Animal trials received approval from the Animal Ethics Committee, Anhui University of Chinese Medicine (Hefei, China; ethical clearance number AHUCM-rats-2023126) and were conducted as per the ARRIVE instructions. Twenty-four male Sprague-Dawley rats (specific pathogen-free, SPF, Anhui Qingyuan Biotechnology Co., Ltd, Hefei, China) with a weight of 230–260 g and aged 8–9 weeks were equally assigned to four groups in a random manner: sham control, M3 (muscone 3 mg/kg), M9 (muscone 9 mg/kg), and M18 (muscone 18 mg/kg). Pharmaceutical-grade muscone (S25595, Yuanye Biotechnology, Shanghai, China) was delivered intravenously via the tail vein. Following a 24-h post-administration period, anesthetization of rats was conducted with 5% isoflurane (R510-22-10, RWD Life Science Co., Ltd, Shenzhen, China) and 2.5% isoflurane was used to maintain them. Terminal blood collection (5 mL) was performed via abdominal aortic puncture prior to euthanasia by cervical dislocation under sustained anesthesia. The blood specimens were spun at 1800 g for 15 min to isolate serum components. Hepatic and renal biomarkers (ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; Cr, creatinine) were quantified using a HITACHI 7600-020 Automatic Analyzer (Hitachi High-Technologies Corporation, Tokyo, Japan) and following standardized protocols.

Sprague-Dawley rats (n = 60) with the age of 8–9 weeks (230–260 g body weight) were classified into five experimental cohorts in a random manner: sham-operated controls, MCAO/R model group, MCAO/R with low-dose muscone (3 mg/kg), MCAO/R with high-dose muscone (9 mg/kg), and MCAO/R with edaravone (3 mg/kg) treatment. The MCAO/R model was prepared as illustrated in previous work [35, 36]. A median cervical incision was made to expose the left common (CCA), internal (ICA), and external carotid arteries (ECA). The CCA was clamped, the distal ECA ligated, and a micro-incision made at the proximal ECA. A silicone-coated thread embolus (907-00029-01, RWD Life Science Co., Ltd) was inserted via the ECA, advanced into the ICA through the CCA bifurcation, and stopped upon resistance. After securing the embolus, ischemia was sustained for 1 h, followed by thread withdrawal. The proximal ECA was ligated, the CCA clamp removed, and reperfusion initiated. A laser Doppler flowmeter (PF5010, Perimed Co., Ltd, Stockholm, Sweden) was utilized for monitoring the regional cerebral blood flow (rCBF). A $>$75% rCBF lessening throughout ischemia indicated effective middle cerebral artery occlusion, and recovery to $>$75% of baseline confirmed reperfusion. Blood flow restoration was achieved through careful suture removal, with subsequent tail vein administration of muscone. The preparation and administration steps of the MCAO/R model have been completed. The animal’s body temperature throughout the experiment and until awaking from anesthesia was about 36.5 °C.

The Longa score scale was utilized to evaluate the neurological deficit in MCAO/R rats as described previously [36]. The neurological deficit evaluation was performed at the 24th hour after blood perfusion was restored in MCAO/R rats.

Following the establishment of the AIS model, MCAO/R rats were euthanized by rapid decapitation, and their whole brains extracted, and then cutting them was conducted into slices of 2 mm thickness on ice. Staining the slices was then conducted (37 ℃, 20 min) with 2% TTC dye solution (BB-44549, Bestbio Co., Ltd, Shanghai, China) and photographed. Infarct areas were identified as pale regions in the cerebral hemisphere, and the percentage of infarct size was calculated using Image J software (v1.52a; NIH, Bethesda, MD, USA).

Total RNA extraction was conducted from both healthy and infarcted cortical tissues via Trizol reagent (15596026CN, Invitrogen, Carlsbad, CA, USA). Following library preparation and alignment, genes were classified as differentially expressed genes (DEGs) if they exhibited a $|$log2(Fold-Change)$|$ $>$1 and a p-value 0.05. Gene Ontology (GO) analysis was carried out on DEGs to detect their biological process (BP), cellular component (CC), and molecular function (MF), with heatmaps illustrating expression differences between groups.

After adequate fixation, the tissues/cells were stained with HE staining kits (C0105S, Beyotime Biotechnology, Shanghai, China). The degree of damage in the cortex of rats was then assessed via an inverted microscope (IX81, Olympus, Tokyo, Japan).

In ice-cold RIPA buffer (89900, Thermo Fisher Scientific, Waltham, MA, USA), cortical tissue and PC12 cells were lysed. Beyotime Biotechnology’s BCA kit (P0012, Shanghai, China) was used to measure total protein. After being separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), protein samples weighing 30 µg were transferred to polyvinylidene fluoride (PVDF) membranes with a thickness of 0.45 µm. Before incubating with primary and secondary antibodies, a 1-h blockage of membranes was conducted with 5% bovine serum albumin at room temperature. Located in Wuhan, China, ProteinTech is the source of the anti-Snap25 antibody (1:5000, 14903-1-AP). Abcam (Cambridge, UK) provided the following antibodies: GAPDH (1:5000, ab181602), horseradish peroxidase (HRP, 1:5000, ab6721), and ab6789, which are goat anti-rabbit and mouse IgG H&L. The enhanced chemiluminescence (ECL) substrate (34580, Pierce, Thermo Fisher Scientific) was used for band visualization. The gray value of the protein band was calculated by using the Image J software (v1.52a; NIH, Bethesda, MD, USA). The original western blots can be found in the Supplementary Material.

The PC12 cell viability was strictly assessed with the CCK-8 assay kit (C0037, Beyotime Biotechnology) as per the manufacturer’s guidelines. Subsequently, cell viability was evaluated via an enzymatic reader at 450 nm. The formula used to determine cell viability was: % cell viability = OD (treatment)/OD (control) $\times$ 100%.

PC12 cells were brought from the American Type Culture Collection (Catalogue no. CRL-1721.1TM, ATCC, Manassas, VA, USA) and cultivated in RPMI 1640 medium (Catalogue no. 11875093, Invitrogen) with 5% fetal bovine serum (Catalogue no. A5670701, Invitrogen). The cells underwent regular authentication through morphological examination and short tandem repeat (STR) authentication and were checked for Mycoplasma contamination. The mycoplasma contamination detection kit (Beyotime Biotechnology, Cat. No. C0297S) was used to test cell lines for mycoplasma, and they were found to be negative. PC12 cells were cultivated into 6-well plates (2 $\times$ 10⁵ cells/well) and then incubation was conducted at 37 °C with 5% CO₂. Before oxygen-glucose deprivation, PBS was employed to rinse the cells that were then cultivated in glucose-free RPMI 1640 medium and incubated in an anaerobic chamber at 37 °C with 5% CO₂ and 95% N₂ for 6 h. Subsequently, replacing the medium was conducted with regular medium containing different drug concentrations, and the incubation of cells was conducted under normal circumstances at 37 °C for 24 h to simulate reperfusion.

ROS concentrations in cortical tissue were assessed via a ROS detection kit (BB-470515, BestBio Co., Ltd.) as per the guidelines of the manufacturer. Fluorescence was detected at 510 nm (excitation) and 610 nm (emission) via a SpectraMax iD3 microplate reader (Molecular Devices, Sunnyvale, CA, USA). Intracellular ROS in PC12 cells was evaluated with the Cell ROS Assay Kit (S0033S, Beyotime Biotechnology). Cells (3 $\times$ 10⁴/well) were seeded in 96-well plates, and fluorescence was assessed at 485 nm (excitation) and 520 nm (emission).

The TBARS Assay Kit (S0131M, Beyotime Biotechnology, Shanghai, China) was utilized to measure lipid peroxidation in cortical tissue and PC12 cells. For this analysis, cortex tissue samples (25 mg) and cell suspensions (1 $\times$ 10⁶ cells) were homogenized or lysed, respectively, using the lysis buffer provided in the kit. Following centrifugation at 4 ℃ for 10,000 g and 10 min, the supernatant was gathered for subsequent evaluation. The supernatant, thiobarbituric acid (TBA) detection reagent, and color-developing reagent were added to a 96-well plate at a ratio of 1:1:8. After mixing, heating for 15 min at 100 ℃, and cooling rapidly to room temperature with water, the absorbance was calculated at 532 nm via a microplate reader (MK3, Thermo Fisher Scientific, Waltham, MA, USA).

The brain cortex tissue of rats, as well as cultured PC12 cells, were stabilized at 4 ℃ for 2 h with 2.5% glutaraldehyde (Catalogue no. G5882, Sigma-Aldrich, St. Louis, MO, USA). Subsequently, the mitochondrial morphology was examined by TEM (JEM-1400 PLUS, Tokyo, Japan).

The LIVE/DEAD^TM imaging kit from Thermo Fisher Scientific (L3224, Invitrogen) was used as recommended by the manufacturer to assess cell mortality across all experimental groups. PC12 cells were photographed at 100$\times$ magnification using an inverted phase microscope (IX81, Olympus).

GSH content and GPX4 activity were measured in homogenized rat cortex and PC12 cell supernatants using the GSH test kit (ab138881, Abcam) and GPX4 Activity Assay Kit (E-BC-K883-M, ElabScience Biotechnology Co., Ltd, Wuhan, China), respectively, as recommended by the guidelines of the manufacturer.

An iron content detection kit (I291, Dojindo Laboratories, Kumamoto, Japan) was used to measure the concentrations of ferrous and total iron ions in rat cortical tissue and PC12 cells, as recommended by the manufacturer. Microplate readers from Thermo Fisher Scientific were employed to calculate absorbance at 593 nm, which was subsequently used to calculate the concentrations of iron ions.

Autodock Vina 1.1.2 software (Scripps Research Institute, San Diego, CA, USA) was employed to conduct molecular docking simulation analysis. The crystal structure of Snap25 protein (Protein Data Bank (PDB) ID: 1N7S) was gained from the RCSB Protein Data Bank (https://www.rcsb.org/, USA) in PDB format. The molecular structure of muscone (ID 10947) was retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov) and transformed into PDB format with the PyMOL Molecular Graphics System (v 2.5.0, Schrödinger, Inc., New York, NY, USA). Finally, Autodock Vina software was employed to perform the docking study.

SPR detection was employed to assess the interaction between Snap25 protein (HY-P73634, MedChemExpress, Monmouth Junctio, NJ, USA) and muscone. Immobilization of the Snap25 protein was conducted on a Series S CM5 Sensor Chip (GE Healthcare, Washington, DC, USA) via amino coupling. Muscone solutions were prepared at numerous doses (0.3125, 0.625, 1.25, 2.5, 5, and 10 µM) using 1$\times$ PBS-P+ buffer containing 5% DMSO. Interaction analysis was conducted with a Biacore 1K instrument (Cytiva, Danahe Group, Washington, D.C., USA). The dissociation (K_D), association rate (Ka), and dissociation rate constants were subsequently determined via Biacore Insight evaluation software (version 3.0.11.15423, Cytiva, Danahe Group).

MD simulations were conducted via GROMACS (v 5.1.4, Uppsala University, Uppsala, Sweden). Two systems were constructed: with ligand (Snap25 protein + muscone), and without ligand (Snap25 protein only). Topological files for the molecular structures were generated via the CHARMM-GUI (http://www.charmm-gui.org) web interface. Both systems were positioned in a cubic simulation box with applying periodic boundary circumstances, using the gmx solvate tool in GROMACS (https://www.gromacs.org, GROMACS 2023.3, KTH Royal Institute of Technology, Stockholm, Sweden). Upon completion of the simulation, various metrics were computed and analyzed.

Synthetic siRNA was used to perform RNAi targeting Snap25. Three Snap25-specific siRNAs (Snap25-siRNA-1, Snap25-siRNA-2, Snap25-siRNA-3) and a negative control (NC) siRNA were synthesized by Hanbio Corporation (JY20220810WY-SI02, Shanghai, China). The siRNAs and NC sequences were as follows: Snap25-siRNA-1, UUGCUCAUCCAACAUAACCTT; Snap25-siRNA-2, UUAUUGAUUUGGUCCAUCCTT; Snap25-siRNA-3, UUCUGCUUCUUUCAUAUCCTT; NC, ACGUGACACGUUCGGAGAATT. Transfection was carried out via Lipofectamine™ 3000 (L3000015, Invitrogen, Carlsbad) with Snap25-siRNA-1, Snap25-siRNA-2, Snap25-siRNA-3, or NC siRNA (100 nM for each). PC12 cells were transfected for 24-h prior to further experiments.

Snap25 overexpression plasmids were prepared by Hanbio Corporation (JY20220811WY-PC01). The pcDNA3.1-cytomegalovirus (CMV)-Snap25 plasmids were introduced into PC12 cells via Lipofectamine™ 3000 (L3000015, Invitrogen). A transfection of 1 $\times$ 10⁵ PC12 cells was conducted with 10 µg of the Snap25 overexpression plasmid. For the NC group, a transfection of 1 $\times$ 10⁵ PC12 cells was conducted with 10 µg of the empty pcDNA3.1-CMV plasmid. The PC12 cells were transfected for 24-h prior to further experiments.

SPSS software (v 20.0) (IBM Corp., Armonk, NY, USA) was employed for the statistical analysis of data. GraphPad Prism (v 8.0, GraphPad Software, Inc., San Diego, CA, USA) was utilized to create all graphical images. The Kruskal-Wallis non-parametric test or one-way ANOVA compares two or more groups, and then post-hoc analysis with Tukey’s multiple comparisons test was conducted. A p-value of 0.05 is significant.

To select the appropriate dose of muscone for experimentation, the liver and kidney functions in rats were first evaluated by serum biochemical indicators. Liver and kidney functions showed no significant damage in the M3 and M9 groups, but substantial liver and kidney toxicity were found in the M18 group, as shown in Fig. 1A (ALT: M18 vs. sham, p 0.01; AST: M18 vs. sham, p 0.001) and Fig. 1B (BUN: M18 vs. sham, p 0.01; Cr: M18 vs. sham, p 0.01). Therefore, concentrations of 3 and 9 mg/kg were chosen for the next trials. Edaravone (3 mg/kg) was chosen as a positive control since it has been reported to ameliorate damage in MCAO/R rats [37, 38]. Neurological deficit scores and infarct volumes were significantly lessened in the M9 and Eda groups compared to the model group, as shown in Fig. 1C (p 0.05) and Fig. 1D (p 0.0001), respectively. HE staining revealed the cortical tissue space in the model group was loose, with unclear tissue structure. In contrast, cortical structures in the M3, M9 and Eda groups exhibited marked improvement (Fig. 1E). Collectively, these outcomes indicate that muscone exerts neuroprotective effects on the brain tissue of rats subjected to MCAO/R damage.

Fig. 1.

Muscone mitigates acute cerebral ischemic injury in middle cerebral artery occlusion/reperfusion (MCAO/R) rats. (A) The impact of different concentrations of muscone (M3 group, 3 mg/kg; M9 group, 9 mg/kg; M18 group, 18 mg/kg) on liver function in rats, as detected by the serum biochemical index. (B) The impact of different concentrations of muscone on kidney function in rats, as detected by the serum biochemical index. (C) Muscone enhanced the neurological deficit scores in MCAO/R rats. The protective influence of muscone on the ischemic cortex in MCAO/R rats was observed by 2,3,5-Triphenyl-Tetrazolium Chloride (TTC) staining (Scale bar = 1 cm) (D) and hematoxylin-eosin (HE) staining (Scale bar = 100 µm) (E). Data are signified as the mean $\pm$ SEM. ^*⁣*⁣**p 0.0001, ^*⁣**p 0.001, ^**p 0.01 vs. the sham group; ^#p 0.05, ^{#⁢#⁢#⁢#}p 0.0001 vs. the model group. ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; Cr, creatinine.

TEM illustrated the disrupted mitochondrial cristae and membranes in the cortical tissue of model rats, which were markedly preserved in the M3, M9, and Eda groups (Fig. 2A). GPX4 activity (Fig. 2B) and GSH levels (Fig. 2C) were significantly higher in these groups vs. the model (M3: p 0.01, p 0.05; M9: p 0.001, p 0.001; Eda: p 0.05, p 0.05). Total iron (Fig. 2D), ferric iron (Fig. 2E), ROS (Fig. 2F), and MDA levels (Fig. 2G) were significantly lower in the M9 and Eda groups vs. the model (M9: p 0.01, p 0.05, p 0.01, p 0.01; Eda: p 0.05, p 0.05, p 0.001, p 0.001). These results indicate that muscone effectively mitigates ferroptosis in the ischemic cortex of MCAO/R-injured rats.

Fig. 2.

Muscone attenuates ferroptosis in the ischemic cortex of MCAO/R rats. (A) Transmission electron microscopy (TEM) shows reduced mitochondrial damage in the cortex after muscone treatment. The red arrows indicate mitochondria. Muscone increased glutathione peroxidase 4 (GPX4) activity (B) and glutathione (GSH) levels (C), and decreased total iron (D), ferric iron (E), reactive oxygen species (ROS) (F), and malondialdehyde (MDA) concentrations (G) in MCAO/R rats. Mean $\pm$ SEM represents the data; one-way ANOVA. ^*⁣*⁣**p 0.0001, ^*⁣**p 0.001, vs. sham; ^#⁢#⁢#p 0.001, ^#⁢#p 0.01, ^#p 0.05 vs. model. Scale bar: 500 nm.

Cortical tissues from the sham and model groups (n = 3 per group) were subjected to RNA sequencing. DEGs were subsequently identified and analyzed using bioinformatics approaches. The results of GO analysis revealed that Snap25 was significantly enriched in multiple signaling pathways associated with BP (Fig. 3A), CC (Fig. 3B), and MF (Fig. 3C). These results suggest that differential expression of Snap25 may have a crucial function in cortical damage in MCAO/R rats. Heatmaps (Fig. 3D) show a significant downregulation of Snap25 expression in the model group compared to the sham group. WB results shown in Fig. 3E confirm that Snap25 protein expression in the cortex illustrated a significant decrease in the model group (sham vs. model, p 0.001), but increased significantly following administration of muscone (M9 vs. model, p 0.05). These outcomes establish that Snap25 expression decreased significantly under AIS damage-like circumstances, and that administration of muscone could significantly increase Snap25 expression.

Fig. 3.

Decreased Synaptosome-associated protein 25 kDa (Snap25) expression in the ischemic cortical tissue of MCAO/R rats. Cortical tissues from the sham and model groups (n = 3 per group) underwent RNA sequencing to identify differentially expressed genes (DEGs). These were subsequently analyzed using Gene Ontology (GO) enrichment analysis, with the Snap25 gene found to be prominently involved in several signaling pathways associated with biological process (BP)s (A), cellular component (CC)s (B), and molecular function (MF)s (C). Heatmap visualization (D) revealed a noticeable reduction in Snap25 expression in the model group relative to the sham group. Additionally, western blotting (WB) analysis (E) revealed that the Snap25 protein level in the rat cortex showed a significant lessening in the model group compared to the sham group, but elevated markedly following muscone treatment. One-way ANOVA compared the means of several groups. Data are shown as the mean $\pm$ SEM. ^*⁣**p 0.001 vs. sham group; ^#p 0.05 vs. model group.

The impact of muscone on PC12 cells exposed to OGD/R was examined using CCK8 assay and live/dead cell staining. CCK8 analysis revealed the M300 group (300 ng/mL muscone) had substantially better cell viability compared to the control group (Fig. 4A; control vs. M300, p 0.01). In contrast, the M600 (600 ng/mL muscone) and M1200 (1200 ng/mL muscone) groups showed markedly reduced cell viability relative to the control group (Fig. 4A; control vs. M600, p 0.05; control vs. M1200, p 0.05). Fig. 4B shows that OGD/R damage caused a significant lessening in PC12 cell viability (control vs. model, p 0.0001). The M100, M300, and ferrostatin-1 (Fer-1) groups showed significantly higher cell viability than the model group (Fig. 4B; p 0.05, p 0.001, and p 0.0001, respectively). Cell death rates were also significantly lower in the M300 and Fer-1 groups (Fig. 4C; both p 0.001 vs. model). Collectively, these outcomes suggest that muscone exerts a substantial protective effect on PC12 cells that have been subjected to OGD/R-induced damage.

Fig. 4.

Muscone reduces mortality in PC12 cells following oxygen-glucose deprivation/reperfusion (OGD/R) damage. (A) CCK-8 assay shows the effects of muscone (100–1200 ng/mL) on PC12 cell viability. (B) Muscone increased viability in OGD/R-injured cells; Ferrostatin-1 served as a positive control. (C) Live/dead staining indicates muscone (100 and 300 ng/mL) reduced cell mortality after OGD/R. Data are mean $\pm$ SEM; one-way ANOVA. ^*⁣*⁣**p 0.0001, ^**p 0.01, ^*p 0.05 vs. control; ^{#⁢#⁢#⁢#}p 0.0001, ^#⁢#⁢#p 0.001, ^#p 0.05 vs. model. Scale bar: 50 µm.

TEM analysis revealed substantial mitochondrial damage in the model group, such as decreased cristae and disrupted membranes (Fig. 5A). Damage was significantly reduced in the M300 and Fer-1 groups. GPX4 activity (Fig. 5B) and GSH levels (Fig. 5C) were significantly higher in the M100, M300, and Fer-1 groups vs. the model (M100: p 0.01; M300: p 0.0001, p 0.001; Fer-1: p 0.05, p 0.01). Total iron (Fig. 5D), ferrous iron (Fig. 5E), ROS (Fig. 5F), and MDA (Fig. 5G) levels were significantly lower in M300 and Fer-1 groups (M300: p 0.01, p 0.05, p 0.01, p 0.01; Fer-1: p 0.05, p 0.05, p 0.001, p 0.001). Western blot (Fig. 5H) showed Snap25 downregulation in the model group (p 0.001), which was restored in M300 and Fer-1 groups (both p 0.05). These findings indicate muscone alleviates ferroptosis in OGD/R-damaged PC12 cells and enhances Snap25 expression.

Fig. 5.

Muscone decreases ferroptosis in PC12 cells exposed to OGD/R. (A) Muscone attenuates mitochondrial damage in OGD/R-injured PC12 cells. The red arrows indicate mitochondria. Scale bar: 500 nm. Muscone increased GPX4 activity (B) and GSH levels (C), while decreasing total iron (D), ferric iron (E), ROS (F), and MDA levels (G). (H) Snap25 protein expression, reduced by OGD/R, was restored by muscone. Data are mean $\pm$ SEM; one-way ANOVA. ^*⁣*⁣**p 0.0001, ^*⁣**p 0.001 vs. control; ^#p 0.05, ^#⁢#p 0.01, ^#⁢#⁢#p 0.001, ^{#⁢#⁢#⁢#}p 0.0001 vs. model.

We next investigated the Snap25 function in the ferroptosis pathway during OGD/R injury by using Snap25-specific siRNAs and NC siRNA. WB results established effective silencing of Snap25 in the Snap25-siRNA-1 group compared to the NC group (Fig. 6A; Snap25-siRNA-1 vs. NC, p 0.0001). Compared to the NC group, both the Snap25-si and Snap25-si+muscone groups exhibited significant decreases in Snap25 expression (Fig. 6B; Snap25-si vs. NC, p 0.05; Snap25-si+muscone vs. NC, p 0.05), GPX4 activity (Fig. 6C; p 0.001, p 0.01), GSH level (Fig. 6D; p 0.01, p 0.05), and cell viability (Fig. 6G; p 0.05, p 0.01). Significant increases in the levels of ROS (Fig. 6E; p 0.01, p 0.01) and MDA (Fig. 6F; p 0.01, p 0.05) were observed in the Snap25-si and Snap25-si+muscone groups, respectively, compared to the NC group. Representative TEM images showed more severe ultrastructural damage of mitochondrial crista in the Snap25-si and Snap25-si+muscone groups compared to the model group (Fig. 6H). Notably, the level of damage between the Snap25-si and Snap25-si+muscone groups was not statistically different. These results indicate that silencing of Snap25 expression increases the ferroptosis level in OGD/R-injured PC12 cells. Moreover, Snap25 silencing eliminated the anti-ferroptosis impact of muscone in OGD/R-injured PC12 cells, thereby demonstrating target specificity for the interaction between muscone and Snap25 protein.

Fig. 6.

Snap25 silencing increased the level of ferroptosis and eliminated the anti-ferroptosis impact of muscone in OGD/R-damaged PC12 cells. (A) Effective knockdown of Snap25 in PC12 cells. WB analysis established the effective reduction of Snap25 expression in PC12 cells. (B) Snap25 expression following knockdown and muscone treatment of PC12 cells exposed to OGD/R. GPX4 activity (C) and GSH (D) levels after Snap25 knockdown and muscone administration in OGD/R-injured PC12 cells. In addition, the ROS (E), MDA levels (F), and cell viability (G) were assessed after Snap25 silencing and muscone treatment in PC12 cells subjected to OGD/R injury. (H) Representative TEM images after Snap25 silencing and muscone treatment of PC12 cells exposed to OGD/R. The red arrows indicate mitochondria. Scale bar, 500 nm. One-way ANOVA compares the means of multiple groups. Data shown are the mean $\pm$ SEM. ^*⁣*⁣**p 0.0001, ^*⁣**p 0.001, ^**p 0.01, ^*p 0.05 vs. the negative control (NC) group; ns, no statistical significance, Snap25-si group vs. Snap25-si+muscone group.

WB analysis established the effective Snap25 upregulation in the Snap25-over expression (OE) group in comparison to the NC group (Fig. 7A; p 0.01). When compared to the NC group, both the Snap25-OE and Snap25-OE+muscone groups showed significant increases in Snap25 expression (Fig. 7B: Snap25-OE vs. NC, p 0.05; Snap25-OE+muscone vs. NC, p 0.001), GPX4 activity (Fig. 7C: p 0.01, p 0.0001), GSH level (Fig. 7D: p 0.001, p 0.0001), and cell viability (Fig. 7G: p 0.05, p 0.001). Additionally, marked decreases in ROS (Fig. 7E: p 0.01, p 0.0001) and MDA levels (Fig. 7F: p 0.0001, p 0.0001) were observed in the Snap25-OE and Snap25-OE+muscone groups. TEM images showed significantly improved mitochondrial cristae ultrastructure in the Snap25-OE and Snap25-OE+muscone groups compared to the model group (Fig. 7H). Compared to the Snap25-OE group, the Snap25-OE+muscone group showed substantially higher levels of Snap25 protein expression (Fig. 7B: p 0.05), GPX4 activity (Fig. 7C: p 0.05), GSH concentration (Fig. 7D: p 0.01), and cell viability (Fig. 7G: p 0.05), but notable reductions in ROS (Fig. 7E: p 0.05) and MDA (Fig. 7F: p 0.05) levels. Additionally, the mitochondrial ultrastructure exhibited significant improvement in the Snap25-OE+muscone group (Fig. 7H). Together, these outcomes indicate that Snap25 overexpression mitigates ferroptosis and enhances the protective effect of muscone against ferroptosis in OGD/R-injured PC12 cells, thus highlighting the specific interaction between muscone and the Snap25 protein.

Fig. 7.

Overexpression of Snap25 reduced the ferroptosis level and strengthened the anti-ferroptosis effects of muscone in OGD/R-injured PC12 cells. (A) WB analysis confirmed the Snap25 overexpression in PC12 cells. (B) Following Snap25 overexpression and muscone administration in PC12 cells exposed to OGD/R damage, significant increases were observed in GPX4 activity (C) and GSH levels (D). Furthermore, Snap25 overexpression combined with muscone treatment led to decreases in ROS (E) and MDA levels (F), increased cell viability (G), and attenuation of mitochondrial injury (H) in PC12 cells subjected to OGD/R. The red arrows indicate mitochondria. Scale bar, 500 nm. One-way ANOVA compared the means of multiple groups. Data shown are the mean $\pm$ SEM. ^*⁣*⁣**p 0.0001, ^*⁣**p 0.001, ^**p 0.01, ^*p 0.05 vs. NC group; ^#⁢#p 0.01, ^#p 0.05, Snap25- over expression (OE) group vs. Snap25-OE+muscone group.

Molecular docking experiments predicted the binding sites of muscone and Snap25. This revealed 8 potential amino acid residues (Ser-25, Ser-28, Thr-29, Met-32, Ile-129, Phe-133, and Ile-134) involved in the interaction between muscone and Snap25 protein (Fig. 8A, $\mathrm{\Delta }$G = –6.5 kcal/mol). The results of SPR technology revealed a significant and direct binding effect between muscone and Snap25 protein (Fig. 8B, K_D = 7.14 $\times$ 10^-6; Ka = 5.81 $\times$ 10⁴; Kd = 4.15 $\times$ 10^-1), thus confirming the results of the molecular docking experiments. MD simulations were subsequently carried out to assess the effect of muscone binding on the structural dynamics of Snap25. Analysis of root mean square deviation (RMSD), root mean square fluctuation (RMSF) and radius of gyration (Rg) revealed that binding of muscone enhanced the stability (Fig. 8C), reduced the flexibility (Fig. 8D), and increased the overall structural compactness (Fig. 8E) of Snap25 protein. Solvent accessible surface area (SASA) analysis found that binding of muscone reduced the overall solvent-accessible surface area of the Snap25 protein (Fig. 8F). Hydrogen bond analysis revealed that binding of muscone to the Snap25 protein exerted a protective effect on the stability of the internal hydrogen bond network (Fig. 8G). The Gibbs free energy analysis revealed that binding of muscone resulted in a more concentrated distribution of free energy and promoted a more stable conformational state (Fig. 8H). Visualization analysis of protein structure showed that binding of muscone strongly maintained the overall conformation of the Snap25 protein (Fig. 8I). These findings indicate the interaction between muscone and Snap25 is characterized by a moderate level of free energy and strong binding stability, thereby enhancing the structural stability and compactness of the Snap25 protein.

Fig. 8.

Muscone binds to Snap25 protein and enhances its structural stability. Molecular docking tests were conducted to predict the binding sites between muscone and Snap25. (A) Stereoview of the binding mode for Snap25 with muscone. (B) Surface plasmon resonance (SPR) analysis showing the direct binding effect between muscone and Snap25 protein (K_D = 7.14 $\times$ 10^-6; Ka = 5.81 $\times$ 10⁴; Kd = 4.15 $\times$ 10^-1). Analysis of the root mean square deviation (RMSD) (C), root mean square fluctuation (RMSF) (D), radius of gyration (E), solvent accessible surface area (SASA) (F), hydrogen bond analysis (G), Gibbs free energy (H), and visualization analysis of the protein structure (I). Together, these analyses revealed that binding of muscone to Snap25 is characterized by moderate free energy and robust binding stability, thus significantly enhancing the structural stability and compactness of the Snap25 protein.