Study on Neuroprotective Mechanism of Houshiheisan in Ischemic Stroke Based on Transcriptomics and Experimental Verification

Houshiheisan (HSHS), a classic prescription in traditional Chinese medicine (TCM), has shown outstanding efficacy in treating stroke. This study investigated various therapeutic targets of HSHS for ischemic stroke using mRNA transcriptomics. Herein, rats were randomly separated into the sham, model, HSHS 5.25 g/kg (HSHS5.25), and HSHS 10.5 g/kg (HSHS10.5) groups. Rats suffering from stroke were induced by permanent middle cerebral artery occlusion (pMCAO). After seven days of HSHS treatment, behavioral tests were conducted, and histological damage was examined with hematoxylin-eosin (HE). The mRNA expression profiles were identified using microarray analysis and quantitative real-time PCR (qRT-PCR) validated gene expression changes. An analysis of gene ontology and pathway enrichment was conducted to analyze potential mechanisms confirmed using immunofluorescence and western blotting. HSHS5.25 and HSHS10.5 improved neurological deficits and pathological injury in pMCAO rats. The intersections of 666 differentially expressed genes (DEGs) were chosen using transcriptomics analysis in the sham, model, and HSHS10.5 groups. The enrichment analysis suggested that the therapeutic targets of HSHS might regulate the apoptotic process and ERK1/2 signaling pathway, which was related to neuronal survival. Moreover, TUNEL and immunofluorescence analysis indicated that HSHS inhibited apoptosis and enhanced neuronal survival in the ischemic lesion. Western blot and immunofluorescence assay indicated that HSHS10.5 decreased Bax/Bcl-2 ratio and suppressed caspase-3 activation, while the phosphorylation of ERK1/2 and CREB was upregulated in a stroke rat model after HSHS treatment. Effective inhibition of neuronal apoptosis by activating the ERK1/2-CREB signaling pathway may be a potential mechanism for HSHS in the treatment of ischemic stroke.


Introduction
Stroke is a signifcant cause of death and morbidity among adults. To date, 80% of stroke cases occur due to thromboembolic occlusion. Neuron death and brain atrophy following a sudden drop in regional cerebral blood fow can cause permanent neurological damage [1,2]. Intravenous thrombolysis and mechanical thrombectomy are efective methods for ischemic stroke in clinics [3]. However, they have some application limitations, including a narrow therapeutic time window, strict evaluation criteria, and the risk of hemorrhage [4,5]. Additionally, more than 80% of stroke survivors sufer from motor impairment of the upper extremities and 50% still have it four years after the stroke [6]. Terefore, neuroprotective strategies bring the greatest hope for stroke survivors, while neuronal protection and regeneration have been the main focus to efectively rescue functional brain defcits [7].
Traditional Chinese medicine (TCM) has a long history and unique advantage in treating ischemic stroke [8], which has been one of the essential sources for new drug development to treat ischemic stroke [9]. As the frst classic prescription for stroke, HSHS can promote recovery of limb and language function in patients with ischemic stroke and improve clinically the quality of their lives [10,11]. HSHS plays a neuroprotective efect by decreasing infammatory factor expression [12] and reducing amyloid precursor protein accumulation 24 and 72 h after cerebral ischemia [13]. Additionally, HSHS improved axon growth by inhibiting Nogo-A/RhoA/ROCK2 and Netrin-1/rac1/Cdc42 pathways [14] and promoted angiogenesis by regulating HIF1α/VEGF and Ang-1/Ang-2 pathways to alleviate neurological damage after seven days of stroke [15]. However, the mechanisms and pathways underlying the multitargeted efects of HSHS on ischemic stroke have been incompletely elucidated.
Using advanced omics technology to study applying TCM is an efcient and comprehensive method that links traditional Chinese medicine and Western medicine [16,17]. As an essential part of systems biology, transcriptomics technology is an efective tool to detect the expression changes of global RNA in corresponding proteins [18,19]. Transcriptomics analysis can determine the precise therapeutic targets and their interactions, which is essential to clarify the multifaceted mechanism of traditional Chinese medicine prescriptions [20]. Highthroughput RNA-seq and microarray analyses have been widely used to reveal molecular mechanisms of Chinese herbal medicines for diseases such as stroke, cancer, and hypertension [21][22][23].
Herein, we aimed to explore the neuroprotective efect of HSHS on cerebral ischemia in a rat stroke model induced by pMCAO. Furthermore, a deliberate strategy was conducted that integrates transcriptomics methods and experimental verifcation to investigate the potential mechanisms of HSHS on ischemic stroke. Figure 1 shows the experimental fow chart of this study.

Animals.
In total, forty-eight male Sprague-Dawley rats (280-320 g) were supplied from Beijing Vital River Laboratory Animal Technology Co. Ltd., China. Tey were kept (three rats/cage) in the specifc pathogen-free animal room of the Animal Center of Capital Medical University, China. All animal protocols were approved by the Institution Animal Care and Use Committee of Capital Medical University (No. AEEI-2019-001). (Table S1), all obtained from Beijing Tongren-Tang Chinese Medicine Co. Ltd. and authenticated by associate professor Jia Li at Capital Medical University, Beijing. All herbs were mixed and immersed in the 10 × volume of 30% ethanol for 2 h, extracted at 40°C with ultrasound-assisted extraction for 1 h. Afterward, the precipitate was soaked in 8 × 30% ethanol at 40°C with ultrasound-assisted extraction for 40 min. Using a rotary evaporator, the two obtained fltrates were mixed and concentrated into the fnal extract (1.2 g/mL). Additionally, the chemical compositions of the extract were subjected to quality control [24].

Experiment Design.
All rats were randomly separated into four groups after adaptive feeding: the sham, model, HSHS5.25 (HSHS 5.25 g/kg), and HSHS10.5 (HSHS 10.5 g/ kg) groups (HSHS 10.5 g/kg was the clinical equivalent daily dose in rats). Te pMCAO model was prepared [25]. Rats were anesthetized with isofurane (5% for induction and 2% for maintenance) in a 2 : 1 N 2 O : O 2 atmosphere during surgery. After operation for seven days, the ischemic regions of the cortex were frozen in liquid nitrogen and stored at −80°C until further use.

Neurological Functional Assessment.
Te neurological dysfunction was assessed on postoperative days 1, 3, 5, and 7. Te test of neurological defcit was scored as follows [25]: (0) no evident symptoms, (1) unable to fully extend the left forepaw, (2) crawling while spinning to the left side, (3) fall to the left side while crawling, and (4) unable to walk or unconscious.
Te beam walking test was used to assess the motor coordination function of rats on days 3 and 7 after pMCAO. Before the operation, each rat was trained to ensure it could habituate to walking on the beam (80 cm long by 3 cm wide, located 60 cm high). Te test was scored as follows [26]: (0) cannot stay on the beam, (1) just stay on the beam but not move, (2) try to traverse the beam but fell, (3) traverse the beam with ≥50% hind-limb foot slips, (4) traverse the beam with <50% hind-limb foot slips, (5) traverse the beam and only one hind-limb slip, and (6) traverse the beam with no slips.

Histological
Assessment. Seven days after surgery, rats were anesthetized and transcardially perfused with 4% paraformaldehyde. Te brains were then routinely embedded in parafn, sectioned at 4 μm, and stained with HE. Light microscopy was used to observe pathological changes (Nikon, Japan).

Transcript Profle Analysis.
Total RNA from the periinfarct cortex of each group for three rats (sham, model, and HSHS10.5 groups) was extracted with TRIZol reagent (Life Technologies, USA). Ten, an RNeasy mini kit (Qiagen, USA) was used to purify total RNA from infarcted tissue. According to Afymetrix protocol, 250 ng total RNA was used to conduct biotinylated cDNA by Ambion ® WT Expression Kit. Te cDNA fragments were hybridized using a Clariom D assay (rat, Afymetrix) for 16 h at 45°C. Afymetrix Fluidics Station 450 was used to wash and stain GeneChips. All arrays were scanned using GeneChip ® Scanner 3000 7G with Afymetrix ® GeneChip Command Console (AGCC).

Diferentially Expressed Genes (DEGs) Analysis.
Te moderated F-statistic was used to choose the multigroup DEGs between model vs. sham groups and HSHS10.5 vs. model groups using the R package "limma" (version 3.36.5). P values were corrected using limma R Empirical Bayes moderating with Benjamini-Hochberg for multiple test corrections. Te threshold set for DEGs was as follows: fold change > 1.2, P < 0.05, and false discovery rate (FDR) <0.05. Among sham, model, and HSHS groups, Venn diagrams were used to determine the overlapped DEGs, performed by hierarchical clustering using the R package "heat-map" (version 1.0.12).

Functional Enrichment
Analysis. An enrichment analysis of KOBAS-i (https://kobas.cbi.pku.edu.cn/) was performed on the overlapping DEGs using gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) [27]. Te bubble charts of GO and KEGG pathway enrichment were plotted using a free online data analysis and visualization platform (https://www.bioinformatics.com.cn/). Pathways and GO terms were considered markedly enriched at P < 0.01.

Quantitative Real-Time PCR Validation.
Te total RNA of the ischemic cortex was obtained with a TRIZol reagent (Life Technologies, USA). Real-time PCR was conducted using a one-step qRT-PCR kit (Toyobo, Japan) and quantifed using the Bio-Rad CFX with a 20 μL system (Bio-Rad, United States). Relative quantifcation of mRNAs was performed using the 2−ΔΔ Ct method, and each sample was normalized. Table S2 lists the PCR primers.

TUNEL Assay.
Te parafn slices were dewaxed and hydrated. Brain slices were washed in PBS containing proteinase K (20 μg/mL), and stained with TUNEL detection reagent (G1501, Servicebio, China) at 37°C for 1 h. Te sections were collected using a fuorescent microscope (Nikon, Japan). ImageJ was utilized to quantify the number of TUNEL-positive cells.

Data Analysis and Statistics.
Results were expressed as mean ± standard error (SEM) and analyzed using GraphPad Prism 8.0.2 software. Te comparison of data between groups was analyzed using one-way analysis of variance (ANOVA) with the least signifcant diference test for multiple comparisons. Statistical signifcance was defned as P < 0.05.

HSHS Improved Neurological Defcits and Pathological
Injury in pMCAO Rats. To assess the neuroprotective efect of HSHS on pMCAO rats, neurological tests and hematoxylin and eosin (HE) staining were conducted. Compared to the model group, neurological defcit scores of HSHS10.5 group rats were reduced on days 3 and 5∼7 after the operation (P < 0.05 or P < 0.01), while the HSHS5.25 group showed a nonsignifcant decrease (Figure 2(a)). Te beam walking test suggested that rats in the treated group performed a better motor function after pMCAO. Te balancing beam scores in the model group decreased compared to the sham group on the 3 rd and 7 th day (P < 0.001) (Figure 2(b)). Compared to the model group, the scores in HSHS10.5 group increased on the 3 rd and 7 th day (P < 0.05 or P < 0.01). Balance beam scores in the HSHS5.25 group increased, but with a nonsignifcant diference.
HE staining revealed that neurons were disorderly arranged, the cell membrane was vague, the cell body was shrunk, the nucleus was stained with pyknosis, and neurons were missing in the ischemic brain. Te treatment with HSHS decreased the pathological abnormalities of the ischemic brain in pMCAO rats (Figure 2(c)).

HSHS Altered Gene Expression Profles in pMCAO Rats.
To further investigate the molecular mechanisms of HSHS, gene expression profles in pMCAO rats were analyzed using high-throughput microarray technology. In total, 8128 DEGs were identifed in the cortex of pMCAO rats between the model and sham group (Figure 3(a)). Tere were 868 DEGs in the cortex of pMCAO rats between the HSHS10.5 and model groups (Figure 3(b)). We obtained 666 DEGs that overlap for further analysis using the Venn plot (Figures 3(c)  and 3(d)). Te results of qRT-PCR confrmed the reliability of microarray data (Figure 3(e)).

Functional Enrichment Analysis of DEGs.
We imported the 666 DEGs into the KOBAS-i database for GO and KEGG pathway analyses to explore functional distribution in the DEGs. Tese genes were associated with multiple biological processes (BP) (P < 0.01) (Figure 4(a)). BP terms were mainly enriched in positive regulation of the neuronal apoptotic process, positive regulation of the apoptotic process, apoptotic process, and negative regulation of the ERK1 and ERK2 cascade, suggesting that HSHS exerts benefcial efects on ischemic stroke by regulating the apoptotic process.
Pathway annotation suggested that these genes were involved in 50 pathways (P < 0.01). Te top 21 pathways were performed (Figure 4(b)). Among these pathways, the PI3K/AKT signaling, mTOR signaling, and MAPK signaling pathways were highly associated with the target genes.

HSHS Prevented Neuronal Apoptosis in pMCAO Rats.
In pMCAO rats, the peri-infarct cortex cell apoptosis was evaluated by TUNEL staining. Te model group showed signifcantly more apoptotic cells that emit green fuorescence than the sham group, while the number of apoptotic neuronal cells in the HSHS5.25 and HSHS10.5 groups was signifcantly reduced compared to the model group (P < 0.01, P < 0.001) (Figures 5(a) and 5(c)). Te immunofuorescence method measured neuronal-specifc marker NeuN ( Figure 5(c)). According to the quantitative analysis, NeuN immunoreactivity in the model group signifcantly decreased (P < 0.001) ( Figure 5(d)). Te number of NeuN-positive cells signifcantly increased in HSHS5.25 (P < 0.05) and HSHS10.5 groups (P < 0.001) compared to the model group ( Figure 5(d)).

HSHS Regulated Apoptosis-Related Proteins in pMCAO
Rats. Bax, Bcl-2, and cleaved caspase-3 are major apoptosisrelated proteins. Te expressions of Bax, Bcl-2, and cleaved caspase-3 were examined using immunofuorescence (Figures 6(a)-6(c)). Te expressions of Bax and cleaved caspase-3 in the peri-infarct cortex were markedly upregulated compared to that of the model group (P < 0.001) (Figures 6(d) and 6(e)), but the expression of Bcl-2 was signifcantly downregulated (P < 0.01) (Figure 6(f)). In pMCAO rats that received HSHS10.5 treatment, the expressions of Bax (P < 0.01) and cleaved caspase-3 (P < 0.001) were signifcantly elevated, and Bcl-2 expression was increased (P < 0.05) in the peri-infarct cortex compared to the model group. Cleaved caspase-3 expression in the peri-infarct cortex was reduced in the HSHS5.25 group compared to the model group (P < 0.01). Furthermore, the cortex around the infarction was examined using a western blot to confrm the regulation of HSHS for apoptosis-related proteins in pMCAO rats (Figures 6(g)-6(i)). We observed a signifcant increase in the Bax/Bcl-2 ratio in  Evidence-Based Complementary and Alternative Medicine the model group compared to the sham group (P < 0.01) and a substantial decrease after HSHS10.5 treatment (P < 0.05) (Figure 6(h)). Furthermore, rats treated with HSHS5.25 and HSHS10.5 showed less cleaved caspase-3 protein level than that of the model group (P < 0.05, P < 0.01) (Figure 6(i)), consistent with the result of immunofuorescence.

HSHS Increased Expression of ERK1/2-CREB Signaling
Pathway-Related Proteins in pMCAO Rats. tTe expressions of ERK1/2-CREB signaling-related proteins were examined to further investigate the possible mechanisms of HSHS on ischemic stroke using western blot. Compared to the sham group, the p-ERK1/2/ERK1/2 ratio in the model group was downregulated (P < 0.05). Te rats in the HSHS5.25 and HSHS10.5 groups showed a signifcant increase in the p-ERK1/ 2/ERK1/2 ratio compared to the model group (P < 0.01) (Figures 7(a) and 7(b)). Compared to the sham group, the p-CREB/CREB ratio in the model group was downregulated (P < 0.05), while the p-CREB/CREB ratio in the HSHS10.5 group increased compared to the model group (P < 0.05) (Figures 7(a) and 7(c)). Te p-CREB/CREB ratio in the HSHS5.25 group was also upregulated (Figure 7(c)).

Discussion
Ischemic stroke is mainly a disorder of blood supply to the brain resulting from various causes, contributing to a clinical syndrome characterized by hypoxic-ischemic damage to brain tissue [28]. Besides vascular recanalization, TCM has shown remarkable neuroprotective efects and gained great attention in treating ischemic stroke [29]. In TCM, HSHS was created by Zhang et al. to treat stroke, following the pathogenesis of defciency of genuine qi and excess of pathogenic factor [12]. Furthermore, HSHS has been used to treat stroke for approximately 2000 years and is safe and efective. However, the molecular mechanism of action has not yet been fully elucidated. Herein, transcriptome analysis and in vivo experiments were used to systematically investigate the pharmacological mechanisms of HSHS in treating ischemic stroke. Herein, HSHS exerted neuroprotective activity on pMCAO rats by improving the symptoms of neurological impairment and pathological injury. Ten, the high-throughput sequencing technology of the microarray chip was conducted to explore the therapeutic mechanism of HSHS for ischemic stroke from the whole transcriptome level. We identifed 8128 DEGs between the model and sham groups and 868 DEGs between the HSHS10.5 and model groups. We obtained 666 intersection DEGs between the sham, model, and HSHS10.5 groups. Furthermore, GO enrichment analysis on total intersection DEGs showed that the efects of HSHS on ischemic stroke were associated with positive regulation of neuron apoptotic process, positive regulation of the apoptotic process, apoptotic process, and negative regulation of ERK1 and ERK2 cascade. Moreover, the KEGG pathways analysis demonstrated that the intersection of DEGs was mainly associated only with the PI3K-Akt signaling, mTOR signaling, and MAPK signaling pathways. Accordingly, the neuroprotective efect of HSHS in stroke rats was related to the regulation of neuronal apoptosis.
Apoptosis plays a vital role in ischemia-induced neuronal death in ischemic stroke [30]. In the infarct core, excitotoxicity and neuronal necrosis occur in several minutes [31]. However, many dormant or semidormant nerve cells in the ischemic penumbra mainly occur in delayed death in the form of apoptosis [32]. Tese cells are the most possible and valuable to be rescued in clinics [30]. Subsequently, preventing neuronal apoptosis in the penumbra and improving its dysfunction is vital to treat ischemic stroke [33]. Herein, TUNEL and NeuN staining results showed that HSHS reduced the number of cell apoptosis to protect neurons in pMCAO rats. Meanwhile, treatment of HSHS decreased the expression of proapoptotic proteins Bax and cleaved caspase-3 and increased antiapoptotic protein Bcl-2 expression. Te previous results indicated that HSHS signifcantly increased the number of surviving neurons by preventing apoptosis in the peri-infarct of pMCAO rats.
survival in ischemic stroke models in vivo and in vitro [37,38]. ERK1/2 acts as a neuroprotective agent by inhibiting postischemic oxidative stress and mitochondriadependent apoptosis of neural cells [39,40]. ERK1/2 activation can promote the phosphorylation of CREB, increase the expression of prosurvival protein Bcl-2, inhibit ischemiainduced neuronal apoptosis, and enhance neuronal survival [41,42]. As a post-translationally activated transcription factor, cyclic AMP response element binding protein (CREB) participates in many brain functions, such as promoting neuronal survival mainly by increasing the expression of neurotrophic factors and antiapoptotic genes [43,44]. Hypoxia and ischemia increase the phosphorylation of CREB in brain tissue. However, inhibiting the phosphorylation of CREB reduces Bcl-2 expression [45]. Herein, the phosphorylation of ERK1/2 and CREB was downregulated in pMCAO rats, whereas HSHS treatment protected neurons and increased ERK1/2 and CREB phosphorylation. Tese results indicated that ERK1/2-CREB pathway activation might play a vital role in the neuroprotection of HSHS on ischemic stroke. Tese fndings provide a solid theoretical basis for the clinical application of HSHS in ischemic stroke.

Conclusion
Tis study used transcriptome analysis and in vivo experiments to systematically investigate the neuroprotective mechanisms of HSHS in ischemic stroke. Te results suggested that HSHS may prevent neuronal apoptosis by activating the ERK1/2-CREB signaling pathway, providing a novel insight for treating ischemic stroke.

Data Availability
Te data that support the fndings of the study are available in this article.

Ethical Approval
Te study was approved by the Institution Animal Care and Use Committee of Capital Medical University (No. AEEI-2019-001).

Conflicts of Interest
Te authors declare that they have no conficts of interest.  . Results were presented as mean ± SEM, n � 3. # P < 0.05 vs. sham group, and * P < 0.05 and * * P < 0.01 vs. model group. 10 Evidence-Based Complementary and Alternative Medicine