reperfusion injury by QiShenYiQi via neuroinflammatory network mobilization

Biomedicine & Pharmacotherapy 125 (2020) 109945

Contents lists available at ScienceDirect

Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha

Protection against acute cerebral ischemia/reperfusion injury by QiShenYiQi via neuroinflammatory network mobilization

T

Yule Wanga,b, Guangxu Xiaoa,b, Shuang Hea,b, Xinyan Liua,b, Lin Zhua,b, Xinyue Yanga,b, Yiqian Zhangc, John Orgaha,b, Yuxin Fenga,b, Xiaoying Wangd, Boli Zhanga, Yan Zhua,b,* a

State Key Laboratory of Component-based Chinese Medicine, Tianjin University of Traditional Chinese Medicine, Beihua South Road, JingHai District, Tianjin 301617, China b Research and Development Center of TCM, Tianjin International Joint Academy of Biotechnology & Medicine, 220 Dongting Road, TEDA, Tianjin 300457, China c State Key Laboratory of Core Technology in Innovative Chinese Medicine, Tianjin Tasly Holding Group Co., Ltd., Tianjin, China d Neuroscience Program, Neuroprotection Research Laboratory, Department of Neurology and Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States

ARTICLE INFO

ABSTRACT

Keywords: QiShenYiQi Cerebral ischemia/reperfusion injury Neuroinflammatory response TCM-based network pharmacology

Cerebral ischemia/reperfusion injury (CI/RI) is a common feature of ischemic stroke, involving a period of impaired blood supply to the brain, followed by the restoration of cerebral perfusion through medical intervention. Although ischemia and reperfusion brain damage is a complex pathological process with an unclear physiological mechanism, more attention is currently focused on the neuroinflammatory response of an ischemia/reperfusion origin, and anti-inflammatory appears to be a potential therapeutic strategy following ischemic stroke. QiShenYiQi (QSYQ), a component-based Chinese medicine with Qi-tonifying and blood-activating property, has pharmacological actions of anti-inflammatory, antioxidant, mitochondrial protectant, antiapoptosis, and antiplatelet aggregation. We have previously reported that the cardioprotective effect of QSYQ against ischemia/reperfusion injury is via improvement of mitochondrial functional integrity. In this research work, we aimed to investigate the possible mechanism involved in the neuroprotection of QSYQ in mice model of cerebral ischemia/reperfusion injury based on the inflammatory pathway. The cerebral protection was evaluated in the stroke mice after 24 h reperfusion by assessing the neurological deficit, cerebral infarction, brain edema, BBB functionality, and via histopathological assessment. TCM-based network pharmacology method was performed to establish and analyze compound-target-disease & function-pathway network so as to find the possible mechanism linking to the role of QSYQ in CI/RI. In addition, RT-qPCR was used to verify the accuracy of predicted signaling gene expression. As a result, improvement of neurological outcome, reduction of infarct volume and brain edema, a decrease in BBB disruption, and amelioration of histopathological alteration were observed in mice pretreated with QSYQ after experimental stroke surgery. Network pharmacology analysis revealed neuroinflammatory response was associated with the action of QSYQ in CI/RI. RT-qPCR data showed that the mice pretreated with QSYQ could significantly decrease IFNG-γ, IL-6, TNF-α, NF-κB p65, and TLR-4 mRNA levels and increase TGF-β1 mRNA level in the brain compared to the untreated mice after CI/RI (p < 0.05). In conclusion, our study indicated the cerebral protective effect of pretreatment with QSYQ against CI/RI, which may be partly related to its potential to the reduction of neuroinflammatory response in a stroke subject.

Abbreviations: QSYQ, QishenYiqi; CI/RI, cerebral ischemia/reperfusion injury; ASCVD, atherosclerotic cardiovascular disease; CVDs, cardiovascular diseases; TCM, traditional Chinese medicine; MCAO, middle cerebral artery occlusion; CCA, common carotid artery; ECA, external carotid artery; ICA, internal carotid artery; BBB, blood-brain barrier; EB, Evans Blue; TTC, 2,3,5-triphenyl-2H-tetrazolium chloride; H&E, hematoxylin & eosin; IFNG-γ, interferon-γ; IL-1, interleukin-1; IL-6, Interleukin-6; TNF-α, Tumor necrosis factor-alpha; NF-κB, nuclear factor κB; TLRs, Toll-like receptors; ROS, reactive oxygen species; HSPs, heat-shock proteins; HMGB1, high mobility group box 1; NO, nitric oxide; MMPs, matrix metalloproteinases ⁎ Corresponding author at: State Key Laboratory of Component-based Chinese Medicine, Tianjin University of Traditional Chinese Medicine, Beihua South Road, JingHai District, Tianjin 301617, China. E-mail address: [email protected] (Y. Zhu). https://doi.org/10.1016/j.biopha.2020.109945 Received 28 September 2019; Received in revised form 22 January 2020; Accepted 23 January 2020 0753-3322/ © 2020 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

1. Introduction

pathogenesis of CI/RI, anti-inflammation appears to be a potential therapeutic strategy for neuroinflammatory injury after ischemic stroke [31]. Neuroinflammatory reactions are mediated by the production of several crucial cytokines, chemokines, secondary messengers (NO), reactive oxygen species (ROS) and matrix metalloproteinases (MMPs) [32]. Many of these mediators are produced by resident immune cells (including microglia and astrocytes) of the central nervous system, endothelial cells, and peripherally derived immune cells [33]. Among these inflammatory mediators, three major pro-inflammatory cytokines: interleukin-1 beta (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α) are highly concerned because of their influence on neuroinflammatory response [34]. Alternatively, transforming growth factor-β1 (TGF-β1), as an accepted anti-inflammatory cytokine, can suppress the levels of pro-inflammatory cytokines thereby inhibiting inflammation and ischemic damage [35]. Nuclear factor κB (NF-κB), a typical transcription factor, is largely associated with the activation of inflammatory responses [36]. It has been reported that down-regulation of pro-inflammatory cytokines expression via reducing NF-κB p65 signaling could attenuate brain damage induced by experimental stroke [37]. Besides, toll-like receptor 4 (TLR4) also plays a vital role in CI/RI, and the reduction of TLR4-mediated neuroinflammation can exert a protective effect on brain damage caused by ischemia/reperfusion [38]. Based on the previous investigations, we proposed that QSYQ may also contribute to cerebral protection in addition to its known role in cardioprotection. In the present study, we aimed to test our hypothesis that pretreatment of QSYQ is beneficial to protect against focal ischemic stroke in mouse model of CI/RI. TCM-based pharmacology network and experimental verification combination method was performed to decipher the potential link between QSYQ and neuroinflammation in response to CI/RI.

According to the 2019 update of heart disease and stroke statistics by American Heart Association (AHA), ischemic stroke is still one of the leading causes of death and serious long-term disability worldwide [1]. At present, intravenous thrombolysis and intra-arterial thrombectomy are the standard clinical treatment strategies for patients with acute ischemic stroke [1,2]. Although recombinant tissue plasminogen activator (rt-PA, alteplase) is well-documented for its effect on improving the prognosis of acute ischemic stroke, the relatively narrow therapeutic time window and a severe hemorrhagic risk limit its further clinical application [3,4]. Meanwhile, the experienced stroke centers and qualified neurointerventional doctors are indispensable to operate rapid mechanical thrombectomy [5]. Moreover, even though restoration of the blood supply by reperfusion can reduce the infarct size and tissue dysfunction caused by ischemia, rapid reperfusion may also result in a more deleterious effect on functional outcomes [6]. There is no doubt that cerebral ischemia/reperfusion injury (CI/RI) has become an increasingly critical challenge for a stroke patient [7]. Up to date, a series of researches indicated that the main underlying mechanisms of CI/RI included inflammatory response, oxidative/nitrosative stress, mitochondrial dysfunction, calcium overload, activation of apoptotic and autophagic pathways, platelet activation and aggregation, leukocyte infiltration, and complement activation [8–12]. Due to the complicated pathological processes of cerebral ischemia/reperfusion, current therapeutic strategies have not achieved the expected effects. Therefore, seeking novel and alternative therapeutics for patients with acute ischemic stroke is still a significant task worldwide. In China, traditional Chinese medicine (TCM), especially compound TCM preparations, has been widely applied in clinic for the treatment of stroke for thousands of years [13]. Based on the theory of TCM, “Qi deficiency and blood stasis” is one of the primary pathogenesis characteristics of ischemic stroke, compound TCM preparations, which belong to tonifying Qi and activating blood, can protect against CI/RI via effecting on multiple targets and multiple pathways [14]. Qishen Yiqi (QSYQ) formula, which is composed of Astragalus membranaceus (Huangqi), Salvia miltiorrhiza (Danshen), Panax notoginseng (Sanqi) and Dalbergia odorifera (Jiangxiang), exerts the function of Qi-tonifying and blood-activating. QSYQ dripping pills is a typical representative of component-based Chinese medicine, which comprises standard extracts from the above four Chinese herbal medicine, approved by the State Food and Drug Administration of China in 2003. QSYQ formula and its dripping pill have been widely used in the clinic to treat cardiovascular diseases (CVDs), including angina pectoris of coronary heart disease origin with Qi deficiency and blood stasis syndrome [15], and secondary prevention of myocardial infarction [16]. Pharmacological studies have reported that QSYQ is capable of cardioprotection against ischemia/reperfusion injury via regulation of energy metabolism, reduction of oxidative stress and inflammation, as well as prevention of apoptosis and excessive autophagy [17,18]. Additionally, we have further found the major bioactive ingredients in QSYQ, such as Astragaloside IV, Calycosin, Ginsenoside Rg1, Ginsenoside Rb1, Notoginsenoside R1, Salvianolic acid A, Salvianolic acid B, and Tanshinone IIA [19,20]. Interestingly, a series of literatures have also reported that the above active components of QSYQ had the potential to protect brain from ischemia/reperfusion injury via multifunctional mechanisms [21–28] (details summarized in Supplemental Table S1). Besides, a recent study by Han and co-workers reported that T541, which is a compound medicine consisting of total Astragalus saponins, total salvianolic acids, and total Panax notoginseng saponins, could exert the function of protecting the integrity of blood-brain barrier (BBB) [29]. These findings imply that in addition to its known role in cardioprotection, QSYQ may also play a role in cerebral protection in a specific pathological setting. Neuroinflammation is defined as an inflammatory response within the brain or spinal cord [30]. Since it contributes prominently to the

2. Materials and methods 2.1. Drugs and regents QiShenYiQi (QSYQ) (NMPN Z20030139) was obtained from Tasly Pharmaceutical Group Co., Ltd. (Tianjin, China), which was prepared from water-ethanol extracts of Astragalus membranaceus (Huangqi), Salvia miltiorrhiza (Danshen), Panax notoginseng (Sanqi) and volatile oils from Dalbergia odorifera (Jiangxiang) by water extraction and distillation. QSYQ formula (Huangqi: Danshen:Sanqi: Jiangxiang oil = 148.01:70.35:70.35:11.97) was prepared and dissolved in ultrapure water to make a solution at concentrations of 5.69 mg/mL, 11.38 mg/mL and 22.75 mg/mL for experiments. Edaravone injection (drug approval number: H20031342) was purchased from Nanjing Xiansheng Dongyuan Pharmaceutical Co., Ltd. (Nanjing, China). TTC (2, 3, 5-Triphenyl-2H-Tetrazolium Chloride) was purchased from Solarbio (T8170−5 g, Beijing, China). Evans Blue was purchased from Solarbio (E8010−1 g, Beijing, China). Paraformaldehyde (4 %) was purchased from Solarbio (P1110−500 mL, Beijing, China. Isoflurane Lot No. B506 was purchased from Ruiwode Lifescience Co., Ltd. Shenzhen, China. Omnipaque 300 mg I/mL, drug approval number: H20000595, Batch No. 12,596,129 was purchased from GE Pharmaceutical Shanghai Co., Ltd. Shanghai, China. TRIzol® reagent was obtained from Invitrogen (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Transcriptor First Strand cDNA Synthesis Kit was obtained from Roche (Mannheim, Germany). Bestar®Sybr Green qPCR Master Mix was obtained from DBI®Bioscience (Ludwigshafen, Germany). 2.2. Ethical use of the laboratory animal Male 6-week-old Institute of Cancer Research (ICR) mice (22 ± 2 g) were purchased from Beijing Vital River Lab Animal Technology Co. Ltd. (Beijing, China, Certificate No.: SCXK Jing 2014-0013). This study was performed in accord with the recommendations by the Ministry of 2

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Science and Technology of China’s Guidance for the Care and Use of Laboratory Animals. The protocol approved by the Laboratory Animal Ethics Committee of Tianjin University of TCM (Permit Number: TCMLAEC2016034 and TJU20160024). The animals were housed in cages with an ambient temperature of 22 ± 2 °C and humidity of 40 ± 5 %, under a 12 h light/dark cycle, standard mouse’s chow and water were provided ad libitum. Animals were fasted for 12 h before the surgery and provided with free access to water only. The mice were also randomized into several groups. The experimental procedures were in accord with the European Union (EU) adopted Directive 2010/63/EU. All animal experiments were carried out following the guidelines of Tianjin University of TCM Animal Research Committee (TCMLAEC2016034) and approved by the animal care and use committee of Tianjin International Joint Academy of Biotechnology (No. TJU20160021). Efforts were made to minimize the number of animals used and the suffering of experimental animals.

contralateral side); score 3, severe focal neurological deficit (falling to the contralateral side); score 4, unable to walk spontaneously with a depressed level of consciousness or death. 2.6. TTC staining and measurement of cerebral infarct After 24 h reperfusion, the mice were euthanized following anesthetization. Then, the brains were quickly removed and freshly dissected into coronal slices (five 2 mm-thick, n = 6 for each group) in a rodent brain matrix. Subsequently, the cerebral slices were stained with 2 % TTC solution in the dark for 15 min at 37 °C. White infarct area in brain tissue was distinguished from red non-infarct area. Stained brain slices were photographed, and cerebral infarct area was calculated using ImageJ image processing software (ImageJ 1.42, Wayne Rasband, National Institutes of Health, United States). Ratios of infarct volume, which revealed the severity of CI/RI, was presented as a percentage (%) of (total cerebral infarct volume/total brain volume) × 100. Total infarct volume for each brain was calculated as summation of the infarcted areas of all brain slices (area of infarct in square millimeters × thickness [2 mm]) from the same hemisphere [43].

2.3. Animal grouping and drug administration Mice were divided into six different groups, including Sham, I/R, Positive control, QSYQ low dose (5.69 mg/mL)+I/R, middle dose (11.38 mg/mL)+I/R, and high dose (22.75 mg/mL)+I/R groups. Seven days before I/R surgery, the mice in treatment groups were orally administered with QSYQ in ultrapure water once daily for 7 days. Positive control drug-edaravone (2 mg/mL) was administered via intravenous injection once after 50 min of ischemia. Animals in shamoperated and I/R model groups received 0.9 % normal saline via gavage at the dose of 10 mL/kg for 7 days.

2.7. Assessment of brain edema via micro-CT scan Anesthetized mice were intra-arterially injected with 0.5 mL CT contrast agent (Omnipaque) over a period of 15 s, at 24 h post-reperfusion. Then, the mouse brain was imaged with a small animal micro-CT imager (Quantum FX; PerkinElmer, United States) with the following main imaging parameters: voltage for 90 kV, current for 180 μA, field of view (FOV) for 20 mm, and scan for 4.5 min. The 3D reconstruction of whole-cerebral slide images was performed using “VOL EDIT” module in Analyze 12.0 image analysis software (Analyze Direct, Overland Park, KS, USA). Next, the ROI measure module of software was used to calculate the volumes of the left and right cerebral hemisphere respectively. According to the left cerebral hemisphere volume as well as the ratio of the bilateral cerebral hemisphere (left cerebral hemisphere volume/right cerebral hemisphere volume), we can indirectly evaluate the severity of cerebral edema. Using the ratio of the bilateral cerebral hemisphere volume, the degree of the cerebral edema is evaluated.

2.4. MCAO surgical procedure Surgery was executed by transient middle cerebral artery occlusion (tMCAO) technique as described elsewhere [39–41]. In brief, anesthesia was administered with 4 % isoflurane inhalation and then maintained with 1.5 % isoflurane in 70 % nitrous oxide (N2O)/30 % oxygen (O2) with the help of a laboratory anesthesia system (Matrix VIP 3000; Midmark, USA). We put a drop of ophthalmic lubricant on each eye using a sterile swab to prevent them from drying during the procedure. After shaving and swabbing the incision area with 70 % ethanol, the animal was placed on a warm surgical surface under a stereomicroscope. Then we performed a shallow midline incision in the neck and carefully separated fatty and connective tissue to expose the trachea. We exposed the left common carotid artery (CCA) without damaging nerves and major veins bundled with the CCA, followed by exposing the top "Y" branching at the external cerebral artery (ECA) and the internal cerebral artery (ICA). A silicone-coated 4-0 nylon monofilament (Guangzhou Jialing Biotechnology Co., Ltd., Guangzhou, China) was inserted into the CCA and gently advanced through the ICA until its tip occluded the origin of the middle cerebral artery (MCA). Cerebral blood flow was monitored using PeriFlux System 5000 during MCA occlusion, and regional cerebral blood flow dropped to 20 % of baseline was recorded. For regaining reperfusion, the monofilament suture was carefully withdrawn after 60 min MCAO. Animals with complications such as excessive bleeding, filament displacement into the pterygopalatine artery, died during or less than 24 h after surgery, were excluded from the experiment.

2.8. Evaluation of the blood-brain barrier Integrity The integrity Blood-Brain Barrier (BBB) was assessed by Evans Blue (EB) stain and fluorescent imaging detection using Caliper IVIS Lumina K Series III system [44]. Briefly, the mice were injected with 0.6 % EB solution in 0.9 % normal saline (30 mL/kg) via tail vein under anesthesia at the beginning of reperfusion. After 24 h reperfusion, the animals were intracardially perfusion under deep anesthesia with 0.9 % normal saline to wash intravascular EB dye away from the circulation. The brains were excised and placed in the matrix for 2 mm-thick coronal slices as well as photographed to record the EB stained cerebral sections. Then, Living Image® Software (version 4.3.1) was used to quantify the concentrations of EB dye leakage according to the value of total radiant efficiency ([p/s]/[μW/cm2]). 2.9. Hematoxylin & Eosin staining

2.5. Neurological deficit score

In order to observe the histopathological alterations in brain tissues of different groups, Hematoxylin & Eosin (H&E) analysis was conducted. After 24 h reperfusion, the mice were euthanized following anesthetization. Brains were quickly removed and cut into coronal slices and fixed in 4 % paraformaldehyde for 24 h, and then dehydrated and embedded in paraffin blocks to be dissected into 5 μm coronal sections. Subsequently, sections were deparaffinized, hydrated with graded levels of alcohol and stained with H&E using the automatic dyeing machine (Gemini, Thermo Ltd., United States). Then, the

Examination of neurological deficits of the experimental animals was performed after 24 h reperfusion. The neurological deficits were blindly scored on a five-point scale described previously by Longa et al. [42] with a minor modification: score 0, no neurological deficit (showing normal and spontaneous movements); score 1, mild focal neurological deficit (failure to fully extend contralateral forelimb); score 2, moderate focal neurological deficit (repetitive circling to the 3

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Table 1 Primer sequences for qPCR. Primer name

Forward primer (5′–3′)

Reverse primer (5′–3′)

qMouse qMouse qMouse qMouse qMouse qMouse qMouse qMouse

ATGAACGCTACACACTGCATC GCAACTGTTCCTGAACTCAACT TAGTCCTTCCTACCCCAATTTCC CCCTCACACTCAGATCATCTTCT AGGCTTCTGGGCCTTATGTG CTCCCGTGGCTTCTAGTGC ATGGCATGGCTTACACCACC TGGTGAAGCAGGCATCTGAG

CCATCCTTTTGCCAGTTCCTC ATCTTTTGGGGTCCGTCAACT TTGGTCCTTAGCCACTCCTTC GCTACGACGTGGGCTACAG TGCTTCTCTCGCCAGGAATAC GCCTTAGTTTGGACAGGATCTG GAGGCCAATTTTGTCTCCACA TGCTGTTGAAGTCGCAGGAG

Ifng Il1b Il6 Tnf Rela Tgfb1 Tlr4 Gapdh

automatic sealing machine (ClearVue, Thermo, Ltd., United States) was used to mount the sections. H&E stained images were photographed using an inverted microscope with 100X and 200X magnification (CKX41; Olympus, Japan).

2.11. RT-qPCR verification Brain samples from different groups were collected from the ischemic hemispheres of mice after 24 h reperfusion, snap-frozen in liquid nitrogen and stored at −80 °C until use. The expressions of predicted following core genes: IFNG (IFNG-γ), IL1B (IL-1β), IL6 (IL-6), TNF (TNF-α), RELA (NF-κB p65), TGFB1 (TGF-β1), and TLR4 (TLR-4) were validated by reverse transcription-quantitative real-time polymerase chain reaction (RT-qPCR). Total RNA samples from the brain tissues were extracted using TRIzol® reagent (Invitrogen, Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer’s protocol. Subsequently, RNA samples were reverse transcribed into complementary DNA (cDNA) using Transcriptor First Strand cDNA Synthesis Kit (Roche, Mannheim, Germany) to acquire cDNA. The Bestar®Sybr Green qPCR Master Mix (DBI®Bioscience, Ludwigshafen, Germany) was used in RT-qPCR to quantify the level of expression of the target genes with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as an endogenous reference gene. The gene-specific primers were synthesized by Sangon Biotech (Shanghai, China) and primer sequences were described in the following Table 1. The amplification and analysis were performed using LightCycler®480 Software Version1.5.0.39 (Roche, Mannheim, Germany) for 45 cycles. The relative mRNA expression levels of predicted key genes were calculated using2−ΔΔCTmethod, following normalization to the housekeeping gene GAPDH.

2.10. Network pharmacology analyses 2.10.1. Database construction The main source of diseases related targets for CI/RI was collected from IPA (http://www.ingenuity.com) database [45,46]. Additional databases, such as OMIM [47], NCBI (gene) [48], TCMSP [49], CAD Gene [50], GeneCards [51], MalaCards [52] and PubMed, were manually searched to replenish the omissive diseases-related targets of IPA. Information on QSYQ (Huangqi, Danshen, Sanqi and Jiangxiang) ingredients were obtained from several online TCM databases, including TCMSP [53], TCMID [54], TCM-ID [55], TCM Database@Taiwan [56], TCMGeneDIT [57] and BATMAN-TCM [58]. PubChem [59] was also used to retrieve and verify ingredients of TCM. Literature mining based on PubMed was indispensable [60–62]. Since TCM ingredients-gene targets relationship were limited in IPA database, additional NCBI (gene) [48], TCMSP [49], TCMID [50], PubChem [59] and PubMed were also mined to supplement QSYQ ingredients corresponding targets. Details including web links to the databases were shown in Supplemental Table S2. 2.10.2. Network establishment and analysis Three datasets, ① Cerebral I/R injury-related targets, ② Huangqi, Danshen, Sanqi and Jiangxiang ingredients, and ③ QSYQ major ingredients and their corresponding targets, were constructed and imported into IPA system for data visualization and detailed analysis. Specifically, ① CI/RI relevant targets as well as the relationship between QSYQ (Huangqi, Danshen, Sanqi and Jiangxiang) components and the diseases related targets were discovered via employing “Build-Path Explorer” module,②“Build-Grow-Diseases& Functions” module was applied to uncover diseases and functions that corresponding targets of QSYQ major ingredients involved in, ③the canonical pathways, which corresponding targets of QSYQ major ingredients referred to, were generated by using “Overlay-Canonical Pathway” module. Ultimately, an integrated compound-target-disease & function-pathway network was established. In order to further acquire the correlation degree between corresponding targets of QSYQ major ingredients and diseases, functions as well as pathways, “Core analysis” module was performed to analyze the above network. Apart from acquisition of top diseases, top functions and top pathways, “Core Analysis-Upstream Analysis” module was employed to obtain top upstream regulators so as to illuminate the causal inference of upstream biological causes and probable downstream effects on cellular and organismal biology [63]. “BuildConnection” module was used to show the interactive relationship of targets-targets and gain the core targets among QSYQ major ingredients corresponding targets based on the significance of targets in interaction network. “Path designer” module was used to polish the network images and graphs. The algorithm of the network analysis was based on Fisher’s exact test with the enrichment score of P-values.

2.12. Statistical analysis Data from different experiments were presented as mean ± SD as indicated. Statistical analysis was carried out using Student’s two-tailed t-test for comparison between two groups and using one-way analysis of variance (ANOVA) followed by Dunnett’s t-test for comparisons between multiple groups. For the neurological deficit score, data were presented as median and nonparametric statistical analysis was used using Mann-Whitney test for comparison between two groups. p < 0.05 was defined as statistically significant and p < 0.01 was regarded to be highly significant. All data graphs were generated by GraphPad Prism 7 software (GraphPad Software, Inc., La Jolla, CA, United States). 3. Results A combined pharmacology network analysis and experimental verification approach were established in this study, to further reveal the action mechanisms of QSYQ in CI/RI, which involves five steps in a workflow shown in Fig. 1: (1) efficacy of a prescription (QSYQ) in preventing CI/RI was determined with several typical indicators; (2) ingredients of the prescription along with their corresponding diseases associated targets were identified by mining various databases; (3) the relationship between compound-target-pathway-disease & function were constructed by interaction networks and top diseases & functions and top canonical pathways were uncovered by network analysis; (4) key targets were determined via integrating upstream analysis, inflammatory response and neuroinflammation signaling pathway; (5) 4

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 1. Workflow for cerebral protective effect of QSYQ in a CI/RI model.

5

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 2. QSYQ pretreatment ameliorated neurological deficit and brain infarction. (A) Effects of QSYQ on the neurological deficit score. QSYQ/edaravone pretreatment obviously improved the neurological deficit score (n＝12). (B) Representative images of TTC-stained brain sections of each group. The infarct areas were colored white, whereas normal areas presented a red color. (C) Quantitative analysis of cerebral infarction volumes of each group (n＝6). QSYQ/edaravone pretreatment significantly reduced the infarct volumes. Neurological deficit score data are reported as the median and TTC staining data are reported as the means ± SD. #p < 0.05, ##p < 0.01, and ###p < 0.001 versus the sham group; *p < 0.05, **p < 0.01, and ***p < 0.001 versus the model group. I/R, ischemia/reperfusion; edaravone; QSYQ-L, 5.69 mg/mL (low dose); QSYQ-M, 11.38 mg/mL (middle dose); QSYQ-H, 22.75 mg/mL (high dose).

the possible mechanism of the prescription in cerebral protection was validated, and the accuracy of network analysis was guaranteed by experimental verification.

also significantly reduced the infarction volumes of MCAO mice [(F (4, 25) = 21.606, p = 0.000)]. 3.3. QSYQ pretreatment attenuated cerebral edema

3.1. QSYQ pretreatment ameliorated neurological deficits

Next, we performed a micro-CT scan to acquire the whole-cerebral slide images and estimated the severity of cerebral edema in different treatment groups using Analyze 12.0 image analysis software. The results of the representative coronal image in each group were shown in Fig. 3A. When the offset distance of midline in the cerebral coronal image was compared, the most obvious midline offset, caused by cerebral edema, was observed in the model group. On the other hand, similar to that of positive drug edaravone, pretreatment with different doses of QSYQ also reduced midline offset. 3D reconstruction of wholecerebral slide images (flow diagram shown in Fig. 3B) and quantitation of the bilateral cerebral hemisphere volume were realized. The value of left cerebral hemisphere volume and statistical data were shown in Fig. 3C, meanwhile, the value of the ratio of the bilateral cerebral hemisphere volume and statistical data were shown in Fig. 3D. Compared with the sham group, left cerebral hemisphere volume and the ratio of the bilateral cerebral hemisphere volume increased because of suffering from left MCAO surgical operation (F = 2.353,t=-3.275, p = 0.011;F = 14.776, t=-10.472, p = 0.000).However, pretreatment with edaravone obviously decreased left cerebral hemisphere volume [F (4,20) = 2.504, p = 0.027 vs. model] and the ratio of the bilateral cerebral hemisphere volume [F(4,20) = 21.530, p = 0.000 vs. model].Surprisingly, pretreatment with QSYQ dose-dependently reduced the ratio of the bilateral cerebral hemisphere volume [from 1.365 ± 0.06507 to 1.266 ± 0.06052, F(4,20) = 21.530, p = 0.012 vs. model; 1.247 ± 0.02491, F(4,20) = 21.530, p = 0.003 vs. model; and 1.171 ± 0.04540, F(4,20) = 21.530, p = 0.000 vs. model]. Besides, pretreatment with QSYQ could also reduce the value of left cerebral hemisphere volume of MCAO mice to some extent, but did not reach significance [F(4,20) = 2.504, p > 0.05 vs. model].In conclusion, the left cerebral hemisphere volume as well as

In order to assess the protective effect of QSYQ on neurological deficits, we made the transient MCAO (1 h occlusion followed by 24 h reperfusion) and a pretreatment regimen. The criterion of Longa’s score was adopted to estimate neurological deficits. As shown in Fig. 2A, mice subjected to sham operation moved normally without neurological deficit symptoms. However, cerebral ischemia strongly altered neural behavior of mice in the model group as indicated by the highest neurological deficit score among six groups. Compared with the model group, pretreatment with a positive-control drug edaravone significantly decreased neurological deficit score with better neural behavior (p = 0.005). Meanwhile, mice pretreated with QSYQ at a middle dose (11.38 mg/mL) and high dose (22.75 mg/mL) also significantly reduced neurological deficit score compared with the MCAO mice in model group (p = 0.046 and p = 0.021). 3.2. QSYQ pretreatment reduced cerebral infarction volume In addition to the neurological deficit score, the cerebral protective effects of QSYQ pretreatment were assessed with regards to infarction volume in MCAO mice by using TTC staining. The representative TTCstained cerebral slices in each group were shown in Fig. 2B, and the corresponding infarction volumes and statistical data were shown in Fig. 2C. As expected, mice in the sham group exhibited no infarction. Conversely, the transient MCAO model group showed obviously increased infarct volume (37.66 ± 3.10 % of the cerebral volume). Compared with the model group, the positive control group (edaravone at a dose of 9 mL/kg) significantly inhibited I/R-induced infarction [F (4, 25) = 21.606, p = 0.000)]. At the same time, pretreatment with QSYQ at a middle dose (11.38 mg/mL) and high dose (22.75 mg/mL) 6

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 3. QSYQ pretreatment reduced the cerebral edema caused by CI/RI. (A) Representative pictures of cerebral coronal images from each group mice. yellow line used to ensure the cerebral midline and red arrow used to indicate offset distance of midline. (B) The flow diagram to display the brief process about how to realize 3D reestablishment and quantization of brain hemisphere. (C)Quantitative analysis of left cerebral hemisphere volume from each group mice. Edaravone pretreatment significantly stopped higher values of left cerebral hemisphere volume caused by cerebral edema compared with the model mice. (D)Quantitative analysis of the ratio of bilateral brain hemisphere volume from each group mice. Pretreatment with QSYQ dose-dependently decreased the raised values of the ratio of bilateral brain hemisphere volume because of left MCAO surgical operation compared with the model mice. Data are expressed as mean ± SD. #p < 0.05, ##p < 0.01, and ###p < 0.001 versus the sham group; *p < 0.05, **p < 0.01, and ***p < 0.001 versus the model group, n＝5. I/R, ischemia/reperfusion; Eda, edaravone; QSYQ-L, 5.69 mg/mL (low dose); QSYQ-M, 11.38 mg/mL (middle dose); QSYQ-H, 22.75 mg/mL (high dose).

the ratio of the bilateral cerebral hemisphere volume could indirectly reveal the severity of cerebral edema. QSYQ pretreatment attenuated cerebral edema in MCAO mice.

and uniformly arranged, exhibiting a normal structural form: the cell outline was clear, the normal morphology was intact, and the nucleolus was clearly visible. However, ischemia /reperfusion caused severe damage to the cerebral tissues/cells. Numerous vacuolated spaces were seen in brain tissues. The cells became sparse and disorderly-arranged, the neurons were swollen or shrunken, and the nucleolus disappeared. Compared with the model group, pretreatment with different dosage of QSYQ as well as edaravone could relieve the pathological abnormalities caused by CI/RI at a certain degree.

3.4. QSYQ pretreatment reduced BBB disruption in MCAO mice It is known to us that BBB breakdown is a common pathophysiological feature in CI/RI. Therefore, the protective effect on the integrity of BBB is a crucial aspect of cerebral protection. As a marker of albumin effluxion, EB dye is used to estimate BBB permeability. During BBB disruption after I/R, EB can easily permeate BBB and stain brain tissue in blue color. Hence, we used EB to investigate whether QSYQ pretreatment could inhibit BBB disruption. Fig. 4A shows the result of EBstained slices. We compared EB leakage in the ipsilateral hemisphere among all groups. The sham group had intact BBB contrarily to EB leakage in the MCAO (model) group. In contrast, pretreatment with QSYQ markedly diminishes the EB extravasation as same as edaravone. In order to quantify the EB leakage, a rapid and convenient method of fluorescent imaging mode detection of Caliper IVIS Lumina K Series III was employed. The representative fluorescent images in each group are shown in Fig. 4B. The corresponding value of total radiant efficiency and statistical data are shown in Fig. 4C. Compared with the sham group, the MCAO caused serious disruption of BBB in the model (F = 10.353, t=-8.737, p = 0.000).As expected, pretreatment with edaravone significantly decreased EB extravasation [F(4,20) = 13.996, p = 0.000 vs. model].Interestingly, pretreatment with QSYQ dose-dependently reduced EB leakage [from 2.10E9 ± 4.53E8 to 1.48E9 ± 3.31E8 [p/s]/ [μW/cm2], [F(4,20) = 13.996, p = 0.007 vs. model; 1.29E9 ± 1.80E8 [p/s]/ [μW/cm2], [F(4,20) = 13.996, p = 0.001 vs. model; and 1.05E9 ± 9.28E7 [p/ s]/ [μW/cm2], [F(4,20) = 13.996, p = 0.000 vs. model].

3.6. Identification of target-disease & function-pathway of QSYQ components in prevention and treatment for CI/RI Among the total of 452 CI/RI-related targets recognized in Ingenuity Pathway Analysis (IPA, Supplemental Table S3), 220 of them (48.7 %) were found to be associated with QSYQ ingredients (Huangqi, Danshen, Sanqi, Jiangxiang, and QSYQ). To decipher the multi-target pharmacological mechanisms of QSYQ in prevention and treatment for CI/RI, compound-target combined with target-disease & functionpathway networks were established. Firstly, a fifty-five QSYQ ingredient network was established, which comprised fourteen Huangqi compounds, fifteen Sanqi compounds, twenty Danshen compounds, and six Jiangxiang compounds. modulated the two hundred and twenty CI/ RI related targets. In order to display the possible cellular locations and multi-target biological processes of QSYQ ingredients, we integrated the compound-target network according to extracellular space, plasma membrane, cytoplasm, and nucleus. Meanwhile, the characteristic of traditional Chinese medicine in prevention and treatment diseases through the multidimensional interactions between multiple ingredients and multiple targets was visually presented in Fig. 6A. Secondly, with the intention of clarifying the main biological processes and molecular mechanisms of QSYQ acting on the 220 CI/RI related targets, an integrated target-disease & function-pathway network containing the most associative diseases, most correlative functions, and the most interrelated pathways were built via taking reasonable advantage of various distinctive modules of IPA, such as “Core Analysis”, “Canonical Pathway”, “Diseases & Functions”, and so on. Based on the Fisher’s

3.5. Histopathological assessment of QSYQ pretreatment on mice brain As shown in Fig. 5, the cerebral tissues/cells in the sham operation group were normal. The brain tissues remain undamaged, and no obvious vacuolated spaces were observed. The nerve cells were densely 7

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 4. QSYQ pretreatment protected the BBB integrity against stroke-related damage. (A) Representative pictures of coronal slices from each group of the mice brain stained with EB dye after ischemic-reperfusion. (B) The corresponding fluorescence images of EB extravasation of each group of the mice cerebral coronal sections. (C) Quantitative analysis of EB leakage of each group of the mice cerebral coronal slices. QSYQ/edaravone pretreatment markedly decreased EB leakage caused by BBB breakdown compared with the MCAO. Data are expressed as mean ± SD. #p < 0.05, ##p < 0.01, and ###p < 0.001 versus the sham group; *p < 0.05, **p < 0.01, and ***p < 0.001 versus the model group, n＝5. I/R, ischemia/reperfusion; Eda, edaravone; QSYQ-L, 5.69 mg/mL (low dose); QSYQ-M, 11.38 mg/mL (middle dose); QSYQ-H, 22.75 mg/mL (high dose).

exact test algorithm of IPA, the diseases, functions, and pathways were ranked respectively according to corresponding p-value scores so as to distinguish the relatedness or importance among the 220 targets and diverse diseases, functions as well as pathways. The most obvious diseases that are affected by QSYQ in the order of descending –log (pvalue) score were, rheumatism, Alzheimer’s, dementia, diabetes mellitus, atherosclerosis, lymphocytic cancer, gastroenteritis, colitis, obesity, and myocardial infarction. Among them, rheumatic ranked the highest with a –log (p-value) score of 85 (Fig. 6B). At the same time, QSYQ was predicted to influence multiple functions including, in the

order of descending –log (p-value) score: angiogenesis, leukocyte migration, necrosis, apoptosis, vasculogenesis, inflammatory response, fibrosis, synthesis of reactive oxygen species, autophagy, and synthesis of nitric oxide. Among them, the angiogenesis ranked the highest with a –log (p-value) score of 125 and the inflammatory response ranked the sixth with a –log (p-value) score of 108 (Fig. 6C). On the basis of the result of core analysis and canonical pathways analysis aiming at 220 targets, targets-related major pathways were acquired. Neuroinflammation signaling-pathway was considered the most-essential pathway, with the highest –log (p-value) score of 72 (Fig. 6D).

Fig. 5. Effect of QSYQ pretreatment on pathological changes of the brain cortex after CI/RI (H&E Staining, ×100; scale bar: 100 μm and H&E Staining, ×200; scale bar: 50 μm) (n = 3). Pretreatment with different dosage of QSYQ as well as edaravone could reduce the histopathological alteration after CI/RI at a certain degree. I/ R, ischemia/reperfusion; Eda, edaravone; QSYQ-L, 5.69 mg/mL (low dose); QSYQ-M, 11.38 mg/mL (middle dose); QSYQ-H, 22.75 mg/mL (high dose). 8

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 6. Analysis of QSYQ’s compound-target-disease & function-pathway. (A) QSYQ ingredient-disease related target network. 14 Huangqi compounds (yellow), 15 Sanqi compounds (leaf-green), 20 Danshen compounds (light-blue), and 6 Jiangxiang compounds (pink) were presented, which cooperatively modulated a total of 220 CI/RI related targets. The bottom right corner diagram displayed the unique numbers of ingredients from QSYQ. (B, C, and D) Diseases, functions, and pathways affected by QSYQ were ranked according to the Fisher’s exact test algorithm. Top 10 diseases, top 10 functions, and top 20 pathways were displayed in descending order based on –log (p-value) score. 9

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 7. Determination of key targets modulated by QSYQ. (A) Upstream network analysis. Top 20 upstream regulators (red) and their corresponding downstream targets were shown. (B)135 targets correlated with inflammatory response picked out from 220 targets regulated by QSYQ with the help of IPA software. (C)The relationship between inflammatory response-related targets (green) and neuroinflammation signaling pathway-related targets (blue). The common targets (yellow) shared by the inflammatory response and neuroinflammation signaling pathway were displayed. The more important was that 7 core upstream regulators (pink) in line with the common targets were found. (D) The PPI network between the seven key upstream regulators. The result of cross-talk proteinprotein interaction (PPI) between the seven upstream regulators was displayed. (E) The cooperative actions of 42 QSYQ ingredients on the seven key upstream regulators. This sub-network was separated from QSYQ ingredient-target network (Fig. 6B), including 11 Huangqi (yellow), 15 Danshen (blue),13 Sanqi (green), and 3 Jiangxiang (pink) components.

10

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

3.7. Determination of QSYQ-modulated target networks in neuroinflammation

infarct size, BBB breakdown and brain edema induced by experimental stroke [67,68]. Similar to edaravone, the present study demonstrated that pretreatment with middle and high dose QSYQ for 7 days significantly decreased CI/RI-induced disability, cerebral infarct volume and histological damage (Figs. 2 and 5). The BBB plays vital roles in maintaining homeostatic microenvironment of the central nervous system and effective protection of brain tissue from outside stimuli [69]. BBB dysfunction, characterized by structural breakdown of tight junction protein complexes leading to increased permeability, ion transporter dysfunction causing cerebral edema, and further inflammatory damage caused by infiltrating leukocytes, is a remarkable pathological characteristic of ischemic stroke [70]. It is worth noting that ischemia-induced BBB disruption in the early stages of stroke largely correlates with hemorrhagic transformation following rt-PA thrombolysis treatment [71]. In the present research, luminance imaging and micro-CT were respectively used to assess the BBB permeability and cerebral edema. The results clearly showed that pretreatment with QSYQ dose-dependently attenuated the BBB breakdown and the severity of brain edema caused by ischemia and reperfusion (Figs. 3 and 4). It is known that brain damage induced by ischemia and reperfusion is a complex pathophysiological process with various mechanisms [6]. In recent years, network pharmacology analyses, which illuminate the synergistic effects and the underlying pharmacological mechanisms of multicomponent and multitarget agents from TCM formulae, proved to be a powerful paradigm to provide a better understanding of the underlying intricate relationships between TCM herbal ingredients and diseases through the bridge “targets” [72–74]. We have previously used this approach to reveal the endothelial inflammation as a common mechanism for stroke and coronary artery disease treatment by Danhong injection [45]. In the present study, similar analysis of a more complex Chinese medicine QSYQ that also contains Danshen (4 vs. 2 herbal components) revealed yet a more complex but still synergistic compound-target-disease & function-pathway network. We found that inflammatory response and neuroinflammation signaling pathway were closely correlative with 220 CI/RI related targets which could be regulated by the identified ingredients of QSYQ (Fig. 6). This result suggested that neuroinflammation could account at least in part for cerebral protection of QSYQ. According to the integration of inflammatory response, neuroinflammation signaling pathway and upstream regulators, we further acquired the core targets associated with neuroinflammatory response, including IFNG (IFNG-γ), IL1B (IL-1β), IL6 (IL-6), TNF (TNF-α), RELA (NF-κB p65), TGFB1 (TGF-β1), and TLR4 (TLR-4) (Fig. 7). Interferon-γ (IFNG-γ), an inflammatory mediator released from CD4+ and CD8+ T-lymphocytes as well as natural killer (NK) cells, is considered a crucial regulator of immune and inflammatory responses after CI/RI [75]. IFNG-γ mRNA expression was elevated in rat brain after permanent middle cerebral artery occlusion (pMCAO) [76]. A group of studies revealed the multipotent inflammatory properties of IFNG-γ that initiated several pathways participated in CI/RI such as induction of adhesion molecule expression, stimulation of NAD(P)H oxidase, and activation of microglia-macrophages [77–79]. Interleukin1 beta (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α) are three major pro-inflammatory cytokines, and are generally considered to be main contributors to acute inflammation in ischemic stroke [32]. In experimental ischemic stroke, elevated IL-1β level led to more serious brain edema and brain infarction [80]. Similarly, knockdown of IL-1β could improve pathological changes induced by hypoxia/ ischemia in cell and animal model [81]. IL-6 expression was enhanced continuously up to 24 h reperfusion after ischemia onset [82]. In acute ischemic stroke patients, increased IL-6 levels of blood samples and larger infarct volume as well as less favorable prognosis were reported [83]. However, ambivalent data on IL-6 also exists. Clark et al. did not observe smaller lesion size and better neurological function in IL-6 deficient mice after brain ischemia surgery [84]. In ischemic brain

In order to confirm the core targets affected by QSYQ, upstream genes of the core targets were first determined and the top 20 upstream regulators and their corresponding targets are shown (Fig. 7A, Supplemental Table S4). Disease & function analysis revealed a close correlation between QSYQ targets and inflammatory response, consistent with the notion that inflammation as one of the most significant mechanisms involved in the pathogenesis of acute ischemic stroke and CI/ RI. The vital role of inflammatory response in QSYQ protective action on CI/RI was further demonstrated by separating the 135 inflammation-related targets from the total 220 QSYQ ingredient-interacting genes (Fig. 7B). The close relationship involved in multiple shared and unique targets between inflammatory response and neuroinflammation signaling pathway was also verified in the network (Fig. 7C). Interestingly, 7 targets belong to the top 20 upstream regulators. Among them, four cytokines, IFNG(IFNG-γ), IL1B (IL-1β), IL6 (IL-6), TNF (TNF-α), one transcription regulator, RELA (NF-κB p65), one growth factor, TGFB1 (TGF-β1), and one transmembrane receptor, TLR4 (TLR-4) were shown to have a cross-talking protein-protein interaction (PPI) (Fig. 7D). Finally, the 42 QSYQ ingredients synergistically regulating these seven inflammatory response-related molecules are illustrated (Fig. 7E). Based on above analysis, these 7 targets were considered as the most crucial targets in integrated inflammatory response, neuroinflammatory signaling pathway and their upstream regulators and were chosen for further experimental validation next. 3.8. RT-qPCR validation of QSYQ-modulated targets in neuroinflammation effects on regulating core targets by QSYQ To validate the regulatory role of QSYQ on the core inflammatory targets forecasted above, we compared the expression of the following genes, IFNG-γ, IL-1β, IL-6, TNF-α, NF-κB p65, TGF-β1, and TLR-4, in brain samples from cerebral I/R injury model, and model treated with QSYQ (22.75 mg/mL) by RT-qPCR (Fig. 8). Compared with the I/R group, the IFNG-γ, IL-6, TNF-α, NF-κB p65, and TLR-4 mRNA levels were significantly decreased in the QSYQ-treated group (for IFNG-γ, F = 0.149, p = 0.000; IL-6, F = 1.469, p = 0.000; TNF-α, F = 3.986, p = 0.000; NF-κB p65, F = 2.203, p = 0.000; TLR-4, F = 0.829, p = 0.018; Fig. 8A-8E). Meanwhile, an increase in expression of TGF-β1 mRNA was significant in QSYQ-treated group when compared to the model (F = 0.034, p = 0.002; Fig. 8F). Besides, pretreatment of QSYQ down-regulated IL-1β mRNA levels to some extent, but did not reach significance (F = 0.752, p = 0.216; Fig. 8G). 4. Discussion Ischemic stroke is one of the leading causes of morbidity and mortality in humans.CI/RI is a typical feature of ischemic stroke, which occurs during the restoration of blood supply after a period of cerebral vessel occlusion [6]. Nowadays, because of the limitations of rt-PA in clinical use, more and more compound TCM preparations applied to treat ischemic stroke with advantage of less side effects and multifunctional pathways [64,65]. QSYQ is one of the representative component-based Chinese medicine to treat CVDs. Since a large clinical trial approved that QSYQ is effective to reduce thrombotic vascular disease risk, it has been widely used for the secondary prevention of myocardial infarction in China [16]. We have previously reported the cardioprotection against ischemia/reperfusion injury by QSYQ and its bioactive ingredients [19,20]. In this study, the cerebral protective effect of QSYQ was observed for the first time in experimental mouse model of CI/RI. Edaravone was chosen as positive drug in this study because of its role as an antioxidant and hydroxyl radical scavenger to significantly improve the degree of neurological impairment of patients during acute cerebral infarction [66]. Besides, it has also been shown to reduce 11

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

Fig. 8. Regulation of inflammatory genes by pretreatment with QSYQ in I/R mouse brain. Relative expression levels of cerebral IFNG-γ (A), IL-6 (B), TNF-α (C), NFκB p65 (D), TGF-β1 (E), TLR-4 (F), and IL-1β (G) mRNA were evaluated by RT-qPCR. Results were normalized to GAPDH. Data are expressed as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001 versus I/R group, n＝3. QSYQ-H, 22.75 mg/mL (high dose).

tissues of rats, TNF-α mRNA was upregulated from 3 h to 24 h after stroke [85]. Brain ischemia also increased TNF-α protein in cerebral tissue at 6 and 24 h after MCAO [86]. Transforming growth factor-beta (TGF-β) is thought to play both neuroprotective and anti-inflammatory roles in stroke [87]. In the acute phase of cerebral ischemia, upregulation of TGF-β can promote regulatory T-cells (Treg cells) development to produce the anti-inflammatory cytokine IL-10 as well as reduce Th1 and Th2 responses [88]. Ronaldson et al. proved that administration of exogenous TGF-β1 decreased BBB permeability during inflammatory pain [89]. Furthermore, overexpression of TGF-β1 could inhibit ischemic brain damage via downregulating the expression of some inflammatory proteins [90]. As a typical transcription factor, nuclear factor κB (NF-κB) is well known to be activated in CI/RI. Brain samples from postmortem ischemic stroke patients further illustrated higher levels of NF-κB in the ischemic infarct area and necrotic infarct core [91]. Xiang et al. reported that down-regulation of NF-κB p65 activation involved in selectively modulating microglia M1/M2 polarization so as to play the anti-inflammatory effect [92]. In addition, prevention of cerebral NF-κB p65 activation and reduction of inflammation response could contribute to protecting the blood brain-barrier (BBB) as well as to attenuate brain damage induced by ischemia-reperfusion [93]. Among Toll-like receptors (TLRs) family, accumulating evidence showed the importance of TLR4 in cerebral ischemic injury. Increased expression of brain TLR4 was found in both cell and animal models of stroke. Hyakkoku et al. reported that TLR4 knockout mice had significantly smaller infarct volume and better neurological function compared to wild-type mice after undergoing brain injury induced by ischemia and reperfusion

[94]. Meanwhile, Caso et al. also observed that TLR4 deficient mice had minor infractions and less inflammatory response after experimental stroke [95]. In order to confirm the reliability of network analysis, RT-qPCR was performed to test the mRNA expressions of neuroinflammation related targets. Results of the present study showed that the mice pretreated with QSYQ could significantly decrease IFNG-γ, IL-6, TNF-α, NF-κB p65, and TLR-4 mRNA levels and increase TGF-β1 mRNA level in brain tissue compared to the untreated mice after 1 h cerebral ischemia and 24 h reperfusion. Besides, pretreatment of stroke subject with QSYQ down-regulated IL-1β mRNA levels to some extent, but showed no significance as compared to the model group. The data suggest that the beneficial effects of QSYQ pretreatment on brain damage caused by experimental stroke is partly through reduction of neuroinflammatory response. 5. Limitation and future directions Despite of interesting new findings in this study, some limitations exist. The following directions and prospects should be considered in future investigation: firstly, our current study mainly used the method of TCM-based pharmacology network to explore the possible mechanism involved in the neuroprotection of QSYQ in mice model of CI/ RI based on the neuroinflammatory pathway. Furthermore, we did an extensive gene profiling of the major pharmacological targets which confirmed the positive therapeutic application of QSYQ in CI/RI. All these is to some extent, adequate for the present study. Of course, multiple omics approaches, especially proteomics or transcriptome 12

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

References

analysis, should be exploited to acquire the overlapping core targets so as to further complement and confirm the results. Secondly, the sensitive and accurate techniques of ingredient identification such as UHPLC–MS/MS and UHPLC-QTOF-MS/MS can be used to verify the neuroinflammatory related bioactive components in QSYQ [96,97]. Besides, the establishment of BBB organoids vitro model offer us a highthroughput and cost-effective approach to accelerate therapeutic discovery [98]. Maybe we can also employ BBB organoids vitro model to filtrate the bioactive components in QSYQ. Thirdly, although we focused on neuroinflammatory response as one of the important underlying mechanism of QSYQ against ischemia/reperfusion injury, on the basis of the properties of multi-components, multi-targets, multiple therapeutic pathways, the other potential mechanisms, including oxidative stress, production of ROS, synthesis of nitric oxide (NO), mitochondrial function, apoptosis and autophagy, still need more studies so as to further account for cerebral protection action of QSYQ in acute ischemic stroke. Meanwhile, according to the present results, it is significant to continue the next research on pharmacodynamic effects of QSYQ in the subacute and long-term phases of focal CI/RI as well as their corresponding neuroinflammatory and other underlying mechanisms.

[1] E.J. Benjamin, P. Muntner, A. Alonso, et al., Heart disease and stroke Statistics-2019 update: a report from the american heart association, Circulation 139 (10) (2019) e56–e528, https://doi.org/10.1161/CIR.0000000000000659. [2] Y. Fan, X. Liao, Y. Pan, et al., Intravenous Thrombolysis Is Safe and Effective for the Cryptogenic Stroke in China: Data From the Thrombolysis Implementation and Monitor of Acute Ischemic Stroke in China (TIMS-China), J. Stroke Cerebrovasc. Dis. 28 (1) (2019) 220–226, https://doi.org/10.1016/j. [3] J.M. Wardlaw, V. Murray, E. Berge, et al., Recombinant tissue plasminogen activator for acute ischaemic stroke: an updated systematic review and meta-analysis, Lancet 379 (9834) (2012) 2364–2372, https://doi.org/10.1016/S0140-6736(12) 60738-7. [4] P.M. Chen, B. Lehmann, B.C. Meyer, et al., Timing of symptomatic intracerebral hemorrhage after rt-PA treatment in ischemic stroke, Neurol. Clin. Pract. 9 (4) (2019) 304–308, https://doi.org/10.1212/CPJ.0000000000000632. [5] W.T. Yang, Y. Wang, Y.H. Shi, et al., Herbal compatibility of ginseng and rhubarb exerts synergistic neuroprotection in cerebral Ischemia/Reperfusion injury of rats, Front. Physiol. 10 (2019) 1174, https://doi.org/10.3389/fphys.2019.01174. [6] L. Lin, X. Wang, Z. Yu, Ischemia-reperfusion injury in the brain: mechanisms and potential therapeutic strategies, Biochem. Pharmacol. (Los Angel) 5 (4) (2016) 213, https://doi.org/10.4172/2167-0501.1000213. [7] M. Shvedova, Y. Anfinogenova, E.N. Atochina-Vasserman, et al., C-jun N-Terminal kinases (JNKs) in myocardial and cerebral Ischemia/Reperfusion injury, Front. Pharmacol. 9 (2018) 715, https://doi.org/10.3389/fphar.2018.00715. [8] T. Kalogeris, C.P. Baines, M. Krenz, et al., Ischemia/Reperfusion, Compr. Physiol. 7 (1) (2017) 113–170, https://doi.org/10.1002/cphy.c160006. [9] T. Kalogeris, C.P. Baines, M. Krenz, et al., Cell biology of Ischemia/Reperfusion injury, Int. Rev. Cell Mol. Biol. 298 (2012) 229–317, https://doi.org/10.1016/ B978-0-12-394309-5.00006-7. [10] J. Pan, A.A. Konstas, B. Bateman, et al., Reperfusion injury following cerebral ischemia: pathophysiology, MR imaging, and potential therapies, Neuroradiology 49 (2) (2007) 93–102, https://doi.org/10.1007/s00234-006-0183-z. [11] W.B. Gorsuch, E. Chrysanthou, W.J. Schwaeble, et al., The complement system in ischemia-reperfusion injuries, Immunobiology 217 (11) (2012) 1026–1033, https://doi.org/10.1016/j.imbio.2012.07.024. [12] A. Mizuma, M.A. Yenari, Anti-inflammatory targets for the treatment of reperfusion injury in stroke, Front. Neurol. 8 (2017) 467, https://doi.org/10.3389/fneur.2017. 00467. [13] K. Sun, J. Fan, J. Han, Ameliorating effects of traditional Chinese medicine preparation, Chinese materia medica and active compounds on ischemia/reperfusioninduced cerebral microcirculatory disturbances and neuron damage, Acta Pharm. Sin. B 5 (1) (2015) 8–24, https://doi.org/10.1016/j.apsb.2014.11.002. [14] J.Y. Han, Q. Li, Z.Z. Ma, et al., Effects and mechanisms of compound Chinese medicine and major ingredients on microcirculatory dysfunction and organ injury induced by ischemia/reperfusion, Pharmacol. Ther. 177 (2017) 146–173, https:// doi.org/10.1016/j.pharmthera. [15] H. Cao, P. Wang, N. Li, et al., Practice of comparative effectiveness research to identify treatment characteristics of similar chinese patent medicine for angina pectoris, Evid. Complement. Alternat. Med. 2017 (2017) 7062714, https://doi.org/ 10.1155/2017/7062714. [16] H. Shang, J. Zhang, C. Yao, et al., Qi-shen-yi-qi dripping pills for the secondary prevention of myocardial infarction: a randomised clinical trial, Evid. Complement. Alternat. Med. 2013 (2013) 738391, , https://doi.org/10.1155/2013/738391. [17] J.Y. Han, Q. Li, C.S. Pan, et al., Effects and mechanisms of QiShenYiQi pills and major ingredients on myocardial microcirculatory disturbance, cardiac injury and fibrosis induced by ischemia-reperfusion, Pharmacol. Res. (2019) 104386, , https:// doi.org/10.1016/j.phrs.2019.104386. [18] J.R. Chen, J. Wei, L.Y. Wang, et al., Cardioprotection against ischemia/reperfusion injury by QiShenYiQi Pill® via ameliorate of multiple mitochondrial dysfunctions, Drug Des. Devel. Ther. 9 (2015) 3051–3066, https://doi.org/10.2147/DDDT. S82146. [19] J. Yu, W. Zhang, Y. Zhang, et al., A critical courier role of volatile oils from Dalbergia odorifera for cardiac protection in vivo by QiShenYiQi, Sci. Rep. 7 (1) (2017) 7353, https://doi.org/10.1038/s41598-017-07659-x. [20] Y. Zhang, J. Yu, W. Zhang, et al., An integrated evidence-based targeting strategy for determining combinatorial bioactive ingredients of a compound herbal medicine Qishen Yiqi dripping pills, J. Ethnopharmacol. 219 (2018) 288–298, https:// doi.org/10.1016/j.jep.2018.02.041. [21] H.L. Wang, Q.H. Zhou, M.B. Xu, et al., Astragaloside IV for experimental focal cerebral ischemia: preclinical evidence and possible mechanisms, Oxid. Med. Cell. Longev. 2017 (2017) 8424326, , https://doi.org/10.1155/2017/8424326. [22] Y. Wang, X. Dong, Z. Li, et al., Downregulated RASD1 and upregulated miR-375 are involved in protective effects of calycosin on cerebral ischemia/reperfusion rats, J. Neurol. Sci. 339 (1-2) (2014) 144–148, https://doi.org/10.1016/j.jns.2014.02.002. [23] W. Xie, P. Zhou, Y. Sun, et al., Protective effects and target network analysis of ginsenoside Rg1 in cerebral ischemia and reperfusion injury: a comprehensive overview of experimental studies, Cells 7 (12) (2018) 270, https://doi.org/10. 3390/cells7120270. [24] W. Chen, Y. Guo, W. Yang, et al., Protective effect of ginsenoside Rb1 on integrity of blood-brain barrier following cerebral ischemia, Exp. Brain Res. 233 (10) (2015) 2823–2831, https://doi.org/10.1007/s00221-015-4352-3. [25] S. Zou, M. Zhang, L. Feng, et al., Protective effects of notoginsenoside R1 on cerebral ischemia-reperfusion injury in rats, Exp. Ther. Med. 14 (6) (2017) 6012–6016, https://doi.org/10.3892/etm.2017.5268.

6. Conclusion In summary, the present study provides the first evidence for cerebral protective effect of pretreatment with QSYQ against CI/RI, which may be related to its potential to inhibit neuroinflammatory response via down-regulation of IFNG-γ, IL-6, TNF-α, NF-κB p65, and TLR-4 mRNA levels and up-regulation of TGF-β1 mRNA level in ischemic brain tissues. The fact that QSYQ could prevent the brain from damage caused by ischemia and reperfusion requires further verification, the more definite functional mechanism, and pharmacodynamic material basis are needed to illuminate so as to provide more scientific and reliable experimental data support for QSYQ application in clinic. Author contributions Y.L.W and Y.Z. conceived and designed the study and wrote the manuscript. Y.Q.Z prepared the extracts of QSYQ. Y.L.W and G.X.X performed drug administration, MCAO surgery, and brain tissue TTC staining. Y.L.W performed micro-CT and IVIS imaging studies. X.Y.L performed H&E staining and pathological analysis. Y.L.W and S.H performed RT-qPCR experiment. L.Z and X.Y.Y performed and participated in neurological deficit score and database construction. Y.L.W performed the compound-target-disease and function-pathway network establishment and analysis. Y.Z., J.O, Y.X.F, B.L.Z and X.Y.W reviewed and edited the final manuscript. Conflict of Competing Interest The authors declare that there are no conflicts of interest. Acknowledgments This work was supported by grants from the National Science Foundation of China NSFC81873037, the Major National Science and Technology Project (2018YFC1704502), and the Major National Science and Technology Projects of China (2018ZX01031301). We would like to thank Drs. Rui Shao and Jian Yang for discussion and suggestions; the members of our laboratory, particularly Jie Ren, Zihao Chen, and Hongxia Du, for technical assistance in some experiments. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.biopha.2020.109945. 13

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al. [26] S. Hou, M.M. Zhao, P.P. Shen, et al., Neuroprotective effect of salvianolic acids against cerebral Ischemia/Reperfusion injury, Int. J. Mol. Sci. 17 (7) (2016) 1190, https://doi.org/10.3390/ijms17071190. [27] H. Lv, L. Wang, J. Shen, et al., Salvianolic acid B attenuates apoptosis and inflammation via SIRT1 activation in experimental stroke rats, Brain Res. Bull. 115 (2015) 30–36, https://doi.org/10.1016/j.brainresbull.2015.05.002. [28] C. Tang, H. Xue, C. Bai, et al., The effects of Tanshinone IIA on blood-brain barrier and brain edema after transient middle cerebral artery occlusion in rats, Phytomedicine. 17 (14) (2010) 1145–1149, https://doi.org/10.1016/j.phymed. 2010.03.017. [29] Q.F. Chen, Y.Y. Liu, C.S. Pan, et al., Angioedema and hemorrhage after 4.5-Hour tPA (TissueType plasminogen activator) thrombolysis ameliorated by T541 via restoring brain microvascular integrity, Stroke 49 (2018) 2211–2219, https://doi. org/10.1161/STROKEAHA.118.021754. [30] D.J. DiSabato, N. Quan, Godbout JP.Neuroinflammation: the devil is in the details, J. Neurochem. 139 (Suppl 2) (2016) 136–153, https://doi.org/10.1111/jnc.13607. [31] T. Peng, Y. Jiang, M. Farhan, et al., Anti-inflammatory effects of traditional chinese medicines on preclinical in vivo models of brain ischemia-reperfusion-Injury: prospects for neuroprotective drug discovery and therapy, Front. Pharmacol. 10 (2019) 204, https://doi.org/10.3389/fphar.2019.00204. [32] M. Kawabori, M.A. Yenari, Inflammatory responses in brain ischemia, Curr. Med. Chem. 22 (10) (2015) 1258–1277, https://doi.org/10.2174/ 0929867322666150209154036. [33] R.L. Jayaraj, S. Azimullah, R. Beiram, et al., Neuroinflammation: friend and foe for ischemic stroke, J. Neuroinflammation 16 (1) (2019) 142, https://doi.org/10. 1186/s12974-019-1516-2. [34] K.L. Lambertsen, K. Biber, B. Finsen, Inflammatory cytokines in experimental and human stroke, J. Cereb. Blood Flow Metab. 32 (9) (2012) 1677–1698, https://doi. org/10.1038/jcbfm.2012.88. [35] J.Y. Kim, M. Kawabori, M.A. Yenari, Innate inflammatory responses in stroke: mechanisms and potential therapeutic targets, Curr. Med. Chem. 21 (18) (2014) 2076–2097, https://doi.org/10.2174/0929867321666131228205146. [36] P.A. Baeuerle, T. Henkel, Function and activation of NF-kappa B in the immune system, Annu. Rev. Immunol. 12 (1994) 141–179, https://doi.org/10.1146/ annurev.iy.12.040194.001041. [37] Y. He, X. Ma, D. Li, et al., Thiamet G mediates neuroprotection in experimental stroke by modulating microglia/macrophage polarization and inhibiting NF-κB p65 signaling, J. Cereb. Blood Flow Metab. 37 (8) (2017) 2938–2951, https://doi.org/ 10.1177/0271678X16679671. [38] H. Zhao, Z. Chen, L.J. Xie, et al., Suppression of TLR4/NF-κB Signaling Pathway Improves Cerebral Ischemia-Reperfusion Injury in Rats, Mol. Neurobiol. 55 (5) (2018) 4311–4319, https://doi.org/10.1007/s12035-017-0552-0. [39] L. Bertrand, L. Dygert, M. Toborek, Induction of ischemic stroke and ischemia-reperfusion in mice using the middle artery occlusion technique and visualization of infarct area, J. Vis. Exp. (120) (2017) 54805, https://doi.org/10.3791/54805. [40] G.P. Morris, A.L. Wright, R.P. Tan, et al., A comparative study of variables influencing ischemic injury in the longa and koizumi methods of intraluminal filament middle cerebral artery occlusion in mice, PLoS One 11 (2) (2016) e0148503, , https://doi.org/10.1371/journal.pone.0148503. [41] J. Koizumi, Y. Yoshida, T. Nakazawa, et al., Experimental studies of ischemic brain edema 1.a new experimental model of cerebral embolism in rats in which recirculation can be introduced in the ischemic area, Jpn J Stroke 8 (1986) 1–8, https://doi.org/10.3995/jstroke.8.1. [42] E.Z. Longa, P.R. Weinstein, S. Carlson, et al., Reversible middle cerebral artery occlusion without craniectomy in rats, Stroke. 20 (1) (1989) 84–91, https://doi. org/10.1161/01.STR.20.1.84. [43] T.N. Lin, Y.Y. He, G. Wu, et al., Effect of brain edema on infarct volume in a focal cerebral ischemia model in rats, Stroke 24 (1) (1993) 117–121, https://doi.org/10. 1161/01.STR.24.1.117. [44] J.O. Orgah, J. Yu, T. Zhao, et al., Danhong injection reversed cardiac abnormality in brain-heart syndrome via local and remote β-Adrenergic receptor signaling, Front. Pharmacol. 9 (2018) 692, https://doi.org/10.3389/fphar.2018.00692. [45] M. Lyu, C.L. Yan, H.X. Liu, et al., Network pharmacology exploration reveals endothelial inflammation as a common mechanism for stroke and coronary artery disease treatment of Danhong injection, Sci. Rep. 7 (2017) 15427, https://doi.org/ 10.1038/s41598-017-14692-3. [46] S. Ghosh, C.A. Loffredo, P.S. Mitra, et al., PCB exposure and potential future cancer incidence in Slovak children: an assessment from molecular finger printing by Ingenuity Pathway Analysis (IPA®) derived from experimental and epidemiological investigations, Environ. Sci. Pollut. Res. Int. 25 (17) (2018) 16493–16507, https:// doi.org/10.1007/s11356-017-0149-1. [47] J.S. Amberger, A. Hamosh, Searching online mendelian inheritance in man (OMIM): a knowledgebase of human genes and genetic phenotypes, Curr. Protoc. Bioinformatics 58 (2017), https://doi.org/10.1002/cpbi.27 1.2.1-1.2.12. [48] G.R. Brown, V. Hem, K.S. Katz, et al., Gene: a gene-centered information resource at NCBI, Nucleic Acids Res. 43 (Database issue) (2015) D36–D42, https://doi.org/10. 1093/nar/gku1055. [49] G. Yu, W. Wang, X. Wang, et al., Network pharmacology-based strategy to investigate pharmacological mechanisms of Zuojinwan for treatment of gastritis, BMC Complement. Altern. Med. 18 (1) (2018) 292, https://doi.org/10.1186/s12906018-2356-9. [50] M. Ghatge, J. Nair, A. Sharma, et al., Integrative gene ontology and network analysis of coronary artery disease associated genes suggests potential role of ErbB pathway gene EGFR, Mol. Med. Rep. 17 (3) (2018) 4253–4264, https://doi.org/10. 3892/mmr.2018.8393. [51] S. Fishilevich, R. Nudel, N. Rappaport, et al., GeneHancer:genome-wide integration

[52] [53] [54] [55] [56]

[57] [58] [59] [60]

[61] [62]

[63] [64] [65] [66] [67]

[68] [69]

[70] [71]

[72] [73] [74]

[75] [76]

[77]

14

of enhancers and target genes in GeneCards, Database (Oxford). 2017 (2017), https://doi.org/10.1093/database/bax028. N. Rappaport, S. Fishilevich, R. Nudel, et al., Rational confederation of genes and diseases: NGS interpretation via GeneCards, MalaCards and VarElect, Biomed. Eng. Online 16 (Suppl 1) (2017) 72, https://doi.org/10.1186/s12938-017-0359-2. J. Ru, P. Li, J. Wang, et al., TCMSP: a database of systems pharmacology for drug discovery from herbal medicines, J. Cheminform. 6 (2014) 13, https://doi.org/10. 1186/1758-2946-6-13. L. Huang, D. Xie, Y. Yu, et al., TCMID 2.0: a comprehensive resource for TCM, Nucleic Acids Res. 46 (D1) (2018) D1117–D1120, https://doi.org/10.1093/nar/ gkx1028. X. Chen, H. Zhou, Y.B. Liu, et al., Database of traditional Chinese medicine and its application to studies of mechanism and to prescription validation, Br. J. Pharmacol. 149 (8) (2006) 1092–1103, https://doi.org/10.1038/sj.bjp.0706945. W. Dai, H.Y. Chen, C.Y. Chen, A network pharmacology-based approach to investigate the novel TCM formula against huntington’s disease and validated by support vector machine model, Evid. Complement. Alternat. Med. 2018 (2018) 6020197, , https://doi.org/10.1155/2018/6020197. Y.C. Fang, H.C. Huang, H.H. Chen, et al., TCMGeneDIT: a database for associated traditional Chinese medicine, gene and disease information using text mining, BMC Complement. Altern. Med. 8 (2008) 58, https://doi.org/10.1186/1472-6882-8-58. Z. Liu, F. Guo, Y. Wang, et al., BATMAN-TCM: a bioinformatics analysis tool for molecular mechANism of traditional chinese medicine, Sci. Rep. 6 (2016) 21146, https://doi.org/10.1038/srep21146. S. Kim, J. Chen, T. Cheng, et al., PubChem 2019 update: improved access to chemical data, Nucleic Acids Res. 47 (D1) (2019) D1102–D1109, https://doi.org/10. 1093/nar/gky1033. Y. Zhang, J. Yu, W. Zhang, et al., An integrated evidence-based targeting strategy for determining combinatorialbioactive ingredients of a compound herbal medicine Qishen Yiqi dripping pills, J. Ethnopharmacol. 219 (2018) 288–298, https://doi. org/10.1016/j.jep.2018.02.041. S. Zheng, Y. Zhang, Y. Qiao, The mechanism research of Qishen Yiqi Formula by module-network analysis, Evid. Complement. Alternat. Med. 2015 (2015) 497314, , https://doi.org/10.1155/2015/497314. Y. Han, M. Zhou, L. Xing, et al., Identification of NF-κB inhibitors in Qishenyiqi dropping pills for myocardial infarction treatment based on bioactivity-integrated UPLC-Q/TOF MS, Biomed. Chromatogr. 29 (10) (2015) 1612–1618, https://doi. org/10.1002/bmc.3468. A. Kramer, J. Green, J. Pollard Jret al., Causal analysis approaches in ingenuity pathway analysis, Bioinformatics. 30 (4) (2014) 523–530, https://doi.org/10. 1093/bioinformatics/btt703. H. Zhang, Y. Xing, J. Chang, et al., Efficacy and safety of NaoShuanTong Capsule in the treatment of ischemic stroke: a meta-analysis, Front. Pharmacol. 10 (2019) 1133, https://doi.org/10.3389/fphar.2019.01133. H. Wang, S. Ren, C. Liu, et al., An overview of systematic reviews of danhong injection for ischemic stroke, Evid. Complement. Alternat. Med. 2016 (2016) 8949835, , https://doi.org/10.1155/2016/8949835. Z. Sun, Q. Xu, G. Gao, et al., Clinical observation in edaravone treatment for acute cerebral infarction, Niger. J. Clin. Pract. 22 (Oct 10) (2019) 1324–1327, https:// doi.org/10.4103/njcp.njcp_367_18. P. Zhang, W. Li, L. Li, et al., Treatment with edaravone attenuates ischemic brain injury and inhibits neurogenesis in the subventricular zone of adult rats after focal cerebral ischemia and reperfusion injury, Neuroscience 201 (2012) 297–306, https://doi.org/10.1016/j.neuroscience.2011.11.005. J. Liu, Y. Jiang, G. Zhang, et al., Protective effect of edaravone on blood-brain barrier by affecting NRF-2/HO-1 signaling pathway, Exp. Ther. Med. 18 (4) (2019) 2437–2442, https://doi.org/10.3892/etm.2019.7859. W. Abdullahi, D. Tripathi, Ronaldson PT.BLood-brain barrier dysfunction in ischemic stroke: targeting tight junctions and transporters for vascular protection, Am. J. Physiol., Cell Physiol. 315 (3) (2018) C343–C356, https://doi.org/10.1152/ ajpcell.00095.2018. X. Jiang, A.V. Andjelkovic, L. Zhu, et al., Blood-brain barrier dysfunction and recovery after ischemic stroke, Prog. Neurobiol. 163-164 (2018) 144–171, https:// doi.org/10.1016/j.pneurobio.2017.10.001. X. Jin, J. Liu, Liu W.EArly ischemic blood brain barrier damage: a potential indicator for hemorrhagic transformation following tissue plasminogen activator (tPA) thrombolysis, Curr. Neurovasc. Res. 11 (3) (2014) 254–262, https://doi.org/ 10.2174/1567202611666140530145643. H. Yuan, Q. Ma, H. Cui, et al., How can synergism of traditional medicines benefit from network pharmacology, Molecules 22 (7) (2017) 1135, https://doi.org/10. 3390/molecules22071135. S. Li, Zhang B.TRaditional Chinese medicine network pharmacology: theory,methodology and application, Chin. J. Nat. Med. 11 (2) (2013) 110–120, https:// doi.org/10.1016/S1875-5364(13)60037-0. T. Wang, M. Lu, Q. Du, et al., An integrated anti-arrhythmic target network of a Chinese medicine compound, Wenxin Keli, revealed by combined machine learning and molecular pathway analysis, Mol. Biosyst. 13 (5) (2017) 1018–1030, https:// doi.org/10.1039/C7MB90035J. G. Yilmaz, T.V. Arumugam, K.Y. Stokes, et al., Role of T lymphocytes and interferon-gamma in ischemic stroke, Circulation. 113 (17) (2006) 2105–2112, https:// doi.org/10.1161/CIRCULATIONAHA.105.593046. H.L. Li, N. Kostulas, Y.M. Huang, et al., IL-17 and IFN-gamma mRNA expression is increased in the brain and systemically after permanent middle cerebral artery occlusion in the rat, J. Neuroimmunol. 116 (2001) 5–14, https://doi.org/10.1016/ S0165-5728(01)00264-8. W. Zhou, A. Liesz, H. Bauer, et al., Postischemic brain infiltration of leukocyte

Biomedicine & Pharmacotherapy 125 (2020) 109945

Y. Wang, et al.

[78] [79] [80] [81] [82] [83] [84] [85]

[86] [87] [88] [89]

subpopulations differs among murine permanent and transient focal cerebral ischemia models, Brain Pathol. 23 (1) (2013) 34–44, https://doi.org/10.1111/j. 1750-3639.2012.00614.x. S. Pawate, Q. Shen, F. Fan, et al., Redox regulation of glial inflammatory response to lipopolysaccharide and interferongamma, J. Neurosci. Res. 77 (4) (2004) 540–551, https://doi.org/10.1002/jnr.20180. K.L. Lambertsen, R. Gregersen, M. Meldgaard, et al., A role for interferon-gamma in focal cerebral ischemia in mice, J. Neuropathol. Exp. Neurol. 63 (9) (2004) 942–955, https://doi.org/10.1093/jnen/63.9.942. Y. Yamasaki, N. Matsuura, H. Shozuhara, et al., Interleukin-1 as a pathogenetic mediator of ischemic brain damage in rats, Stroke 26 (4) (1995) 676–680, https:// doi.org/10.1161/01.STR.26.4.676. S. Liu, S. Zhu, Y. Zou, et al., Knockdown of IL-1β improves hypoxia-ischemia brain associated with IL-6 up-regulation in cell and animal models, Mol. Neurobiol. 51 (2) (2015) 743–752, https://doi.org/10.1007/s12035-014-8764-z. T. Yoshimoto, K. Houkin, M. Tada, H. Abe, Induction of cytokines, chemokines and adhesion molecule mRNA in a rat forebrain reperfusion model, Acta Neuropathol. 93 (2) (1997) 154–158, https://doi.org/10.1007/s004010050596. U. Waje-Andreassen, J. Kråkenes, E. Ulvestad, et al., IL-6: an early marker for outcome in acute ischemic stroke, Acta Neurol. Scand. 111 (6) (2005) 360–365, https://doi.org/10.1111/j.1600-0404.2005.00416.x. W.M. Clark, L.G. Rinker, N.S. Lessov, et al., Lack of interleukin-6 expression is not protective against focal central nervous system ischemia, Stroke 31 (7) (2000) 1715–s, https://doi.org/10.1161/01.STR.31.7.1715. R. Berti, A.J. Williams, J.R. Moffett, et al., Quantitative real-time RT-PCR analysis of inflammatory gene expression associated with ischemia-reperfusion brain injury, J. Cereb. Blood Flow Metab. 22 (9) (2002) 1068–1079, https://doi.org/10.1097/ 00004647-200209000-00004. M. Haddad, H. Rhinn, C. Bloquel, et al., Anti-inflammatory effects of PJ34, a poly (ADP-ribose) polymerase inhibitor, in transient focal cerebral ischemia in mice, Br. J. Pharmacol. 149 (1) (2006) 23–30, https://doi.org/10.1038/sj.bjp.0706837. A. Dobolyi, C. Vincze, G. Pal, et al., The neuroprotective functions of transforming growth factor beta proteins, Int. J. Mol. Sci. 13 (7) (2012) 8219–8258, https://doi. org/10.3390/ijms13078219. C. Iadecola, J. Anrather, The immunology of stroke: from mechanisms to translation, Nat. Med. 17 (7) (2011) 796–808, https://doi.org/10.1038/nm.2399. P.T. Ronaldson, K.M. Demarco, L. Sanchez-Covarrubias, et al., Transforming growth

[90] [91] [92]

[93]

[94]

[95] [96]

[97]

[98]

15

factor-beta signaling alters substrate permeability and tight junction protein expression at the blood-brain barrier during inflammatory pain, J. Cereb. Blood Flow Metab. 29 (6) (2009) 1084–1098, https://doi.org/10.1038/jcbfm.2009.32. L. Pang, W. Ye, X.M. Che, et al., Reduction of inflammatory response in the mouse brain with adenoviral-mediated transforming growth factor-ss1 expression, Stroke 32 (2) (2001) 544–552, https://doi.org/10.1161/01.STR.32.2.544. A. Nurmi, P.J. Lindsberg, M. Koistinaho, et al., Nuclear factor-kappaB contributes to infarction after permanent focal ischemia, Stroke 35 (4) (2004) 987–991, https:// doi.org/10.1161/01.STR.0000120732.45951.26. B. Xiang, C. Xiao, T. Shen, et al., Anti-inflammatory effects of anisalcohol on lipopolysaccharide-stimulated BV2 microglia via selective modulation of microglia polarization and down-regulation of NF-κB p65 and JNK activation, Mol. Immunol. 95 (2018) 39–46, https://doi.org/10.1016/j.molimm.2018.01.011. W. Zhang, J.K. Song, X. Zhang, et al., Salvianolic acid A attenuates ischemia reperfusion induced rat brain damage by protecting the blood brain barrier through MMP-9 inhibition and anti-inflammation, Chin. J. Nat. Med. 16 (3) (2018) 184–193, https://doi.org/10.1016/S1875-5364(18)30046-3. K. Hyakkoku, J. Hamanaka, K. Tsuruma, et al., Toll-like receptor 4 (TLR4), but not TLR3 or TLR9, knock-out mice have neuroprotective effects against focal cerebral ischemia, Neuroscience 171 (1) (2010) 258–267, https://doi.org/10.1016/j. neuroscience.2010.08.054. J.R. Caso, J.M. Pradillo, O. Hurtado, et al., Toll-like receptor 4 is involved in brain damage and inflammation after experimental stroke, Circulation 115 (12) (2007) 1599–1608, https://doi.org/10.1161/CIRCULATIONAHA.106.603431. W. Chen, Y. Zheng, L. Yan, et al., Validated UHPLC-MS/MS method for simultaneous determination of four triterpene saponins from Akebia trifoliata extract in rat plasma and its application to a pharmacokinetic study, Biomed. Chromatogr. 33 (10) (2019) e4585, https://doi.org/10.1002/bmc.4585. Q. Qian, N. Zhou, P. Qi, et al., A UHPLC-QTOF-MS/MS method for the simultaneous determination of eight triterpene compounds from Poria cocos (Schw.) Wolf extract in rat plasma: application to a comparative pharmacokinetic study, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 1102-1103 (2018) 34–44, https://doi.org/10. 1016/j.jchromb.2018.10.011. S. Bergmann, S.E. Lawler, Y. Qu, et al., Blood-brain-barrier organoids for investigating the permeability of CNS therapeutics, Nat. Protoc. 13 (12) (2018) 2827–2843, https://doi.org/10.1038/s41596-018-0066-x.