2.1 Grouping and treatment of animals
The rats were randomly divided into three groups: control group, model group and SGB group, with 10 rats in each group. Model group and SGB group were intraperitoneally injected with 0.6 mg/kg LPS to induce acute brain injury in rats. Six hours after injection, rats in SGB group were anesthetized by intraperitoneally injected with pentobarbital sodium, fixed on the operating table, and punctured at the spines of the 7th cervical spine of rats. After no blood and no cerebrospinal fluid were extracted, 0.2 mL of 1% lidocaine was injected . The SGB model was successfully established in rats with obvious Horner syndrome, i.e., ptosis of upper eyelid and wide eye fissure , while the control group was only injected with equal volume of normal saline.
2.2 Morris Water Maze experiment 
A water maze was prepared and platform quadrant was set. After surgery, the experimental rats were placed in the water maze for adaptive free swimming for 5 min. The rats were trained to find a platform at a fixed point every day for 4 days. On day 5, the time needed to find the platform in each group was recorded as the escape incubation period. If the rats failed to find the platform within 90 s, the escape incubation period was recorded as 90 s, that is, the learning ability of rats in each group was tested. The platform was removed for escape latency experiment from the 7th day after surgery, and the times of crossing the original platform quadrant of rats in each group were recorded to test the memory ability of rats.
2.3 Sample Collection
2 h after SGB block, rats in each group were decapitated and their brains were taken out. After rinsed with pre-cooled normal saline, hippocampal tissue was quickly separated on ice, part of which was fixed in 4% paraformaldehyde for tissue section preparation, and part of which was stored in a refrigerator at -80℃ for later use.
2.4 Histopathological observation of hippocampus
The hippocampal tissues fixed in 4% paraformaldehyde were routinely dehydrated, transparent, paraffin embedded, sliced, stained with HE and sealed with neutral gum. Pathological morphological changes of hippocampal tissues were observed under a microscope, and 5 fields were randomly selected for photography.
2.5 Neuronal apoptosis in hippocampus detected by TUNEL method
The sections of hippocampus prepared in 2.3 were dewaxed and hydrated, protease K was dropped in, and TUNEL reaction solution was added, DAB was used for color rendering, hematoxylin was used to counterstain, observed under the microscope and photographed. Positive neurons, namely apoptotic neurons, were brown, and normal neurons were blue and purple. The apoptosis rate of neurons (number of positive neurons/total number of neurons ×100%) was calculated.
2.6 Bax protein expression in hippocampus detected by immunohistochemistry
Sections of hippocampal tissue prepared in 2.3 were taken for dehydration and antigen repair, sealed with 5% goat serum, and incubated overnight with Bax antibody (1:1000) at 4℃. Then, enzymed-labeled goat anti-rabbit IgG was dropped in, followed by DAB for color rendering. Pictures were taken under microscope, and Image J Image analysis software was used for semi-quantitative analysis. The positive rate of protein (number of positive neurons/total number of neurons ×100%) was calculated.
2.7 Expression of Bax mRNA in hippocampus detected by QRT-PCR
Sections of hippocampal tissue stored in the refrigerator were taken to extract total RNA by TRIzol method and cDNA was obtained by reverse for testing. The reaction conditions were 90℃60 s, 90℃15 s, 65℃30 s and 72℃30 s, with a total of 40 cycles and 60 s extension at 75℃. Related expression of target genes was calculated by 2-ΔΔCt method. Primers required for PCR were: Bax upstream sequence 5 '-GGTCCCGAAGTAGGAAAGGA-3', downstream sequence 5 '-GTCCCGAAGTAGga-AAGGA-3'; GADPH upstream sequence of was 5 '-AccACAGTCCATGCCATcac-3', downstream sequence was 5 '-TCCAccACCCTGTTGCTGTA-3'.
2.8 Expression of Bax and Bcl-2 proteins in hippocampus detected by Western blot
The hippocampal tissue was taken from the refrigerator at -80℃, and the protein was lysed by tissue lysate, and the supernatant was centrifuged. The protein concentration was determined by BCA method. 30 μg protein was sampled, and SDS-PAGE electrophoresis was performed to isolate the protein. Then the protein was transferred to PVDF membrane by wet transfer method, and sealed with 5% goat serum for 1 h. Bax and Bcl-2 primary antibody (1:1000) were added after washing the membrane, and incubated at 4℃ overnight. Primary antibody (1:1000) was added after washing the membrane, and incubated at 37℃ for 2 h without light. After washing the membrane, ECL was developed, and gray values were scanned and analyzed by gel imaging system. Using GAPDH as internal reference, the relative expression levels of target proteins were calculated.
2.9 Statistical Analysis
Graph Prism 8.0 statistical software was used to analyze and process the experimental data, and the measurement data were expressed as mean ± standard deviation (x̄ ± s). One-way AVOVA was used for comparison between multiple groups, and LSD-t test was used for comparison between groups. P < 0.05 was considered statistically significant.