Magnetic targeting enhances the neuroprotective function of human mesenchymal stem cell-derived iron oxide exosomes by delivering miR-1228-5p

Abstract

Background

Treating mitochondrial dysfunction is a promising approach for the treatment of post-stroke cognitive impairment (PSCI). HuMSC-derived exosomes (H-Ex) have shown powerful therapeutic effects in improving mitochondrial function, but the specific effects are unclear and its brain tissue targeting needs to be further optimized.

Results

In this study, we found that H-Ex can improve mitochondrial dysfunction of neurons and significantly enhance the cognitive behavior performance of MCAO mice in OGD/R-induced SHSY5Y cells and MCAO mouse models. Based on this, we have developed an exosome delivery system loaded with superparamagnetic iron oxide nanoparticles (Spion-Ex) that can effectively penetrate the blood-brain barrier (BBB). The research results showed that under magnetic attraction, Spion-Ex can more effectively target the brain tissue and significantly improve mitochondrial function of neurons after stroke. Meanwhile, we further confirmed that miR-1228-5p is a key factor for H-Ex to improve mitochondrial function and cognitive behavior both in vivo and in vitro. The specific mechanism is that the increase of miR-1228-5p mediated by H-Ex can inhibit the expression of TRAF6 and activate the TRAF6-NADPH oxidase 1 (NOX1) pathway, thereby exerting protective effects against oxidative damage. More importantly, we found that under magnetic attraction, Spion-Ex exhibited excellent cognitive improvement effects by delivering miR-1228-5p.

Conclusions

Our research found that H-Ex has a good therapeutic effect on PSCI by increasing the expression of miR-1228-5p in PSCI, while H-Ex loaded with Spion-Ex exhibited more excellent effects on improving mitochondrial function and cognitive impairment under magnetic attraction, which can be used as a novel strategy for the treatment of PSCI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-024-02941-3.

Introduction

Post-stroke cognitive impairment (PSCI) is one of the most common complications among stroke survivors, which is mainly manifested as memory decline and affects about one-third of stroke patients. It not only significantly increases the disability rate and mortality rate of patients [1], but also brings high nursing costs. Although many studies have tried to treat PSCI, the improvement of most patients’ conditions is still not ideal [2]. Therefore, it is crucial to explore new treatment methods and clarify their specific mechanisms for the treatment of PSCI [3].

Mitochondrial dysfunction refers to defects or abnormalities in the energy supply process of mitochondria, leading to cellular energy metabolism disorders, thereby causing an increase in intracellular oxidative stress and the production of a series of harmful substances such as free radicals [4]. These harmful substances will further damage cells and lead to adverse outcomes [5]. Studies have shown that mitochondrial dysfunction plays a key role in neuron death induced by ischemic stroke [6]. The increase in oxidative phosphorylation metabolism and phagocytosis in neurons may jointly promote brain tissue remodeling and neural function recovery [7]. In mPFC photo-thrombotic-induced PSCI mice, improving mitochondrial homeostasis can promote the expression of synaptic markers and the improvement of cognitive impairment [8]. In addition, in PSD mice, electroacupuncture treatment improves depression behavior and cognitive impairment by activating CB1R to promote mitochondrial biogenesis [9]. Therefore, targeting mitochondrial dysfunction may provide a new direction for the treatment of PSCI [10].

In recent years, exosomes derived from human umbilical cord mesenchymal stem cells (HuMSC-Exo, H-Ex) have been fully proven to cross the blood-brain barrier (BBB) and target the active ingredients (such as proteins, miRNAs, etc.) of HuMSCs to the damaged areas of the central nervous system, becoming a research hotspot in the field of neurotherapy [11]. Multiple studies have reported that H-Ex plays a therapeutic role in neurological diseases (Parkinson’s disease, Alzheimer’s disease) by improving mitochondrial dysfunction [12], revealing its good reversal effect on mitochondrial dysfunction. However, in practical in vivo applications, due to the significant loss of H-Ex by the liver, its brain targeting is greatly reduced, seriously reducing the therapeutic effect in vivo. It is worth noting that exosomes, as naturally formed vesicles, have excellent modifiability [13]. Magnetic materials can develop and synthesize Ex-like delivery nano-platforms through surface modification strategies to increase the accumulation of Ex in target organs [14]. Superparamagnetic iron oxide nanoparticles (Spions) are a kind of nanomaterial with superparamagnetism, excellent biocompatibility, and high saturation magnetization, which can effectively bind to Exo and enhance the migration behavior and organ-specific accumulation of magnetized Exo under the action of magnetic force [15]. Therefore, Spions modification may further improve the targeting of Exo by enhancing magnetization, and may have a more significant improvement effect on mitochondrial dysfunction and cognitive impairment in PSCI.

In this study, we have demonstrated the excellent effect of H-Ex on improving mitochondrial dysfunction and correcting cognitive impairment in MCAO mice in both in vitro and in vivo ischemia-hypoxia models [16]. We further constructed an exosome-delivery system loaded with superparamagnetic iron oxide nanoparticles (Spion-Ex) [17]. We aimed to prove that Spion-Ex could further enhance the brain targeting and therapeutic effect of H-Ex, and to elucidate the specific mechanism of H-Ex treatment for cognitive impairment caused by MCAO, providing new treatment methods and targeted optimization strategies for the treatment of PSCI.

Materials and methods

Subjects and sample preparation

Blood samples from healthy control (HC) subjects and PSCI patients were recruited from the Affiliated Hospital of Jiangsu University. The clinical characteristics of all participants in this study are provided in Table 1. The study was approved by the Ethics Committee of the Affiliated Hospital of Jiangsu University, Zhenjiang, China, and written informed consent was obtained from all participants or their legally authorized representatives (JDFY2021041).

Table 1.

Patient information for this study

Parameters	PSCI (n = 40)	HC (n = 40)	P value
Age(year)	56.59 ± 4.34	55.96 ± 5.22	0.561
Gender (male/female)	29/11	26/14	0.324
Education(years)	9.18 ± 1.90	8.58 ± 2.00	0.177
NIHSS	7.42 ± 1.51	/	/
MoCA	16.83 ± 1.82	27.27 ± 0.91	< 0.001

The inclusion criteria were as follows: (i) participant’s age of 40–65 years (ii) stroke patients in accordance with the diagnostic criteria established by the fourth National Cerebrovascular Disease Academic Conference in 1995 confirmed by a brain CT or MRI (iii) a National Institutes of Health Stroke Scale (NIHSS) score ≤ 8 and Patients with a MoCA score < 26. The exclusion criteria were as follows: (i) patients who presented significant neurological deficits such as drowsiness, aphasia, or limb weakness and were, therefore, unable to complete the cognitive function test or contraindication to MRI (ii) patients with a history of seizures and obvious cognitive impairment before stroke, mental disorders or significant emotional problems (iii) any neuropsychiatric comorbidity and affective disorder that could influence the test outcomes (iiii) any other factors that could affect cognitive assessments and treatments.

Cell culture and establishment of the oxygen-glucose-deprived reoxygenation (OGD/R) model

SHSY5Y cells obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The SHSY5Y cells were nurtured in DMEM (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin/streptomycin (Gibco, USA). Human umbilical cords were obtained from the Fourth Affiliated Hospital of Jiangsu University. The HuMSC were cultured in α-MEM (Hyclone, USA) supplemented with 10% FBS (Gibco, Australia), and cells between the 3rd and 5th generations were used for subsequent experiments. All cells were cultured with 5% CO2 and 95% air at 37 °C.

When the SHSY5Y cells reached 80% adherence, the culture medium was substituted with serum-free and glucose-free DMEM and placed into a tri-gas incubator (Heal Force Bio-meditech Holdings, China) for 6 h, After that, The SHSY5Y cells were incubated in normal complete medium for 12 h to simulate the reperfusion.

Synthesis of Spion-Ex

Briefly, 10 mL Fe3O4 aqueous solution and 2 mL of dopamine aqueous solution (5.6 mg/mL) were added to 40 mL trimethylol aminomethane aqueous solution (10 mm) with pH = 8.5. The solution was placed on a shaking table at room temperature for 4.5 h and then removed. In order to remove large-sized impurities, the solution first centrifuge at 2000r for 5 min to remove the supernatant and remove the precipitation. The supernatant was centrifuged at 8000r for 15 min to precipitate, and finally homogeneous poly-dopamine-coated Fe3O4 was obtained. MSCs grown to 80% confluency were incubated with Fe3O4 (50 µg/mL) for 16 h, washed three times with PBS, and the distribution of Spion-Ex was observed under electron microscope.

H-Ex identification and identification

The HuMSC of passages 3–6 were cultured with Ex-free serum for 48 h. The supernatant was collected and centrifuged at 3000×g for 20 min to remove cells. The resulting supernatant were transferred to the ultrafiltration device (Millipore, USA) and then centrifuged at 3000×g at 4 °C for 50 min. The supernatant was collected and mixed with total Ex isolation kit (Umibio, UR52121) for 2 h, followed by centrifugation at 10,000 g at 4 °C for 1 h. Finally, the precipitates was resuspended in 50 µL PBS (Gibco, USA) and stored at − 80 °C.

The Ex morphology was observed under transmission electron microscopy (TEM) (HITACHI H-7000FA, Japan) The Ex particle size were analyzed by nanoparticle tracking analysis (NTA) (Particle Metrix, Meerbusch, Germany). The surface markers of H-Ex were determined by western blot analysis.

The uptaking of Ex was observed by confocal microscopy (LSM880, Zeiss, Germany). After incubated with PKH26 (Sigma-Aldrich, St. Louis, Mo, USA) at room temperature for 30 min, Ex were incubated with SHSY5Y for 24 h. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Gibco, USA), and cells were observed under confocal laser scanning microscopy (Leica, Germany).

Labelling and internalization of H-Ex and Spion-Ex

H-Ex and Spion-Ex were labelled with fluorescent 1,1’-dioctadecyltetramethyl indotricarbocyanine iodide (DiR) (D12731, Invitrogen, Life Technologies) according to the manufacturer’s recommendations. Briefly, H-Ex and Spion-Ex were incubated in 5 µM DiR for 30 min at 37 °C and were then ultracentrifuged (12000×g, 30 min) to remove unbound dye. Then the pellet was then resuspended in PBS. Subsequently, DiR-H-Ex and DiR-Spion-Ex were injected via the tail vein of mice respectively. In vivo fluorescence imaging machine (Xenogen IVIS Spectrum, USA) was used for fluorescence intensity analysis the uptake of mice on 0, 6, 12, 24, 48, 72 h respectively.

Cell counting kit-8 (CCK-8) assay

Cell viability was determined using Cell Counting Kit-8 (Gibco, USA) assays. Briefly, SHSY5Y were seeded in 96-well plates. After overnight incubation, CCK-8 solution per well was added to the growing cells and were incubated for an additional 1–4 h. Cell growth was analyzed by measuring the optical density (OD) at 450 nm using a microplate reader (Bio-Rad Laboratories Inc, USA).

Quantitative real-time PCR

Total RNA was extracted by Trizol reagent (ThermoFisher, USA), and transcribed to cDNA with the First-Strand cDNA Synthesis Kit (Vazyme, China). The qRT-PCR was implemented via SYBR Green (Vazyme, China). qPCR primers were showed in Table 2. The 2 − ΔΔCt method was used to calculate the relative quantification of the target gene.

Table 2.

Primers for quantitative RT-PCR

Gene	Forward	Reverse
MiR-1228-5p	5’-ATTATAATATGTGGGCGGGGGCAG-3’
TRAF6	5’-GAGACAGGTTTCTTGTGACAAC-3’	5’-TGGCAACCAAAAGTACTGAATG-3’
NOX1	5’-GGAATTAGGCAAACTGGGTTTT-3’	5’-CAGGTGGCCTTGTCAAAGTTTAA-3’
HiF1α	5’-GCACAGTTACAGTATTCCAGCAGAC-3’	5’-ATCAGTGGTGGCAGTGGTAGTG-3’
β-actin	5’-GTGCTATGTTGCTCTAGACTTCG-3’	5’-ATGCCACAGGATTCCATACC-3’
U6	5′-CTCGCTTCGGCAGCACA-3′
Universal reverse	5′-GTGCAGGGTCCGAGGT-3′

Cell transfection and coculture system

The Lipofectamine 2000 reagent (Millipore, USA) was used for transfection as described in our previous experiments [18]. MSCs were transfected with miRNA inhibitor (Ribobio, China) or negative controls (NC) and then collected H-Ex for H-Ex-INH or H-Ex-INH NC groups. The H-Ex-INH or H-Ex-INH NC were cocultured with OGD/R-induced SHSY5Ycells for subsequent experiments. The MSCs were transfected with Cy3-labeled miR-1228-5p mimic and then cocultured with OGD/R-induced SHSY5Y cells for subsequent experiments. The sequences were showed in Table 3.

Table 3.

Sequences of the Inhibitor, mimics and si-TRAF6 and the corresponding negative control

Gene	Forward	Reverse
Inhibitor	5’-CACACACCUGCCCCCGCCCAC-3’
Negative control	5’-UUCUCCGAACGUGUCACGUTT-3’	5’-ACGUGACACGUUCGGAGAATT-3’
Mimics	5’-GUGGGCGGGGGCAGGUGUGUG-3’	5’-CACACCUGCCCCCGCCCACUU-3’
Inhibitor NC	5’-CAGUACUUUUGUGUAGUACAA-3’
Si-TRAF6	5’-GAGACAUCUUGAGGAUCAUTT-3’	5’-AUGAUCCUCAAGAUGUCUCTT-3’

Western blot

Proteins from SHSY5Y and tissues were extracted and determined for concentration, and then heated at 95 °C for 5 min with loading buffer. After that, proteins were resolved by 10% SDS-PAGE gels and transferred to the PVDF membrane (Merck KGaA, Germany). Subsequently, the membranes were sealed with 5% skim milk and then nurtured with the following primary antibodies at 4 °C overnight: anti-TRAF6 (1 : 1500, ab40675, Abcam) and anti-NOX1 (1 : 2000, 27035,abcept). And then incubating the membranes with goat anti-rabbit antibody 1:2000 (bc002409, proteintech, China) at 37 °C for 2 h, after washing which with tris-buffered saline tween (TBST) 3 times for 10 min. The gray levels of the protein bands were examined by Image J software.

ROS staining assay

Superoxide production was detected using 2’,7’-dichlorofluorescein diacetate (S0033S, Beyotime, China). In brief, SHSY5Y cells were seeded into 6-well plates, washed by PBS and then stained with DCFDA (10 µM) for 30 min under the dark. Next, fluorescence microscope (Leica, Germany) was used to examining the fluorescence of DCFDA.

Malondialdehyde

The levels of MDA were measured using commercial assay kits (Gibco, USA) according to the manufacturer’s instructions. The color reaction was measured at 532 nm and 600 nm.

Mitochondrial membrane potential (MMP)

The alterations in mitochondrial membrane potential (Ψm) was measured by JC-1 detection kit (Gibco, USA). In brief, after appropriate treatment, SHSY5Y cells were stained with JC-1 (10 µg/ml in medium) at 37 °C for 20 min in the darkness. Subsequently, cells were washed twice with JC-1 staining buffer and the fluorescence signal was analyzed by fluorescence microscope (Leica, Germany).

Determination of ATP level

ATP levels were determined with an ATP assay kit (Abcam, USA) according to the manufacturer’s instructions. Briefly, the cells were incubated with ATP substrate solution for 5 min in darkness. The luminescence of each well was determined with a Tecan Infinite M200 microplate reader (Tecan, USA). ATP contents were measured in triplicate.

Oxygen consumption rate (OCR)

the SHSY5Y cells were seeded in Seahorse XFp cell cultured miniplates and subject to specific treatments for 24 h. OCR was acquired in basal conditions by XFp Extracellular Flux Analyzer (Seahorse Biosciences, United States) followed by in the presence of 1 µM Oligomycin, 500 nM carbonylcyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP) and 0.5 μm as rotenone/antimycin A manufacturer’s instructions.

EDU assay

SHSY5Y cells were incubated with EDU for 24 h in a constant temperature incubator and discard it. After washing third with PBS, add paraformaldehyde (4%) at room temperature for 15 min. Add glycine to decolor, and wash the cells with PBS. Under dark conditions, stain with Apollo staining reaction solution (Ruibo, China) for 30 min. Add penetrant (PBS containing 0.5% TritonX-100) to permeate the cells, and then add 1×Hoechst33342 reaction solution incubate for 30 min. EDU assay was conducted by a fluorescence microscope (Leica, Germany).

Immunofluorescence assay

SHSY5Y cells were fixed in 4% paraformaldehyde, blocked in 5% goat serum, and then incubated in primary antibody β3 tubulin (1:50, 5666, Cell Signaling) for 12 h followed by incubation with secondary antibody dylight 594 goat anti-rabbit IgG (1:20, SF134, Solarbio) at 37 °C for 60 min. Images were taken with confocal microscopy.

Animal and test group

C57BL/6 mice (males, 5–6 weeks old, weighing 18–25 g) were used for the experiment. The animals were maintained at room temperature and had free access to food and water. Seventy mice were individually caged and grouped into seven groups.

Middle cerebral artery occlusion model (MCAO)

A total of 70 adult C57BL/6J mice were provided from Anima Center of JiangSu University (Zhenjiang, China). All animal producers were approved by the Institutional Animal Care and Use Committee of Anima Center of Jiangsu University (Zhenjiang, China). Briefly, mice were anesthesia with 4% isoflurane. A 2-cm length of a 4 − 0 nylon monofilament (Yushun Bio Technology Co. Ltd, China) was inserted into the carotid artery to occlude the middle cerebral artery for 60 min. After 60 min of occlusion, reperfusion were performed by removing the suture. During surgery, the mice were placed on a warm pad to maintain the animal temperature at 37 °C. Triphenyl tetrazolium chloride (TTC, Gibco, USA) staining was used to assess the infarct volume of MCAO mice 72 h after surgery.

After that, the modeled mice were randomly sighed into six group: PBS injection group, H-Ex injection group, Spion-Ex injection group, Spion-Ex/MF injection group, INH NC injection group, INH injection group (200 µg Ex in 300 µL of PBS). All groups were injected by tail vein every two days, respectively. The protocol was approved by the Animal Use and Care Committee of Jiangsu University (2021310). In the Spion-Ex/MF group, the magnet helmet was applied.

In vivo magnetic resonance imaging (MRI)

The mice anesthetized with isoflurane (the induction concentration was 3%, the maintenance concentration was 2–2.5%) in air, fixed on the scanning bed, and injected with Spion-Ex through the tail vein. T2 images were acquired using a 7T MRI system (Biospec 70/30; Bruker, Ettlingen, Germany) with the following parameters: repetition time/echo time = 5000/56 ms; resolution = 256 × 256 pixels; slice thickness = 5 mm; and field of view = 30 × 30 cm2.

Nissl staining

The brain sections were dewaxed, rehydrated, and then treated with Nissl staining kit (Solarbio, China) according to the manufacturerʹs instruction. The hippocampus were visualized with an optical microscope (Olympus, Japan).

TUNEL assay

Determination of cell death was evaluated by Cell Death Detection Kit (Roche, Germany) as specified by the manufacturers protocol. Briefly, brain sections were deparaffined and rehydrated. Next, the slides were incubated with 20 g ml-1 Proteinase K working solution for 30 min. The sections were then rinsed with PBS and permeabilized in citrate buffer for 5 min and then incubated with blocking solution for 30 min. After the sections were washed with PBS for 5 min, they were incubated with a TUNEL reaction mixture for 60 min in the dark. Finally, the sections were rinsed 2 times with PBS and analyzed with a fluorescence microscope (Leica, Germany). In some sections, double label with NeuN were conducted as described above. The first antibody used was goat anti-NeuN (1:100, ab177487, Abcam). The second antibody used was donkey anti-goat IgG conjugate Texas Red (1:20, SF134, Solarbio) for 2 h at room temperature.

Hematoxylin and eosin (H&E) staining and Perls Prussian blue staining

The tissues were fixed in buffered formaldehyde solution, and embedded in paraffin, and were cut in 5 μm sections. HE staining was performed for histologic examination. Perls Prussian blue stain was used to detect the presence of iron on light microscopy.

Ultrastructural observation

The fresh hippocampus tissue (1 mm×1 mm×1 mm) was fixed in 2.5% glutaraldehyde for 2 h, and then samples were prepared after fixation, dehydration and embedding, Finally, the samples were observed under a TEM (H-7650, Hitachi, Japan).

Immunohistochemistry staining

After deparaffinization, antigen repair and blocking, the sections were incubated with primary antibodies: anti-TRAF6 (1:200, ab108319, Abcam) overnight at 4 °C, followed by secondary antibodies incubation. Then, the sections were visualized using diaminobenzidine (DAB, ZSGB-Bio) staining. Finally, sections were observed using microscope (Leica, Germany).

Neurological deficit evaluation

Neurological deficit was evaluated by the modified Neurological Severity Score (mNSS), which included motor, sensory, reflex, and balance tests. The 0 score means no deficit and 14 score means maximal deficit.

Open-field test (OFT)

The OFT was used to assess the general locomotivity and cognition of mice after MCAO within an open field cubic box (50 × 50 × 30 cm3). Mice were allowed to move freely for 10 min after habituation. To measure general activity variables, we measured the total distance explored and distance in the central area, time spent in the central area. Data were obtained and analyzed using the Techman software Behavior analyzing system.

Morris water maze (MWM) test

Cognitive function was tested with the MWM test. The circular stainless steel pool had a diameter of 100 cm, an altitude of 50 cm and 25 cm deep (Shanghai Xinsoft Information Technology Co, Ltd). The platform was divided into four equal quadrants. The mice were tested forth daily for 70 s for the first 6 d and were allowed to remain on the platform for 10 s after each test. On day 7, a probe trial was performed by removing the platform and allowing the mice to swim freely in the pool for 70 s or until the mouse arrives on the platform. The occupancy and crossing of animals in the proximity of the target quadrant were recorded. Finally, the data were exported and analyzed using Morris water maze system (Chengdu Techman Software Co, Inc).

Novel object recognition (NOR)

Short-term learning and working memory were tested with the NOR test. Each mouse was placed in a rectangular box (20 cm × 20 cm × 30 cm) for 5 min to adapt to the environment. Two identical objects were placed in the box during the training process, and mice were allowed to explore them for 5 min. Two hours later, one of the training objects was replaced with a novel object and the mouse were allowed to explore them for 5 min. The total exploration time was recorded and analyzed by the Any-maze video-tracking system.

Statistical analysis

Graph Pad Prism 8.0 software was used for statistical analyses. All data were expressed as means ± standard deviation (mean ± SD). Student’s t-test was used to assess the statistical significance between the two groups. One-way analysis of variance (ANOVA) was used to analyze differences between multiple groups, while Tukey test was used to analyze differences between two groups. Values of P < 0.05 were considered statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, and P > 0.05, not significant).

Results

Spion-Ex/MF alleviated oxidative damage and mitochondrial dysfunction in SHSY5Y cells after OGD/R

Our previous study found that H-Ex could improve neural injury, however the targeting ability of H-Ex towards brain damaged tissue was suboptimal. Thus, engineered Spion-Ex with enhanced targeting efficiency had drawn our attention. As shown in Fig. 1A, H-Ex was packaged into Spion-Ex. Next, the characteristics of H-Ex and Spion-Ex were compared. The results showed no significant changes in structure (Fig. 1B), particle size (Fig. 1C) and surface markers (Fig. S1A) between two groups, implying that the physical and chemical properties of H-Ex remain unchanged after modification. Additionally, after being stored at 4℃ at pH 7.4 for 48 h, the relative content and shape of Spion-Ex showed no difference, indicating the stability of Spion-Ex (Fig. S1B-C). Then, the functions of H-Ex and Spion-Ex were further explored. After co-culturing H-Ex or Spion-Ex with OGD/R cell model, it was observed that PKH-26-labeeled Ex were internalized by SHSY5Y cells, while Spion-Ex/MF showed a higher internalization efficiency (Figs. 1D and S1D). Moreover, the H-Ex and Spion-Ex/MF markedly improved impaired cell proliferation (Fig. S1E-F) and rescued the decrease in neurite length (Figs. 1E and S1G). The H-Ex and Spion-Ex/MF significantly reduced oxidative damage in the OGD/R cell model, as did the levels of MDA (Figs. 1F and S1H-I). The above results indicated that H-Ex have the potential to inhibit oxidative stress induced by OGD/R, and the Spion-Ex/MF showed a stronger effect. Studies show that mitochondrial dysfunction and oxidative stress play a crucial role in cognitive disorders, and highlight mitochondria as key regulators of molecular signaling and cell fate. Then, we noted that mitochondria exhibited severe swelling and cavitation after OGD/R exposure, and this effect was reversed by an admixture of H-Ex and Spion-Ex/MF (Fig. 1G). Similarly, H-Ex and Spion-Ex/MF obviously increased the MMP (Figs. 1H and S1J) and OCR (Fig. 1I) and ATP synthesis (Fig. S1K). Undoubtedly, Spions modification (Spion-Ex group) evidently increased the homing and accumulation, and this effect was further enhanced by MF (Spion-Ex/MF group). Those evidence confirmed that H-Ex partially reversed nerve damage and repaired mitochondrial dysfunction in OGD/R, and Spion-Ex/MF exerted a stronger effect by enhancing H-Ex accumulation.

Fig. 1.

Spion-Ex/MF alleviated oxidative damage and mitochondrial dysfunction in SHSY5Y cells after OGD/R. (A). Schematic illustration of preparation of Spion-Ex. (B–C). TEM imaging (B) and NTA of particle size (C) of H-Ex and Spion-Ex. Scale bar = 200 nm (D). Co-culture flow chart. (E). Total neurite length was imaged by β3-tubulin staining. Scale bar = 20 μm, n = 3 independent experiments. (F). Intracellular DCFDA staining. Scale bar = 250 μm, n = 3 independent experiments. Green: DCFDA staining. Blue: DAPI-labeled nucleus. G. Mitochondrial morphology under TEM. Scale bar = 200 nm, n = 3 independent experiments. H. JC-1 staining was used to detect the changes of mitochondrial membrane potential. Scale bar = 250 μm, n = 3 independent experiments, Red: JC-1 aggregates. Green: JC-1 monomers. I. The analysis of OCR. n = 3 independent experiments. Spion-Ex/MF: Spion-Ex under magnetic field

Spion-Ex/MF safely improved cognitive and memory function in stroke

Growing evidence has shown that stroke is associated with declines in global cognition, episodic memory and visuospatial abilities and that these symptoms (often typical symptoms of PSCI) are mediated by neurological deficits [19]. Given the therapeutic potential of Spion-Ex/MF for nerve cell damage in vitro, we investigated the effects on cognitive function and memory in vivo. First, middle cerebral artery occlusion (MCAO) [19], a widely used animal model, was adopted to simulate experimental cerebral ischemia. Seventy-two hours after reperfusion, the results of brain slices and modified neurological severity score (mNSS) data clearly demonstrated that the MCAO mouse model was successfully established in this study (Fig. S2A-B). To assess the effect of Spion-Ex/MF on cognition disorder, interventions were administered to mice via the tail vein (Fig. 2A). The total distance travelled in the open field test (OFT) did not differ among the groups (Fig. S2C), which indicates that all mice had normal motor activity. The time spent and distance travelled in the central area were greater in the H-Ex and Spion-Ex/MF group than in the PBS group (Figs. 2B and S2D). In addition, H-Ex and Spion-Ex/MF treated mice exhibited improved spatial learning and memory abilities, as evidenced by a significantly decreased latency to reach the platform, increased time spent in the target quadrant and increased number of platform region crossings in the Morris water maze (MWM) test (Figs. 2 C and S2E-F). We also observed increased preference for new objects in the New Object Recognition (NOR) test in H-Ex and Spion-Ex/MF groups (Figs. 2D and S2G). These results confirmed that Spion-Ex/MF improved ischaemic stroke-induced cognitive and memory impairment.

Fig. 2.

Spion-Ex/MF safely improved cognitive and memory function in stroke. (A) Schematic illustration of preparation of Spion-Ex. (B) Representative track traces of each respective group and time spent in the central area in the open field test, n = 10 independent experiments. (C) The representative swim path showing sample paths of mice from training trials on day 7 and crossing platform times in the target quadrant in the MWM, n = 10 independent experiments. (D) Representative traces from each group in the novel object recognition (NOR) and preference of different groups of mice for new objects on the testing day of the NOR experiment, n = 10 independent experiments. *P < 0.05,**P < 0.01,***P < 0.001,****P < 0.0001, Spion-Ex/MF: Spion-Ex under magnetic field

Spion-Ex/MF mitigated oxidative damage to neurons in vivo

There is a general consensus that the hippocampal neuronal function is vital for cognitive and memory impairment. To explore the effects of H-Ex and Spion-Ex/MF on neuronal function in hippocampus, oxidative stress in hippocampus tissues were assessed. In Nissl staining, we observed that the irregular and missing neurons in the hippocampal DG region were rescued in the treated group (Fig. S2I). Then, the cells were labelled with 2′,7′-dichlorodihydrofluorescein diacetate (DCFDA), and the resulting fluorescence signals were obviously attenuated in the H-Ex and Spion-Ex/MF-treated groups (Fig. 3A and S2H), indicating that H-Ex and Spion-Ex/MF mitigated oxidative stress. Similarly, in the TUNEL/NeuN validation experiment, the infarcted area exhibited less cellular colocalization of apoptotic cells, which indicated the potential neuroprotective effect of H-Ex and Spion-Ex/MF (Fig. 3B). Next, the mitochondria were observed that exhibited more distinct crista breakage, outer membrane destruction, and swelling in MCAO mice (Fig. 3C). In H-Ex and Spion-Ex/MF-treated MCAO mice, the mitochondria were less swollen with well-organized cristae (Fig. 3C), which verified the role of H-Ex and Spion-Ex/MF in restoring mitochondrial integrity. Synapses are key structures in neural networks involved in learning and memory in the central nervous system [20]. Further, the synaptic ultrastructure of hippocampal neurons was assessed by TEM. The synaptic ultrastructure became more normal, with more uniform synaptic vesicles and clear gaps in the H-Ex and Spion-Ex/MF-treated group (Fig. 3D). Consistent with the results of Fig. 1, the results of Fig. 3 also confirm the superior targeting therapeutic effect of Spion-Ex/MF compared with H-Ex. Taken together, these data indicated that H-Ex could improve the adverse effects of oxidative stress on the structure of hippocampal neurons, and Spion-Ex/MF amplified the therapeutic effect of H-Ex.

Fig. 3.

Spion-Ex/MF mitigated oxidative damage to neurons in vivo. (A) Representative pictures of ROS production. Scale bar = 250 μm. (B) Image representing colocalization of NeuN (green) and Tunel (red) in dentate gyrus of hippocampus brain section. (C–D). TEM images showed mitophagy (C) and (D) synaptic structures in hippocampal neuron. Scale bar = 200 nm. **P < 0.01,****P < 0.0001, Spion-Ex/MF: Spion-Ex under magnetic field

Spion-Ex/MF effectively increased the targeted delivery of HuMSC-derived exosomes to injured neurons

Magnetic H-Ex can be effectively manipulated by magnetic guidance, endowing them with exceptional targeting capabilities for treatment. To confirm whether fluorescence imaging could be used to trace Spion-Ex and whether fluorescence intensity measurements could be used to estimate brain-targeting ability, we labelled both H-Ex and Spion-Ex with the fluorescence marker DiR. The results showed that the Spion-Ex/MF group exhibited increased fluorescence signals in the ischaemic region (Fig. 4B), suggesting that MF increased the amount of H-Ex that crossed the BBB and localized to the ischaemic penumbra. Spion-based magnetic resonance imaging (MRI) visualization is a very sensitive, repetitive, noninvasive method for in vivo magnetic labelling detection [21]. As shown by the T2-weighted MR images, Spion-Ex deposited medially in the ischaemic region were hypointense, and the hypointense area was enlarged after MF treatment, whereas injured animals treated with H-Ex did not show low signal intensities (Fig. 4 C). These results suggested that Spion-Ex have remarkable BBB permeability and that MF can effectively increase the targeted delivery of Spion-Ex in vivo. Prussian blue staining also confirmed that the Spion-Ex/MF group had a greater number of iron-laden cells in hippocampal lesions (Fig. 4A). As expected, the NIRF images clearly showed that the Spion-Ex/MF group exhibited enhanced fluorescence signal intensity in the brain, and a small amount of Spion-Ex was still retained in the brain at 48 h (Figs. 4D-E and S2J), which demonstrated that Spion-Ex/MF had a marked targeting ability. Based on these findings, we concluded that Spion-Ex/MF strongly increased H-Ex-targeted therapy by enriching H-Ex in ischaemic lesions via magnetic guidance.

Fig. 4.

Spion-Ex/MF effectively increased the targeted delivery of HuMSC-derived exosomes to injured neurons. (A) Experimental design and Prussian blue reaction (B) Image of H-Ex uptake in mouse hippocampus. (C) Mice were followed by injected in H-Ex、Spion-Ex and Spion-Ex/MF, administered via tail vein injection. Coronal sections of T2-weighted images showed the labeled Spion-Ex as a hypointense area (arrowhead). (D and E). Noninvasive NIRF imaging of Dir-labeled H-Ex in mice after intravenous injection with H-Ex and Spion-Ex. Spion-Ex/MF: Spion-Ex under magnetic field

Mir-1228-5p ameliorated oxidative stress and cognitive impairment induced by ischaemic stroke

Accumulating evidence indicates that H-Ex and their constituents, particularly miRNAs, play crucial roles in the pathological process of ischemic stroke, suggesting their potential as additional biomarkers and therapeutic targets for PSCI [22]. To further investigate the role of miRNA in PSCI pathophysiology, miRNA expression in serum from HC subjects and PSCI patients was analyzed (Guangzhou Ribo Bio Co., Ltd.) (Fig. 5A). A volcano plot revealed 248 significantly differentially expressed miRNAs, 65 of which were upregulated and 183 of which were downregulated in PSCI patients (Fig. 5B). Among the miRNAs validated through qRT‒PCR, miR-1228-5p was the only one that passed qRT‒PCR validation (Fig. S3A-H), indicating its potential role as a messenger in PSCI pathogenesis. Notably, miR-1228-5p was expressed at low levels in OGD/R-induced SHSY5Y cells, which was reversed by cocultivation with H-Ex (Fig. 5 C). The OGD/R-induced SHSY5Y cells cocultured with Cy3-labeled miR-1228-5p transfected MSC cells, which internalized the Cy3-labeled miR-1228-5p, leading to an increase in miR-1228-5p levels (Fig. 5D). To probe the potential effect of miR-1228-5p on neuron repair, we manipulated the level of miR-1228-5p in OGD/R-induced SHSY5Y cells by applying corresponding miRNA inhibitors in H-Ex (Fig S3I). As expected, following transplantation of H-Ex with miR-1228-5p silencing (H-Ex-INH group), the neurite length significantly decreased, cell proliferation decreased, and the degree of oxidative stress increased in OGD/R-induced SHSY5Y cells (Fig. 5E-G and Fig. S3J-L). Moreover, the mitochondria were more fragmented and swollen than those in the control group (Fig. 5I). Interestingly, the results revealed that the MMP (Figs. 5H and S3M-O) and OCR (Fig. 5J) decreased, accompanied by a decrease in ATP (Fig. S3P). In summary, the protective effect of H-Ex against OGD/R-induced injury relies on the transport of miR-1228-5p.

Fig. 5.

miR-1228-5p ameliorated oxidative stress and cognitive impairment induced by ischaemic stroke. (A) Partial heatmap of differentially expressed serum miRNAs between PSCI patients and HC subjects. (B) The volcano maps. (C) Screening purpose miRNA. (D) MSC transfected with the CY3-miR-1228-5p mimic (red fluorescence) were plated in the upper chamber and co-incubated with SHSY5Y in the lower chamber in a coculture system with a 0.4 μm pore membrane. Red fluorescence was detected in the MSC recipient cells. (E) The proliferation. Scale bar = 250 μm, n = 3 independent experiments. (F) Total neurite length. (G) Intracellular DCFDA staining. Scale bar = 250 μm, n = 3 independent experiments. Green: DCFH-DA staining. Blue: DAPI-labeled nucleus. (H) Representative JC-1 staining pictures. Scale bar = 250 μm. n = 3 independent experiments. Red: JC-1 aggregates. Green: JC-1 monomers. (I) TEM images of mitochondrial ultrastructure. Scale bar = 200 nm. n = 3 independent experiments. (J) The analysis of OCR. n = 3 independent experiments. **P < 0.01,****P < 0.0001

Regulation of the oxidative stress pathway by miR-1228-5p by targeting TRAF6/NOX1

To predict the potential target genes of miR-1228-5p, the intersection of target genes was screened from the TargetScan database and mitochondria-associated genes. The 3’-UTR of TRAF6 mRNA was identified as a miR-1228-5p binding site (Fig. 6A). To assess the interaction between TRAF6 and miR-1228-5p, dual-luciferase reporter assays were conducted. Our findings revealed that the miR-1228-5p mimic significantly decreased the luciferase activity of the TRAF6-WT reporter, whereas no such effect was observed for the mutant reporter (Fig. 6B). Furthermore, the mRNA and protein expression levels of TRAF6 were downregulated in the mimic group and upregulated in the inhibitor group (Fig. 6 C and S4A-B), indicating that miR-1228-5p targets and negatively regulates TRAF6. The activation of the NADPH oxidase 1 (NOX1) enzyme has been shown to be responsible for TRAF6, and we subsequently examined NOX1 expression. The expression of NOX1 was positively correlated with that of TRAF6 (Fig. 6D-F and S4C-D). Therefore, we speculated that miR-1228-5p might regulate ROS production by targeting TRAF6/NOX1. To confirm the biological functions of miR-1228-5p and whether TRAF6 affects the function of miR-1228-5p, rescue assays were performed. Not surprisingly, upon treated with si-TRAF6, the cell proliferation rate and neurite length were increased, and the oxidative stress response and ROS damage levels were decreased in the SHSY5Y cells (Figs. 6G-I and S4E-I). Similarly, the mitochondria in the si-TRAF6-treated group were relatively less fragmented and swollen (Fig. 6K) and exhibited increased MMP (Fig. 6J and S4J), OCR (Fig. 6L) and ATP (Fig. S4K) levels, indicating the recovery of mitochondrial function. Together, the above data illustrated that H-Ex exerted neuroprotective effects by regulating the miR-1228-5p/TRAF6/NOX1 pathway.

Fig. 6.

Regulation of the oxidative stress pathway by miR-1228-5p by targeting TRAF6/NOX1. (A and B) Dual luciferase assays of miR-1228-5p targeting effects on TRAF6 3’-UTR. Data are presented as means ± SEM, n = 3 independent experiments. (C and D) The protein expression levels of TRAF6 and quantified. β-actin was used as a loading control. (E and F)The level of TRAF6 (E) and NOX1(F). Data are presented as means ± SEM, n = 3. (G). The proliferation. Scale bar = 50 μm, n = 3 independent experiments. (H). Total neurite length. (I). Intracellular ROS production. Scale bar = 250 μm, n = 3 independent experiments. (J). Representative JC-1 staining pictures. Scale bar = 250 μm, n = 3 independent experiments. (K). TEM images of mitochondrial ultrastructure. Scale bar = 250 μm, n = 3 independent experiments. (L). The analysis of OCR. n = 3 independent experiments. *P < 0.05,**P < 0.01,***P < 0.001, Spion-Ex/MF: Spion-Ex under magnetic field

Spion-Ex-embedded miR-1228-5p protects against memory and oxidative impairment in vivo

To elucidate whether H-Ex-miR-1228-5p could restore oxidatively damaged neural function in MCAO mice, we manipulated the level of miR-1228-5p in Spion-Ex by applying corresponding miRNA inhibitors (Fig. 7A-C and S5A-B). As expected, the miR-1228-5p inhibitor Spion-Ex (INH) significantly increased the mNSS (Fig. 7D) and resulted in a longer time spent finding the hidden platform or in the margin during training days (Fig. 7E-L). Similarly, the INH group exhibited reduced novel object preference. These results indicated that miR-1228-5p may improve neural deficits and spatial learning and memory in MCAO mice (Fig. 7 M-O). ​Immunofluorescent labelling assays demonstrated that the levels of ROS damage (Fig. 7P and S5C) and TUNEL-positive lesions (Fig. 7Q and S5D) in the hippocampal DG region were also sharply increased. Furthermore, TEM images demonstrated that the integrity of mitochondria (Fig. 7R) and synapses (Fig. 7S) was disrupted in the INH group. The results suggested that miR-1228-5p in Spion-Ex could decrease ROS damage and cognitive dysfunction in MCAO mice and may be a potential target for PSCI treatment.

Fig. 7.

Spion-Ex-embedded miR-1228-5p protects against memory and oxidative impairment in vivo. (A) The level of miR-1228-5p in the hippocampus. (B) Representative images of immunohistochemistry results for TRAF6 protein. (C) The protein expression levels of TRAF6 (B) and NOX1 (C) and quantified. (D) Neurological deficit scores, n = 10 independent experiments. (E) Representative track traces in the OFT. (F) Quantitative analysis of total distance in the OFT, n = 10 independent experiments. (G–H). Time spent (G) and distance (H) in the central area in the OFT, n = 10 independent experiments. (I). Traces on the probe trial on day 7 in the MWM. (J–L). The time spent (J), crossing platform (K) and latencies (L) times in the target quadrant in the MWM, n = 10 independent experiments. (M). The preference of different groups of mice for new objects on the training day of the NOR experiment, n = 10 independent experiments. (N). The preference of different groups of mice for new objects on the testing day of the NOR experiment, n = 10 independent experiments. (O). Representative traces in the NOR. (P–Q). Representative pictures of ROS production (P) and TUNEL staining (Q) in the hippocampus. Scale bar = 50 μm. (R–S). TEM images showed mitophagy and synaptic structures in hippocampal neuron. Scale bar = 200 nm. *P < 0.05,**P < 0.01,***P < 0.001,****P < 0.0001,Spion-Ex: exosome delivery system loaded with superparamagnetic iron oxide nanoparticles

Abnormal expression of miR-1228-5p in PSCI patients

To determine the clinical relevance of miR-1228-5p in PSCI, we explored the endogenous expression pattern of miR-1228-5p in the serum of PSCI patients. Based on clinical symptoms, the patients were identified as having ischaemic stroke, and MRI images more precisely delineated the intracranial lesions (Fig. 8A). Among these 80 samples (from PSCI patients, N = 40; from HC, N = 40), miR-1228-5p expression was significantly lower in PSCI patients than in HC, as determined by PCR (Fig. 8B). Moreover, increased TRAF6 and NOX1 protein levels were observed in the PSCI group (Fig. 8 C and S5E-F). Acute stroke is characterized by impaired oxygen and cellular metabolism, leading to ischaemic injury. Magnetic resonance spectroscopy (MRS) has been widely acknowledged as a potentially powerful tool for noninvasive metabolic imaging. By monitoring metabolic substances in the brain through MRS, we found that the NAA/Cr ratio was obviously lower in PSCI patients than in HC. However, the Cho/Cr, and Glu/Cr ratios were not significantly different between the two groups (Fig. 8D-E, P > 0.05). Spearman’s correlation analysis revealed a positive correlation between miR-1228-5p and MoCA score (Fig. 8F). Taken together, these data support the hypothesis that miR-1228-5p-dependent regulation of TRAF6 and NOX1 is associated with the prognosis of PSCI patients (Fig. 8G).

Fig. 8.

Abnormal expression of miR-1228-5p in PSCI patients. (A) The MRI representative image of patients with PSCI. (B) The miR-1228-5p levels in serum. (C) The expression of TRAF6 and NOX1. (D–E). The MRS of hippocampus in PSCI patients. (F). Positive correlation between serum miR-1228-5p levels and MoCA scores in PSCI patients (n = 40) (Pearson analysis). (G). The proposed model illustrating the property of Spion-Ex/MF. ***P < 0.001,****P < 0.0001, PSCI: Post-stroke cognitive impairment

Discussion

In this work, a BBB-targeted miRNA-based nanosystem was developed and aimed at improving cognition by promoting active brain targeting of H-Ex leading to increased neuronal uptake. In mechanism, we uncovered a previously unrecognized role for H-Ex, which could attenuate PSCI-induced oxidative damage and neurobehavioral dysfunction. These effects were primarily mediated through miR-1228-5p/TRAF6/NOX1 signaling pathway-mediated oxidative stress and mitochondrial damage induced by PSCI.

Our previous study and available study have confirmed the therapeutic effect of Ex [23]. In this study, the Ex were confirmed that hold dual therapeutic functions (promoting mitochondrial functional recovery and resisting oxidative impairment), resulted in a much better improvement in impaired cognition in PSCI. However, the desirable therapeutic effect of Ex is hindered by its poor bioavailability and subeffective concentration in brain. Consistent with this, we found that most H-Ex localized in the liver and few H-Ex accumulated at the brain after intravenous injection. Therefore, improving the targeting ability of H-Ex and increasing their application in the treatment is urgently needs to consider. Spions, the only clinically approved metal oxide nanoparticles, conjugate with Ex has many advantages, including magnetic targeted functionalization, magnetic thermotherapy, and delivery of agents. Spion-Ex have been successfully used in enhanced targeting hippocampal part of brain and significantly improved cognitive in AD [23]. In our study, the uptake of Ex in brain was significantly elevated by the coupling with the encapsulation of Spions. Moreover, Spion-decorated Ex as vehicles to deliver into the hippocampus in the presence of a MF resulted in more effective therapeutic concentration of Ex. This strategy to promote active brain targeting resulted in about 3-fold increase in the uptake by neuron, which can be further potentiated for the potential magnetic targeting in in vivo models. In addition, Spion-Ex/MF showed no toxic effect to normal organs such as impaired liver, kidney and heart functions, suggesting that we have developed a dual-functional Ex-based Spion cluster as a targeted and safe agent for therapy. Thus, it can be concluded that the Spion-Ex delivery system can significantly increase the concentration of Ex at the treatment site and decrease the needed dosage, cost of treatment and adverse effects on the body.

Previous studies showed that ROS enhancement is the primary cause of neuron injury in the process of stroke and cognition [24]. Therefore, we explored whether and how H-Ex could reduce neuroinflammation by inhibiting the oxidative stress caused by ROS production. We found that H-Ex treatment accompany with decreased ROS levels and improved cognitive. Moreover, mitochondria are the main source of ROS production, and mitochondrial dysfunction is related to the excessive increase of ROS. Our TEM results showed that the ruptured mitochondrial membrane and disappeared cristae were significantly recovered while treated with H-Ex. Interestingly, not only was the morphology of mitochondria protected by H-Ex but also its function. Thus, our results suggest that H-Ex may reduce oxidative stress and repair cognition after may improve mitochondrial function and reduce ROS production. To further explore the potential mechanism, we conducted the differential expression analysis between HC and PSCI patients, and subsequently the miR-1228-5p came into our eyes. Sonoda T et al. identified miR-1228-5p was significantly associated with stroke, and also participated in the regulation of mitochondrial homeostasis and function. Consistently, our data showed that inhibition of miR-1228-5p in H-Ex markedly rescued mitochondrial damage and decreased ROS production, which revealed miR-1228-5p as a key molecule in H-Ex treatment. It is well known that miRNAs play a biological role by regulating the expression or translation of target genes [25]. The downstream regulation target of miR-1228-5p was also detected by combining bioinformatic prediction and binding site certification TRAF6. Simultaneously, our data also confirmed that miR-1228-5p directly targets the TRAF6. In our study, we found that knockdown of TRAF6 significantly exacerbated mitochondrial dysfunction and oxidative stress. These data strongly supported the notion that H-Ex upregulated miR-1228-5p level and targeted inhibition of TRAF6 expression, thereby mediating mitochondrial damage and oxidative stress.

Collectively, these finding demonstrated that Spion-Ex/MF plays a fundamental role in neuronal mitochondrial reverse and oxidative stress by targeted enhanced delivery of miR-1228-5p to inhibit TRAF6.

Conclusions

In this study, miR-1228-5p was loaded into Spion-modified Ex to improve mitochondrial function and oxidative stress through negative regulation of TRAF6/NOX1 to reverse the cognition both in vitro and in vivo. Our study provides a favorable strategy to improve the concentration in the targeted area which is valuable for the application PSCI therapeutics. Therefore, it can provide an efficient strategy for the treatment of improve cognition with the potential for clinical translation.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Wei-Jia Hu, Hong Wei and Li-Li Cai contributed equally to this work.W-JH: data curation, investigation, methodology, writing-original draft. HW: revise the manuscript, investigation, methodology, writing-review & editing. L-LC: data curation, investigation, methodology, writing-original draft. Y-HX: investigation, methodology, writing-review & editing. QZ and RD: data curation, investigation. X-LZ and Y-FL: investigation: conceptualization, funding acquisition, investigation, project administration, supervision, writing-original draft, writing-editing. All authors read and approved the final manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All experiments were carried out in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals, and approved by the Experimental Animal Care and Use Committee of Jiangsu University (2021310). The study was approved by the Ethics Committee of the Affiliated Hospital of Jiangsu University, Zhenjiang, China, and written informed consent was obtained from all participants or their legally authorized representatives (JDFY2021041).

Consent for publication

All authors read and approved the final manuscript.

Competing interests

The authors declare no competing interests.