Male Sprague Dawley (SD) rats weighing (8–10 weeks old, 250 ± 10 g) were obtained from SPF (Beijing) Biotechnology Co., Ltd. (Beijing, China), permit number: SCXK (Jing) 2019-0010. All animals were handled in keeping with institutional guidelines and ethics. Rats were housed at 22 ± 2°C with a 12 h light/12 h dark cycle. The animals had free access to food and water. Randomization was used to allocated animals to various experimental groups and the data analysis were performed by a blinded investigator. The experimental processing is approved by ethics committee of Affiliated Hospital of Chengdu University of Traditional Chinese Medicine, number: DL2019002.

The l-borneol was purchased from Luodian, Guizhou, China, which is a high-quality production area for l-borneol. According to the method specified in the Pharmacopoeia of the People’s Republic of China, 88.6% of the l-borneol in this study meets the requirement of no less than 85.0%. Nimodipine was obtained from Yabao Pharmaceutical Group Co., Ltd. (Shanxi, China). Hematoxylin (batch number, G1140) and Eosin (G1100) was purchased from Solarbio co., Ltd. (Beijing, China). 2,3,5-Triphenyltetrazolium chloride (TTC) was purchased from Kelong co. (Chengdu, China). Brain derived neurotrophic factor (BDNF) (Lot No. 9ZFHS54AQT), Tie2 (Lot No. M6RQRBWEEN), Matrix metalloprotein 9 (MMP 9) (Lot No. YU5E7JW4UR) and Ang-1 (Lot No. JNS67D5Z7W) kits were purchased from Elabscience Biotech Co., Ltd. (Wuhan, China). Vascular endothelial growth factor (VEGF) (Lot No. A38381252) and TGF-β1 (Lot No. A18181155) kits were purchased from MultiSciences (Lianke) Biotech Co., Ltd. (Hangzhou, China).

The borneol dosing was calculated according to the dose conversion table of experimental animals. The 10 times of an adult’s (60 kg) daily consumption (0.3 g) is as equivalent dosing (0.05 g/kg). The 1, 2, four time(s) of equivalent dose were used as low dosing (0.05 g/kg), medium dosing (0.1 g/kg) and high dosing (0.2 g/kg). Every group were dissolved in 5% tween 80 solution. The concentration of nimodipine was 0.012 g/kg, which is dissolved in pure water.

Rats were adaptively fed for 5 days. A total of 54 male rats were divided into two parts of the experiment. The first part included the sham operation group, model group, vehicle group, nimodipine group, and l-borneol 0.05 g/kg group, l-borneol 0.1 g/kg group, l-borneol 0.2 g/kg group, with six rats in each group. The second part includes sham operation group, vehicle group, l-borneol 0.1 g/kg group, l-borneol 0.2 g/kg group, with three rats in each group. The rats were subjected to permanent MCAO surgery as described in previous study (Dong et al., 2018; Ma et al., 2020). The rat body temperature was kept at 37°C during surgery. The skin incision and blood vessel dissection only were performed in the sham operation (Control) group, and the rest of the operation was the same as the other groups. Perform neurobehavioral testing by the classic neurologic deficit score method Zea-Longa to determine the neurological deficit after MCAO (Table 1). For the first part of the experiment, then grade 2 of MCAO animals were randomly divided into six group: model group, vehicle group, nimodipine group, l-borneol 0.05 g/kg group, l-borneol 0.1 g/kg group, l-borneol 0.2 g/kg group. All of rats were treated by intragastric administration with 10 ml/kg. The control group and model group were given the same volume of pure water. The vehicle group was given the same volume of 5% tween 80 solutions. The nimodipine group, l-borneol 0.05, 0.1, 0.2 g/kg group were given the same volume of the corresponding drug solution. All rats were given continuous intervention for 3 days, and neurobehavioral tests were performed at 24, 48, and 72 h after MACO. In the second part of the experiment, rat brain tissue is used for IHC and PCR.

After neurological test, the rats were anesthetized and decapitated under anesthesia. Thereafter, the serum was obtained from abdominal aortathe and centrifuged 3,000 rpm after standing. The supernatant was collected and frozen at −80°C. The rat brains were promptly obtained and the olfactory bulb, lower brain stem and cerebellum were removed.

The brain was frozen at −20°C for 15 min. Five coronal brain sections of 2 mm thickness were stained with TTC at 37°C for 30 min in darkness. After TTC staining, those pieces were fixed with 4% formaldehyde for 24 h in the dark. The cerebral infarction rate was analyzed.

The supernatant serum had been obtained. levels of VEGF, Ang1, BDNF, Tie2, TGF-β1, MMP9 were detected according ELISA kits complied with the manufacturer's protocol.

The brain tissue was fixed in 4% paraformaldehyde and embedded in paraffin. The paraffin-embedded brain tissue was cut into slices. The slices were dehydrated and dewaxed and then stained with HE for morphological evaluation. Each slice was observed under ×200 and ×400 microscopy. According to the scoring standards of semi-quantitative injury and repair in Table 2 and Table 3, the pathological conditions of brain tissue in each group were analyzed.

According to the degree of pathological changes from light to heavy, the semi-quantitative results are as follows: slight or very small "-" marked 0, slight or small "+" marked 1, moderate "+" marked 2, severe or large "+ +" marked 3.

According to the degree of repair from bad to good, the semi-quantitative results are as followed: slight or very small "-" marked 0, slight or small "+" marked 1, moderate or medium "+ +" marked 2.

Total RNA was obtained with the Trizol Reagent, which was reverse transcribed to cDNA with the Servicebio^® RT First Strand cDNA Synthesis Kit (Servicebio. Co., Ltd., G3330, Wuhan, China). The CFX96 real-time PCR detection system (Bio-Rad Laboratories Ltd., Hertfordshire, United Kingdom) was used to perform qRT-PCR. The specific primers and according sequences were displayed in Table 4. In the reaction, 1 μL cDNA of each sample was mixed with Servicebio^® 2 × SYBR Green qPCR Master Mix (None ROX) (Servicebio. Co., Ltd., G3320, Wuhan, China) according to the manufacture’s protocol. The PCR conditions were listed as follow: 30 s at 95°C, then 40 cycles at 95°C for 15 s, followed by 60°C for 30 s. Results were normalized to Actin mRNA level and presented as the fold change (2^–ΔΔCt).

IHC was used to determine the expression of anti-angiotensin-converting enzyme (ACE) and CD34 protein. In brief, the brain sections with paraffin-embedded (4 μm) were rinsed with 3% H₂O₂ for 10 min to block endogenous peroxide activity and incubated with 10% goat serum albumin for 30 min to block nonspecific binding. Then, the sections were incubated with the primary antibodies (ACE, abcam, No. ab254222, United Kingdom, 1:100) and anti-CD34 (1:200) overnight at 4°C, after incubation, sections were incubated with corresponding secondary antibody (anti-rabbit IgG-HRP, 1:200). Counterstaining was performed using hematoxylin. For the semiquantitative analysis of the immunohistochemical results, three sections from each brain, with each section containing three microscopic fields from the ischemic cerebral cortex, were digitized under a ×200 objective. The immunoreactivity of the target proteins was quantified based on the integrated optical density of immunostaining per field using Image J software.

The brains were prepared, washed with cold phosphate buffer solution (PBS), resuspended in a lysis buffer, and sonicated the lysate. The proteins were separated on 10% sodium dodecyl sulfate (SDS) gels and transferred to poly-vinylidene fluoride membranes. After incubation with 1:1,000 primary antibody–ACE dilution buffer for 1 h, Goat Anti-Rabbit IgG (H + L) HRP (1:3,000, absin) was used as secondary antibody and developed by enhanced chemiluminescence.

Molecular docking analysis between l-borneol and were performed using Discovery Studio 3.5. The human X-ray crystal structures of VEGF, Ang1, BDNF, Tie2, TGF-β1, MMP9 were obtained from the Protein Data Bank (PDB) archives and used as target for molecular docking. Table 5 showed that the information about those targets PDB ID. The structure of l-borneol was drawn by ChemDraw. The active site was searched via “Receptor - Ligand interactions - Define and Edit Bind Site”. The interactions between l-borneol and target proteins were detected to obtain the score using LibDock function.

Data are expressed as mean ± SD. GraphPad Prism 8 (GraphPad Software Inc., San Diego, CA, United States) was used to analyze all data. Two or multiple groups were compared using Student's unpaired t-test or one-way ANOVA followed by Bonferroni post hoc test with F at p < 0.05 and no significant variance inhomogeneity. p < 0.05 was considered statistically significant.

Neurological function score was performed by Zea - Longa methods. The results showed in Figure 2. The trend of neurological score after MCAO is shown in Figure 2A, which suggested l-borneol post treatment could gradually improve the neurological function. In the day of MCAO, the rats in the control group had no neurological damage, so the score was 0, and the rats in the other groups were scored 2, indicating a successful modeling. The model and vehicle group had no significant difference, it is suggested that 5% Tween solution as a solvent has no significant effect on neurological function after cerebral ischemia. Compared with model group, nimodipine could significantly improve neurological function scores after 24 h, 48 h and 72 h (p < 0.05 or p < 0.01). Compared with vehicle group, l-borneol 0.2 g/kg could significantly reduce the neurological function scores after 72 h (p < 0.05). The other time points had the trend of improving the neural function, but there was no significant difference (p > 0.05). The results suggest that l-borneol can significantly reduce neurological deficits caused by cerebral ischemia.

For the assessment of infarction, TTC staining was used. TTC is converted to red formazone pigment by NAD and dehydrogenase and there of stained the viable cells deep red. The infarcted cells have lost the enzyme and cofactor and thus remained unstained dull yellow or white. The results showed in Figure 3, the rats in the control group had no significant infarct volume, the infarction of the model group and vehicle group showed significantly different from that of the control group (p < 0.01). The model and vehicle group had no significant difference. Cerebral infarct size was obviously reduced in the nimodipine group compared with the model group (p < 0.01). Compared with the vehicle group, l-borneol could reduce the cerebral infarction rate (p < 0.05 or p < 0.01), showing the dose dependent relationship.

The levels of VEGF, Ang1 and Tie two in the serum were performed by ELISA methods. The results showed in Figure 4. The expressions of angiogenic growth factors including VEGF and Ang-1 in the model and vehicle group were significantly decreased, while the expression of Tie two was significantly increased as compared with those in the control group (p < 0.01). The model and vehicle group had no significant difference. Compared with the vehicle group, VEGF and Ang-1 expression levels significantly increased in the l-borneol 0.2 g/kg groups (p < 0.01 or p < 0.05), Tie two expression in the l-borneol 0.1 and 0.2 g/kg group was significantly decreased (p < 0.05 or p < 0.01). These results demonstrate that l-borneol could promote angiogenesis in a dose-dependent manner against ischemic brain injury.

The levels of TGF-β1, BDNF, MMP9 in the serum were performed by ELISA methods. The results showed in Figure 5. Compared with control group, the expressions TGF-β1 and MMP9 were significantly increased in the model and vehicle group, the expressions level of BDNF was significantly decreased in the model and vehicle group (p < 0.01). The model and vehicle group had no significant difference. While, the expressions of TGF-β1 in the l-borneol 0.2 g/kg group was significantly decreased (p < 0.01), and the BDNF was significantly increased as compared with those in the vehicle control group (p < 0.05), and presenting dosing dependent. The expressions level of MMP9 was significantly decreased in l-borneol 0.1 g/kg group (p < 0.01). It suggested l-borneol could promote neurogenesis against ischemic brain injury.

HE staining was performed to obverse the pathological changes under light microscope with ×200 and ×400 magnification. The morphology and structure of the cerebral cortex were normal without obvious pathological changes in the control group. And there were many blood vessels full of red blood cells in some fields (Figure 6A). There was obvious infarct, the homogenized cells, liquefied and necrotic nerve tissue to form a cribriform focus in the vehicle group. And shows a large number of red neurons, which is a sign of acute ischemic injury. What’s more, the cell density was obviously reduced and there was swelling phenomenon, which is different from control group. Infarct size were found in l-borneol 0.1 and 0.2 g/kg, while the degree of ischemic damage was relatively light (Figure 6A). It was found that there was the decrease of liquefied necrosis area, rare red neurons, more capillaries, higher cell density and obvious phagocytosis of proliferative glial cells than the vehicle group. According to the results of semi-quantitative scoring of tissue injury and tissue repair (Figures 6B,C), the brain injury was significant reduced in l-borneol 0.1 and 0.2 g/kg group (p < 0.01, p < 0.05), l-borneol 0.2 g/kg group had more brain repair effect than vehicle group (p < 0.01), presenting vessels production and glial cells clear out necrotic neurons.

The ACE, CD34, HIF 1α mRNA expression was detected by qRT-PCR (Figure 7). Compared with control group, the expression of ACE mRNA was obviously increased in the vehicle group (p < 0.01), the expression of HIF1αmRNA showed an increasing trend, and the expression of CD34 mRNA showed a decreasing trend. Compared with vehicle group, l-borneol 0.1 and 0.2 g/kg could significantly reduce ACE expression (p < 0.01) and l-borneol 0.2 g/kg also increased CD 34 expression (p < 0.05). The expression of HIF1α mRNA in the l-borneol 0.1 and 0.2 g/kg groups showed a downward trend.

The effect of l-borneol interventions on blood vessels was measured by IHC staining using ACE antibody to mark vessels. The results showed that the number of ACE -positive cells was markedly elevated after MCAO than those of control group, However, the increased number of ACE -positive cells were attenuated by l-borneol (Figure 8A). The results of western blot also showed, l-borneol had a trend of reducing ACE expression compared with vehicle group (Figure 8B).

The effect of l-borneol interventions on vascular endothelial growth factor receptor (VEGFR) was measured by IHC staining using VEGFR antibody. The results showed that the number of VEGFR -positive cells was markedly reduced after MCAO than those of control group, However, the number of VEGFR -positive cells were increased by l-borneol (Figure 9A). In addition, we use CD34 antibody to mark blood vessels by IHC staining. CD34 staining was performed as an indicator of microvessel density (MVD). The results showed the number of CD34-positive cells was increased after MCAO than that of control group. This suggests that ischemia activates the endogenous angiogenesis compensatory improvement of ischemic damage. The number of CD34-positive cells was still markedly increased after MCAO with l-borneol interventions than that of control group (Figure 9B), suggesting l-borneol increased the number of the MVD, further promotes angiogenesis.

The heat map can visualize the data, and integrate the indicators to intuitively compare the strength of their expression. The heat map shows the overall regulation of l-borneol on cerebral ischemia and mechanism factors (Figure 10). l-borneol promotes angiogenesis and neurogenesis by increasing the levels of Ang 1, VEGF and BDNF after cerebral ischemia. The increase of CD34 protein confirms this result. l-borneol can also inhibit neuronal apoptosis and enhance the stability of neovascularization by inhibiting the levels of ACE, MMP9, HIF1α, TGF-β1 and Tie2, thereby improving the neurological deficit in ischemic rats, reducing the rate of cerebral infarction, and exerting neuroprotective effects.

The 3D and 2D binding graphs of the small molecule and the target protein after docking are shown in Figure 11, and the binding scores are shown in Table 6. The results showed that l-borneol has certain binding effect with VEGF, Tie2, Ang1, TGF - β1, BDNF, MMP9, mainly through hydrogen bonding.