BYHWD preparation

The original prescription of BYHWD is available in the ‘TCM Prescriptions Dictionary’, which lists the following seven Chinese herbs: Astragalus mongholicus Bunge, Angelicae sinensis (Oliv.) Diels, Ligusticumi chuanxiong Hort, Paeonia lactiflora Pall, Prunus persica (L.) Batsch, Carthamus tinctorius L, and Pheretima aspergillum (E. Perrier) (Table 1). The raw herbs were sourced from the Xiangya Hospital of Central South University (Changsha, China). The authenticity of the plant materials was confirmed by Professor Suiyu Hu from the Department of Chinese Herbal Medicine at Central South University in Changsha, China. Voucher specimens of the plants were deposited at Xiangya Hospital, affiliated with Central South University (Changsha, China). These herbs were decocted twice and dissolved in distilled water to a final concentration. Based on the preliminary experimental results, mice with ICH were intragastrically administered 43.29 g/kg raw herbs in the BYHWD group.

Table 1 Composition and corresponding ratio of BYHWDAnimals

Adult male C57BL/6 specific pathogen-free mice (25–30 g) were provided by the Animal Experimental Center of Central South University, China. The animals were housed under controlled conditions (temperature, 23 °C ± 2 °C; relative humidity, 50% ± 10%; 12 h light/darkness) and had free access to standard nutrition throughout the study period.

Experimental groups

The mice were randomly divided into the sham, ICH, and BYHWD groups. Before surgery, the animals were deeply anesthetized by sodium pentobarbital solution (0.3%, 0.22 mL/10 g). Then the tails of the mice were transected at a 3-mm diameter, and 15 μL blood was collected. Subsequently, the mice were mounted on a stereotactic headframe in the prone position. A scalp incision was made, and a small cranial burr was drilled in the skull near the right coronal suture. Blood was injected slowly (over 5 min) via a sterile microliter syringe into the right globus pallidus, according to the following coordinates relative to bregma: 0.5 mm posterior, 2 mm lateral, and 4 mm ventral to the cortical surface. As a precaution against backflow, the needle remained in place for an additional 10 min and was then slowly removed. Mice in both the ICH and BYHWD groups received the autologous blood injection procedure described above. The sham-operated group had the same type of empty needle inserted in the same way at the same site. Mice in the BYHWD group were administered BYHWD (43.29 g/kg) once a day over for 7 days. An equal volume of double-distilled water was used as the control for mice in the ICH and sham groups. Body weight was assessed before and on days 1, 3, and 7 post-modelling. Finally, all animals were sacrificed on day 7, and the whole brain or brain tissue surrounding the hemorrhagic region was harvested and stored for experimental evaluation. Failed models and mice that died were excluded.

Behavioural test

The survival status of the animals was recorded daily and Kaplan–Meier curves were used to express the survival rate. Neurologic deficits were assessed using the modified neurological severity score (mNSS) and foot fault test. The mNSS is based on motor, reflex, and balance tests, which are rated on a scale from 0 to 18. The higher the neurological function score, the more severe the neurological deficit. The lowest score was taken after three independent replicate experiments for each animal. To evaluate sensorimotor function, the foot fault test was performed 1, 3, and 7 days after modelling. Data are presented as the percentage of left foot faults per the total number of both left and right steps (normal score, 0%; maximal deficit score, 100%). The results of the pre-modelling behavioural tests were used as baseline levels, with higher scores indicating more severe neurological impairments.

Haematoxylin and eosin (H&E) staining

The paraffin sections of brain tissue were baked in an oven at 60 °C for 1 h. Next, they were soaked in xylene solution for 15 min and then replaced with clean xylene and soaked again for 15 min, shaking continuously throughout to ensure complete wax removal. For hydration, the sections were immersed in a gradient of concentrations containing alcohol, in the order of 100%, 100%, 95%, 85%, and 75%, for 5 min, respectively. The haematoxylin staining solution (Servicebio, China) was used to immerse the sections for approximately 10 min, followed by rinsing under running water for 5 min to remove floating colours. Next, the tissue sample was placed in hydrochloric acid ethanol for 3 s till it turned blue. Afterward, we immersed the sections in an eosin staining solution (Servicebio, China) for approximately 5 min. The sections were then dried naturally and sealed with neutral resin. Histopathological examinations were performed using a light microscope (Axio Image M2, ZEISS, Germany) at 25 °C according to the manufacturer’s instructions.

Nissl staining

For preparation, 4-μm coronal brain sections were deparaffinised in xylene, rehydrated with ethanol, and stained in Nissl Staining Solution (Servicebio, China) for 5 min at 60 ℃. Each section was deparaffinized in xylene. A light microscope (Axio Image M2, ZEISS, Germany) was used for observation.

mRNA microarray

Transcriptomic experiments were performed using microarrays to measure mRNA abundance. Briefly, total RNA from the right globus pallidus was extracted and purified using the RNeasy Mini Kit (Qiagen, USA). Tissues were prepared, and chips were hybridised using the Agilent Gene Expression Hybridisation Kit (Agilent Technology, USA). After washing, the arrays were scanned with an Agilent chip reader and analysed using the Agilent feature extraction software (version 10.5.1.1.1.1.1). Comparing the ICH group with the sham group and the BYHWD with the ICH group, differentially expressed transcripts and mRNAs were revealed (Fold Change, FC ≥ 1.3, P < 0.05). The mRNA expression in 15 mice (three groups with five replicates) was detected using microarrays.

Bioinformatics analysis

Gene Ontology (GO; containing biological process, cellular component, and molecular function and Kyoto Encyclopaedia of Genes and Genomes (KEGG)) was conducted to determine the biological annotation of targets using Metascape (http://metascape.org).

Construction of the network pharmacology

Candidate bioactive ingredients of seven herbs were screened from the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database (http://lsp.nwu.edu.cn/tcmsp.php) and the BATMAN-TCM (http://bionet.ncpsb.org/batman-tcm/) database and searched in the PubMed database. Only chemicals with an oral bioavailability ≥ 30 and a drug-likeness ≥ 0.18 that met the criteria recommended by the TCMSP database were retained. The name of the identified compound was input into the TCMSP database, SEA database (SEA Search Server bkslab.org), and Swiss Target Prediction (http://www.swisstargetprediction.ch/) to obtain its target name. The target name was used to convert to the target gene and confine the species to Homo sapiens in the UniProt (http://www.uniprot.org/) database. Information on ICH-associated target genes was collected from GeneCards (http://www.genecards.org/), the OMIM database (http://www.omim.org/), Therapeutic Target Database (TTD) (http://db.idrblab.net/ttd/), and Drugbank (https://go.drugbank.com/) database. The drug-protein-target and the disease-drug-target networks were constructed using Cytoscape 3.6.1.

Real-time quantitative polymerase chain reaction (RT-qPCR)

RT-qPCR assays were used to detect the amplification of specific genes. The total RNA from the basal ganglia of the mouse brain was reverse transcribed into cDNA using a first-strand synthesis kit (Vazyme, China) according to the manufacturer’s instructions. A CFX Connect fluorescent quantitative PCR system (Bio-Rad, USA) was used for RT-qPCR amplification. The cycling procedure was as follows: incubation at 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s and 55 °C for 30 s. Relative expression was calculated as 2−ΔΔC. GAPDH was used as the internal reference. Specific primer sequences are listed in Table 2.

Table 2 Sequences of primers for RT-qPCR validationImmunofluorescence

After immersion in xylene and alcohol, antigen repair was performed using sodium citrate buffer (pH = 6.0) for thermal repair, 0.1 M phosphate-buffered saline (PBS, pH 7.4) for three washes, 3% BSA to block non-specific antigen, followed by overnight incubation of the sections with anti-ATG5 (1:5000, Huabio, China) at 4 °C. Subsequently, the sections were incubated with the secondary antibody donkey anti-mouse Cy3 (1:1000, Jackson ImmunoResearch, USA) for 1 h at 25 °C. Following incubation, the cell nuclei were stained with DAPI (1:50, Solarbio, China). Finally, paraffin sections were covered with an anti-fluorescent quencher (Servicebio, China).

Western blotting

The protein concentrations of the samples were determined using a BCA Protein Assay Kit (Thermo, USA). Then, the samples were mixed with a fourfold loading buffer, boiled for 5 min, and stored at − 80 °C until use. Equal amounts of protein (30 μg) were electrophoresed on 10% polyacrylamide gel with sodium dodecyl sulphate (Applygen, China). After gel electrophoresis, the target proteins were transferred to polyvinylidene difluoride membranes (Millipore Corp., Billerica, USA) using WB transfer buffer (Applygen, China) and fixed with 5% skimmed milk on a shaker for 2 h. The membranes were then incubated with the primary antibody overnight at 4 °C. After washing three times with Tris-BufferedSalineTween20, the membranes were incubated with an HRP-labelled secondary antibody for 1 h at 25 °C. The PVDF membrane was then immersed in an electrochemiluminescence solution (ECL, Vazyme, China), avoiding light. Finally, the intensity of the bands was quantified using ImageJ software (Rawak Software, Germany). The ratio between the grey scale values of the target band and the internal reference band was used as the relative protein expression. The following primary antibodies were used: anti-ATG5 (1:5000, Huabio, China), anti-Beclin 1 (1:5000, Huabio, China), anti-cathepsin B (1:1000, Huabio, China), anti-p62 (1:1000, Huabio, China), and lc3b rabbit pAb (1:1000, Zenbio, China), with GAPDH (ABCAM, UK) as an internal reference. HRP-conjugated goat anti-rabbit IgG (1:10,000, Jackson ImmunoResearch, USA) and HRP-conjugated goat anti-mouse IgG (1:10,000, Jackson ImmunoResearch, USA) were used as secondary antibodies. All primary antibodies were diluted with WB primary antibody diluent (Beyotime, China). The reaction was performed using ECL reagents. The band density was visualised and analysed using Quantity One software (Bio-Rad, USA).

Transmission electron microscopy (TEM)

Brain tissue samples were fixed overnight in an electron microscope fixation liquid (Servicebio, China). After washing with PBS, samples were post-fixed in 1% osmium tetroxide in PBS for 2 h. Then, the samples were briefly washed with PBS, dehydrated in an ethanol series (30–100%), and embedded in London Resin (LR) white resin (Taab, Aldermaston, Berks, UK) for polymerisation. Ultrathin Sects. (60–80 nm thick) were cut using a diamond knife on a Reichert Ultracut ultramicrotome (Reichert Company, Vienna, Austria). Subsequently, sample morphology was identified using a transmission electron microscope (Hitachi, Japan) operating at 80 kV.

Statistical analysis

All quantitative statistical results are presented as the mean ± standard error. A two-way ANOVA was used to analyse the results of the behavioural tests; the quantitative results of the Nissl staining and WB were analysed by one-way ANOVA, followed by the Student–Newman–Keuls test and Dunnett’s test (for comparison of multiple experimental treatments to a common control value). Survival analyses were conducted using the log-rank test. All statistical analyses and plots were performed using GraphPad Prism 8.0. A p-value < 0.05 indicated statistical significance.