All animal experiments were carried out following the Guideline for the Care and Use of Laboratory Animals and authorized by the Experimental Animal Ethics Committee of Liaoning University of Traditional Chinese Medicine (No. 21000042023094). All cell lines were validated by STR and tested negative for mycoplasma. The BV2 cell is a microglia cell line derived from mice and is used widely in study of IS [24, 25]. BV2 cells (mouse microglia cells; Cat. No. iCell-m011) were obtained from iCell Bioscience Inc. (Shanghai, China) and cultured in an incubator (5% CO₂, 37 °C) in Dulbecco’s modified Eagle medium (DMEM; Servicebio, Wuhan, Hubei, China) containing 10% fetal bovine serum. Primary cortical neurons were obtained from five mice at 16 embryonic days. Embryonic day 16 mice were prepared from timed-pregnant C57BL/6J mice. The pregnant mice were euthanized with CO₂. Fetuses were taken out and their brains were removed rapidly. The cerebral cortices were dissected out. After digestion with trypsin (Sigma, St. Louis, MO, USA), the neurons were cultured in Neurobasal medium supplemented with 1% B27 in plate, then cells were cultured in an incubator (5% CO₂ at 37 °C). The cytarabine (10 µM; Aladdin, Shanghai, China) was added to eliminate glial cells in the next day. The frequency of half-culture solution replacement was kept every 3 days.

The cells were cultured in medium without glucose in an incubator with 95% N₂ and 5% CO₂ for 4 h to establish the Oxygen-Glucose Deprivation (OGD) model, as described previously by Zhou et al. [26]. Next, cells were incubated with AL (25, 50, and 100 µg/mL, Cat. No. HY-N0222; MedChemExpress, Monmouth Junction, NJ, USA) or exosome (20 µg) for 24 h under normal conditions (5% CO₂, 37 °C; corresponding medium).

BV2 cells were used to established OGD model and then treated with AL (100 µg/mL, Cat. No. HY-N0222; MedChemExpress) under normal conditions. Twenty-four h after AL treatment, the supernatant was collected and centrifuged multiple times to obtain exosomes (AL-Exo). Exosomes isolated from BV2 cells without AL treatment (BV2-Exo) were regarded as contrast. Morphology and size distribution of exosomes were observed using an HT-7700 transmission electron microscope (TEM, Hitachi, Tokyo, Japan) and a ZetaVIEW nanoparticle tracking analyzer (NTA, PARTICLE METRIX, Munich, Germany). Western blot analysis was used to detect the markers of exosomes (CD63, CD81, and CD9).

The mice were purchased from Huachuang Sino (Taizhou, Zhejiang, China). Male C57BL/6J mice (8 weeks old) were housed in a 12-h light/12-h dark cycle environment with appropriate temperature and humidity, and given free access to food and water for 1 week of acclimation. One hundred and fifty mice were randomly assigned to five groups using a computer-generated random-number table. The grouping set in our paper included, sham group (N = 30) received sham surgery without the monofilament block and sodium carboxymethyl cellulose (CMC-Na, Cat. No.C835846; Macklin, Shanghai, China) by oral gavage, MCAO/R group (N = 30) received blood occlusion and CMC-Na by oral gavage, MCAO/R+1.25 mg/kg/day avicularin (MCAO/R+AL-L) group (N = 30) received blood occlusion and 1.25 mg/kg/day AL by oral gavage, MCAO/R+2.50 mg/kg/day avicularin (MCAO/R+AL-M) group (N = 30) received blood occlusion and 2.50 mg/kg/day AL by oral gavage, and MCAO/R+5.00 mg/kg/day avicularin (MCAO/R+AL-H) group (N = 30) received blood occlusion and 5.00 mg/kg/day AL by oral gavage.

The MCAO/R model was established as previously described and isoflurane (Shenzhen Rayward Life Technology Co., Ltd., Shenzhen, China; 3% for anesthesia induction and 2% for anesthesia maintaining) was used for anesthesia [27]. Briefly, monofilament was used to obstruct the origin of the left middle cerebral artery. One h after occlusion, the monofilament was removed to start reperfusion. Sham mice received the same surgery without the monofilament block.

AL (Cat. No. HY-N0222; MedChemExpress) was dissolved in 0.5% CMC-Na and administered by oral gavage once a day for 7 days at the doses of 1.25, 2.5, or 5 mg/kg, starting 3 h after reperfusion. The treatment time and doses of AL used in our study were determined by the procedures of a previous study [28]. Sham mice and MCAO/R mice received an equal volume of 0.5% CMC-Na by oral gavage. Twenty-four h after the last administration, mice underwent the behavioral test and were then euthanized with CO₂.

The modified neurological severity score (mNSS) test was used to evaluate the neurological deficit (normal score, 0; maximum-deficit score, 18), and conducted as described in previous studies [29, 30], with slight modifications, on day 8, by observers blind to the experiment. The higher the score, the more serious the injury.

Brain samples were harvested from the sacrificed mice, frozen for 1 h and cut into slices. The brain slices were immersed in Triphenyltetrazolium Chloride (TTC, Cat. No. T8170; Solarbio, Beijing, China) and incubated for 15 min at 37 °C. With TTC, the normal tissue was stained red and the infarct-damaged tissue was white.

After the mice were euthanized with CO₂ (Shenyang Jingquan Gas Plant, Shengyang,China), the brains were removed, weighed (wet weight) and then dried at 105 °C overnight to obtain the dry weight. Brain-water content (%) = (wet weight – dry weight)/wet weight $\times$100%.

Fresh ischemic penumbra (cortex) tissue was removed from the brain, and then fixed, dehydrated, infiltrated, embedded, and cut into 5-µm slices. After dewaxing, hydration, permeabilization, and antigen retrieval, slices were incubated with TdT-mediated dUTP nick-end labeling (TUNEL) reactive solution (Cat. No. 11684795910; Roche, Basel, Switzerland) for 60 min in the dark. After 30 min of blocking with bovine serum albumin (BSA; Cat. No. A602440-0050; Sangon Biotech (Shanghai) Co., Ltd., Shanghai, China), slices were incubated with NeuN polyclonal antibody (1:50; Cat. No. 26975-1-AP; Proteintech Group, Inc., Rosemont, IL, USA) at 4 °C overnight. Next, they were incubated with Cy3–conjugated secondary antibody (1:200; Cat. No. SA00009-2; Proteintech Group, Inc.) for 60 min. Finally, cell nuclei were re-stained with 4^′,6-diamidino-2-phenylindole (DAPI; Cat. No. D106471; Aladdin) and examined with a BX53 microscope (Olympus, Tokyo, Japan).

For TUNEL analysis, cell slide was made and permeabilized. The cell slide was incubated with TUNEL reactive solution for 60 min in the dark. After counterstaining with DAPI, the cell slide was photographed using a DP73 camera system (Olympus, Tokyo, Japan).

Twenty-four h after AL treatment, the BV2 cells were collected, centrifuged, and resuspended in phosphate buffered saline (Cat. No. B548117; Sangon Biotech (Shanghai) Co., Ltd.). CD86 antibody conjugated with phycoerythrin fluorescent dye (Cat. No. PE-65068; Proteintech Group, Inc.) and CD206 antibody conjugated with fluorescein isothiocyanate (Cat. No. 141703; FITC; Biolegend, San Diego, CA, USA) were added to cells. The co-incubation lasted for 30 min in the dark. A NovoCyte flow cytometer (Agilent Technologies, Santa Clara, CA, USA) was used for detection.

Lactate Dehydrogenase (LDH) assay was performed to analyze the death rate of primary neurons using an LDH assay kit according to manufacturer’s instructions. The kit was purchased from Jiancheng Bioengineering Institute (Cat. No. A020; Nanjing, Jiangsu, China). Absorbance was measured using a microplate reader ELX-800 (BioTek, Winooski, VT, USA).

The levels of IL-1$\beta$ (Cat. No. EK201B), IL-6 (Cat. No. EK206), IL-18 (Cat. No. EK218), and TNF-$\alpha$ (Cat. No. EK282) in tissue homogenate or cell supernatant were quantified using Enzyme Linked Immunosorbent Assay (ELISA) kits (MultiSciences Biotech, Hangzhou, Zhejiang, China) by following the manufacturer’s instructions. Protein concentration was detected by bicinchoninic acid (BCA) protein quantitative kit (Beyotime, Shanghai, China). Absorbance was measured using a microplate reader ELX-800 (BioTek, Winooski, VT, USA).

Tissue slices were treated for double immunofluorescence. After antigen retrieval and sealing, slices were incubated with the following primary antibodies (1:100) at 4 °C overnight: Ionized calcium binding adapter molecule (IBA1) antibody (Cat. No. Ab283319; Abcam, Cambridge, UK), CD86 antibody (Cat. No. DF6332; Affinity, Changzhou, Zhejiang, China), CD206 antibody (Cat. No. DF4149; Affinity), and NOD-like receptor thermal protein domain associated protein 3 (NLRP3) antibody (Cat. No. DF7438; Affinity). After being rinsed, slices were incubated with FITC-conjugated or Cy3-conjugated secondary antibody (1:200; Cat. No. SA00003-2/SA00009-1; Proteintech Group, Inc.) for 90 min. The cell slide was used to measure the polarization of microglia, and permeated with tritonX-100 (Cat. No. ST795; Beyotime) for 30 min then blocked with BSA for 15 min. After incubation with INOS antibody (1:100; Cat. No. AF0199; Affinity) or Arginase 1 (ARG1) antibody (1: 100; Cat. No. DF6657; Affinity), the cell slide was incubated with Cy3-conjugated secondary antibody (1:200; Cat. No. SA00009-2; Proteintech Group, Inc.) for 60 min. All slices and slides were redyed by DAPI and photographed using a BX53 microscope.

The protein was obtained by radio-immunoprecipitation-assay lysis buffer (Cat. No. PR20001; Proteintech Group, Inc.), and its concentration was measured using the BCA protein quantitative kit (Cat. No. PK10026; Proteintech Group, Inc.). Next, proteins were separated by gel electrophoresis and transferred to polyvinylidene fluoride (PVDF) membranes (Cat. No. LC2005; Thermo Scientific, Pittsburgh, PA, USA). After incubation in blocking buffer (Cat. No. PR20011; Proteintech Group, Inc.), the membranes were blotted using the following primary antibody at 4 °C overnight: caspase 3 antibody (1:1000; Cat. No. AF6311; Affinity); cleaved caspase 3 antibody (1:1000; Cat. No. AF7022; Affinity); NLRP3 antibody (1:1000; Cat. No. DF7438; Affinity); cleaved caspase 1 antibody (1:1000; Cat. No. AF4005; Affinity); PYD And CARD Domain Containing (ASC) antibody (1:1000; Cat. No. DF6304; Affinity); CD63 antibody (1:1000; Cat. No. AF5117; Affinity); CD81 antibody (1:1000; Cat. No. DF2306; Affinity); CD9 antibody (1:1000; Cat. No. AF5139; Affinity); and $\beta$-actin (1:20,000; Cat. No. 66009-1-Ig; Proteintech Group, Inc.). The membranes were incubated with the secondary antibody (Affinipure Goat Anti-Mouse or Anti-Rabbit IgG; 1:10,000; Cat. No. SA00001-1/SA00001-4; Proteintech Group, Inc.) conjugated with horseradish peroxidase and then were visualized using enhanced chemiluminescence (Cat. No. PK10003; Proteintech Group, Inc.). Subsequently, the protein bands were obtained in films and scanned with EPSON perfection v19 II (Epson Co., Ltd., Nagano, Japan). The Gel-Pro-Analyzer was used for result analysis.

GraphPad Prism 9.0.0 (Dotmatics, Boston, MA, USA) was used to perform statistical analysis and differences were analyzed using ordinary one-way analysis of variance (ANOVA) followed by the Tukey’s multiple comparisons test or Kruskal-Wallis test. Data followed by the Dunn’s multiple comparisons test were shown as mean $\pm$ standard deviation (SD). p 0.05 was considered to be a statistically significant difference.

To confirm whether AL treatment ameliorated IS injury, the MCAO/R mouse model was established and the pharmacological effects of AL treatment were evaluated. As shown in Fig. 1A, after the MCAO/R procedure, an increase in mNSS in the model group (surgery only) indicated significant ischemic injury (p 0.05), and AL treatment caused lower scores (p 0.05), suggesting that AL treatment prevented or reduced some neurological damage in the MCAO/R mice. The TTC staining results showed differences in brain-infarct size of the mice in modeling group and control group (p 0.05), suggesting that our MCAO/R model was established successfully. AL resulted in a lower infarct volume in the brains of treated mice (p 0.05), than in those of the MCAO/R mice (Fig. 1B). Water content of the brains of MCAO/R mice was significantly higher than in the brains of sham mice (p 0.05). AL treatment disturbed MCAO-produced increase of brain-water content (Fig. 1C, p 0.05). Double immunofluorescence staining showed that TUNEL-positive neurons were sparse in the cortex of sham mice, whereas increased neuronal death was observed in MCAO/R mice, as evidenced by more TUNEL-positive cells and higher cleaved caspase 3 expression (p 0.05). Compared with the model mice, AL treated mice had a lower number of apoptotic cells and a reduced level of cleaved caspase 3 (Fig. 1D,E, the original figures of Western Blot can be found in the Supplementary Materials-original western blot images-Fig1, p 0.05). These results indicated that AL administration effectively inhibited neuronal death produced by MCAO/R. ELISA was used to measure the inflammatory responses that occurred in the cortex of mice with MCAO/R. Apparent downregulation of IL-6 and TNF-$\alpha$ was observed in AL-treated MCAO/R mice, but not in the untreated MCAO/R group (Fig. 1F, p 0.05). These findings suggest that AL exerts a neuroprotective effect after IS.

Fig. 1.

AL treatment markedly decreased ischemic lesion in MCAO/R mice. (A) After seven days of AL treatment, the MCAO/R mice was accepted behavioral test to evaluate degree of nerve damage. Brain samples were harvested from the sacrificed and used for the following experiments. (B) TTC staining was conducted to measure the cerebral infarct volume. (C) The wet weight and dry weight of brain tissue of mice were weighed to calculate brain water content. (D) Neuronal apoptosis of ischemic penumbra was detected by using double immunofluorescence (NeuN: red; TUNEL: green; DAPI: blue; Bars = 50 µm). (E) Western blotting analysis was used to measure the protein expression of total caspase 3 and cleaved caspase 3. (F) IL-6 and TNF-$\alpha$ levels was confirmed by ELISA. Data are expressed as mean $\pm$ SD (n = 6, *p 0.05). AL, avicularin; MCAO/R, middle cerebral artery occlusion/reperfusion; TTC, 2,3,5-Triphenyltetrazolium chloride; TUNEL, TdT-mediated dUTP nick-end labeling; DAPI, 4^′,6-diamidino-2-phenylindole; IL-6, interleukin-6; TNF-$\alpha$, tumor necrosis factor-$\alpha$; ELISA, enzyme linked immunosorbent assay; SD, standard deviation; mNSS, modified neurological severity score.

In order to elucidate the effect of AL on microglia polarization against a background of ischemia, polarization markers were detected using immunofluorescence. After IS, the numbers of CD86-marked M1 microglia and CD206-marked M2 microglia were higher those in sham mice (p 0.05), indicating that inflammation was enhanced which, in turn, activated the anti-inflammatory phenotype to respond to the body’s self-protection mechanism after IS. Treatment with AL significantly decreased the inflammatory M1 phenotype and increased the anti-inflammatory M2 phenotype (Fig. 2, p 0.05). Based on these results, we concluded that AL regulates microglia polarization after IS development.

Fig. 2.

AL regulated microglial polarization after MCAO/R. Double immunofluorescence (IBA1: red; CD86 or CD206: green; DAPI: blue; The images in the dashed rectangles are amplified on the right = 50 µm) was used to assay the microglial polarization in ischemic penumbra of MCAO/R treated with AL or untreated (n = 6, *p 0.05). The images in the dashed rectangles are amplified on the right, showing the representative double staining. The bars of the right image were also 50 µm. AL, avicularin; MCAO/R, middle cerebral artery occlusion/reperfusion; DAPI, 4^′,6-diamidino-2-phenylindole; IBA1, Ionized calcium binding adapter molecule.

Considering the above results, we described an important role for AL in suppressing inflammation. Next, the pathway mediating the function of AL was explored. Double immunostaining results showed that the NLRP3 inflammasome was activated in the cortex of MCAO/R mice but not in cortex of the sham group mice (p 0.05). AL treatment reduced NLRP3-positive neurons to a lower level than in model mice (Fig. 3A, p 0.05). Protein levels of NLRP3, cleaved caspase 1, and ASC were higher in model mice than in sham mice, whereas AL treatment decreased this elevation induced by the MCAO/R procedure (Fig. 3B, the original figures of Western Blot can be found in the Supplementary Materials-original western blot images-Fig1, p 0.05). The ELISA results showed that production of IL-1$\beta$ and IL-18 was higher in the cortex after MCAO/R than it was in sham mice (p 0.05). AL treatment inhibited the level of IL-1$\beta$ and IL-18 in the MCAO/R mice (Fig. 3C, p 0.05). These findings indicate that AL inhibits the NLRP3 inflammasome activation that is seen in the cortex of MCAO/R mice.

Fig. 3.

AL treatment interfered with the activation of NLRP3 inflammasome after MCAO/R. (A) Representative double immunofluorescence staining for IBA1 (red) and NLRP3 (green). The nucleus was stained blue (Bars = 50 µm). The images in the solid rectangles are magnified representative cell. (B) The protein expression of NLRP3, cleaved caspase 1 and ASC in ischemic penumbra of mice was measured by Western blot analysis. (C) IL-1$\beta$ and IL-18 levels in ischemic penumbra of mice was detected by ELISA. Data are presented as mean $\pm$ SD (n = 6, *p 0.05). AL, avicularin; NLRP3, NOD-like receptor thermal protein domain associated protein 3; MCAO/R, middle cerebral artery occlusion/reperfusion; ASC, apoptosis-associated speck-like protein containing a CARD; IL-1$\beta$, interleukin-1$\beta$; ELISA, enzyme linked immunosorbent assay; SD, standard deviation.

To verify that the inflammatory response was mediated by polarized microglia, an OGD/R model was established in vitro (Fig. 4A). ELISA results showed that the levels of IL-6 and TNF-$\alpha$ in BV2 cells were higher after the OGD/R procedure, and AL treatment significantly decreased the production (Fig. 4B, p 0.05), suggesting that the OGD/R procedure induced inflammatory responses and AL played an anti-inflammatory role. The polarization of microglia was examined using immunofluorescence and flow-cytometry assays. The number of INOS⁺ cells and ARG1⁺ cells was higher after OGD/R treatment (p 0.05). Compared with OGD/R cells, strength of INOS was weakened, and ARG1 was enhancive in the presence of AL (Fig. 4C, p 0.05). Furthermore, polarization markers showed similar changes. OGD/R induced an increase of the M1 phenotype (CD86⁺CD206^–) and the M2 phenotype (CD86^–CD206⁺), whereas AL reduced M1 polarization, and increased M2 polarization (Fig. 5, p 0.05). Overall, our study demonstrated that AL treatment suppresses inflammation by attenuating M1 polarization and enhancing M2 polarization.

Fig. 4.

AL treatment ameliorated the inflammatory response of OGD/R BV2 cells by affecting cell polarization. (A) OGD/R model was established to detect the effect of AL on microglial polarization in vitro. (B) ELISA was used to measure the levels of IL-6 and TNF-$\alpha$ of BV2 cells. (C) Marker of M1 polarization (INOS: red) or M2 polarization (ARG1: red) was showed using immunofluorescence (Bars = 50 µm). The nucleus was stained blue. The dotted box represents representative INOS or ARG1 labeled cells. Data are showed as mean $\pm$ SD (n = 3, *p 0.05). AL, avicularin; OGD/R, oxygen–glucose deprivation and reoxygenation; ELISA, enzyme linked immunosorbent assay; IL-6, interleukin-6; TNF-$\alpha$, tumor necrosis factor-$\alpha$; INOS, inducible nitric oxide synthase; ARG1, arginase-1; SD, standard deviation.

Fig. 5.

AL affected cell polarization of OGD/R BV2 cells. Flow cytometry was conducted to assay the expression of CD86 and CD206. Data are showed as mean $\pm$ SD (n = 3, *p 0.05). AL, avicularin; OGD/R, oxygen–glucose deprivation and reoxygenation; SD, standard deviation; FITC, fluorescein isothiocyanate.

We examined whether NLRP3 was a mediator of AL in inflammation intervention. The expression of NLRP3-inflammasome-related protein was detected by Western blot analysis. As shown in Fig. 6A (the original figures of Western Blot can be found in the Supplementary Materials-original western blot images-Fig6), OGD/R upregulated NLRP3, cleaved caspase 1 and ASC protein levels (p 0.05). AL treatment decreased these NLRP3-inflammasome-associated proteins that had been increased by low glucose and low oxygen processing (p 0.05). The ELISA results showed that the levels of IL-1$\beta$ and IL-18 were increased by OGD/R, and that AL markedly reduced this increase (Fig. 6B, p 0.05). These results indicated that AL weakens the activation of NLRP3 in vitro.

Fig. 6.

AL inhibited NLRP3 pathway in vitro. (A) NLRP3, cleaved caspase 1 and ASC protein expression in BV2 cells was measured by western blotting analysis. (B) Downstream factors (IL-1$\beta$ and IL-18) of NLRP3 activation were detected by ELISA. Results are presented as mean $\pm$ SD (n = 3, *p 0.05). AL, avicularin; NLRP3, NOD-like receptor thermal protein domain associated protein 3; ASC, apoptosis-associated speck-like protein containing a CARD; IL-1$\beta$, interleukin-1$\beta$; ELISA, enzyme linked immunosorbent assay; SD, standard deviation.

Microglial exosomes are known to play an important role under ischemic conditions. However, the question of whether AL treatment affected the size or function of exosomes had not been studied, so our study investigated this issue (Fig. 7A). As shown in Fig. 7B,C, the microglial exosomes were observed using a transmission electron microscope and their size was measured by nanoparticle-tracking analysis. In the OGD/R condition, the average diameters of AL-treated cells and control cells were 156.9 and 157.4 nm, respectively, which were not significantly different. Western blot analysis showed that exosomes derived from cells co-expressed CD63, CD81, and CD9 (Fig. 7D, the original figures of Western Blot can be found in the Supplementary Materials-original western blot images-Fig7), which are all recognized as representative markers of exosomes. These results suggest that AL treatment appears to have no effect on the appearance of exosomes.

Fig. 7.

Exosomes derived from AL-treated OGD/R BV2 cells reduced neuron death. (A) After OGD/R, the microglial exosomes were obtained with and without AL treatment. These exosomes were used to culture primary cortical neurons that had undergone OGD processing. Characterization of microglial exosomes was assayed by TEM (Bars = 1 µm) (B), NTA (C) and Western blot analysis (D). (E) Neuronal death was evaluated by the measurement of LDH. (F) Representative images of TUNEL staining (green) for primary cortical neurons. The nucleus was stained blue (Bars = 50 µm). Results are showed as mean $\pm$ SD (n = 3, *p 0.05). AL, avicularin; OGD/R, oxygen–glucose deprivation and reoxygenation; TEM, transmission electron microscope; NTA, nanoparticle tracking analysis; FCM, flow cytometry; LDH, lactate dehydrogenase; TUNEL, TdT-mediated dUTP nick-end labeling; SD, standard deviation; WB, western blot; DMEM, Dulbecco’s modified Eagle medium.

Next, primary cortical neurons were used to detected exosome function. OGD/R-treated cells released more LDH than did control cells (p 0.05). Exosomes from AL-treated cells had an inhibitory effect on LDH release that was stronger than that of exosomes from AL-untreated cells (Fig. 7E, p 0.05). TUNEL staining was used to analyze apoptosis. As showed in Fig. 7F, the number of apoptotic cells was higher in the model group (OGD/R only), whereas exosomes weakened the TUNEL fluorescence. Notably, the inhibition of TUNEL was more by exosomes obtained from AL-cultured BV2 cells than by uncultured cells (p 0.05). These data suggested that exosomes from AL-treated OGD/R BV2 cells play a role in suppressing neuronal apoptosis.