MENG Qing?wen, LIU Hua?jiang, DING Shun, HUANG Shan, YANG Yang, ZHANG Yu?zhuo, YANG Shan?shan, ZUO Qi, XIE Yi?qiang

1. Department of Cardiovascular Medicine, the First Affiliated Hospital of Hainan Medical University

2. Department of Interventional Vascular Surgery, the First Affiliated Hospital of Hainan Medical University

3. Hainan Medical University, Haikou 570102, China

Keywords:Astragalus Diabetic cardiomyopathy Network pharmacology

ABSTRACT Objective: To explore the potential active ingredients and targets of Astragalus, and also to predict the targets and mechanisms of Astragalus in the treatment of diabetic cardiomyopathy.Based on the predicted results, the key signaling pathways were validated in a diabetic cardiomyopathy model mouse.Methods: Compounds and targets in Astragalus were retrieved from Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform. The protein names to corresponding "Gene Symbol ID" was convert by STRING database.We obtained targets of diabetic cardiomyopathy data from DisGeNET datasets. The protein?protein interaction network (PPI network) was established using STRING database.Cytoscape 3.6.0 was used to construct a disease?drug?target gene network map and to screen the 10 closest target genes by Cytohuba plug?in. The overlapping genes were then subjected to gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)?based enrichment analysis.Finally, the key molecules of the MAPK signaling pathway were validated by in vitro experiments. Animal experiments were performed using 21 Kunming mice randomly divided into normal group, model group, and Chinese herbal medicine Astragalus group, with seven mice in each group. The myocardium of mice in each group was stained with HE to compare the pathological morphological changes, and Western Blot was also used to compare the key molecules of MAPK signaling pathway, ERK1 and p?p38. Results:Astragalus contained 20 active ingredients with 188 corresponding targets, 220 targets related to diabetic cardiomyopathy and 37 targets acting in conjunction with Astragalus. The common targets were imported into the STRING database to obtain a PPI network graph of overlapping genes, with 37 nodes and 391 edges. The PPI network map was imported into Cytoscape 3.6.0 software, and the most significant top 10 hub genes were obtained using the MCC algorithm in the cytoHubba plugin, namely AKT1,TP53,CASP3,MMP9,EGF,IL?10,CXCL8,IL?1β,VEGFA,PPARG. GO functional enrichment analysis yielded 40 entries for biological process (BP), 23 entries for cellular component (CC), 22 entries for molecular function (MF) and 94 entries for KEGG pathway enrichment screening, mainly involving PI3K?AKT, MAPK, HIF?1, FOXO, TNP pathway and other inflammation or apoptosis regulatory pathways. Animal experiments showed that Astragalus can improve the inflammatory state of myocardial tissue in mice with diabetic cardiomyopathy, and the expression of ERK1 and p?p38 protein in myocardial tissue of mice in the model group was higher than that in the normal group (P<0.05, P<0.01), and after the intervention with Astragalus, the expression of ERK1 and p?p38 protein was significantly lower than that in the model group, and the difference was statistically significant (P<0.05, P<0.01). Conclusion:Astragalus has multi?target, multi?component and multi?pathway action characteristics in the treatment of diabetic cardiomyopathy, which can exert anti?inflammatory and anti?oxidative stress effects by regulating protein expression of MAPK signaling pathway ERK1, p?p38.

1. Introduction

According to the statistics, the rising incidence of diabetes has become a social issue of great concern, China already ranks first in the world in terms of the total number of people with diabetes,which is expected to reach 147 million by 2045 [1]. Diabetic cardiomyopathy (DCM) is a chronic complication of diabetes mellitus and is one of the leading causes of death in diabetic patients[2]. The pathology of DCM is characterized by myocardial fibrosis and myocardial hypertrophy, which ultimately affects myocardial systolic and diastolic functions and causes the development of heart failure. The pathogenesis of DCM is complex, with multiple mechanisms such as polyol pathways, glycosylation end products,inflammatory responses, growth factors, and oxidative stress all involved in the pathogenesis of the disease. The progression of DCM is currently controlled clinically by controlling blood glucose,nourishing the myocardium, and improving circulation, but the disease is difficult to identify in its early stages and the adverse effects of Western drugs make the clinical benefit of DCM patients limited.DCM belongs to the category of "xiaoke", "xiongbi" and "xindan"in Chinese medicine. Its pathogenesis is mainly due to the deficiency of spleen and kidney yin and fluid, and the predominance of dryness and heat, resulting in the deficiency of qi and blood and the deficiency of yin and dryness and heat. Long?term disease enters the ligaments and stagnates the blood vessels, resulting in chest paralysis. DCM has a long duration of illness and is characterized by deficiency of qi, combined with qi stagnation, phlegm obstruction,blood stasis, dampness and heat [3-4]. Astragalus is the root of Astragalus mongolica, a leguminous plant with a sweet taste and slightly warm nature, which belongs to the spleen and lung meridians and has the effect of tonifying qi and raising yang,detoxifying and draining pus, draining diuresis and strengthening the body's resistance. In recent years, studies have shown that Astragalus can reduce the expression of inflammatory factors such as TNF?α as well as IL?6, delaying the process of myocardial fibrosis and ultimately improving the systolic and diastolic functions of the heart[5]. Astragalus polysaccharide can inhibit the expression of apoptosis target genes such as Bax and enhance the expression of SOD2 and MAPK proteins, thus improving oxidative stress and achieving therapeutic effects in DCM [6]. However, there is a lack of in?depth reports on the mechanism of Astragalus in the treatment of DCM.Therefore, this study intends to clarify the specific components and action targets of Astragalus and its mechanism of action in DCM through a network pharmacological approach, and provide a scientific basis and reference direction for the application of Astragalus in DCM.

2. Materials and methods

2.1 Screening of active compounds and targets of Astragalus

The pharmacological database of TCMSP (http://lsp.nwu.edu.cn/tcmsp.php) was searched with "Astragalus" as the keyword, and the ADME parameters OB≥30% and DL≥0.18 were set as the screening basis to screen the active compounds of Astragalus The target protein name was converted to "gene symbol" by setting the species category in the STRING database (http://www.string?db.org).

2.2 Screening for the targets of DCM diseases

Using the English term "diabetic cardiomyopathy" as the search term, disease targets were identified through the disease?related gene and mutation loci database (DisGeNET, http://www.disgenet.org).

2.3 Prediction of potential targets of Astragalus for DCM

The targets of Astragalus major pharmacological components and the corresponding target genes of DCM were imported into Venny2.1 (http://bioinfogp.cnb.csic.es/tools/venny/index.html), and the overlap obtained was the potential target genes of Astragalus for DCM treatment.

2.4 Astragalus active compound, DCM and overlapping gene network construction

The active compounds, DCM and common genes screened by Astragalus were imported into Cytoscape 3.6.0 and constructed into a network diagram to understand the correlation of the three more visually.

2.5 Core target screening and network graph construction

The obtained common target genes were imported into STRING database, and the species "homo sapiens" was selected with a minimum interaction score of 0.40 at the highest confidence level,and then the overlapping gene PPI network map was obtained. The PPI network map was imported into Cytoscape 3.6.0 software, and the top 10 most significant hub genes were obtained using the MCC algorithm in cytoHubba plugin.

2.6 GO and KEGG pathway enrichment analysis

The DAVID database (http://david.abcc.ncifcrf.gov) was used for GO functional analysis and KEGG pathway enrichment of the common targets of Astragalus and DCM to obtain the closely related biological process (BP), cellular component (CC) and molecular function (MF) and signaling pathways of Astragalus against DCM.We obtained the biological process (BP), cellular component (CC)and molecular function (MF) and signaling pathways related to Astragalus and DCM.

2.7 Animal experiment validation

2.7.1 DrugAstragalus granules were purchased from Haikou Chinese Medicine Hospital, and the manufacturer was Guangdong Fang Fang Chinese Medicine Granules Co.

2.7.2 Reagents and instruments

Streptozotocin (STZ) was purchased from SIGMA, USA, and STZ was dissolved in pH 4.5 citric acid 0.1 mol/L buffer, ready to use; ERK1 and β?actin primary antibodies were purchased from abcam, p?p38 antibody was purchased from CST, USA; horseradish peroxidase?labeled secondary antibody was purchased from Shanghai Biyuntian Company.

2.7.3 Animal grouping and moldelingTwenty?one male Kunming mice, 8?10 weeks old, weighing 18?22 g, were purchased from Changsha Tianqin Biotechnology Co., Ltd.and housed at room temperature 20?24℃, relative humidity 45?50%,and light and dark cycles (12h/12h). Common and high?fat diets,purchased from Guangdong Medical Laboratory Animal Center.After 1 week of adaptive feeding, 7 mice were randomly selected as the normal group and fed with normal chow; the remaining 14 mice were used for modeling as follows: after being fed with high?fat chow for 4 weeks, STZ was injected intraperitoneally at 30 mg/kg after 12 hours of fasting, and blood glucose was detected after 16 hours of fasting after 3 days, and fasting blood glucose >16.7 mmol/L was considered as the modeling of diabetes. The success criterion was fasting blood glucose >16.7 mmol/L.

Seven mice were randomly selected as the model group and seven mice were selected as the Astragalus group after successful modeling. Astragalus granules were converted to the conventional human dose, and the gavage concentration was 0.25 g/ml. The model group and the normal group were given the same amount of pure water for gavage once a day, 0.2 ml once a day, for 4 weeks. The experimental protocol was approved by the ethics committee of the First Affiliated Hospital of Hainan Medical College.

2.7.4 Pathological sectionsAfter the end of perfusion, the mice were anesthetized with chloral hydrate and then executed, and the heart tissues were taken, fixed with 10% paraformaldehyde, embedded in paraffin, stained with hematoxylin?eosin stain (HE stain), and the pathological changes were observed by light microscopy and image analysis.

The whole court applauded her with hands and tails; and for a moment her heart felt quite gay, for she knew she had the loveliest voice25 of any on earth or in the sea

2.7.5 Western?Blot detection of ERK1, p?p38 protein expression

Cut small pieces of heart tissue, add RIPA lysis solution to lyse myocardial tissue, extract total protein, and operate according to the BCA kit instructions for protein concentration determination. Prepare the electrophoresis gel required for WB, perform sample addition,electrophoresis, transmigration by electrotransference instrument,5% skimmed milk for room temperature closure, add ERK1, p?p38 primary antibody (1:1000, TBST for dilution) for incubation, shake at 4℃ overnight, recover the primary antibody, wash the membrane 3 times by TBST, dress the secondary antibody at room temperature,wash the membrane 3 times again, luminescence color development by developer, and exposure imaging by gel imager. The strips were analyzed for grayscale using Image J software and processed for semi?quantitative analysis.

2.7.6 Statistical methodsPrism software was used for data processing and analysis, and one?way ANOVA was used for comparison among the three groups, and if there was a statistical difference, Tukey was used for further two?by?two comparison between the two groups, and P<0.05 was used as the difference was statistically significant.

3. Results

3.1 Main potent compounds and target genes of Astragalus

By searching the TCMSP database, we obtained 20 main potent compounds of Astragalus and 188 corresponding targets from the TCMSP database. The information of the main potent compounds is shown in Table 1 below.

3.2 Excavation of DCM targets by Astragalus for treatment

220 DCM?related target genes were screened from the DisGeNET database. After finding the intersection with the targets corresponding to the active compounds of Astragalus by Venn diagram, 37 common targets were obtained, which were GSK3B, PPARG,DPP4, AKT1, BCL2, AHSA1, CASP3, MAPK8, XDH, HMOX1,NR1I2, MAPK14, ADRB2, SIRT1, EGFR, VEGFA, CASP9,MMP9, MAPK1, IL10, EGF, TP53, CASP8, ERBB2, GJA1, IL1B,CXCL8, NOS3, TGFB1, SERPINE1, NFE2L2, PARP1, PPARA,CRP, CXCL10, SPP1, IGF2. The Venn diagram results are shown in Figure 1.

Tab 1 The main medicinal compounds of Astragalus and their corresponding ADME parameters

3.3 Construction of the "disease?compound?target" network of Astragalus for DCM

3.4 Protein Interaction Network Map Construction

The common target interactions data of Astragalus treatment DCM were imported into STRING database to obtain the overlapping gene PPI network graph (Figure 3). There were 37 nodes and 391 edges.The PPI network map was imported into Cytoscape 3.6.0 software,and the most significant top 10 hub genes were obtained using the MCC algorithm in cytoHubba plugin, which were AKT1, TP53,CASP3, MMP9, EGF, IL?10, CXCL8, IL?1β, VEGFA, and PPARG(Figure 4).

3.5 GO functional enrichment

GO functional analysis and KEGG pathway enrichment were performed for the common targets of Astragalus and DCM using DAVID database. The top 10 entries were selected for the GO functional enrichment of BP, mainly related to the cellular response to mechanical stimulation, cellular response to hypoxia, negative regulation of cell growth, etc.; the top 10 entries were selected for the CC, mainly related to the extracellular space, platelet granule lumen, mitochondria, cytoplasm, etc.; the top 10 entries were selected for the MF, mainly related to protein phosphatase binding,transcription factor binding, and cysteine?type endopeptidase activity in the apoptotic process. The above results were entered into the microbiology website (http://www.bioinformatics.com.cn), and the BP, CC and MF bubble maps were obtained (Figure 5).

3.6 KEGG pathway enrichment analysis

The common targets of Astragalus for DCM were enriched in 30 pathways (P < 0.01), mainly including PI3K?AKT, MAPK,HIF?1, FOXO, TNP pathway and other inflammatory or apoptosis regulatory pathways. The results are shown in Figure 6. MAPK signaling pathways were selected for display, where green represents common genes.

3.7 Comparison of myocardial pathomorphology in mice

In the normal group, the myocardium was evenly stained, the cell shape was intact and closely arranged, the nucleus was centered,there were no breaks or defects, and no inflammatory cell infiltration was seen. In the model group, myocardial cells were changed in shape, disordered in arrangement, widened myocardial gaps, nuclei were fixed, some of them were edematous and deformed, and a small amount of inflammatory cell infiltration was seen. In the Astragalus group, the myocardial tissue changed with improved cell shape and arrangement, narrowed myocardial gap, and significantly reduced inflammatory cell infiltration and nucleus consolidation. See Figure 8.

3.8 Comparison of ERK1 and p?p38 in myocardial tissues of mice in each group

As shown in Figure 9, Table 2 and Table 3, the myocardial tissue ERK1, p?p38 protein expressions of mice were statistically different among the three groups (P < 0.5). Further comparison between the two groups, the expression of ERK1 and p?p38 protein was statistically different between the normal and model groups compared (P=0.0153; P=0.0017); the expression of ERK1 and p?p38 protein was statistically different between the model and astragalus groups compared (P=0.0377; P=0.0143).

Tab 2 Statistics of gray?scale values of ERK1 protein expression in myocardial tissue of mice in each group (n=7, ±s)

Tab 2 Statistics of gray?scale values of ERK1 protein expression in myocardial tissue of mice in each group (n=7, ±s)

group mean±Standard deviation F P Control group 0.757 5 ± 0.0385 3 Model group 0.995 9 ± 0.0484 29.378 0.014 2 Astragalus group 0.803 1 ± 0.0359 7

Tab 3 Statistical table of grayscale values of p?p38 protein expression in myocardial tissue of mice in each group(n=7, ±s )

Tab 3 Statistical table of grayscale values of p?p38 protein expression in myocardial tissue of mice in each group(n=7, ±s )

group mean±Standard deviation F P Control group 0.690 0 ± 0.0440 3 Model group 1.024 0± 0.0392 421.09 0.001 9 Astragalus group 0.808 1 ± 0.0245 2

4. Discussion

DCM is one of the most common cardiovascular complications of diabetes mellitus, with an insidious onset and rapid progression that is not easily recognized in the clinical setting. Myocardial cell damage continues and worsens progressively during the course of the disease, and is one of the main factors leading to death in diabetic patients. Studies on the pathogenesis of DCM have focused on disorders of glucolipid metabolism, activation of the renin?angiotensin?aldosterone system, oxidative stress and mitochondrial damage, inflammatory response, endoplasmic reticulum stress,apoptosis, autophagy, and imbalance of calcium homeostasis [7]. At present, the clinical treatment of DCM is mainly based on drugs to control blood sugar and improve heart failure, but the overall efficacy is not significant, or there are more adverse reactions to long?term drug use and other conditions. Chinese medicine has obvious advantages for the early treatment of DCM and the improvement of patients' clinical symptoms. In traditional medicine, it was pointed out in Suwen that "the heart disease, the pulse does not pass", and the term "xindan" was first introduced, referring to diabetes combined with cardiovascular disease. Although the etiology of DCM is complex and varied, and the dialectical focus varies at different stages of the disease, Qi deficiency is always present throughout the overall course of DCM, so Qi benefit is the basic treatment for the disease, and Huang Qi is one of the most widely used Qi benefit drugs in clinical practice. It has been shown that Astragalus can significantly improve cardiac function and reduce myocardial injury in DCM rats through activation of MAPK signaling pathway[8]. The role of Astragalus polysaccharide in DCM may be related to its anti?apoptotic effect [9].

Based on molecular biology, systems biology and pharmacology theories, network pharmacology constructs complex interaction networks based on bioactive compounds, target molecules and biological functions to elucidate the mechanism of action of Chinese medicine. Therefore, the present study finally obtained 20 main compounds of Astragalus and 188 corresponding targets through screening. From the results, it can be seen that a single chemical component can have multiple corresponding targets, and at the same time, different chemical components can intersect on the same target,reflecting the characteristics of Astragalus with multiple targets and rich effects. Among them, quercetin and kaempferol have the most potential targets of action, revealing that quercetin and kaempferol may be the main active components of Astragalus action. Studies have shown that quercetin is able to improve the body's insulin pancreas and lower blood glucose by reducing the expression of skeletal muscle inflammatory factors and macrophage infiltration

[10]. Kaempferol significantly inhibits high glucose?induced inflammatory cytokine expression and ROS production, which ameliorates hyperglycemic damage to cardiomyocytes by inhibiting NF?κB and Nrf?2 activation [11]. Therefore, combining previous studies and TCMSP database results predicts that Astragalus can be used to treat DCM may be related to mechanisms such as lowering blood glucose, modulating inflammatory response and oxidative stress.

The PPI network map targets interact with each other and constitute a complex biological network system, in which the greater the connectivity, the greater the possibility of Astragalus interfering with DCM through this target. The most significant top 10 hub genes were obtained using the MCC algorithm in the cytoHubba plugin, namely AKT1, TP53, CASP3, MMP9, EGF, IL?10, CXCL8, IL?1β, VEGFA,and PPARG. Studies have demonstrated that hyperglycemia causes increased cardiac stress, and cardiac stress leads to cardiomyopathy through increased matrix metalloproteinases (MMPs) activity and extracellular matrix (ECM) remodeling, and MMP9 has been shown to be one of the key matrix metalloproteinases that is activated in diabetes [12]. Diabetes triggers an excessive production of mitochondrial ROS, which contributes to the secretion of inflammatory and chemokines such as IL?10, IL?1β, and TNF?α,leading to myocardial injury [13-14]. Studies have shown that serum concentrations of inflammatory factors such as TNF and IL?6 in diabetic patients are associated with abnormal cardiac diastolic function, and these inflammatory factors can cause cellular damage and inflammation and promote the process of myocardial fibrosis[15].

The GO and KEGG pathway enrichment analysis of the common targets of Astragalus and DCM using DAVID database revealed that the main signaling pathways associated with DCM include PI3K/AKT pathway, MAPK pathway, HIF?1 signaling pathway, FoxO signaling pathway, cancer, Toll?like receptor signaling pathway, etc.The results of the pathway enrichment analysis indicated that the active ingredients of Astragalus act on The results of the pathway enrichment analysis indicate that the core targets of the action of the active ingredients of Astragalus are mapped on different pathways,and can exert the drug effect through multiple pathways together,which indicates the direction for further mechanism research. It has been found that Astragalus polysaccharide can inhibit the activity of MAPK signaling pathway thereby delaying the development of DCM[8]. Zhang Shuchun et al. showed that astragaloside had a protective effect on high glucose?induced H9c2 cardiomyocyte injury, and the possible mechanism was the inhibition of excessive activation of the IKK/NF?κB inflammatory pathway[16]. Foreign scholars have shown through animal experimentsthat insulin affects glucolipid metabolism and contractile function of cardiac myocytes mainly through MAPK pathway and PI3K signaling pathway[17]. Yan Jiamin et al. showed that the pathological process of DCM caused by high glucose and high fat resulted in an inflammatory response mediated through the p38 MAPK pathway [18]. HIF?1α decreased myocardial oxidative stress damage and apoptosis in diabetic myocardial IR injury by increasing the expression of PGC?1α and improving the energy metabolism level of mitochondria[19]. MAPK is stimulated by growth factors, neurotransmitters, and cytokines to undergo phosphorylation and thus activation, which in turn is involved in pathophysiological processes such as cell growth, survival, and apoptosis, and is an important component of the inflammatory and oxidative stress signaling pathways. The key molecules of MAPK,ERK, JNK, and p38MAPK, have been documented to play an important role in diabetes and its complications [20]. ERK1/2 is a major member of the MAPK family, and activation of ERK1/2 regulates cell division and is closely related to growth, proliferation,differentiation, migration, and survival [21]. p38 MAPK signaling pathway activation activation of the p38 MAPK signaling pathway is a key step in the intracellular phosphorylation cascade, which is associated with cellular stress, inflammation, and apoptosis.The p38 MAPK signaling pathway plays an important role in the pathogenesis of DCM microangiopathy, myocardial interstitial fibrosis, and myocardial hypertrophy[22]. The present study showed that inflammatory cell infiltration was seen in myocardial tissue of model mice, and the expression levels of ERK1 and p?p38 proteins were significantly higher than those of the normal group, suggesting that ERK1 and p?p38 were involved in the progression of DCM disease, and after administration of Astragalus, the expression of the above proteins decreased compared with the previous one, and the inflammatory cell infiltration, disordered cell arrangement, and nucleus fixation in myocardial tissue improved compared with the previous one, confirming that Astragalus has a The mechanism of action may be related to the inhibition of MAPK signaling pathway activation, which is also consistent with the results of network pharmacology discussed earlier.

In conclusion, the study of the multifaceted network components between Astragalus and DCM diseases clarified the main potential active components and action targets of Astragalus and related pathways, which provided new ideas for further in?depth exploration of its action mechanism and indicated the research direction for the application of Astragalus in DCM, and further experimental studies in metabolomics and other aspects are still needed at a later stage.