text{value}}{\text{Standard}\text{OD}\text{value}‐\text{blank}\text{OD}\text{value}}\times \text{Standard}\text{concentration}÷\text{Sample}\text{protein}\text{concentration}$

2.16. Detection of cell mitochondrial function

The levels of Ca²⁺, apoptosis, ROS and JC-1 in HBMECs were detected by flow cytometry: Six-well plates were seeded with HBMECs at a density of 1 × 10⁵/mL, and transfected when the cell density was close to 30–50 %. Cell treatment: Following a single PBS wash, the cells were digested with 1 mL of EDTA-free trypsin, terminated with 3 mL of PBS, moved into a 15 mL EP tube, and centrifuged for 5 min at 800 rpm. Index detection: The experimental steps of apoptosis, ROS, JC-1 and Ca²⁺ were the same as above.

High content cell imaging: The cells of HBMECs were planted in PerkinElmer 96-well plates at a density of 4 × 10⁴/mL. After the cell model was established, the medium was discarded, 200 μL of PBS was washed twice per well, 150 μL of 4 % tissue cell fixative per well was fixed for 10 min, and the 96-well plate was placed on a decolorization shaker at a speed of about 40 rpm. 200 μL of 0.1 % Triton per well was washed 4 times/5 min, 200 μL of PBS per well was washed 3 times/3 min, and 100 μL of 5 % goat serum per well was blocked for 30 min. After adding 50 μL of primary antibody to the blocking solution and letting it sit at 4 °C for the entire night, 0.1 % PBST was rinsed five times for 5 min, 50 μL secondary antibody was added to incubate for 1 h in the dark, 0.1 % PBST was washed 5 times/5 min, DAPI was incubated for 5 min, 0.1 % PBST was washed 5 times/5 min, and 63 times of water mirror was selected for detection.

2.17. Detection of cell angiogenesis

Detection of cell migration ability: The cell scratch test was employed to find out if HBMECs may migrate in each group, and the 6-well plate cells were transfected. After modeling and administration, the scratch test was performed. Trypsin was used to break down the cells and seeded at 1 × 10⁵/mL in a new 6-well plate marked with a ruler, with 3 replicates in each group. The cells were cultivated in DMEM complete medium for 24 h. The 10 μL lancet was scratched evenly along the direction perpendicular to the marking line. The medium was sucked, washed 2–3 times with PBS, plus DMEM medium with 2 % FBS was added to exclude the impact of serum on the proliferation of cells. The whole area of the 6-well plate was photographed using a high-content imaging system with a 5-fold air mirror. At this time, the cells were recorded for 0 h, and the uneven scratches were removed. After 24 h of culture, the photographs were taken again. The experiment was repeated three times, and the picture was exported. The scratch area was measured by Image J software. The mobility calculation formula is as follows.

$$\text{Mobility}\left(100\%\right)=\frac{\text{Primary}\text{scratch}\text{area}‐\text{scratch}\text{area}\text{after}24\mathrm{h}}{\text{Primitive}\text{scratch}\text{area}}\times 100\%$$

Detection of cell adhesion ability: Cell pretreatment: In 6-well plates, HBMECs were sown at a density of 1 × 10⁵/mL, and the cells were transfected, modeled and administered according to the steps. Matrigel pretreatment: The whole bottle of matrix glue was placed in a broken ice box and melted overnight in a refrigerator at 4 °C. Gun head, 96-well plate, EP tube 4 °C refrigerator precooling. Adhesion experiment: The ice box is placed on the super clean table, and the subsequent operation needs to be carried out on the ice. The melted matrix adhesive was divided into 1.5 mL EP tubes with a pre-cooled gun head and kept at −80 °C in a refrigerator. The remaining matrix glue required for this experiment, pre-cooled 96-well plates were placed on ice in advance, 50 μL/hole matrix glue, and the surface of the matrix glue was kept smooth when the matrix glue was added, and no bubbles were allowed. The 37 °C incubator was placed for 30 min to solidify the matrix glue. During this period, single cell suspension was prepared, cells in 6-well plates were washed twice with PBS, digested with 1 mL trypsin, centrifuged for 10 min at 1000 rpm, cells were counted, and cell density was adjusted. Cells were sown in 96-well plates with matrix glue at a density of 4 × 10⁴/mL. Each group had six wells, which were then cultivated for 24 h in an incubator. After 24 h, the medium was thrown out and given two PBS washes. After that, 100 μL of complete medium was put into every well, and 10 μL of CCK8 was employed to assess the vitality of cells, reflecting the adhesion of HBMECs.

Tube formation: Cell pretreatment and Matrigel matrix pretreatment were the same as the detection of cell adhesion ability. Matrix glue 50 μL/well, 37 °C incubator for 30 min. In a matrix gel, HBMECs were sown at a density of 2 × 10⁵/mL, cultured in a 5 % CO2 incubator at 37 °C for 6 h, seen using an inverted microscope and captured on camera (100 × ). From each hole, six fields of view were chosen at random, and three duplicate holes were set up for each group of cells. The experiment was repeated three times.

2.18. Statistical analysis

SPSS 20.0 software was used for statistical analysis, and the data were expressed in the form of $\bar{x}$ ±s. The experimental data analysis was first tested for homogeneity of variance (Levene test), followed by one-way analysis of variance (One-Way ANOVA), and the pairwise comparison of the standard deviation values was selected by LSD or Dunnett T3 test. P < 0.05 was considered statistically significant, and P < 0.01 was considered extremely significant.

3. Results

3.1. Effect of 5-HMF on behavior of MCAO rats

As shown in Fig. 1A–C, the results of the open-field test indicated that compared with the Sham group, the movement time and distance of the MCAO group were significantly reduced (P < 0.01); In comparison with the MCAO group, 5-HMF significantly increased the movement time and distance of MCAO rats (P < 0.01). The number of spontaneous activities was detected. As shown in Fig. 1D, the number of spontaneous activities in the MCAO group was significantly lower than that in Sham group (P < 0.01). Compared with MCAO group, 5-HMF greatly increased the number of spontaneous activities in MCAO rats (P < 0.01), and the balance beam score was detected. As shown in Fig. 1E, the balance beam score of the MCAO group was significantly higher than that of the Sham group (P < 0.01). In contrast to the MCAO group, the balance beam score decreased greatly after 5-HMF intervention (P < 0.01). The results showed that compared with the Sham group, the time spent on the rotarod in the MCAO group was markedly reduced (P < 0.01). As opposed to the MCAO group, 5-HMF significantly increased the residence time of MCAO rats on the rotarod (P < 0.01), as shown in Fig. 1F. The above results indicate that 5-HMF can improve the motor function of MCAO rats.

Fig. 1.

Effects of 5-HMF on behavior of MCAO rats. (A) Open field experiment results schematic diagram. (B–C) Quantification of open field experimental results (n = 6). (D) Spontaneous activity of rats (n = 6). (E) Balance beam score of rats (n = 8). (F) Rotarod test of rats (n = 6). Compared with Sham group. ∗∗P < 0.01; Compared with MCAO group. ^##P < 0.01.

3.2. Effects of 5-HMF on neurological function in MCAO rats

The results of small animal ultrasound were shown in Fig. 2A–C. Compared with the Sham group, the average velocity and maximum velocity of middle cerebral artery blood flow in MCAO group were noticeably decreased (P < 0.01). In comparison with the MCAO group, 5-HMF significantly increased the average and maximum velocities of middle cerebral artery blood flow in MCAO rats (P < 0.01), suggesting that 5-HMF can increase the blood flow velocity of the middle cerebral artery in MCAO rats. Fig. 2D shows the results of neurological function score. The results showed that the neurological function score of MCAO group was markedly higher than that of the Sham group (P < 0.01). Contrary to the MCAO group, the neurological function score was greatly decreased after 5-HMF intervention (P < 0.05 or P < 0.01), indicating that 5-HMF can improve the neurological impairment of MCAO rats. The brain coefficient was detected. The experimental results showed that the brain coefficient of the MCAO group was significantly higher than that of the Sham group (P < 0.01). As opposed to the MCAO group, 5-HMF noticeably reduced the brain coefficient of MCAO rats (P < 0.01), as shown in Fig. 2E, suggesting that 5-HMF intervention can reduce the brain coefficient and alleviate cerebral edema in MCAO rats. The results of TTC staining were shown in Fig. 2F and G. The brain tissue of the Sham group appeared rose red, while a distinct white area was observed on the right side of the brain in the MCAO group, with a significantly increased cerebral infarction volume (P < 0.01). In contrast, the cerebral infarction volume in the 5-HMF group was significantly reduced (P < 0.01), suggesting that 5-HMF intervention can reduce the brain injury of MCAO rats. The above results show that 5-HMF can reduce the brain injury of MCAO rats and improve neurological function in MCAO rats.

Fig. 2.

Effects of 5-HMF on neurological function of MCAO rats. (A) Depiction of ultrasonic blood flow. (B) Rats' average cerebral artery blood flow velocity (n = 6). (C) Rats' maximal cerebral artery blood flow velocity (n = 6). (D) Rats' neurological function score (n = 8). (E) Rat brain coefficient (n = 6) (F) Brain tissue TTC staining. (G) The TTC statistical outcome (n = 3). Compared with Sham group. ∗∗P < 0.01; Compared with MCAO group. ^#P < 0.05; ^##P < 0.01.

3.3. Effects of 5-HMF on the pathological structure of brain tissue and hippocampal neurons in MCAO rats

The results of HE staining showed that the cerebral cortex cells in the Sham group were arranged neatly with clear nucleoli and intact structures. In the MCAO group, the cerebral cortex cells were disordered, with vacuolated cytoplasm and pyknotic nucleoli; in the 5-HMF group, the cell structures were gradually intact and arranged neatly. These results are shown in Fig. 3A. The results of Nissl staining showed that the hippocampal neurons in the Sham group were intact, appearing round or oval with obvious nuclei, and a large number of Nissl bodies with uniform staining. The nuclei of hippocampal neurons in MCAO group were fragmented or absent, and the cells were atrophic and necrotic. After 5-HMF intervention, the number of death and atrophy of hippocampal neurons was significantly reduced, and the number of Nissl bodies was increased. The results are shown in Fig. 3B. The above results indicate that 5-HMF exerts a neuroprotective effect, thereby alleviating the damage to the cerebral cortex and the pathological injury of hippocampal neurons in MCAO rats.

Fig. 3.

Effects of 5-HMF on hippocampal neurons and mitochondrial function in MCAO rats. (A) HE staining result diagram. Magnification factor 200 × , arrow pointing to vacuolated cells. (B) Nissl staining result diagram. Magnification factor 400 × , arrow pointing to Nissl bodies.

3.4. Effects of 5-HMF on mitochondrial function in MCAO rats

Flow cytometry was used to detect the apoptosis level of rat primary brain cells. As shown in Fig. 4A, contrary to the Sham group, the early, late and total apoptosis levels of primary brain cells in the MCAO group were markedly increased (P < 0.01), and the cell survival rate was significantly decreased (P < 0.01). In contrast to the MCAO group, 5-HMF can reduce the early, late and total apoptosis levels of primary brain cells in MCAO rats (P < 0.01) and increased the cell survival rate (P < 0.01), indicating that 5-HMF can reduce the apoptosis level of primary brain cells in MCAO rats. The level of ROS in primary brain cells of rats was detected by flow cytometry. The results showed that the level of ROS in primary brain cells of MCAO group was significantly higher than that of Sham group (P < 0.01). As opposed to the MCAO group, 5-HMF greatly reduced the level of ROS (P < 0.01), see Fig. 4B, suggesting that 5-HMF can alleviate oxidative stress in MCAO rats. The level of mitochondrial membrane potential of primary brain cells in rats was detected by flow cytometry. As shown in Fig. 4C, compared with the Sham group, the level of JC-1 polymer in primary brain cells of the MCAO group was noticeably decreased, indicating that the mitochondrial membrane potential was significantly decreased (P < 0.01). In comparison with the MCAO group, 5-HMF could markedly increase the level of JC-1 polymer, decrease the level of JC-1 monomer, and significantly increased the level of mitochondrial membrane potential (P < 0.01), indicating that 5-HMF can increase the level of mitochondrial membrane potential in MCAO rats and alleviate mitochondrial damage. The level of Ca²⁺ in primary brain cells of rats was detected by flow cytometry. The results showed that compared with Sham group, the concentration of Ca²⁺ in primary brain cells of MCAO group was greatly increased (P < 0.01). Contrary to the MCAO group, 5-HMF can significantly reduce the intracellular Ca²⁺ level, as shown in Fig. 4D. The above results indicate that 5-HMF can improve the mitochondrial function of MCAO rats.

Fig. 4.

Effects of 5-HMF on mitochondrial function of MCAO rats. (A) Analysis of the degree of apoptosis in rat primary brain cells using flow cytometry (n = 3). (B) Flow cytometry analysis of ROS in rat primary brain cells (n = 3). (C) Using flow cytometry to analyze the mitochondrial membrane potential in rat primary brain cells (n = 3). (D) Flow cytometry analysis of Ca²⁺ level in rat primary brain cells (n = 3). Compared with Sham group. ∗∗P < 0.01; Compared with MCAO group. ^##P < 0.01.

3.5. The effect of 5-HMF on GluR2 protein in MCAO rats and OGD-R induced HBMECs and the effect of 5-HMF on cell viability and LDH activity of HBMECs after Si-GluR2 silencing induced by OGD-R

The expression of GluR2 protein in MCAO rats was detected. The findings demonstrated that the MCAO group's GluR2 expression level was much lower than the Sham group's (P < 0.01). As opposed to the MCAO group, 5-HMF intervention markedly increased the GluR2 expression level in the ischemic brain tissue of MCAO rats (P < 0.01). The findings are displayed in Fig. 5A. The expression of GluR2 protein in OGD-R-induced HBMECs was detected. The results showed that the expression level of GluR2 in the M group was significantly lower than that in the NC group (P < 0.01). Compared with the M group, 5-HMF medium and high dose group could greatly raise the amount of GluR2 expression (P < 0.01), and 5 μmol/L was determined as the best dose. The results are shown in Fig. 5B. In order to determine the silencing sequence of Si-GluR2, RT-qPCR was used to detect the level of GluR2 mRNA in HBMECs. The results showed that compared with the NC group, there was no significant difference in GluR2 mRNA level between the Si-NC group and the Lipo 2000 group, indicating that neither the silencing sequence nor the transfection reagent Lipo 2000 affected GluR2 mRNA level. In contrast to the Si-NC group, the GluR2 mRNA levels in the Si-GluR2-1, Si-GluR2-2 and Si-GluR2-3 groups were considerably reduced (P < 0.01). The findings are displayed in Fig. 5C. Western Blot analysis was used to determine the amount of GluR2 protein expression in HBMEC, and the results were similar to those of RT-qPCR. The above experimental results showed that the silencing sequence and the transfection reagent Lipo 2000 have no effect on the expression of GluR2. Sequence 1,2,3 can silence GluR2, and sequence 1 has the highest silencing efficiency. Therefore, sequence 1 was chosen for further testing. The results are shown in Fig. 5D. Cell viability and LDH activity were further detected after GluR2 silencing. The findings indicated that in contrast to the NC group, the cell viability of the M group was significantly decreased (P < 0.01), while the LDH activity in the cell supernatant was significantly increased (P < 0.01). As opposed to the M group, the cell viability of 5-HMF group was greatly increased (P < 0.01), and the LDH activity was significantly decreased (P < 0.01). After adding Si-GluR2, the improvement effect of 5-HMF disappeared. The results showed that 5-HMF may improve the survival rate and LDH activity of HBMECs induced by OGD-R by regulating the expression of GluR2. These results are shown in Fig. 5E and F. The above results showed that 5-HMF could increase the expression level of GluR2 protein in animal and cell models, and the specific fragments of silenced GluR2 were identified. After silencing GluR2, the effect of 5-HMF on cell viability and LDH expression was inhibited.

Fig. 5.

The effect of 5-HMF on GluR2 protein in MCAO rats and OGD-R induced HBMECss and the effect of 5-HMF on cell viability and LDH activity of HBMECs after OGD-R induced silencing of GluR2. (A) Western blot was employed to find GluR2 protein expression in MCAO rats (n = 3). In contrast to the Sham group. ∗∗P < 0.01; Compared with MCAO group. ^##P < 0.01. (B) Utilizing a Western blot, the expression of GluR2 protein in HBMECs induced by OGD-R (n = 3). Compared with the NC group, ∗∗P < 0.01; Compared with the M group. ^##P < 0.01. (C) GluR2 expression was found using RT-qPCR (n = 3). (D) Western blot analysis was used to identify GluR2 expression and the measurement of GluR2 (n = 3). In contrast to the Si-NC group. ∗∗P < 0.01. (E) Cell viability of HBMECs (n = 6). (F) LDH viability of HBMECs (n = 6). In contrast to the NC group. ∗∗P < 0.01; In contrast to the M group. ^##P < 0.01; In contrast to the 5-HMF group. ^^P < 0.01.

3.6. Effect of 5-HMF on mitochondrial function of HBMECs induced by OGD-R

The levels of Ca²⁺ and glutamate in OGD-R-induced HBMECs were detected. The experimental results are shown in Fig. 6A. Contrary to the NC group, the levels of Ca²⁺ and glutamate in M group were considerably increased (P < 0.01). In comparison with the M group, the levels of Ca²⁺ and glutamate in 5-HMF group were markedly decreased (P < 0.01). Compared with the 5-HMF group, the levels of Ca²⁺ and glutamate were greatly increased after the addition of Si-GluR2 (P < 0.01) 5-HMF may reduce the levels of Ca²⁺ and glutamate in HBMECs induced by OGD-R by regulating the expression of GluR2. Flow cytometry was used to detect the level of apoptosis. The results showed that compared with the NC group, the early, late and total apoptosis rates of cells in the M group were considerably increased (P < 0.05 or P < 0.01), and the cell survival rate was decreased (P < 0.05). As opposed to M group, the early, late and total apoptosis rates of 5-HMF group were noticeably decreased (P < 0.05 or P < 0.01), and the cell survival rate was increased (P < 0.05). In comparison with the 5-HMF group, the apoptosis level was significantly increased after the addition of Si-GluR2 (P < 0.01), as shown in Fig. 6B, suggesting that 5-HMF may reduce the apoptosis level of HBMECs induced by OGD-R by regulating the expression of GluR2. The mitochondrial membrane potential and ROS levels were detected by flow cytometry. As shown in Fig. 6C and D, the experimental results showed that compared with the NC group, the mitochondrial membrane potential of the M group was markedly decreased (P < 0.01), while the ROS level was significantly increased (P < 0.01). Compared with the M group, the mitochondrial membrane potential of the 5-HMF group was significantly increased (P < 0.01), and the ROS level was greatly decreased (P < 0.01). Contrary to the 5-HMF group, the mitochondrial membrane potential and ROS level were significantly improved after adding Si-GluR2 (P < 0.01), suggesting that 5-HMF may improve the mitochondrial membrane potential and ROS level of HBMECs induced by OGD-R by regulating the expression of GluR2. The expression level of mitochondrial dynamics protein was detected by immunofluorescence. The experimental results are shown in Fig. 6E. Compared with the NC group, the fluorescence intensity of p-DRP1 (S637) and MFN2 was considerably decreased (P < 0.01), while the fluorescence intensity of MFF was noticeably increased (P < 0.01). In contrast to the M group, the fluorescence intensity of p-DRP1 (S637) and MFN2 was significantly increased (P < 0.01), and the fluorescence intensity of MFF was markedly decreased (P < 0.01). In comparison with the 5-HMF group, the improvement effect of 5-HMF on mitochondrial dynamics related indicators decreased after the addition of Si-GluR2. The above results indicate that 5-HMF may alleviate OGD-R-induced mitochondrial dysfunction in HBMECs by regulating the expression of GluR2.

Fig. 6.

Effects of 5-HMF on mitochondrial function of HBMECs induced by OGD-R. (A) Flow cytometry analysis of Ca²⁺ level and glutamate levels in HBMECs induced by OGD-R (n = 3/6). (B) Analysis of the degree of apoptosis in HBMECs caused by OGD-R using flow cytometry (n = 3). (C) Flow cytometry analysis of the mitochondrial membrane potential in HBMECs induced by OGD-R (n = 3). (D) Flow cytometry analysis of ROS in HBMECs induced by OGD-R (n = 3). (E) Outcomes of high content analysis of mitochondrial dynamics related indicators (Magnification factor 200 × )(n = 3). Compared with NC group. ∗P < 0.05,∗∗P < 0.01; Compared with M group. ^#P < 0.05, ^##P < 0.01; Compared with 5-HMF group. ^^P < 0.01.

3.7. The effect of 5-HMF on OGD-R-induced angiogenesis in HBMECs

The effect of 5-HMF on the migration of HBMECs induced by OGD-R was detected. The experimental results are shown in Fig. 7A–C. In contrast to the NC group, the cell migration ability and the fluorescence intensities of CXCL12 and CXCR4 in the M group were markedly decreased (P < 0.01). In opposition to the M group, the cellmigration ability and the fluorescence intensities of CXCL12 and CXCR4 in the 5-HMF group were significantly increased (P < 0.01). Contrary to the 5-HMF group, the cell migration ability was greatly decreased after the addition of Si-GluR2, suggesting that 5-HMF may improve the migration ability of HBMECs induced by OGD-R by regulating the expression of GluR2. The results of cell adhesion assay showed that compared with the NC group, the adhesion ability between cells and matrigel and the fluorescence intensity of adhesion-related proteins ICAM-1 and VCAM-1 in the M group were noticeably decreased (P < 0.01). As opposed to the M group, the cell adhesion ability and the fluorescence intensity of ICAM-1 and VCAM-1 in the 5-HMF group were considerably increased (P < 0.01). In comparison with the 5-HMF group, the cell adhesion ability was markedly decreased after the addition of Si-GluR2, as shown in Fig. 7D and E, suggesting that 5-HMF may improve the adhesion ability of HBMECs induced by OGD-R by regulating the expression of GluR2. The tube formation assay was used to detect the angiogenesis ability of HBMECs induced by OGD-R, and the expression of angiogenesis markers CD31 and α-SMA was detected by high-content cell imaging technique. The experimental results are shown in Fig. 7F–H. Contrary to the NC group, the angiogenesis ability of the cells in the M group decreased, and the fluorescence intensity of angiogenesis blood-related protein CD31 and α-SMA were noticeably decreased (P < 0.01). In contrast to the M group, the angiogenesis ability of the cells in the 5-HMF group was increased, and the fluorescence intensities of angiogenesis blood-related protein CD31 and α-SMA were significantly increased (P < 0.01); in opposition to the 5-HMF group, the ability of cell angiogenesis was greatly reduced after the addition of Si-GluR2. The above results indicate that 5-HMF may improve the angiogenesis ability of HBMECs induced by OGD-R by regulating the expression of GluR2.

Fig. 7.

The effect of 5-HMF on OGD-R-induced angiogenesis in HBMECs

(A) Scratch test results figure (Magnification factor 200 × ). (B) Utilizing high content cell imaging, CXCL12 and CXCR4 expression was found (Magnification factor 200 × ). (C) Mobility quantification and fluorescence intensity quantification of CXCL12 and CXCR4 (n = 3). (D) Using high content cell imaging, ICAM-1 and VCAM-1 expression was found (Magnification factor 200 × ). (E) Cells adhesion rate and fluorescence intensity quantification of ICAM-1 and VCAM-1 (n = 3). (F) Matrigel was used to detect the tube formation ability of HBMECs induced by OGD-R (Magnification factor 100 × ). (G) The expression of CD31 and α-SMA was detected using high content cell imaging (Magnification factor 200 × ). (H) Fluorescence intensity quantification of CD31 and α-SMA (n = 3). Compared with NC group. ∗∗P < 0.01; In contrast to the M group. ^##P < 0.01; Compared with 5-HMF group. ^^P < 0.01.

4. Discussion

Ischemic stroke is mainly a neurological disorder caused by insufficient blood supply to the brain caused by cerebral artery disease. The fifth most common cause of mortality worldwide is stroke. The high incidence and high disability rate of this disease have caused a huge burden on the social medical system. Although China's social health conditions have developed rapidly and some progress has been made in the fight against stroke, the stroke death rate continues to rise annually with the aging of the population, and the prevention and treatment of stroke remain extremely challenging [11].

Modern pharmacological studies have shown that there are two common pathogenesis of ischemic stroke, namely endothelial cell injury and emboli. Endothelial cell damage is usually caused by adverse external stimuli. For example, nicotine in tobacco irritates the vascular intima, leading to atherosclerosis. Substances such as fat, cholesterol and protein in blood vessels continue to accumulate in atherosclerotic plaques. Plaques are constantly squeezed by blood flow. Once the plaque ruptures, it will quickly cause platelet aggregation, promote thrombosis, and lead to the occurrence of ischemic stroke. Another mechanism of ischemic stroke is embolism. Most of the emboli are blood clots generated by heart diseases such as atrial fibrillation and heart valve disease, which reach the brain with blood circulation and block the blood vessels in the brain [12]. Brain tissue is damaged by ischemia and hypoxia, forming an ischemic core. Whether ischemic stroke is caused by endothelial injury or embolism, it is closely related to thrombus formation.

Therefore, the clinical therapy of ischemic stroke is mainly from the perspectives of thrombolysis, anticoagulation, and neuroprotection. The existing treatment methods are severely limited due to the narrow therapeutic window and obvious side effects, which hinders their wide application [13]. At the same time, patients with stroke usually suffer from limb dysfunctions such as hemiplegia, mouth and eye deviation, weakness and numbness of hands and feet, unclear language disorders, and emotional disorders such as slow thinking, depression, and depression. Patients' quality of life significantly declines as a result of these after effects [14]. Therefore, the research and development of new treatment methods and therapeutic drugs for ischemic stroke is imminent. At present, the research and development of new drugs is still facing great challenges. As a splendid treasure left by the long history of mankind, traditional Chinese medicine was gradually developed and enriched in the long-term struggle between the working people and diseases in ancient China. The application of traditional Chinese medicine has a good guiding role in the research and development of modern drugs [15].

GluR2 is one of the subtypes of AMPA ionotropic glutamate receptors, which regulates the transmission of glutamate by regulating the permeability of cell membrane to Ca²⁺. Literature studies have shown that up-regulation of GluR2 is involved in post-stroke neuroprotection and may be a novel target for stroke therapy [16,17]. Can 5-HMF also prevent ischemic stroke by regulating GluR2 expression? In order to solve this problem, the animal MCAO model was used to explore the efficacy of 5-HMF. The results displayed that 5-HMF could enhance the motor coordination ability of MCAO rats, lower the score for neurological function, lessen the cerebral infarction's area, and improve the damage of cerebral cortex and neuronal cells, suggesting that 5-HMF has the effect of anti-ischemic stroke. In MCAO rats, 5-HMF can simultaneously enhance GluR2 expression, decrease GluR2 expression, and increase Ca²⁺ influx. Intracellular Ca²⁺ overload can trigger the apoptosis of mitochondrial pathway, promote the production of mitochondrial ROS and lead to mitochondrial dysfunction. The brain is the organ most dependent on mitochondrial energy supply. Mitochondrial dysfunction seriously affects the survival and normal physiological functions of brain cells. The results displayed that 5-HMF reduced the degree of intracellular Ca²⁺ in MCAO rats, reduced apoptosis and ROS expression, alleviated mitochondrial dysfunction and raised the degree of mitochondrial membrane potential. To find out more about the role of GluR2 in ischemic stroke, the silencing sequence of GluR2 was added to the cell experiment to detect mitochondrial function and angiogenesis-related indicators. Mitochondria are highly dynamic organelles that maintain their shape, subcellular distribution, and function by dividing, fusing with adjacent mitochondria, migrating to energy-demanding regions, and removing dysfunctional mitochondria [[18], [19], [20]]. For mitochondrial biogenesis and the preservation of mitochondrial health, the ratio of fission to fusion must be balanced, and is highly correlated with the pathological processes of various central nervous system diseases, such as stroke and other neurodegenerative diseases [21]. DRP1 is a mitochondrial fission dynein that catalyzes mitochondrial fission. Calcium overload promotes dephosphorylation of the S637 residue of DRP1, facilitates DRP1's location on the mitochondrial surface, and accelerates mitochondrial division. Mitochondrial excessive fission is associated with ischemic stroke [18,22]. The stable recruitment of DRP1 to mitochondria depends on MFF, and the overexpression of MFF leads to fragmentation of the mitochondrial network [23]. MFN2 mediates mitochondrial fusion. Endothelial cell migration driven by VEGF and the capacity to construct vascular structures on Matrigel are both decreased when MFN2 is silenced in endothelial cells [24].

The results revealed that 5-HMF increased the levels of p-DRP1 (S637) and MFN2 by regulating GluR2 expression, reduced MFF expression, and alleviatede mitochondrial dysfunction of HBMECs induced by OGD-R. The act of creating new blood vessels as an extension of preexisting ones, known as angiogenesis, enhances the supply of nutrients and oxygen to tissues. In the process of angiogenesis, endothelial cells need to transform to the phenotype of proliferation and migration [25]. The homeostasis of endothelial cell mitochondria determines the function and fate of vascular endothelial cells to a certain extent [26,27]. The results of angiogenesis experiments showed that 5-HMF promoted the migration and adhesion of endothelial cells by regulating the expression of GluR2 and improved the angiogenesis of endothelial cells. In summary, 5-HMF can alleviate brain injury of ischemic stroke by regulating the expression of GluR2, reducing the concentration of intracellular Ca²⁺, improving mitochondrial function and promoting endothelial cell angiogenesis.

5. Conclusion

5-HMF can alleviate brain damage in MCAO rats and reduce the damage caused by OGD-R to HBMECs by regulating GluR2 expression, which in turn reduces intracellular Ca²⁺ concentration, improves mitochondrial function, and promotes angiogenesis, as shown in Fig. 8.

Fig. 8.

The mechanism of action of 5-HMF against ischemic stroke.

CRediT authorship contribution statement

Yan Zhang: Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis. Peipei Yuan: Writing – review & editing, Methodology, Investigation, Formal analysis. Yaxin Wei: Visualization, Methodology, Formal analysis. Saifei Li: Writing – review & editing, Data curation. Lirui Zhao: Visualization, Data curation. Qingyun Ma: Resources, Data curation. Yiran Huo: Resources, Data curation. Xiaoke Zheng: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization. Weisheng Feng: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Data availability statement

The original contributions presented in the study are included in the manuscript, further inquiries can be directed to the corresponding authors on request.

Funding

This study was supported by the National Key Research and Development Project (2019YFC1708802, 2017YFC1702800), the National Natural Science Foundation of China (82405238), the Central Government Guide Local Science and Technology Development Funds ([2016]149), the Major Science and Technology Projects in Henan Province: Study on the key technology for quality control and the key characteristics of Rehmannia glutinosa, Dioscorea opposita Thunb. and Achyranthes bidentata Blume from Henan Province (171100310500), and the Ph.D. Research Funds of Henan University of Chinese Medicine (RSBSJJ2018-04).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.