Air-dried TFCJ was treated twice with 70% aqueous ethanol (1:25, w/v), mixed and sonicated 60°C for 40 min, then concentrated in a rotary evaporato as described by the previous method (Liu S. et al., 2020; Ma et al., 2021). The crude extract was purified by the D-101 macroporous resin. TFCJ (60.21% content) was measured by UV-vis-spectrophotometer, and rutin was used as a reference compound. TFCJ and Xuesaitong (5 mg/pill, KPC Xuesaitong Pharmaceutical Co., Ltd.) were dissolved in 0.5% Carboxymethylcellulose sodium (CMC) before use.

Experiments were conducted on 7-weeks old-male Sprague-Dawley rats (240–280 g body weight). All the animals were purchased from Huaxing Experimental Animal Farm (Zhengzhou, China), and maintained in standard housing conditions in colony cages 12 h light/dark cycles with free access to food and water. The animal care and experimental procedures were approved by Animal Ethics Committee of Anhui Science and Technology University. Because estrogen has a certain effect on the experimental results, male rats were selected as experimental animals.

Before the operation, the rats were fasted for 12 h to reduce the mortality during and after the operation and anesthetized by intraperitoneal injection of 2% pentobarbital sodium (4 mL⋅kg^–1). We slowly inserted the threaded plug from the left external carotid artery (ECA) through the common carotid artery (CCA) into the internal carotid artery (ICA) until the origin of middle cerebral artery (MCA) was blocked as the previous literature (Longa et al., 1989; Olga Nikolaevna et al., 2020). After ischemia 2 h, the blood supply was restored, and reperfusion was achieved. The reperfusion process continued for 24 h.

After reperfusion for 24 h, the degree of neurological injury was performed by researchers who were blinded to the grouping (Joya et al., 2021). Evaluations were performed as follows: Grade I, no nerve injury symptoms; Grade II, unable to fully extend the left front paw; Grade III, unable to fully extend the left front paw, circle to the left when walking; Grade IV, unable to fully extend the left front paw, hemiplegia to the left when walking; Grade V, loss of consciousness, unable to walk spontaneously.

After neurological evaluation, the rats were sacrificed, and the rat brain was sliced with 2 mm thin slices along the coronal plane. The slices were incubated with 2% TTC dye solution at 37°C for 15 min in dark. After rinsing, the slices were fixed in 4% paraformaldehyde over 24 h. Image pro plus 6.0 software (Media Cybernetics, Rockville, MD, United States) was employed for infarct area analysis. Red stained areas corresponded to unaffected areas, while that stained in white was the infarcted area (Joya et al., 2021). The cerebral infarct area percentage was: (cerebral infarct area/total brain area) × 100%, which is the percentage of the brain area affected by the infarct.

The brain hemisphere ipsilateral to MCAO was homogenized with pre-cooled physiological salt, centrifuged like the blood, and the supernatant was immediately collected and stored in the refrigerator at −80°C until use. SOD, GSH-Px, CAT activities, and the content of MDA (Liao et al., 2020; Homma et al., 2021; Liu et al., 2021)were determined according to manufacturer’s instructions.

The animals were anesthetized and decapitated, and brains were collected. The brain sample was embedded in paraffin, coronally sliced at 5-μm, and measured by HE staining (Zhang L. et al., 2021), and terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labeling (TUNEL) staining (Lee et al., 2021). The areas around five randomly selected injury sites were quantified. Images were acquired by an Olympus CX41 microscope and analyzed by Image-Pro Plus.

The chemical compositions of TFCJ were identified from Traditional Chinese Medicine Systems Pharmacology database and analysis platform (TCMSP¹) (Ru et al., 2014), Integrative Pharmacology-Based Research Platform of Traditional Chinese Medicine (TCMIP²) (Xu et al., 2019), and Bioinformatics Analysis Tool for Molecular Mechanism of Traditional Chinese Medicine (BATMAN-TCM³) (Liu et al., 2016). The compound targets were downloaded from PharmMapper database⁴ (Wang et al., 2017).

The key words ‘‘brain infarction,’’ ‘‘cerebral ischemia,’’ ‘‘cerebral infarction,’’ and ‘‘stroke’’ were input into the following GeneCards database⁵, OMIM database⁶ (Amberger et al., 2015), and DisGeNET database⁷ (Piñero et al., 2020) to search for IS-related targets.

The common TFCJ and IS-related targets were uploaded to the String11.0 database⁸ (Szklarczyk et al., 2019), the type was set to “Homo sapiens,” the confidence level was set to >0.700. Protein–protein interactions (PPI) network and herb-compound-target-disease network were constructed by Cytoscape 3.7.2(Shannon et al., 2003).

The common TFCJ and IS targets were uploaded to the DAVID 6.8 databases⁹ (Huang da et al., 2009) that used to analyze the GO and KEGG pathways enrichment of common targets. In our work, GO and KEGG terms with P < 0.01 were considered as significantly enrichment analyses.

The brain hemisphere ipsilateral to MCAO was homogenized with pre-cooled physiological salt, and the supernatant was collected and stored in the refrigerator at −80°C until use. SOD, GSH-Px, CAT activities, and the content of MDA were determined according to manufacturer’s instructions.

Total RNA of the ischemic penumbra of brain tissue was extracted using RNA isolater Total RNA Extraction Reagent (R401-01, Vazyme, Nanjing, China). The isolated RNA samples were reverse transcribed using HiScript^® QRT SuperMix for qPCR (R122-01, Vazyme, Nanjing, China) following standard techniques.

Data of BCL-2, BAX, and cleaved-Caspase-3 mRNA were normalized to β-actin mRNA (Table 1). RT-PCR was performed using a quantitative ChamQ Universal SYBR qPCR Master Mix (Q711-02, Vazyme, Nanjing, China).

We performed Western blot analysis following previously described protocols (Abbruzzese et al., 2020). The total proteins in the cerebral ischemia-reperfusion brain were purified, centrifuged, and collected in the supernatant. The protein concentrations were quantified with a bicinchoninic acid (BCA) protein assay kit (P0010, Beyotime, Shanghai, China). The protein samples were first loaded into 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene fluoride (PVDF) membranes. After blocking with 5% fat-free milk or BSA for 1 h, the PVDF membranes were incubated with primary antibodies at 4°C overnight (Table 2), following incubation with HRP-conjugated goat anti-mouse secondary polyclonal antibody (1:5000, 66009-1-Ig, Proteintech, Wuhan, China) for 2 h at room temperature. The protein bands were visualized using enhanced chemiluminescence (ECL), β-actin served as a loading control, and densitometric analysis was determined by ImageJ software. The ratio of the OD of the target protein to β-actin was used to represent the relative expression levels of the target proteins.

All data were expressed as mean ± SD; SPSS 25.0 statistical software (IBM SPSS Statistics for Windows, IBM Corp.) was adopted for all statistical analyses. The multiple variables were performed by one-way analysis of variance (ANOVA), and student’s t-tests were used for comparison of variable pairs. P < 0.05 was considered as statistically significant.

The neurological deficit scores and cerebral infarction area were evaluated (Figure 1) after reperfusion for 24 h. The neurological deficit scores in the MCAO group were higher than that in the sham group (Figure 1A, ^##P < 0.01); the neurological deficits in the rats pre-treated with TFCJ (10, 20, and 40 mg⋅kg^–1) were significantly lower than those in the MCAO group (see Figure 1A, *P < 0.05, ^**P < 0.01); the neurological deficit scores in the Xuesaitong group significantly decreased compared to the MCAO group (Figure 1B, ^**P < 0.01).

The sham group had a smaller cerebral infarct area, while the MCAO group had a bigger infarct area (Figure 1B, ^##P < 0.01). Cerebral infarct area in three dose groups of TFCJ was much lower than MCAO group (Figure 1B, ^**P < 0.01). The cerebral infarct area in the Xuesaitong group was significantly decreased than the MCAO group (Figure 1B, ^**P < 0.01).

The activity of SOD, GSH-Px, and CAT in the brain tissue and serum were lower in the MCAO group than in the sham group (Figures 2A–C,E–G, ^##P < 0.01), and the concentration of MDA was higher in the MCAO group than in sham group (Figures 2D,H, ^##P < 0.01), indicating that MCAO induced oxidative stress in the brain tissue and serum.

The activity of SOD, GSH-Px, and CAT in the brain tissue and serum of high and middle dose groups of TFCJ was much greater than the MCAO group (Figures 2A–C,E–G, ^**P < 0.01). The concentration of MDA in the brain tissue and serum of the three-dose groups of TFCJ was significantly lower compared to the MCAO group (Figures 2D,H, *P < 0.05, ^**P < 0.01).

Compared with the MCAO group, the activity of SOD and GSH-Px in the serum of the TFCJ-L group was significantly improved (Figures 2E,F, ^**P < 0.01), and the activity of CAT in the serum and brain tissue was significantly increased (Figures 2C,G, *P < 0.05).

Compared with the MCAO group, the activity of SOD, GSH-Px, and CAT in the brain tissue and serum of the Xuesaitong group was significantly improved (Figures 2A–C,E–G, ^**P < 0.01), and the concentration of MDA was significantly decreased (Figures 2D,H, ^**P < 0.01).

The structure of hippocampal neurons was intact in the sham group; the cells were regularly arranged and abundant in the cytoplasm; the cell nucleolus was large and obvious, and there was no pyknosis in the nucleus. Contrary, compared to the sham group, the neurons were disordered, and the cells were round vacuoles in the MCAO group. The arrangement of neuronal cells in the three-dose groups of TFCJ was regular, and the cell morphology was improved compared to the MCAO group (Figure 3A). Compared to the MCAO group, the neurons were neatly arranged, and the cell morphology was improved in the Xuesaitong group (Figure 3A).

The nucleus was stained in blue, and the TUNEL-positive neurons were brown. Almost no neuronal death was observed in the sham group. The numbers of TUNEL-positive neurons were significantly higher in the MCAO group than in the sham group (Figures 3A,B, ^##P < 0.01), while they were reduced in the TFCJ group (Figure 3B, ^**P < 0.01). Compared to the MCAO group, the numbers of TUNEL-positive neurons were significantly lower in the Xuesaitong group (Figure 3B, ^**P < 0.01).

The 12 compounds [luteolin, quercetin, isorhamnetin, kaempferol, acacetin, eupatorin, 5, 7-dihydroxy-2-(3-hydroxy-4-methoxyphenyl) chroman-4-one, linarin, diosmetin, chryseriol, naringenin, and artemetin in TFCJ are described in Supplementary Table 1]. The results retrieved from the PharmMapper databases were integrated to obtain the 300 targets with the highest matching degree with the 12 active TFCJ compounds. After deduplication, all information were merged, resulting in a database of 395 targets of TFCJ.

IS-related targets were identified from GeneCards, OMIM, and DisGeNET databases. After deduplication, all information were merged, resulting in a database of 5,111 IS-related targets, including 934 common targets (Figure 4A). A total of 247 common targets of TFCJ-anti-IS were identified using a Venn tool. The network was constructed with the String 11.0 database ignored disconnected targets, resulting in the PPI network of only 223 targets (Figure 4B). The top ten targets, selected according to the degree of the PPI network, were MAPK1, Akt1, PIK3R1, SRC, EGFR, MAPK8, HRAS, HSP90AA1, IGF1, and RHOA (Figure 4C). The herb-compound-target-disease network was composed of one herb, 12 compounds, 247 targets, and one disease. A diagram of this network is shown in Figure 4D.

We further analyzed the 247 targets of TFCJ-anti-IS using the David database. A total of 395 biological processes (BP), 50 cellular components (CC), 108 molecular functions (MF), and 110 KEGG pathways met the screening criteria of P-value < 0.05 (Supplementary Tables 2–5). The top ten terms of MF were identical protein binding, enzyme binding, receptor binding, serine-type endopeptidase activity, protein tyrosine kinase activity, kinase activity, steroid hormone receptor activity, RNA polymerase II transcription factor activity, ligand-activated sequence-specific DNA binding, transmembrane receptor protein tyrosine kinase activity, and non-membrane spanning protein tyrosine kinase activity. The top ten terms of CC were cytosol, extracellular exosome, plasma membrane, extracellular region, extracellular space, focal adhesion, extracellular matrix, membrane raft, blood microparticle, and an extrinsic component of cytoplasmic side of the plasma membrane. The top ten terms of BP were proteolysis, negative regulation of the apoptotic process, protein autophosphorylation, peptidyl-tyrosine phosphorylation, transcription initiation from RNA polymerase II promoter, response to hypoxia, steroid hormone mediated signaling pathway, phosphatidylinositol-mediated signaling, peptidyl-tyrosine autophosphorylation, and cellular response to insulin stimulus (Figure 4E).

The top 20 terms of KEGG pathways were pathways in cancer, PI3K-Akt signaling pathway, Ras signaling pathway, proteoglycans in cancer, Rap1 signaling pathway, FoxO signaling pathway, insulin signaling pathway, estrogen signaling pathway, pancreatic cancer, thyroid hormone signaling pathway, prolactin signaling pathway, prostate cancer, colorectal cancer, VEGF signaling pathway, central carbon metabolism in cancer, Fc epsilon RI signaling pathway, adherens junction, chronic myeloid leukemia, PPAR signaling pathway, and non-small cell lung cancer (Figure 4F).

The first part of the article found that TFCJ has a brain-protective effect and was related to the inhibition of oxidative stress and cell apoptosis. As the top 20 of the KEGG pathway, the PI3K-Akt signaling pathway has an important role in apoptosis and oxidative stress. Therefore, we focused on the regulation of TFCJ on the PI3K-Akt signal pathway and its downstream mTOR signal pathway (the 86th position of the KEGG pathway).

The p-Akt and p-mTOR expression in the MCAO control rats were significantly lower than in the sham rats (Figures 5A,B, ^##P < 0.01). The p-Akt and p-mTOR expression in the TFCJ treatment rats were significantly higher than in the MCAO control rats (Figure 5B, ^**P < 0.01). After using LY294002, the TFCJ treatment was reversed (see Figure 5B, ^▲▲P < 0.01). There was no significant change in total Akt and mTOR protein expression in each group (P > 0.05).

Compared with the sham group, the activities of SOD (Figure 6A), GSH-Px (Figure 6B), and CAT (Figure 6C) in the serum were significantly reduced in the MCAO control animals compared with the sham animals, while the level of MDA (Figure 6D) was significantly increased (Figures 6A–D, ^##P < 0.01). In the TFCJ treatment animals, GSH-Px (Figure 6B) and CAT (Figure 6C) activities were higher, while the MDA level (Figure 6D) was lower compared to MCAO control (Figures 6A–D, ^**P < 0.01). After using LY294002, the effect of TFCJ was reversed (Figures 6A–D, ^▲▲P < 0.01). The result showed that TFCJ could inhibit oxidative stress caused by MCAO in the serum through PI3K/Akt/mTOR pathway.

As shown in Figure 7, the BCL-2 protein expression was reduced after IS 24h (Figure 7A, ^##P < 0.01), whereas TFCJ upregulated BCL-2 protein expression (Figure 7A, **P < 0.01). The BAX and cleaved-Caspase-3 expression were increased after IS 24h (Figure 7A, ^##P < 0.01), whereas TFCJ upregulated BAX and cleaved-Caspase-3 protein expression (Figure 7A, **P < 0.01). After using LY294002, the TFCJ treatment was reversed (see Figure 7A, ^▲▲P < 0.01). Moreover, RT-PCR analysis indicated that MCAO reduced the expression of BCL-2 and increased the level of BAX and cleaved-Caspase-3 (Figure 7B, ^##P < 0.01), while TFCJ induced the expression of BCL-2 and decreased the level of BAX and cleaved-Caspase-3 (Figure 7B, **P < 0.01). After using LY294002, the effect of TFCJ was significantly reversed (Figure 7B, ^▲▲P < 0.01).