5.1 AIRE preparation

AIR was obtained from Xin Jiang Yao Du Pharmaceutical Co., LTD (Xinjiang, China, #20190110). The steps for extracting AIRE from AIR have been described in our previous study [23]. Briefly, the dried and ground AIR plant was first soaked in a solution of 60% v/v aqueous ethanol for 6 h. Afterwards, it was extracted by utilizing a 10 L reflux reactor at 90 °C for 2 h, with a liquid–solid ratio of 1:6, using 75% v/v aqueous ethanol as the extracting agent. Following two extraction cycles, the whole extract was concentrated with reduced pressure at 60 °C. This process resulted in a viscous extract with a concentration of about 30 mg/mL. The extract was subsequently purified through a D101 macroporous resin column, using 70% ethanol as the eluent. Finally, after the ethanol was removed, the eluent was vacuum freeze-dried.

5.2 Analysis of the main components of AIRE

To analyze of the chemical components of AIRE, the samples were prepared and subjected to chromatographic and mass spectrometric conditions. For sample preparation, the sample was mixed with 80% methanol and ultrasonicated for 30 min, with a liquid–solid ratio of 40:1. Then, the suspension was centrifuged at 4 °C and 12,000 rpm for 10 min. Subsequently, the supernatant was pipetted into an injection vial for analysis.

The Vanquish Flex UHPLC system, equipped with an ACQUITY UPLC HSS T3 column (2.1 mm × 100 mm, 1.7 μm), was used for chromatographic analysis. The mobile phase comprised an aqueous solution containing 0.1% formic acid (phase A) and acetonitrile (phase B) flowing at a rate of 0.3 mL/min. The column temperature was kept constant at 40 °C. Table 1 displayed the elution gradient.

Table 1 Elution gradient

The Q Exactive hybrid quadrupole-orbitrap mass spectrometer (Thermo Fisher Scientific) with a HESI-II spray probe was used to obtain mass spectrometric data. The parameters were set as follows: positive ion source voltage at 3.7 kV, negative ion source voltage at 3.5 kV, heated capillary temperature at 320 °C, sheath gas pressure at 30 psi, auxiliary gas pressure at 10 psi, and desolvation temperature at 300 °C. Nitrogen was employed as the sheath and auxiliary gas, as well as the collision gas. Data acquisition was performed in "Full scan/dd-MS2" mode. The full scan parameters were set to a resolution of 70000, an auto gain control target of 1 × 106, and a maximum isolation time of 50 ms. The dd-MS2 data were collected with a resolution of 17500, an auto gain control target of 1 × 105, a maximum isolation time of 50 ms, a loop count of the top 10 peaks, an isolation window of m/z 2, collision energies of 10, 30, and 60 V, and an intensity threshold of 1 × 105.

Data analysis was conducted using Progenesis QI 3.0 software (Waters Corp., MA, USA), which included steps for raw data introduction, peak extraction, and adduct deconvolution. Identification was determined based on the mass error of the parent ion, match degree of fragment ions, isotope distribution, and peak area by searching against a theoretical database constructed from literature and public databases.

5.3 Animals

Adult male Sprague–Dawley (SD) rats (SPF grade, weight: 240–280 g) were provided by Beijing HFK Bioscience Co., Ltd. (Beijing, China, Certificate No: SCXK (Beijing) 2019–0008). All experiments were conducted in accordance with protocols approved by the Ethics Committee of the Institute of Materia Medica, Chinese Academy of Medical Sciences, and Peking Union Medical College. All SD rats underwent a 7-day adaptation period to acclimate to laboratory conditions before being randomly assigned to various groups based on experimental requirements.

5.4 Establishment of CIRI model and animal grouping

The CIRI model, as outlined in prior work [46], was utilized with slight adjustments. Rats were anesthetized using 2.5% isoflurane, maintained at 1.5%, and positioned supine. After sterilizing the surgical area, an incision was performed to uncover the common carotid artery (CCA) through blunt dissection from the surrounding tissues. Both the external carotid artery (ECA) and internal carotid artery (ICA) were carefully isolated and exposed. Initially, vascular clips were used to occlude the ICA and CCA. The ECA was then incised approximately 0.5 cm from its bifurcation with the ICA. Following this, a nylon thread was inserted through ECA into the ICA. Afterwards, the vascular clip on the ICA was taken off, allowing the thread to be transferred forward into the middle cerebral artery (MCA). Then, the vascular clip on the CCA was also removed. After 1.5 h of induced ischemia, the thread was removed to initiate the restoration of blood supply. Rats in the sham group received the same surgical procedures, excluding the insertion of a nylon thread.

Besides the sham group, animals that underwent successful operations were randomly allocated into the CIRI model group, the treatment groups receiving 10, 30, and 100 mg/kg AIRE, and the positive control EDA (Simcere, China, #H20031342) group, respectively.

5.5 Evaluation of neurological function

Neurological impairment was assessed using the mNSS, Zea-Longa rating, and grip strength test at 24 h post-ischemia–reperfusion. All neurological assessments were carried out by researchers unaware of the experimental assignments. The mNSS test was evaluated using a range from 0 to 18 and consisted of assessments for motor function, sensory perception, reflexes, and balance. A score of 0 indicated normal condition, while a score of 18 indicated the most severe deficiency. Severe CIRI model rats with scores of 10 to 15 were adopted for this study.

5.6 Measurement of cerebral infarct volume in TTC staining

Deep anesthesia of rats was induced using 2.5% isoflurane. The cerebrum was then harvested and sectioned coronally into 2 mm slices. These slices were then incubated in a 1% solution of TTC for 30 min. Following incubation, the tissues were fixed in 4% formaldehyde and subsequently photographed. The cerebral infarct volume was quantified using ImageJ software (National Institutes of Health, USA). The percentage of cerebral infarct volume was calculated using the formula: [total cerebral infarct volume—(ipsilateral hemisphere volume—contralateral hemisphere volume)] / contralateral hemisphere volume × 100%.

5.7 H&E staining and Nissl staining

Brain damage induced by CIRI was investigated using both H&E staining and Nissl staining. Brain tissue samples, preserved in 4% formaldehyde solution, were encased in paraffin and sliced into 5 μm thin sections. These slices were then used to conduct H&E and Nissl staining to observe tissue or neuronal damage. The stained slices were subsequently scanned using a Pannoramic MIDI scanner (3DHISTECH Ltd., Hungary).

5.8 Metabolomic analysis

Metabolomic analysis was conducted on penumbra brain tissues from the sham group, the model group and the 100 mg/kg treatment group. Tissue samples were homogenized at 15 mg per 1.5 mL in ice-cold lysis/extraction buffer (chloroform: methanol: water at a 2:2:1 ratio). After centrifugation at 12,000 rpm for 15 min at 4 °C, the supernatant from each sample was collected for analysis. The 1H-NMR spectra were acquired using a Bruker 600 MHz Avance III NMR spectrometer and analyzed with MestReNova software (Mestrelab Research). Then signal peaks in the spectra were matched to specific metabolites using the documented literature as well as the NMR database. Finally, the different metabolites were selected to investigate metabolic pathways by MetaboAnalyst 6.0 [47].

5.9 Transcriptomic analysis

From each of the sham group, the CIRI model group, and the 100 mg/kg treatment group, three rats were randomly selected for further study. Initially, total RNA was acquired from the penumbra brain tissues with RNAiso Plus (NovBio, China, #9109), and its purity was enhanced with AMPure XP beads. The RNA quality was subsequently evaluated using an Agilent 2100 Bioanalyzer. Subsequent steps involved generating RNA-seq libraries utilizing the TruSeq RNA Sample Prep Kit (Illumina, USA), followed by sequencing on an Illumina HiSeq2500 system. To standardize gene expression levels, all data were normalized using the fragments per kilobase per million (FPKM) value. Genes exhibiting a |Log2(Fold Change)| more than 1 and P value less than 0.05 were considered to be significantly differentially expressed. Pathways associated with identified DEGs were enriched by KEGG, using the DAVID database [48].

5.10 Reverse transcription-quantitative PCR (RT-qPCR)

Samples obtained from the penumbra region of the sham group, CIRI model group, and 100 mg/kg treatment group were homogenized to extract total RNA using RNAiso Plus (NovBio, China, #9109). cDNA was generated with the HiScript III All-in-one RT SuperMix Perfect for qPCR (Vazyme Biotech). Real-time PCR was conducted on the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems). The primers are shown in Table 2.

5.11 ATP assay

Tissue samples from the penumbra region were homogenized in ice-cold lysis/extraction buffer at 30 Hz for 30 min. Following centrifugation at 4 °C (12000 rpm for 15 min), the supernatant from each sample was gathered. ATP levels were quantified using an ATP test kit (Beyotime, China, #S0026) according to the manufacturer’s instructions.

5.12 Western blot analysis

The penumbra tissues were homogenized in RIPA buffer with added protease and phosphatase inhibitors. Total protein was assessed using a BCA protein assay kit (Applygen, China, #P1511). The proteins were separated by SDS-PAGE and transferred onto PVDF membranes. The membranes were then incubated with primary antibodies from Abcam: anti-DRP1 antibody (1:1000, #ab184247), anti-Mfn2 antibody (1:1000, #ab124773), anti-OPA1 antibody (1:1000, #ab157457), anti-HK1 antibody (1:1000, #ab150423), anti-PFKP antibody (1:2000, #ab204131), anti-Frizzled 7 antibody (1:1000, #ab64636), anti-beta Catenin antibody (1:5000, #ab32572), anti-COX IV antibody (1:2000, #ab16056), anti-Actin antibody (1:20000, #ab198991), and anti-GAPDH antibody (1:20000, #ab128915). The membranes were subsequently treated with secondary antibodies (goat anti-rabbit, Abcam, #ab205718; goat anti-mouse, Abcam, #ab205719) for a duration of 2 h at room temperature. Bands were recognized using an enhanced chemiluminescence system.

5.13 Statistical analysis

The experimental data are displayed as mean ± SEM and processed with GraphPad Prism 7.05. For normally distributed data, statistical significance was evaluated with one-way ANOVA, followed by Tukey's multiple comparisons test. In cases of unequal standard deviations, the Brown-Forsythe and Welch ANOVA tests were applied. For non-normally distributed data, the Kruskal–Wallis test was utilized. A P value below 0.05 was deemed statistically significant.