Systems Pharmacology Dissection of Mechanisms of Dengzhan Xixin Injection against Cardiovascular Diseases

Abstract

Dengzhan Xixin injection (DZXXI), a herbal product prepared from a Chinese herb called Erigeron breviscapus, is a classical and traditional therapeutic for cadiovascular diseases (CVDs), including coronary heart disease (CHD), angina, and stroke, etc. However, its potential pharmacology mechanism against CVDs remains unclear. In this paper, a systems pharmacology-based strategy is presented for predicting drug targets and understanding therapeutic mechanisms of DZXXI against CVDs. The main ingredients were identified by HPLC-diode array detector (DAD). The target fishing was performed on the PharmMapper Server (http://lilab-ecust.cn/pharmmapper/). Potential targets were confirmed by two molecular docking tools, Sybyl-X 1.3 and Ledock to ensure the accuracy. The resulting target proteins were applied as baits to fish their related diseases and pathways from the molecular annotation system (MAS 3.0, http://bioinfo.capitalbio.com/mas3/) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg/). Network generation and topological analysis were performed in Cytoscape 3.6.0. 15 main ingredients from DZXXI were identified. Forty five putative drug targets and 50 KEGG pathways, which have highly relevance to the therapeutic effects of DZXXI against CVDs, were then obtained. The systems analysis suggested that DZXXI could attenuate cardiac fibrosis, regulate cardiac contractility, and preserve heart function in adverse cardiac remodeling; meanwhile DZXXI also could have the function of activating blood circulation and dilating blood vessels. DZXXI exerts its therapeutic effects on CVDs possibly through multi-targets including CMA1, epidermal growth factor receptor (EGFR), phenylalanine-4-hydroxylase (PAH), SRC, F7, etc., and multi-pathways including Focal adhesion, mitogen-activated protein kinase (MAPK) signaling pathway, complement and coagulation cascades, Wnt signaling pathway, vascular endothelial growth factor (VEGF) signaling pathway, Renin-angiotensin system, etc.

Introduction

Cardiovascular diseases (CVDs) are the most prevalent non-communicable diseases worldwide, accounting for an estimated 17.8 million deaths in 2017, of which more than three quarters were occurred in developing countries.¹⁾ Coronary heart disease (CHD), one main form of cardiovascular diseases (CVDs), usually refers to myocardial ischemia (MI) and hypoxia due to atherosclerotic changes of coronary artery causing stenosis or occlusion of vascular lumen. CHD is the leading cause of deaths attribute to CVDs around the world.²⁾

Dengzhan Xixin injection (DZXXI) is a kind of sterile water injection prepared from the extracts of a Chinese herb, Erigeron breviscapus. DZXXI has been approved by the Chinese Food and Drug Administration and listed in the Chinese Pharmacopeia since 2005. DZXXI can activate blood circulation to dissipate blood stasis and relieve pain, and it has been extensively used in the treatment of CVDs including coronary heart disease, angina, acute myocardial infarction, hypertension, hyperlipidaemia, chronic heart failure, and pulmonary heart disease in conjunction with Western medicine in China.³⁾ In one randomized controlled trial,⁴⁾ 106 patients with ST segment elevated acute myocardial infarction after percutaneous coronary intervention (PCI) were randomly allocated into two groups that received DZXXI (20 mL/250 mL of 50g/L glucose injection, intravenous (i.v.) drip, quaque die (qd) combined with standard medication (n = 54) or standard medication alone (n = 52) for 10 d. The outcome showed that 3 months after PCI, compared with the standard medication group, the left ventricular volume, the left ventricular end diastolic volume index and the left ventricular end systolic volume index in the DZXXI group decreased, and the left ventricular ejection fraction increased (p < 0.05). And the incidence of severe arrhythmia, heart failure, cardiogenic shock and mortality during hospitalization were lower than that of the standard medication group (p < 0.05). Besides, DZXXI could treat both unstable angina pectoris⁵⁾ and stable angina pectoris,⁶⁾ and the frequency of use of DZXXI for angina pectoris was usually two or four weeks.^7–9) However, due to the multicomponent nature of herbal drug ingredients, the effect mechanism of DZXXI against CVDs is still not very clear, which is unfavorable to its the clinical application.

Systems pharmacology is a novel research tool that is based on the application of omics and systems biology-based technologies, which is fully suitable for the multi-components and multi-targets characteristics of traditional Chinese Medicines (TCMs). In recent years, systems pharmacology has been successfully applied to TCMs for screening bioactive ingredients,¹⁰⁾ predicting potential drug targets,¹¹⁾ disclosing therapeutic mechanisms,^12–19) understanding rules of drug combination,²⁰⁾ developing new synergistic drug combinations,²¹⁾ which has greatly promoted the modernization of TCMs.

In this paper, to uncover the material basis of DZXXI, a sensitive and reliable HPLC-diode array detector (DAD) method was developed to identify the chemical constituents of DZXXI. Subsequently, the systems pharmacology approach is employed for dissecting the comprehensive effects and mechanisms of multiple components of DZXXI against MI. The present study provides new insights into understanding or digging the therapeutic indications of DZXXI on CVDs.

Experimental

Chemicals and Reagents

E. breviscapus injections were obtained from Yunnan Biovalley Dengzhanhua Pharmaceutical Co., Ltd. (Batch no. 20171147). 3-O-Caffeoylquinic acid(3-CQA), caffeic acid (CA), 4-O-caffeoylquinic acid (4-CQA), 5-O-caffeoylquinic acid (5-CQA), Scutellarin, 1,3-O-dicaffeoylquinic acid (1,3-diCQA), 3,4-O-dicaffeoylquinic acid (3,4-diCQA), 3,5-O-dicaffeoylquinic acid (3,5-diCQA), 4,5-O-dicaffeoylquinic acid (4,5-diCQA) and apigenin-7-O-glucuronide (A-7-O-G) were provided by Shanghai Ronghe Medicine Technology Development Co., Ltd., with a purity of >99%. 4-Caffeoyl-2,7-anhydro-3-deoxy-2-octulopyranosonic acid (4-CDOA), 3-caffeoyl-2,7-anhydro-3-deoxy-2-octulopyranosonic acid (3-CDOA), 9-caffeoyl-2,7-anhydro-3-deoxy-2-octulopyranosonic acid (9-CDOA), Erigeron B, 3,4-O-dicaffeoyl-2,7-anhydro-3-deoxy-2-octulopyranosonic acid (3,4-diCDOA) (or 4,9-diCDOA) were prepared by laboratory and identified by NMR. The purities of all the compounds were determined to be above 98% by HPLC analysis.

HPLC Analysis

HPLC analysis was applied on an Agilent 1290 Infinity LC instrument (Agilent, Waldbronn, Germany) consisting of a binary pump, a diode-array detector, an auto-sampler and a column compartment. Separation was carried out by elution on an Ultimate XB-C18 column (4.6 × 100 mm, 3.5 µm). The mobile phase consisted of deionized water–acetic acid (A; 100 : 0.5, v/v) and acetonitrile (B). The gradient elution was employed as follows: 94% A and 6% B at 0–15 min; 82% A and 18% B at 15–17 min; 80% A and 20% B at 17–28 min; 79% A and 21% B at 28–29 min; 5% A and 95% B at 29–34 min. The re-equilibrium took 9 min, giving a total run time of 43 min. The flow rate was 0.7 mL/min. The column temperature was kept at 30°C. For HPLC analysis, 1 mL DZXXI was accurately diluted to 10 mL with methanol aqueous solution and the volume of sample injected was 10 µL.

Drug Targeting

Accurately predicting compound-target interaction profiles is of vital importance for revealing the underlying mechanisms of drugs.²²⁾

Firstly, we input all the ingredients into Chemspider (http://www.chemspider.com/) to obtain the structure of each compound in DZXXI, and imported these files into ChemBioDraw 14.0 software (PerkinElmer, Inc., Waltham, MA, U.S.A.) to optimize the molecular energy. Next, the output ‘mol2’ format files of compounds were imported into PharmMapper (lilab.ecust.edu.cn/pharmmapper), a web server that uses a pharmacophore mapping approach for potential drug target identification.^23–25) The results were filtered by a comprehensive consideration of fit score, Z score and the correlativity between protein and diseases.²⁶⁾

Secondly, two molecular docking softwares, SYBYL-X 1.3 and Ledock were used as confirmatory tools for the virtually selected compounds and their targets. The Surflex-docking module in SYBYL is based on a “ProtoMol,” a representative protein activity docking pocket, which can be automatically and/or user-defined generated. Before initiating the docking simulations, the co-crystallized ligand and structural water molecules were removed from the crystal structure and the polar hydrogen atoms were added in SYBYL software. LeDock, an easily-used molecular docking program which has comparable accuracy to popular docking programs, like Autodock and GOLD,²⁷⁾ is also employed in this paper. LeDock achieves a pose-prediction accuracy of greater than 90% on the Astex diversity set through simulating annealing and evolutionary optimization. The protein–ligand interactions were viewed by PyMOL viewer.

Thirdly, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways as well as target diseases were predicted via a comprehensive analysis tool, Mas 3.0 (molecule annotation systems, http://mas.capitalbiotech.com/mas3/).

Network Generation and Topological Analysis

Networks, a mathematica²⁸⁾ way to describe and quantify²⁹⁾ connections beneath the complex biological systems, were constructed in Cytoscape (Version 3.6.0).³⁰⁾ Three topological parameters were calculated for each network using the plugin Network Analyzer: degree, betweenness centrality (this property correlates more closely with essentiality than connectivity, exposing critical nodes that usually belong to the group of scaffold proteins or proteins involved in crosstalk between signal pathways (called bottlenecks))²⁹⁾ and Closeness centrality. The importance of a node in a network is indicated by the values of these indices, with higher values indicating greater important nodes.

Results

Main Ingredients in DZXXI

As shown in Fig. 1, DZXXI at least contains 15 ingredients, which have been identified by HPLC with the aids of their standard references. Table 1 lists the structure, formula and weight mass information of the identified main ingredients in DZXXI.

Fig. 1. HPLC Chromatogram of DZXXI

Peaks 1–15 were identified as 5-CQA, 3-CDOA, 3-CQA, 4-CQA, 4-CDOA, Caffeic acid, 9-CDOA,1,3-diCQA, Scutellarin, 3,4-diCQA, Erigoster B, 3,5-diCQA, 3,4-diCDOA (or 4,9-diCDOA), A-7-O-G and 4,5-diCQA, respectively with the aids of their standard substances.

Table 1. The Structure Information of Main Ingredients in DZXXI

All the compounds in Table 1 were considered to be potential active ingredients in DZXXI according to literature. It should be pointed out that caffeoylquinic acids (CQAs), the representative ingredients in DZXXI possess a broad range of pharmacological properties, including, anti-inflammatory,³¹⁾ antimicrobial,³²⁾ neuroprotective,³³⁾ and other biological effects. CQAs also have protective effect against H₂O₂-induced H9c2 cardio myoblast damage.³⁴⁾ Specifically, 5-CQA is one of the major chlorogenic acids with high biological importance because of its antioxidant,³⁵⁾ antimicrobial,³⁶⁾ neuroprotective,³⁷⁾ anticancer,³⁸⁾ anti-inflammatory,³⁹⁾ and others biological effects. 3,4-DiCQA could inhibit angiotensin-II-induced vascular smooth muscle cell proliferation.⁴⁰⁾

Furthermore, CDOAs are also abundant in DZXXI and exhibit multiple biological activities such as antioxidant, vasodilating and antimicrobial effects.⁴¹⁾ Notably, although Scutellarin has low aqueous solubility, poor chemical stability, short biological half-life and rapid elimination rate from the plasma, it also has been proved as drugs in China since 1984 to treat acute cerebral infarction and paralysis induced by cerebral hemorrhage, cerebral thrombosis, and hypertension.⁴²⁾ A-7-O-G, the aglycone of Scutellarin, a candidate drug for cardio-cerebrovascular diseases, was also selected in our further investigation for that this compound has effect on ischemia-reperfusion injury⁴³⁾ According to the above considerations, it was reasonable to list all the compounds as potential active ingredients of DZXXI.

Target Proteins of DZXXI and MI

Finally, 45 potential targets of ingredients in DZXXI related to MI were obtained from PharmMapper. The information of the resulting targets proteins is listed in Table 2. All the ingredients and their potential targets were further confirmed by Sybyl and Ledock (See Supplementary Table S1) and failed docking relationships (Cscore <4 and/or Total score < 0) were deleted. In Fig. 2A, we show Total score of ligand–target interaction and their corresponding Binding free energy. Targets of ingredients with high Total scores usually have been calculated to have low Binding free energy correspondingly. The correlation analysis showed the negative correlation of Total score and corresponding Binding free energy (Pearson Correlation Coefficient = −0.394, p < 0.01). Table 3 lists top 20 Total score of molecular docking and their corresponding Binding free energy. Both Total score and Binding free energy indicates that 4,9-diCDOA is a good ligand for epidermal growth factor receptor (EGFR). As is shown in Figs. 2B and 2C, among the 249 interactions, Binding free energy mainly distributes in −5 to 8.5 kcal/mol, accounting for 73.8% of all interactions; Total scores mainly distributes in 4–8, accounting for 78.2% of interactions. Figure 2D shows the sum of Total score of every target being docked by their ligands. CMA1, EGFR, phenylalanine-4-hydroxylase (PAH), SRC and F7 are of top 5 comprehensive effects of their ligands. Figure 3 shows the pose of the 16 ingredients and the docking position of their targets with the top Cscore and Total score.

Table 2. Potential Targets Information of DZXXI

UniProt ID	Protein name	Gene name	Related diseases or function indications
P84077	ADP-ribosylation factor 1	ARF1	GTP-binding protein
P78536	Disintegrin and metalloproteinase domain-containing protein 17	ADAM17	Inflammation, Rheumatoid arthritis, Cardiac failure, Multiple sclerosis
P63316	Troponin C, slow skeletal and cardiac muscles	TNNC1	Chronic heart failure (CHF)
P53779	Mitogen-activated protein kinase 10	MAPK10	Ischemic stroke, Neurological diseases
P53041	Serine/threonine-protein phosphatase 5	PPP5C	Involved in hydrolase activity
P49841	Glycogen synthase kinase-3 beta	GSK3B	Braininjury, Immunodeficiency, Ischemia
P45983	Mitogen-activated protein kinase 8	MAPK8	Neurodegenerative disease, Inflammation, Ischemia
P43155	Carnitine O-acetyltransferase	CRAT	Carnitineacetylase affects the transport of acetyl-CoA into mitochondria.
P42574	Caspase-3	CASP3	Venous thrombosis
P37231	Peroxisome proliferator-activated receptor gamma	PPARG	Inflammation, Ischemic heart disease
P36897	TGF-beta receptor type-1	TGFBR1	Glaucoma, Inflammation, Angiogenesis disorder, Cancer, Atherosclerosis
P35968	Vascular endothelial growth factor receptor 2	KDR	Vascular development
P35558	Phosphoenolpyruvatecarboxykinase, cytosolic [GTP]	PCK1	Energy production and conversion
P29474	Nitric oxide synthase, endothelial	NOS3	Cardiovascular diseases, Inflammation, Coronary artery disease, Angina
P28845	Corticosteroid 11-beta-dehydrogenase isozyme 1	HSD11B1	Rheumatoid arthritis, Diabetes mellitus, Atherosclerosis
P25774	Cathepsin S	CTSS	Psoriasis, Arteriosclerosis, Inflammation, Multiple sclerosis
P24941	Cyclin-dependent kinase 2	CDK2	Involved in protein kinase activity
P23946	Chymase	CMA1	Asthma, Thrombosis, Inflammation, Congestive heart failure, Chronic obstructive pulmonary disease infection, Cardiovascular disease
P23141	Liver carboxylesterase 1	CES1	Hypercholesterolemia, Atherosclerosis, Hyperlipidemia, Cardiovascular disease
P19971	Thymidine phosphorylase	TYMP	Angiogenesis disorder, Cancer
P18075	Bone morphogenetic protein 7	BMP7	Tooth disease, Motor neurone disease, Cerebrovascular ischemia
P14174	Macrophage migration inhibitory factor	MIF	Cancer, Inflammation, Myocardial infarction
P14061	Estradiol 17-beta-dehydrogenase 1	HSD17B1	Cardiovascular disease.
P12931	Proto-oncogene tyrosine-protein kinase Src	SRC	Hypercalcemia; Cancer, Osteoporosis; Cerebrovascular ischemia, Bone metastases, Metastasis, Solid tumor
P10275	Androgen receptor	AR	Cardiac failure, Reperfusion injury, Plasmodium infection, Atherosclerosis
P09211	Glutathione S-transferase P	GSTP1	Inflammation, Cancer
P08069	Insulin-like growth factor 1 receptor	IGF1R	Involved in protein kinase activity
P07858	Cathepsin B	CTSB	Myocardial infarction, Arthritis, Inflammation, Cancer
P07195	L-Lactate dehydrogenase B chain	LDHB	Parkinsons disease, Cardiovascular disease.
P02647	Apolipoprotein A-I	APOA1	Atherosclerosis
P00797	Renin	REN	Hypertension, Cardiovascular disease, Renal failure
P00749	Urokinase-type plasminogen activator	PLAU	Angiogenesis disorder, Restenosis, Skin ulcer, Cardiovascular disease
P00747	Plasminogen	PLG	Pulminary embolism, Coronary artery thrombosis
P00746	Complement factor D	CFD	Autoimmune disease, Reperfusion injury
P00742	Coagulation factor X	F10	Cardiovascular diseases, Coagulative disorders
P00734	Prothrombin	F2	Thrombosis, Myocardial infarction, Coronary atherosclerosis, Coagulative diseases
P00533	Epidermal growth factor receptor	EGFR	Arteriosclerosis, Ischemic heart disease, Hypercholesterolemia
P00439	Phenylalanine-4-hydroxylase	PAH	Cardiovascular diseases, Hypertension
Q04828	Aldo-keto reductase family 1 member C1	AKR1C1	Cardiovascular diseases
P03372	Estrogen receptor	ESR1	Inflammation, Myocardial infarction, Atherosclerosis, Cardiovascular disease
Q07869	Peroxisome proliferator-activated receptor alpha	PPARA	Hypertriglyceridemia, Hyperlipidemia, Cardiovascular disease, Inflammation
P08246	Leukocyte elastase	ELANE	Cardiovascular disease
P08235	Mineralocorticoid receptor	NR3C2	Metabolic disorder, Hypertension, Cardiovascular disease
P16109	P-Selectin	SELP	Inflammation, Myocardial infarction, Thrombosis
P08709	Coagulation factor VII	F7	Thromboembolism, Cardiovascular disease

Table 3. Top 20 Total Score of Molecular Docking and Their Corresponding Binding Free Energy

Compound	Target	C Score	Total score	Binding free energy (kcal/mol)
4,9-diCDOA	EGFR ↑	5	10.62	−10.53
4,5-diCQA	CFD ↓	4	10.55	−9.52
Scutellarin	EGFR ↑	5	10.24	−9.61
Erigoster B	CES1 ↓	5	9.73	−8.66
3,5-diCQA	CFD ↓	4	9.71	−9.36
4,9-diCDOA	MAPK8 ↓	4	9.69	−9.25
4,5-O-CQA	ADAM17	5	9.66	−8.9
4,5-O-CQA	EGFR ↑	4	9.38	−10.79
1,3-diCQA	PLAU ↓	5	9.33	−4.8
3,4-diCDOA	ADAM17	4	9.2	−9.04
A-7-O-G	EGFR ↑	4	9.1	−9.24
4,9-diCDOA	CFD ↓	5	9.01	−9.58
Erigoster B	PAH ↓	4	8.96	−4.53
3,4-diCDOA	CMA1	5	8.83	−8.53
9-CDOA	MAPK8 ↓	4	8.72	−7.24
3,5-diCQA	ADAM17	4	8.64	−8.02
4,5-diCQA	CMA1	4	8.63	−7.92
Scutellarin	CDK2 ↓	5	8.58	−8.52
4,5-diCQA	HSD17B1 ↑	5	8.57	−8.64
3-CDOA	MAPK10 ↓	4	8.56	−8.81

‘↑’ and ‘↓’ represent the predicted target could be activated and inactivated by the ingredient, respectively.

Fig. 2. (A) the Binding Free Energy and Total Score of Molecular Docking; (B) The Distribution of Binding Free Energy; (C) The Distribution of Total Score; (D)Sum of Total Scores of Targets Being Docked with Ingredients in DZXXI

Fig. 3. Ingredients and the Docking Position of Their Targets with the Top CScore and Total Score

(The yellow dotted line represents hydrogen bond.) (Color figure can be accessed in the online version.)

Drug-Target Network Analysis

Figure 4 shows the drug-target network of DZXXI, revealing its multi-component and multi-target therapy. It consists of 249 drug-target interactions connecting the 16 drugs to 45 targets, which means the average number of target proteins of per ingredients is 15.6.

Fig. 4. The Drug-Target Network

The red circles represent chemical constituents of DZXXI; the green circles represent indirect targets for drugs; The size of nodes are proportional the connective betweenness. (Color figure can be accessed in the online version.)

Target-Disease Network Analysis

Forty five potential targets related diseases were classified into 11 groups, like cardiovascular diseases, immune system diseases, energy production, angiogenesis disorder, autoimmune disease, and so forth. Accordingly, a target-disease network was constructed, containing 75 target-disease interactions (Fig. 5), and about half of the targets (24/45) are involved into multiple diseases, while other targets only are related to cardiovascular disease (21/45).

Fig. 5. The Target-Disease Network

The red circles represent gene name of targets; the green circles represent a disease or function for gene; the grey edges represent gene are involved in related disease or function. The size of nodes are proportional the connective betweenness. (Color figure can be accessed in the online version.)

Target-Pathway Network Analysis

Figure 6 shows target-pathway network consisting of 45 putative drug targets of DZXXI and related 50 KEGG pathways. As shown in the figure, most of the target proteins (35/45) act in a diversity of pathways reflecting the crucial and cross-talk roles that these targets may play in the interactions among different pathways. Similarly, most of the pathways are also modulated through multiple target proteins, including Focal adhesion, mitogen-activated protein kinase (MAPK) signaling pathway, complement and coagulation cascades, Wnt signaling pathway, vascular endothelial growth factor (VEGF) signaling pathway and Renin-angiotensin system. Noteworthily, several targets and pathways of DZXXI against MI have been found for the first time in our work. These findings are expected to help promote full clinical advantage of DZXXI.

Fig. 6. The Target-Pathway Network

The green circles represent gene name of targets; the yellow hexagons represent involved pathways; the grey edges represent gene are mapped in the pathway. The size of each node is proportional the connective degree. All pathways abbreviations are listed in Supplementary Table S2. (Color figure can be accessed in the online version.)

Discussion

In clinic in China, DZXXI (specification: 10 mL/ampoule; containing total scutellarin and caffeates, 0.2–0.3 mg/mL and 0.4–0.6 mg/mL, repectively) is often used to treat CVDs with the rountine regimen (20 mL/250 mL of 50g/L glucose injection, i.v. drip, qd). DZXXI mainly contains 15 ingredients which have been identified as listed in Table 1 in the present study. All the 15 ingredients could possess protective effect against myoblast damage due to their antioxidant, neuroprotective, anti-hyperlipidemic and anti-inflammatory effects. Some of them, like scutellarin and A-7-O-G, were proved to have anit-CVDs potentials.^42,43) Indeed, in our previous study, the 13 ingredients, including 5-CQA, 3-CQA, 4-CQA, 4-CDOA, caffeic acid, 1,3-diCQA, scutellarin, 3,4-diCQA, 3,5-diCQA, erigoster B, 3,4-diCDOA (or 4,9-diCDOA), A-7-O-G and 4,5-diCQA have anti-myocardial ischemia effect. Especially, 11 of them possessed favorable PK profiles (easily detected plasma drug concentrations and available AUCs)⁴⁴⁾ also suggesting that they could be cardiovascular active ingredients. But, as a mixture of these ingredients in DZXXI, it is still lack of the comprehensive knowledge about the effect mechanism of the preparation against CVDs. Even though exploring the complex mechanisms of multiple active ingredients underlying diseases is difficult, target interactions and their roles in corresponding diseases can be quantified in network model of systems pharmacology, which enable us study these connections more intuitively.

Recently, Zhao et al. reported the antiplatelet and neuroprotective mechanism of Dengzhan Xixin injection in the treatment of ischemic stroke using computational systems pharmacology.⁴⁵⁾ They utilized three databases and web servers-STITCH, TCMSP and SwissTargetPrediction to fish putative targets of fifteen DZXXI plasma absorbed compounds for ischemic stroke. Differently, we focused on MI, and we used PharmMapper, a pharmacophore mapping approach for potential drug target identification, to fish potential targets of 15 ingredients, which all were identified by reference products. In addition, Sybyl and Ledock, two molecular docking tools, were used to confirm the targets for the first time. These two complementary approaches could assure that the molecular docking result was more accurate.

Most of these targets predicted in our research are involved in vascular systems. For instances, Cathepsin B (CTSB) is a proteolytic enzyme potentially modulating angiogenic processes and extracellular matrix remodeling⁴⁶⁾; ESR1 mediates the vascular system to promote the functional recovery of vascular injury⁴⁷⁾; Cell-specific TGFBR1 deficiency of smooth muscle attenuates neointimal hyperplasia but promotes an undesired vascular phenotype for injured arteries⁴⁸⁾; F2, F7 and F10 are concerned with thrombosis⁴⁹⁾; PPARG, ADAM17 and GSK3B is relevant to ischemic and inflammation processes^50–52); NOS3 produces nitric oxide (NO) in endothelium, which mediates vasodilation.⁵³⁾ Accumulating evidence suggests that MAPKs (MAPK10 and MAPK8) are related to cardiac diseases, such as myocardial infarction (ischemia), hypertrophy and heart failure.⁵⁴⁾ CASP3 deletion promotes necrosis in atherosclerotic plaques of ApoE knockout mice.⁵⁵⁾ Macrophage migration inhibitory factor (MIF) has been defined as an important chemokine-like function (CLF) chemokine with an essential role in monocyte recruitment and arrest, and adhesion of monocytes to the vessel wall and their transendothelial migration are critical in atherogenesis and many other inflammatory diseases.⁵⁶⁾ Systemic infusion of insulin-like growth factor-1 (IGF-1) exerted anti-inflammatory and anti-oxidant effects and reduces atherosclerotic burden in apolipoprotein E (Apoe) deficient mice.⁵⁷⁾ Elevated oxTrp72-apoA1 levels in subjects presenting to a cardiology clinic were associated with increased cardiovascular disease risk.⁵⁸⁾ P-selectin (SELP) contributes to atherogenesis in humans.⁵⁹⁾ TNNC1 implicates for heart failure and cardiomyopathies.⁶⁰⁾ CMA1 is an ideal candidate for investigating the genetic predisposition to coronary heart disease.⁶¹⁾ BMP7 drives human adipogenic stem cells into metabolically active beige adipocytes.⁶²⁾ LDHB mediates the mitochondrial activity, which was reported to play crucial roles in the pathological progress of energy conversion.⁶³⁾

The topology parameters calculated from the drug-target network can provide us vital information of target-ligand interaction profiles. As shown in Fig. 4, in this network, every ingredient of DZXXI binds to over one target, which signifies polypharmacological phenomena.⁶⁴⁾ One drug binding to multi-targets may further exert the overall effectiveness on organisms, thereby promoting therapeutic effects, or may compete with each other and reduce the effectiveness. Top two highly connected nodes in drug-target network are Scutellarin (degree = 21, betweenness = 0.14) and caffeic acid (degree = 19, betweenness = 0.22). Scutellarin has high affinities with MAPK8, MAPK10, F2, F10, CMA1, which leads to a highly complex pharmacological profile including antioxidant, anti-inflammatory, etc.⁶⁵⁾ Erigoster B interacted with the targets including F2, F10, BMP7 involved in embolism, thrombosis⁴⁹⁾ and ischemia.⁶¹⁾ 4-CDOA, caffeic acid are predicted to mediate NOS3 to increase the nitric oxide biosynthesis, thereby possibly exerting blood vessel dilation, coordination of heart rhythm and regulation of cellular respiration activities.⁵³⁾ Reversely, the protective effect against myocardial infarction of Scutellarin⁶⁶⁾ was predicted to due to its interaction with NOS3, MAPK10 and ADAM17 simultaneously. Inflammatory response is of enormous significance in MI. Three active compounds of DZXXI, including 4-CQA, 5-CQA, and 9-CDOA, may have interactions with PPARG, PLG and MAPK10 which are expected to suppress inflammation associated with ischemia.^67,68) It is also obvious that most of the target proteins are cross-linked together in this network. Hence, enhancing pharmacological synergies may be existed among bioactive ingredients due to the fact that ingredients directed at cross-linked receptor target or physiological system.⁶⁹⁾ For example, caffeic acid (degree = 19, and betweenness = 0.22), as well as caffeoylquinic acids (CQAs) like 3,5-diCQA, 4,5-diCQA and 5-CQA, exert antibacterial, antimutagenic, antioxidant, anti-inflammatory and anti-carcinogenic effects through binding to multi-targets simultaneously. To sum up, the drug-target network analysis provides insights into the drug-target and therapeutic polypharmacology of ingredients.

Target-disease network shows that DZXXI mainly protects human against MI in 2 ways: treating cardiovascular diseases (degree = 34), and treating inflammation (degree = 12). Figure 5 shows that different diseases share common pathological proteins, indicating that different diseases could be cured by components combination.⁷⁰⁾ For example, CES1 has been shown to play a role in metabolic control. It not only can decrease plasma cholesterol and TG levels but also reduce the risk factors of atherosclerosis.⁷¹⁾ In this network, CES1 is a common target for the eleven compounds, including 1,3-diCQA, 3,4-diCDOA, 3-CDOA, A-7-O-G, Erigoster B, etc. to exert intervention effects against various diseases.

In the target-pathway network, focal adhesion kinase (FAK) is a broadly expressed tyrosine kinase implicated in cellular functions such as migration, growth and survival. Emerging data support a role for FAK in cardiac development, hypertrophy and failure.⁷²⁾ Members of the MAPK cascade such as extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK) and p38 are implicated as important regulators of cardiomyocyte hypertrophic growth in culture. Mitogen-activated protein extracellular kinase (MEK)1–ERK1/2 signaling pathway stimulates a physiologic hypertrophy response associated with augmented cardiac function and partial resistance to apoptosis.⁷³⁾ The coagulation serine proteinase cascades have been associated with functions of the immune and cardiovascular systems including venous thrombosis and atherosclerosis.⁷⁴⁾ The Wnt pathway plays multiple roles in regulating cellular behavior including proliferation, differentiation, cell migration, and cell polarity. Numerous studies have shown that the Wnt signaling pathway significantly participates in cardiac fibrosis pathogenesis. A plethora of recent data have shown that Wnt/β-catenin signaling played an important role in many stages of cardiovascular development including progenitor proliferation and myocyte differentiation.⁷⁵⁾ VEGF signaling pathway contributes to normal and pathological regulation of cardiac contractility.⁷⁶⁾ The recognition that VEGF signaling pathways are critical in physiological angiogenesis has led to the concept that these pathways are suitable targets for therapeutic angiogenesis, and the role of VEGF in the pathogenesis of cardiovascular diseases such as atherosclerosis, ischemic heart disease, pulmonary hypertension, and vascular restenosis is important.⁷⁷⁾ In addition, the Renin-angiotensin system is crucial for the pathophysiology of cardiac hypertrophy and failure.

Ultimately, to predict the possible mechanism of DZXXI against MI, an integrated ‘MI-related pathway’ is constructed, in which 45 drug targets with their closely connected proteins and involved pathways are shown in group and the intermediate interactions are removed in order to display the interactions more clearly. As seen in Fig. 7, 45 targets are involved in multi-pathways indicating that multi-components in DZXXI mainly cure MI by attenuating cardiac fibrosis, regulating cardiac contractility, preserving heart function in adverse cardiac remodeling and activating blood circulation and dilating blood vessels.

Fig. 7. MI-Related Pathway

Targets of DZXXI are marked in orange and pathways are marked in blue. Arrows represent activation, T-arrows represent inhibition. (Color figure can be accessed in the online version.)

In this work, we show a systems-level perspective of therapeutic mechanisms of DZXXI against MI, a case of treating diseases by targeting multi-targets with multi-ingredients in the herbal combination. However, this research has three limitations. First of all, due to the difficulty of data acquisition, the human signal network may not include all known protein–protein interactions with direction and/or mode. Secondly, as current tools for target prediction cannot always distinguish whether the target is activated or inhibited by the substance of interest, the assumption is reasonable, to some extent. Thirdly, the above findings are mainly based on dry experiments. More wet experiments as well as potential clinical applications should be conducted to verify these theoretical analyses.

Conclusion

In this paper, we explored the targets and pathways through which DZXXI exerts the therapeutic effects on CVDs using a systems pharmacology-based strategy. Forty five targets including CMA1, EGFR, PAH, SRC, F7, etc., and 50 involved KEGG pathways including Focal adhesion, MAPK signaling pathway, complement and coagulation cascades, Wnt signaling pathway, VEGF signaling pathway, Renin-angiotensin system, etc. were identified accounting for the therapeutic effect for DZXXI against MI. Conclusively, this study provides insights into estimating the comprehensive therapeutic effect and mechanism of DZXXI for MI.

Acknowledgments

The authors gratefully acknowledge the financial supports of the National Nature Science Foundation of China (81303298 and 81973558), the Fujian Provincial Natural Science Foundation (2018J01596 and 2016J01371), Joint Funds for the innovation of Science and Technology, Fujian province (2017Y9123), the Program for New Century Excellent Talents in Fujian Province University (2018), and Training program for excellent scientific research talents of young teachers in Fujian Province University (2017).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

References

• 1) Kaptoge S., Pennells L., De Bacguer D., et al., Lancet Glob Health, 7, e1332–e1345 (2019).
• 2) “GBD Compare.”: ‹http://vizhub.healthdata.org/gbd-compare›, cited 10 June, 2020.
• 3) Gao J., Chen G., He H., Liu C., Xiong X., Li J., Wang J., Front. Pharmacol., 8, 289 (2017).
• 4) Li H., Yang H., Liu H., Chai H., Wu D., Guangdong Med. J., 31, 1611–1613 (2010).
• 5) Li Y., Li J., Wang L., Li X., China Contin. Med. Educ., 8, 183–184 (2016).
• 6) Hu J., Yin J., Qiu Y., Wei Q., Ning F., Xu Y., Zhongguo Zhong Xi Yi Jie He Za Zhi, 24, 655–656 (2004).
• 7) Yang N., Kang T., Pang J., J. Cardiovasc. Surg. (Electron. Ed.), 7, 225 (2018).
• 8) Li C., Wang M., TCM Res., 28, 27–29 (2015).
• 9) Yang Q., Shenzhen J. Integr. Tradit. Chin. West. Med., 29, 107–108 (2019).
• 10) Xue J., Shi Y., Li C., Song H., J. Cell. Biochem., 120, 6431–6440 (2019).
• 11) Hong M., Li S., Wang N., Tan H. Y., Cheung F., Feng Y., Int. J. Mol. Sci., 18, 620 (2017).
• 12) Li H., Liu L., Liu C., Zhuang J., Zhou C., Yang J., Gao C., Liu G., Lv Q., Sun C., Med. Sci. Monit., 24, 5668–5688 (2018).
• 13) Gan D., Xu X., Chen D., Feng P., Xu Z., Med. Sci. Monit., 25, 5700–5716 (2019).
• 14) Zhu B., Zhang W., Lu Y., Hu S., Gao R., Sun Z., Chen X., Ma J., Guo S., Du S., Li P., Biomed. Pharmacother., 105, 159–166 (2018).
• 15) Zhao M., Chen Y., Wang C., Xiao W., Chen S., Zhang S., Yang L., Li Y., Front. Pharmacol., 9, 1448 (2019).
• 16) Cai H., Luo Y., Yan X., Ding P., Huang Y., Fang S., Zhang R., Chen Y., Guo Z., Fang J., Wang Q., Xu J., Sci. Rep., 8, 3104 (2018).
• 17) Li J., Zhao P., Li Y., Tian Y., Wang Y., Sci. Rep., 5, 15290 (2015).
• 18) Wang J. B., Cui H. R., Wang R. L., Zhang C. E., Niu M., Bai Z. F., Xu G. H., Li P. Y., Jiang W. Y., Han J. J., Ma X., Cai G. M., Li R. S., Zhang L. P., Xiao X. H., Sci Rep., 8, 5645 (2018).
• 19) Yu W., Li Z., Long F., Chen W., Geng Y., Xie Z., Yao M., Han B., Liu T., Sci. Rep., 7, 1189 (2017).
• 20) Wen J., Zou W., Wang R., Liu H., Yang Y., Li H., Wei S., Li R., Cai H., Wang J., Zhao Y., J. Ethnopharmacol., 238, 111880 (2019).
• 21) Liu J., Zhu J., Xue J., Qin Z., Shen F., Liu J., Chen X., Li X., Wu Z., Xiao W., Zheng C., Wang Y., Sci. Rep., 7, 16364 (2017).
• 22) Huber K. V. M., Superti-Furga G., Methods Mol. Biol., 1394, 211–218 (2016).
• 23) Liu X., Ouyang S., Yu B., Liu Y., Huang K., Gong J., Zheng S., Li Z., Li H., Jiang H., Nucleic Acids Res., 38 (Suppl. 2), W609–W614 (2010).
• 24) Wang X., Pan C., Gong J., Liu X., Li H., J. Chem. Inf. Model., 56, 1175–1183 (2016).
• 25) Wang X., Shen Y., Wang S., Li S., Zhang W., Liu X., Lai L., Pei J., Li H., Nucleic Acids Res., 45 (W1), W356–W360 (2017).
• 26) Bottomly D., Ryabinin P. A., Tyner J. W., Chang B. H., Loriaux M. M., Druker B. J., McWeeney S. K., Wilmot B., Genome Med., 5, 103 (2013).
• 27) Wang Z., Sun H., Yao X., Li D., Xu L., Li Y., Tian S., Hou T., Phys. Chem. Chem. Phys., 18, 12964–12975 (2016).
• 28) Xu Q., Fan Q., Olavi P., “Network Pharmacology and Traditional Chinese Medicine,” ed. by H. Sakagami, “Alternative Medicine,” InTech, 2012, pp. 277–298.
• 29) Azuaje F. J., Zhang L., Devaux Y., Wagner D. R., Sci. Rep., 1, 52 (2011).
• 30) Shannon P., Markiel A., Ozier O., Baliga N. S., Wang J. T., Ramage D., Amin N., Schwikowski B., Ideker T., Genome Res., 13, 2498–2504 (2003).
• 31) dos Santos M. D., Chen G. J., Almeida M. C., Soares D. M., de Souza G. E. P., Lopes N. P., Lantz R. C., Nat. Prod. Commun., 5, 733–740 (2010).
• 32) Fiamegos Y. C., Kastritis P. L., Exarchou V., Han H., Bonvin A. M. J. J., Vervoort J., Lewis K., Hamblin M. R., Tegos G. P., PLoS ONE, 6, e18127 (2011).
• 33) Basnet P., Matsushige K., Hase K., Kadota S., Namba T., Biol. Pharm. Bull., 19, 1479–1484 (1996).
• 34) Yu B. W., Li J. L., Guo B. B., Fan H. M., Zhao W. M., Wang H. Y., Acta Pharmacol. Sin., 37, 1413–1422 (2016).
• 35) Jeszka-Skowron M., Sentowska A., Pyrzyíska K., DePeña M. P., Eur. Food Res. Technol., 242, 1403–1409 (2016).
• 36) Zhao M. M., Wang H. Y., Yang B., Tao H., Food Chem., 120, 1138–1142 (2010).
• 37) Mikami Y., Yamazawa T., Life Sci., 139, 69–74 (2015).
• 38) Hou N., Liu N., Han J., Yan Y., Li J., Anticancer Drugs, 28, 59–65 (2017).
• 39) de Magalhães P. M., Dupont I., Hendrickx A., Joly A., Raas T., Dessy S., Sergent T., Schneider Y.-J., Food Chem., 134, 864–871 (2012).
• 40) Chiou W. F., Chen C. C., Wei B. L., Evid. Based Complement. Alternat. Med., 2011, 634502 (2011).
• 41) Barroso M., Barros L., Rodrigues M. A., Sousa M. J., Santos-Buelga C., Ferreira I. C. F. R., Ind. Crops Prod., 90, 49–55 (2016).
• 42) Tang H., Tang Y., Li N., Shi Q., Guo J., Shang E., Duan J., Pharmacol. Biochem. Behav., 118, 51–59 (2014).
• 43) Fialovaa S., Veizerova L., Nosalova V., Drabikova K., Tekelova D., Grancai D., Sotnikova R., Nat. Prod. Commun., 10, 937–940 (2015).
• 44) Huang J., Su Y., Yang C., Li S., Wu Y., Chen B., Lin X., Huang L., Yao H., Shi P., RSC Adv., 9, 25309–25317 (2019).
• 45) Zhao J., Lv C., Wu Q., Zeng H., Guo X., Yang J., Tian S., Zhang W., Pharmacol. Res., 147, 104365 (2019).
• 46) Noda S., Asano Y., Akamata K., Aozasa N., Taniguchi T., Takahashi T., Ichimura Y., Toyama T., Sumida H., Yanaba K., Tada Y., Sugaya M., Kadono T., Sato S., PLOS ONE, 7, e32272 (2012).
• 47) Burns K. A., Korach K. S., Arch. Toxicol., 86, 1491–1504 (2012).
• 48) Liao M., Yang P., Wang F., Berceli S. A., Ali Y. H., Chan K. L., Jiang Z., Physiol. Rep., 4, e13056 (2016).
• 49) Ten Cate H., Hackeng T. M., de Frutos P. G., Thromb. Haemost., 117, 1265–1271 (2017).
• 50) Villapol S., Cell. Mol. Neurobiol., 38, 121–132 (2018).
• 51) Qian M., Shen X. Q., Wang H. H., Cell. Mol. Neurobiol., 36, 471–482 (2016).
• 52) Takahashi-Yanaga F., Biochem. Pharmacol., 86, 191–199 (2013).
• 53) Groves J. T., Wang C. C., Curr. Opin. Chem. Biol., 4, 687–695 (2000).
• 54) Javadov S., Jang S., Agostini B., Pharmacol. Ther., 144, 202–225 (2014).
• 55) Grootaert M. O., Schrijvers D. M., Hermans M., Van Hoof V. O., De Meyer G. R., Martinet W., Oxid. Med. Cell. Longev., 2016, 3087469 (2016).
• 56) Tillmann S., Bernhagen J., Noels H., Front. Immunol., 4, 115 (2013).
• 57) Higashi Y., Sukhanov S., Shai S. Y., Danchuk S., Tang R., Snarski P., Li Z., Lobelle-Rich P., Wang M., Wang D., Yu H., Korthuis R., Delafontaine P., Circulation, 133, 2263–2278 (2016).
• 58) Huang Y., DiDonato J. A., Levison B. S., et al., Nat. Med., 20, 193–203 (2014).
• 59) Zhang N., Liu Z., Yao L., Mehta-D’souza P., McEver R. P., Arterioscler. Thromb. Vasc. Biol., 36, 1114–1121 (2016).
• 60) Li M. X., Hwang P. M., Gene, 571, 153–166 (2015).
• 61) Jin C. N., Ma H., Lin Y., Wang J. A., Xiang M. X., J. Zhejiang Univ. Sci. B, 12, 660–667 (2011).
• 62) Okla M., Ha J. H., Temel R. E., Chung S., Lipids, 50, 111–120 (2015).
• 63) Chen Y. J., Mahieu N. G., Huang X., Singh M., Crawford P. A., Johnson S. L., Gross R. W., Schaefer J., Patti G. J., Nat. Chem. Biol., 12, 937–943 (2016).
• 64) Boran A. D., Iyengar R., Curr. Opin. Drug Discov. Devel., 13, 297–309 (2010).
• 65) Boots A. W., Haenen G. R., Bast A., Eur. J. Pharmacol., 585, 325–337 (2008).
• 66) Wang L., Ma Q., Pharmacol. Ther., 190, 105–127 (2018).
• 67) Das R., Ganapathy S., Settle M., Plow E. F., Blood, 124, 679–688 (2014).
• 68) Tureyen K., Kapadia R., Bowen K. K., Satriotomo I., Liang J., Feinstein D. L., Vemuganti R., J. Neurochem., 101, 41–56 (2007).
• 69) Spinella M., Altern. Med. Rev., 7, 130–137 (2002).
• 70) Lin Q. C., Chin. J. Integr. Tradit. West. Med., 4, 242–246 (1998).
• 71) Xu J., Xu Y., Xu Y., Yin L., Zhang Y., Sci. Rep., 7, 17845 (2017).
• 72) Chu M., Iyengar R., Koshman Y. E., Kim T., Russell B., Martin J. L., Heroux A. L., Robia S. L., Samarel A. M., Cardiovasc. Res., 92, 409–419 (2011).
• 73) Bueno O. F., De Windt L. J., Tymitz K. M., Witt S. A., Kimball T. R., Klevitsky R., Hewett T. E., Jones S. P., Lefer D. J., Peng C. F., Kitsis R. N., Molkentin J. D., EMBO J., 19, 6341–6350 (2000).
• 74) Oikonomopoulou K., Ricklin D., Ward P. A., Lambris J. D., Semin. Immunopathol., 34, 151–165 (2012).
• 75) Tian Y., Cohen E. D., Morrisey E. E., Pediatr. Cardiol., 31, 342–348 (2010).
• 76) Rottbauer W., Just S., Wessels G., Trano N., Most P., Katus H. A., Fishman M. C., Genes Dev., 19, 1624–1634 (2005).
• 77) Rajalakshmi D. C., Gopalakrishnan A. R. K., Kartha C. C., “VEGF Signaling: a therapeutic target for cardiovascular disease,” ed. by Srivastava A. K., Anand-Srivastava M. B., “Signal transduction in the cardiovascular system in health and disease,” Springer, U.S.A., 2008, pp. 301–326.