Preparation of HSHS
HSHS formular is composed of 13 herbs: Flos Chrysanthemi(Juhua), Radix Saposhnikoviae(Fangfeng), Ramulus Cinnamomi(Guizhi), Rhizoma Chuanxiong(Chuanxiong), Radix et Rhizoma Asari(Xixin), Radix Platycodonis (Jiegeng), Rhizoma Atractylodis Macrocephalae(Baizhu), Poria(Fuling), Rhizoma Zingiberis(Ganjiang), Radix Angelicae Sinensis(Danggui), Radix et Rhizoma Ginseng(Rensheng), Radix Scutellariae(Huangqin) and Concha Ostreae(Muli). All the herbs were purchased from Beijing Tongren-Tang Chinese Medicine Co. Ltd. (Beijing, China), and verified by Pro. Li Jia in keeping with the Chinese Pharmacopoeia 2020. The herbs were extracted with 10×30% ethanol (ten fold weight of the thirteen components) for 2 h. Then, sediment was immersed with 8×30% ethanol for 1.5 h. Vacuum extraction at 50°C to take out ethanol and got the HSHS extract. 1.2 g herbs generated one milliliter of the extract[7].
Animal experiments and samples collection
All male Sprague-Dawley (SD) rats (verified SPF, weighing 280-320g) were supplied by the Vital River Laboratory Animal Technology Co. Ltd., Beijing, China (license No. SCXK(Jing) 2016-0011). Rats were randomly distributed into sham group, model group and HSHS group with 10 rats in each group. All rats were anesthetized with a face mask inhalation of 1.5–2.5% isoflurane in a 2:1 N2O:O2 atmosphere during surgeries to minimize suffering. The preparation of MCAO model was performed as we described[15]. Additionally, the clinical equivalent daily dose of HSHS in rats was 10.5g/kg, which was the optimal dosage against MCAO in our previous study[16]. Seven days after MCAO, the ischemic core region of brain cortex tissues was frozen in liquid nitrogen immediately.
The above experiments were approved by the Institution Animal Care and Use Committee of Capital Medical University [Ethical License No.AEEI-2019-001], and all animal experimental manipulations conformed to the Guidance Suggestions for the Care and Use of Laboratory Animals published by the Ministry of Science and Technology of China.
Behavioral testing
Neurological assessment
The behavioral function was assessed at 1,3,5 and 7 days after MCAO. Elimination standard: rats scored 0, unable to walk or unconscious. Score scale: 0=no evident symptom; 1=can not wholly extend the left forepaw; 2=spin to the left; 3=failure to the left; 4=only walk when irritated.
Motor function assessment
The beam test was carried out at 3 and 7 days after MCAO[17]. Trained rats in advance to ensure they could cross a bar (100 cm × 3 cm, located 60 cm above the floor). After MCAO, test was conducted with the scoring system: 0, drop from the bar; 1, stand on the bar but no moving; 2, try to pass the bar but fall off; 3, pass the bar with ≥ 50% left posterior slips; 4, pass the bar with < 50% left posterior slips; 5, pass the bar with only one left posterior slip; 6, pass the bar and no slips.
Neuropathological assessment
7 days after cerebral ischemia, 3 rats from each group were transcardially perfused with 500 mL of physiological saline, and then changed to 4% paraformaldehyde(PFA). After that, immersed the brain tissue in 4% PFA to fix them, then made them dehydrated, embedded in paraffin and sliced into sections. The sections were baked, dewaxed, dehydrated, washed and stained with hematoxylin-eosin(HE). Pathological alterations were observed by a light microscope (Nikon, Tokyo, Japan).
Microarray analysis
RNA preparation and microarray detection
Total RNA was extracted from the infarcted tissue using Trizol reagent (Life Technologies, Carlsbad, CA, USA) and purified with an RNeasy mini kit (Qiagen, V alencia, CA, USA). Biotinylated cDNA was prepared in line with the standard Affymetrix protocol from 250ng total RNA by using Ambion® WT Expression Kit. Fragmented cDNA was hybridized for 16 h at 45℃ on Clariom D Assay(rat, Affymetrix). GeneChips were washed and stained in the Affymetrix Fluidics Station 450. All arrays were scanned by Affymetrix® GeneChip Command Console (AGCC) which was installed in GeneChip® Scanner 3000 7G.
DEmRNAs and DElncRNAs analysis
In microarrays, we used moderated F-statistic to select the multi-group DEmRNAs and DElncRNAs among groups (model vs sham and HSHS vs model) by the limma R package (vension:3.36.5). Limma R Empirical Bayes moderation was used to correct the P values. Benjamini-Hochberg was used for multiple tests correction (FDR was used to adjust the P values for multiple comparisons). The threshold set for up- and down-regulated mRNAs and lncRNAs was fold change > 1.2, P value < 0.05 and FDR < 0.05. The intersections of DEmRNAs and DElncRNAs between model vs sham and HSHS vs model were determined by Venn diagrams. Hierarchical clustering was performed based on overlapped mRNAs and lncRNAs using the R package pheatmap (vension:1.0.12).
Bioinformatics analysis of the coding genes
Gene Ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway analysis
Based on the GO database, analyzed the primary function of the DEmRNAs from molecular function, biological process and cellular component[18]. Moreover, according to the KEGG database, we found out the significant pathways of the DEmRNAs[19]. Pathway enrichment analysis indicated the main metabolic and signaling pathways. Through the Cytoscape plug-in ClueGO and CluePedia, we constructed the enriched pathways interaction networks.
Protein-protein interaction (PPI) network
We used the STRING database to investigate functional interactions between proteins, including prediction and experimental interaction data[20]. By setting the criteria with combined scores ≥0.9, we generated PPIs among the DEmRNAs, and used Cytoscape to visualize the network. Furthermore, we used the Molecular Complex Detection plug-in (MCODE) in Cytoscape to screen out significant modules from the PPI network with the conditions: degree cut-off = 2, k-core = 2, node score cut-off = 0.2, and max depth = 100. MCODE score ≥8 and node ≥10 were considered as functional enrichment analysis of the modules.
Bioinformatics analyses of the non-coding genes
LncRNA-mRNA co-expression network
LncRNAs regulate genes expression positively or negatively at the level of transcription and post-transcription. After normalizing the signal intensity between the DElncRNAs and DEmRNAs, the R function cor.test (Hmisc and corrplot) was used to calculate the Pearson's correlation coefficient(PCC) of lncRNA-mRNA pairs[21]. The significant correlation pairs and the correlation value cutoff was 0.99. The visualization of the network was built by the program Cytoscape3.8.3 software(http://cytoscape. github.io/). The degree was a measurement that indicates a lncRNA or mRNA centrality in a network. Core lncRNAs or mRNAs were determined by the degree differences between two groups.
LncRNA target pathway network
LncRNA target pathway network was constructed according to the relationships of significant pathways and lncRNAs[22]. In the network, a circle means a pathway, a square means a lncRNA, and relationships between them are indicated by edges. We evaluated the regulatory status of lncRNAs and pathways with graph theory; the standard were the degrees of lncRNAs and pathways in the network. The degree of each lncRNA was the number of pathways regulated by that lncRNA, and the degree of each pathways was the number of lncRNAs which regulated that pathway. Prime lncRNAs and pathways in the network had the highest degrees.
Quantitative real-time Polymerase Chain Reaction (qRT-PCR)
Total RNA was isolated from about 40 mg of cerebral ischemic cortex in rats with TRIZol(Sigma-Aldrich). RNA samples with an OD260/OD280 ratio of 1.9-2.1 and an OD260/OD230 ratio greater than 2.0 were used for analysis. The expression level of the destination mRNAs and lncRNAs was confirmed with a one-step qRT-PCR kit (Toyobo, Osaka, Japan). β-actin was used to normalize the mRNAs and lncRNAs. The expression level of mRNAs and lncRNAs was quantified with the Bio-Rad CFX real-time system and analyzed by the CFX management software v2.0 (BioRad, Hercules, CA). Relative quantification of mRNAs and lncRNAs was performed with the 2-ΔΔCt method, and each sample was normalized.
Gene.Symbol
|
forward
|
reverse
|
IL-6
|
5’-AGGATACCACCCACAACAGACC-3’
|
5’-TTGCCATTGCACAACTCTTTTC-3’
|
Wnt4
|
5’-ATCCTGACACACATGCGGGT-3’
|
5’-ATCCGTATGTGGCTTGAACTGC-3’
|
Rock2
|
5’-TGCTATTGGATAAACACGGACA-3’
|
5’-ACCAATCACATTCTCGTCCATAG-3’
|
Rps6kb1
|
5’-GAAATGCTGCTTCTCGTCTTGG-3’
|
5’-AGACCTGGTTGGCACTTTCACT-3’
|
GAPDH
|
5’-CTGGAGAAACCTGCCAAGTATG-3’
|
5’-GGTGGAAGAATGGGAGTTGCT-3’
|
LOC102555751
|
5’-GGAAGTTGGGTCTGGAGGTATT-3’
|
5’-TAGCCCAGTTATGAGCCTCTGT-3’
|
NONMMUG011611
|
5’-AACCTACTACTGTTATTCTGTGCGAG-3’
|
5’-GCTCCCTTAAAGGCCATCTTCT-3’
|
Data analysis and statistics
All data were analyzed with SPSS 16.0 software (SPSS Inc., Chicago, IL, USA), and presented as mean±SEM. Differences among two groups was analyzed by Student's t-test. Statistics in microarray and bioinformatic analysis were performed by One-way analysis of variance (ANOVA) using Affymetrix® Expression Console™ TAC (Affymetrix® Expression Console™), followed by Student's t-test. Using Spearman correlation analysis to analyze co-expression relationships between lncRNAs and mRNAs[23]. P < 0.05 was considered statistically significant.