1School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China

1. 2Guangdong Provincial Key Laboratory of Chinese Medicine Pharmaceutics, Guangzhou, China
2. 3Guangdong Provincial Engineering Laboratory of Chinese Medicine Preparation Technology, Guangzhou, China Correspondence should be addressed to Xiao-Mei Tan; [email protected] Received 17 January 2020; Revised 30 April 2020; Accepted 6 May 2020; Published 25 May 2020 Academic Editor: Olumayokun A. Olajide

Copyright © 2020 Xiao Liang et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

e decoction is an important dosage form of traditional Chinese medicine (TCM) administration. e Mahuang Fuzi Xixin decoction (MFXD) is widely used to treat allergic rhinitis (AR) in China. However, its active compounds and therapeutic mechanisms are unclear. e aim of this study was to establish an integrative method to identify the bioactive compounds and reveal the mechanisms of action of MFXD. LC-MS/MS was used to identify the compounds in MFXD, followed by screening for oral bioavailability. TCMSP, BindingDB, STRING, DAVID, and KEGG databases and algorithms were used to gather information. Cytoscape was used to visualize the networks. Twenty-four bioactive compounds were identified, and thirty-seven predicted targets of these compounds were associated with AR. DAVID analysis suggested that these compounds exert their therapeutic effects by modulating the Fc epsilon RI, B-cell receptor, Toll-like receptor, TNF, NF-κB, and T-cell receptor signaling pathways. e PI3K/AKT and cAMP signaling pathways were also implicated. Ten of the identified compounds, quercetin, pseudoephedrine, ephedrine, β-asarone, methylephedrine, α-linolenic acid, cathine, ferulic acid, nardosinone, and higenamine, seemed to account for most of the beneficial effects of MFXD in AR. is study showed that LC-MS/MS followed by network pharmacology analysis is useful to elucidate the complex mechanisms of action of TCM formulas.

Allergic rhinitis (AR), an immunoglobulin E- (IgE-) mediated inflammatory disease, seriously impairs the quality of life of patients [1]. Epidemiological surveys show that AR affects more than 20% of the world’s population and its incidence has progressively increased in developing countries [2]. Currently, treatments based on Western medicine alleviate the symptoms of AR but do not cure. Once drug treatments stop, the disease relapses [3]. Traditional Chinese medicine (TCM) has long been used as effective therapeutic interventions in Asia, particularly China [4]. TCM is primarily based on the use of compound formulas, utilizing a combination of herbs or their extracts for improved efficacy.

Research into the active and effective compounds of TCM promotes the development and design of new therapeutic drugs.

e decoction is the main form of TCM administration. e identification of the chemical components in a decoction underpins research on the mechanism of action of TCM. Mahuang Fuzi Xixin decoction (MFXD), a classical Chinese herbal formula, is widely used to treat AR. MFXD consists of Mahuang (Herba Ephedrae), Fuzi (Radix aconiti lateralis praeparata), and Xixin (Radix et Rhizoma Asari), which are boiled in water at specific proportions before administration. Pharmacological studies have shown that MFXD exerts anti-inflammatory and antiallergic effects by preventing the release of mediators from macrophages and

mast cells and inhibiting the production of interferon gamma and interleukin (IL)-4 [5, 6]. MFXD also suppresses 2 cytokine production and regulates the balance of 1 and 2 responses [7]. Using collected compounds and targets from databases and references, Tang constructed a network that shows the interactions of components and targets and verified the pharmacological activity of four components (salsolinol, pseudoephedrine, dibutyl phthalate, and herbacetin) of MFXD in vitro [8]. However, to our knowledge, there have been no comprehensive studies of the compounds in the decoction prepared from MFXD and their potential therapeutic mechanism in AR.

Network pharmacology is a new field that integrates pharmacological information, omics, and systems biology [9]. It is based on the concept that targeting multiple nodes in interconnected systems, rather than individual nodes, generates information for the identification of a drug with better efficacy and fewer adverse effects. TCM formulas commonly comprise a mixture of several herbs and ingredients that deliver synergistic effects by targeting multiple targets and pathways and modulating the links between pathways. us, network pharmacology, which elucidates interactions between multiple compounds and targets, is particularly suited for investigating the activities of TCM [10, 11]. Currently, drug discovery is rapidly evolving towards systematic and multipharmacological approaches to address the poor efficacy, loss of efficacy, and development of drug resistance when single compounds are used to target single therapeutic targets [12, 13]. Research is now focused on the simultaneous investigation of targets in the context of entire biological networks [14]. is network pharmacology approach to drug discovery and design may identify effective drugs from herbal medicines.

erefore, the aim of this study was to identify the active compounds in the decoction prepared from MFXD to investigate their potential synergistic effects on cellular signaling pathways that are associated with AR. In this study, the combinatorial approach of LC-MS/MS followed by network pharmacology analysis was established to identify the active chemical compounds in the decoction prepared from MFXD and their potential pharmacological mechanism in AR. is research considers bioactive compounds, protein targets, protein-protein interactions, genes, and signaling pathways to elucidate the mechanisms responsible for the curative effects of MFXD in AR.

1. 2.1. Materials. HPLC-grade acetonitrile and methanol were procured from Merck (Darmstadt, Germany). Formic acid was procured from Sigma-Aldrich (MO, USA). Other chemicals were of analytical grade.

e decocting pieces of Mahuang, Fuzi, and Xixin were purchased from Kangmei Pharmaceutical Co., Ltd. (Guangzhou, China) and met the standards of the Chinese Pharmacopeia. Voucher specimens (no. 20191117) were deposited at the authors’ laboratory at Southern Medical University.

2.2. Preparation of MFXD. MFXD consisted of Mahuang, Fuzi, and Xixin at a weight ratio of 2:31. Mahuang was immersed in distilled water (15 times the total weight) for 30min and boiled for 20min. Fuzi and Xixin were then added to the suspension, which was simmered for another 90min. e liquid extract obtained was concentrated to

1. 1.52 g/mL under reduced pressure.
2. 2.3. LC-MS/MS Analyses. LC-MS/MS was carried out using a UPLC-Orbitrap-HRMS platform ( ermo Fisher) with a Waters ACQUITY UPLC HSS T3 Column (100 mm×2.1mm, 1.8μm). e mobile phases consisted of acetonitrile (A) and 0.1% aqueous formic acid (v/v) (B) using a gradient elution of 0% A at 0−1min, 0−20% A at 1−2min, 20−50% A at 2−12min, 50−95% A at 12−15min, and 95−100% A at 15−20min. e flow rate was set at 0.4mL/ min, and the column temperature was maintained at 30°C. e injection volume was 5μL. e electrospray ionization source was set to positive and negative modes. e mass range scanned was 100−1000m/z. MS data were collected with ermo Xcalibur software (version 4.0).

1. 2.4. Construction of the Chemical Ingredient Database. We processed the chromatogram by matching the data with an in-house Orbitrap Traditional Chinese Medicine Library (OTCML), which provided a list of compounds with their formulas and MS/MS fragment modes. Based on errors less than 5ppm and MS/MS fragment matching, we identified compounds in decoction prepared from MFXD.
2. 2.5. Oral Bioavailability (OB) Screening. OB is an important pharmacokinetic property used to assess the rate and percentage of an orally administered drug that has been absorbed into the blood circulation to produce pharmacological effects. Important parameters of the identified compounds, such as Caco-2 cell permeability, human intestinal absorption (HIA), and OB limits F (F-20%, F-30%), were screened using ADMETlab, a platform for systematic ADMET evaluation based on a comprehensively collected ADMET database [15].
3. 2.6. Potential AR-Associated Targets of the Compounds in MFXD. Predicting whether a compound interacts with intended targets is a critical phase of drug discovery [16]. e targets of compounds in MFXD were obtained from the Traditional Chinese Medicine System Pharmacology (TCMSP, http://lsp.nwu.edu.cn/tcmsp.php) and BindingDB (https://www.bindingdb.org/bind/index.jsp) databases. After the deletion of redundant hits, the remaining protein targets were standardized by their ID and gene symbols in the UniProtKB database (http://www.uniprot.org/) [17].

Information on AR-associated targets was retrieved from the therapeutic target disease (http://db.idrblab.net/ttd/), DrugBank (https://www.drugbank.ca), and DisGeNET (https://www.disgenet.org/) databases and standardized using the UniProtKB database.

RT : 0.00–20.00

100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10

1. 2
2. 3
3. 4
4. 5
5. 6
6. 7
7. 8
8. 9

Relative abundance

18 17

15 14

1. 12
2. 13

10

19

16

20

21

24

11

22 23

1

5 0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Time (min)

Figure 1: Total ion current chromatogram of LC-MS/MS of MFXD.

1. 2.7. Construction of the Target Protein-Protein Interaction (PPI) Network. To determine whether the therapeutic targets of MFXD are associated with AR, we intersected the drug targets and disease targets to obtain the common targets, which were considered potential therapeutic targets.
2. 2.8.TargetPathwayandEnrichmentAnalysis. To analyze the gene ontology (GO) functional annotations and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment of genes and their roles in signal transduction, the Database for Annotation, Visualization, and Integrated Discovery (DAVID 6.8) was employed. DAVID is an online analytical program that provides a comprehensive set of functional annotation tools to explore the biological functions of genes from gene lists [18]. DAVID can identify and describe the biological processes, cellular components, molecular functions, and pathways that are associated with genes of interest.
3. 2.9. Construction of Compound-Target-Pathway Networks. To understand the underlying mechanism of MFXD in AR, a large-scale combinatorial network was established by integrating data obtained on drug information, drug-target interaction, and target-related pathway interactions. rough the hierarchical network, we characterized the relationships and pathways targeted by MFXD in AR.

1. 3.1. Analysis of Chemical Compounds from MFXD. e total ion chromatogram of MFXD is shown in Figure 1. is was matched with the OTCML database, jointly constructed by ermo Scientific and Tsinghua University as a reference for compound identification. Using electrostatic field orbital trap high-resolution MS for fragment MS acquisition, more than 1,200 reference compounds of TCM have been collected in this database. More than 7,000 second-order MS images of the highest quality are available, enabling the rapid and accurate characterization of TCM components and natural products. A total of 24 compounds in MFXD were identified through chromatogram matching. Table 1 shows retention time, experimental and calculated m/z values, molecular formulas, errors in parts per million (ppm), and the major MS/MS fragments. e mass error of all identified compounds was less than 5ppm.
2. 3.2. Construction and Analysis of the Component-Target Interaction Network. e TCMSP database contains 499 Chinese herbs with 19,384 compounds, 3,311 targets, and 837 associated diseases [20]. BindingDB is an experimental protein-small molecule database for virtual compound screening based on maximal chemical similarity and support

Table 1: Results of LC-MS/MS analysis of MFXD. No. Retention time eoretical m/z Measured m/z Error ppm Molecular formula Major MS/MS fragments Compound

1. 1 3.78 170.0215 170.0211 −2.3526 C7H6O5
2. 2 4.02 151.0997 151.0992 −3.3091 C9H13NO
3. 3 4.03 165.1154 165.1158 2.4226 C10H15NO
4. 4 4.11 271.1208 271.1199 3.3196 C16H17NO3
5. 5 4.12 165.1154 165.1149 −3.0282 C10H15NO
6. 6 4.30 179.1310 179.1316 3.3495 C11H17NO
7. 7 4.48 594.1585 594.1578 −1.1781 C27H30O15
8. 8 4.75 290.0790 290.0796 2.0684 C15H14O6
9. 9 5.25 578.1636 578.1623 −2.2485 C27H30O14
10. 10 5.42 432.1057 432.1051 −1.3886 C21H20O10
11. 11 5.57 164.0474 164.0482 4.8766 C9H8O3
12. 12 5.97 302.0427. 302.0492 2.1520 C15H10O7
13. 13 6.33 272.0685 272.0679 2.2053 C15H12O5
14. 14 6.35 162.0317 162.0312 −3.0858 C9H6O3
15. 15 6.98 589.2887 589.2889 3.3939 C31H43NO10
16. 16 7.67 218.1671 218.1665 −2.7502 C15 H22 O
17. 17 7.83 146.0368 146.0365 −2.0543 C9H6O2

119.0488 103.0538 Coumarin

91.0538

Table 1: Continued. No. Retention time eoretical m/z Measured m/z Error ppm Molecular formula Major MS/MS fragments Compound

1. 18 8.00 194.0579 194.0581 1.0306 C10H10O4
2. 19 10.28 615.3044 615.3030 2.2753 C33H45NO10
3. 20 11.23 629.3200 629.3193 −1.1123 C34H47NO10
4. 21 11.83 232.1463 232.1457 −2.5846 C15H20O2
5. 22 12.84 250.1569 250.1560 −3.5977 C15H22O3
6. 23 13.88 208.1099 208.1094 −2.4026 C12H16O3
7. 24 15.94 278.2246 278.2241 −1.7971 C18H30O2

109.1009 95.0852 Linolenic acid 81.0696

Table 2: ADME profile of the compounds from MFXD.

Compounds Caco-2 HIA F20 F30 Gallic acid −5.767 0.431 0.673 0.616 Cathine −4.783 0.877 0.851 0.869 Ephedrine −4.956 0.902 0.853 0.879 Higenamine −5.293 0.538 0.226 0.386 Pseudoephedrine −4.956 0.902 0.853 0.879 Methylephedrine −4.477 0.809 0.701 0.707 Vicenin-2 −6.624 0.226 0.447 0.253 Catechin −6.495 0.400 0.488 0.404 Kaempferitrin −6.364 0.399 0.584 0.325 Vitexin −6.317 0.263 0.494 0.287 p-Coumaric acid −4.892 0.745 0.699 0.536 Quercitrin −6.469 0.155 0.515 0.242 Naringenin chalcone −5.198 0.472 0.596 0.519 Umbelliferone 4.601 0.797 0.534 0.441 Benzoylmesaconine −6.087 0.301 0.332 0.41 Germacrone −4.378 0.768 0.661 0.625 Coumarin −4.142 0.848 0.288 0.257 Ferulic acid −4.943 0.635 0.680 0.509 Hypaconitine −5.513 0.343 0.265 0.350 Deoxyaconitine −5.469 0.349 0.261 0.355 Atractylenolide II −4.35 0.866 0.564 0.549 Nardosinone −4.353 0.826 0.633 0.585 β-Asarone −4.379 0.763 0.722 0.681 Linolenic acid −4.729 0.805 0.568 0.396

vector machine methods [21]. We screened the targets from the TCMSP and BindingDB databases for the 24 active compounds. A total of 198 targets were obtained after the deletion of redundant hits (113 from TCMSP and 138 from BindingDB).

To visualize the relationship between compounds and their targets, we constructed a compound-target interaction network (Figure 2). is network had a density of 0.022 with a characteristic path length of 3.396 and an average number of 4.918 neighbors. In the compound-target network, multiple compounds could act on the same target protein and a single compound could be associated with multiple target proteins. For instance, prostaglandin G/H synthase 2 (PTGS2-P35354) and sodium-dependent noradrenaline transporter (SLC6A2-P23975) represented the hub nodes with a high degree of distribution, whereas peripheral nodes, such as gallic acid, interacted with the progesterone receptor (PGR-P06401) and represented a lower degree of distribution. ese results were consistent with the common characteristics of TCM, in which multiple compounds and multiple targets are observed.

3.3. Construction and Analysis of the Target PPI Network. rough the comparative analysis of 198 component targets and AR-related disease targets, we identified 37 common potential targets for MFXD in AR (Table S1). e STRING database is a commonly used tool for predicting PPI and producing integrated and objective association networks [22]. Using the common protein targets as input for network visualization, a diversified PPI network was created (Figure 3(a)). e network systematically summarized the interactions of MFXD targets associated with AR treatment. PSIP1 was not analyzed in the PPI network, as it does not interact with other proteins. e network shows viable protein target nodes (n 36) connected by edges (n 135)

P34969

P35462 P14416

O95622

P28233

P28335

Q99720

P21731

P08908 P11229

Q05940

P28223

P21728

P08235 P80365

P01100

1. P13945
2. Q15822

Q16445 P04150

P31645

P23975

Q04760

Hypaconitine

Deoxyaconitine

P35368 Q16853

P34903 Nardosinone

P20309

P36544

P08172 P17538

Germacrone

Q01959

P47869

Higenamine

P25100

Q96RJ0

P10275

1. P18089

P08588

Methylephedrine

Cathine

P42262

Q92731 P03372

P05023

β−Asarone

P14867

Q9Y2D0

Benzoylmesaconine

P09960 Pseudoephedrine

Atractylenolide II

Q16236

O60218

P07550 P53985

P51843

P35348

P23280

Ephedrine

P22303 Q9H4B7

Q9UNQ0

P11511

p−Coumaric acid

Q14524

P18825

P35218

1. P08709
2. Q8N1Q1

1. P9BV79
2. Q9UNA4

P56817

P08183

Ferulic acid P08913

Umbelliferone

P27487

P21397

P03956

P12821

P35968

Q16678

P30542

P27338

Q9NZK7

Naringenin chalcone P00915

Q9Y253

P35354

Catechin hydrate

Q9ULX7

P00742

Q9NPD5

P06401

P18031

1. P04746 P29474
2. Q16790P00533P00918

Q96BR5

1. P12931
2. Q15596

P23219

P15121

P36269 Q13164

P05067

P00749

P43166

P08238

1. P11388
2. Q13085

P15090 O75469

P08581

P09917

O43570

O75475 Q9Y259

P17612

P35869 P60709

Coumarin

P07477

Q12899

Q9Y6L6 Q86SK9

P05177

Q9H8P0 P37231

P19793

Kaempferitrin

P14679

P02768

P24941

α−Linolenic acid

P48506

Quercetin

P07339

P38936

P22748

Q9UM73

P22301

P48443 P11387

P55851

Q96RR4

1. P09923
2. Q9NPH5

P05121

P04626

P0DMS8 P49763

Gallic acid

P20823

1. P14920
2. Q91819

1. P50750
2. Q04828

O76074

Q01469 P07148

P35236

P08684

P33527

P26358

P47989

Vicenin−2

P49841

P55211

Q5NUL3

O14842

P10721

P09211

P60568

Q8NER1

P11509

P01106

P12004

P11413

Q11130

P51812

Q14534

O14977

Q04206

P31639

P42574

P48736 P10620

P01375

P05231

P09601

P49327

P42330

Q9ULA0

P02766

Q9HB55

with an average node degree of 7.5 and average local clustering coefficient of 0.548. e PPI enrichment p value was less than 1.0 e−16; thus, proteins have more interactions among themselves than would be expected for a random set of proteins of similar size drawn from the genome. Such a significant enrichment indicated that the proteins are at least partially biologically connected as a group.

e genes ranked by the MCC algorithm were selected using the cytoHubba plugin. e predicted top 10 contributing hub genes were TNF, PPARG, ALB, PTGS2, IL-10, ACTB, NR3C1, ACE, SERPINE1, and AHR, which were considered crucial targets of MFXD against AR.

Two significant modules were selected from the PPI network by MCODE (Figures 3(b) and 3(c)). For module 1, the genes were significantly enriched in the nuclear factorκB (NF-κB) signaling pathway and tumor necrosis factor (TNF) signaling pathway. Cluster 2 genes were significantly enriched in the T-cell receptor signaling pathway. ese data suggested that MFXD likely acts on AR through all three signaling pathways.

1. 3.4. DAVID Pathway Analysis. To further investigate the multiple mechanisms of MFXD at a systematic level, 37 common genes were uploaded into DAVID 6.8. Functional

association clustering analysis discovered 14 annotation clusters, and the highest enrichment score was 4.07. e top 20 terms in molecular function, cellular component, and biological process are presented in Figures 4(a)–4(c), respectively.

GO enrichment analysis showed that the target genes were expressed in the plasma membrane, cell surface, cell membrane, and other cell compartments. At the molecular level, the target genes were involved in drug binding, enzyme binding, G-protein acetylcholine receptor activity, and steroid binging. At the cellular level, they were related to cell apoptosis, proliferation, and migration. Moreover, biological processes were also enriched with inflammation-related terms. AR is a chronic inflammatory disease that involves the release of various inflammatory mediators [1]. erefore, the inhibition of the inflammatory response is a therapeutic strategy for AR.

KEGG pathway annotation showed that 31 of the 37 (86.1%) potential target genes were enriched and involved in 153 pathways associated with the immune system, cardiovascular system, cancer, inflammation, and diabetes. As AR is an allergic disease, pathways that are associated with the immune system and inflammation were selected (Figure 4(d)). For example, the Fc epsilon RI, B-cell receptor, T-cell receptor, and Toll-like receptor signaling pathways are

CHRM5

PIK3CG CHRM3 CHRM1

PSIP1 CHRM2

SLC6A3 TBXA2R

PTGS1 ADRA2C

ALOX5

IL10 SLC6A2 SLC6A4

ACTB

GSK3B ADRB1 PTGS2

ACE

TNF

ABCB1 PPARG

ENSG00000196689 HSD11B2 NR3C1

ALB SLCO1B1 DPP4

SERPINE1 HSP90AB1

HSD11B1 AHR NR3C2

PRSS1

SHBG CYP3A4 PGR

(a)

ALOX5

TNF

PPARG

PTGS2

ALB

ABCB1

CYP3A4

(b)

IL10

ACE

AHR

ACTB

DPP4

GSK3B

SERPINE1

NR3C1

(c)

1. Figure 3: e protein-protein interaction (PPI) network and significant module constructed by the STRING database and Cytoscape 3.7.2. Node, target proteins; lines, interactions between proteins. (a) PPI of the common targets related to AR interacting with MFXD molecules. (b, c) Significant module selected from PPI.

involved in the regulation of the immune system and the TNF and NF-κB signaling pathways are involved in inflammation. We also identified the PI3K/AKT, cAMP, and AMPK signaling pathways in this analysis. e PI3K/AKT signaling pathway is important for cell growth, differentiation, metabolism, survival, and apoptosis [23]. Recent research has revealed that the PI3K-AKT signaling pathway regulates mast cell activity and is modulated in AR [24, 25]. Moreover, the central and peripheral regulation of energy homeostasis relies on the cAMP signaling pathway [26]. KEGG pathway analysis showed that the active components of MFXD act on the gene nodes within the PI3K/AKT and cAMP signaling pathways (Figure 5). ese data are consistent with the multiple effects that TCM has on different signaling and cellular pathways [27].

Comparative analysis of these KEGG pathways revealed that, among the 37 potential target genes identified, genes that were repeatedly associated with these pathways were PIK3CG, TNF, PTGS2, CHRM2, CHRM1, GSK3B, ADRB1, CHRM5, CHRM3, and HSP90AB1.

1. 3.5. Compound-Target-Pathway Network Analysis of MFXD. Chinese herbal formulas can contain dozens, or even hundreds, of compounds, and each compound can act on one or multiple targets to exert synergistic therapeutic effects [28]. To directly explore potential synergistic relationships, a compound-target-pathway combination network was constructed

(Figure 6). is network revealed that MFXD contains multiple compounds that have multiple targets within multiple pathways that are associated with AR treatment. For example, quercetin acted on 14 targets (such as P48736, P22301, Q75475, and P08684) and gallic acid acted on three targets (P23219, P35354, and P48736). P48736 was associated with both quercetin and gallic acid. Using network topological analysis, the top 10 compounds that may make major contributions to AR treatment are presented in descending order: quercetin, pseudoephedrine, ephedrine, β-asarone, methylephedrine, α-linolenic acid, cathine, ferulic acid, nardosinone, and higenamine. Quercetin can stimulate the immune system, inhibit the release of histamine, decrease the production of proinflammatory cytokines, increase the synthesis of leukotrienes,restrainthe formationofIgE antibody,and improvethe 1/ 2 balance by participating in multiple signal pathways [29]. All these mechanisms of action contribute to the antiinflammatory and immunomodulatoryproperties of quercetin, which can be effectively utilized in the treatment of AR, as shown in an animal model [30]. Gallic acid alleviates nasal inflammation via the activation of 1 and inhibition of 2 and 17 cells and has immunomodulatory effects [31]. α-Linolenic acid dampens AR through the eosinophilic production of 15-hydroxyeicosapentaenoic acid [32], and ephedrine has been used as a nasal wash for AR treatment [33].

Most of the compounds, such as kaempferitrin and pseudoephedrine, interact with PTGS2 (P35354). PTGS2 is responsible for the production of inflammatory

Target count

2 3 4 5 6 7 8

Drug binding Steroid binding

G-Protein coupled acetylcholine receptor activity Monoamine transmembrane transporter activity Steroid hormone receptor activity

Protease binding Phosphatidylinositol phospholipase C activity

Dopamine: sodium symporter activity Prostaglandin-endoperoxide synthase activity

11-β-hydroxysteroid dehydrogenase activity Alpha-2A adrenergic receptor binding

Epinephrine binding Identical protein binding

Nitric-oxide synthase binding Enzyme binding

Heme binding Receptor binding

Neurotransmitter: sodium symporter activity Activating transcription factor binding Peroxidase activity

0 2 4 6 8 10

–Log10 (P-value)

(a)

Target count

2 3 4 5 6

7 8

Adenylate cyclase-inhibiting GCAR signaling pathway

Phospholipase C-activating GCAR signaling pathway

Response to drug G-protein coupled acetylcholine receptor signaling pathway

Synaptic transmission, cholinergic Monoamine transport

Positive regulation of nitric oxide biosynthetic process

Regulation of blood pressure Transcription initiation from RNA polymerase II promoter

Lipid metabolic process

Negative regulation of synaptic transmission, dopaminergic Steroid hormone mediated signaling pathway

Positive regulation of protein binding Response to glucocorticoid Regulation of vascular smooth muscle contraction G-protein coupled receptor signaling pathway Positive regulation of fever generation Dopamine uptake involved in synaptic transmission

Inflammatory response

Cellular oxidant detoxification

2 3 4 5 6

–Log10 (P-value)

(c)

Target count

0 2 4 6 8 10 12 14 16 18 20

Integral component of plasma membrane Plasma membrane

Asymmetric synapse Axon terminus Membrane ra

Postsynaptic membrane Cell surface

Synapse Aryl hydrocarbon receptor complex

Membrane Neuron projection

Cell junction Organelle membrane Neuronal cell body Endoplasmic reticulum membrane Dendrite Blood microparticle Protein complex Basolateral plasma membrane Extracellular exosome

1 2 3 4 5

–Log10 (P-value)

Target count

2 4 6 8 10

Neuroactive ligand-receptor interaction Calcium signaling pathway Cholinergic synapse

Serotonergic synapse Steroid hormone biosynthesis Arachidonic acid metabolism

cAMP signaling pathway P13K-Akt signaling pathway

T cell receptor signaling pathway

TNF signaling pathway yroid hormone signaling pathway

NOD-like receptor signaling pathway mTOR signaling pathway Fc epsilon RI signaling pathway B cell receptor signaling pathway NF-kappa B signaling pathway HIF-1 signaling pathway

Toll-like receptor signaling pathway Natural killer cell mediated cytotoxicity

AMPK signaling pathway

1 2 3 4 5 6 7

–Log10 (P-value)

(d)

1. Figure 4: GO enrichment and KEGG pathway analysis for targets of MFXD constructed by DAVID 6.8. e order of importance was ranked from the top to bottom by −Log10 (P value) with bar charts. e number of targets sticks into each term with line charts. Go analysis was done under the categories of (a) molecular function, (b) cellular component, (c) biological process, and (d) KEGG enrichment pathway. GCAR stands for G-protein-coupled acetylcholine receptor.

Raf-1

GF RTK Grb2 SOS Ras

+p

Toll-like receptor signaling pathway

Pathogen-associated molecular patterns (PAMPs)

TLR24

IRS1 Rac1

B-cell receptor signaling pathway

PP2A CTMP

+p

GSK3

Antigen

BCR Syk

+p

P13K CD19

PDK1

Class IA

JAK/STAT signaling pathway

BCAP

p

–p

p

PIP3 –p

CylokineR JAk

Cylokine

AKT

+p

Focal adhesion

+p –p

Class IB FAK

HSP90 ECM Cdc37

1. ITGA
2. ITGB P13K PTEN

FOXO

+p

14-3-3

mTORC2

MAGI

PHLPP

+p +p

Chemokine signaling pathway Chemokines, Hormones, Neurotransmitters

BAD Casp9 CREB RXRa NUR77

TCL1

GPCR Gβy

p

(a)

cAMP signaling pathway

ABCC4

HCO3

Ca2+

GPCR

AC

cAMP

Rap1 signaling pathway

p

p

+p

+p

+p

GPCR

p

ϚAMP

(b)

Figure 5: DAVID pathway analysis. (a) PI3K/Akt signaling pathway. (b) cAMP signaling pathway. e stars indicate the targets where the molecules interact.

yroid hormone signaling pathway

NOD-like receptor signaling pathway

TNF signaling pathway

B-cell receptor signaling pathway

T-cell receptor signaling pathway

P12821

P28845

P08238

P37231 P07477

P80365

P04150

P01375

Fc epsilon RI signaling pathway

Arachidonic acid metabolism P08235

P08912

P35869

P04278

Q9Y6L6

Steroid hormone biosynthesis

O75475

mTOR signaling pathway

p-Coumaric acid

Cathine

Methylephedrine

Nardosinone

P49841

P21731

β-Asarone

Atractylenolide II

Kaempferitrin

Ferulic acid

P22301

Gallic acid P08183

Naringenin chalcone Hypaconitine

quercetin

cAMP signaling pathway

NF-kappa B signaling pathway

P23219

Catechin hydrate

Germacrone

Vitexin

Deoxyaconitine

P05121

α-Linolenic acid

Vicenin-2 Higenamine

Umbelliferone

P35354

P60709

Coumarin

Ephedrine Benzoylmesaconine

Pseudoephedrine

P06401

Serotonergic synapse

HIF-1 signaling pathway

P02768

P48736

Q8NER1

P20309

P09917

P11229

Toll-like receptor signaling pathway

PI3K-Akt signaling pathway

P08684

P08588

P31645

P18825 P27487

P08172 Natural killer cell-mediated cytotoxicity

P23975 Q01959

Cholinergic synapse

Calcium signaling pathway

AMPK signaling pathway

Neuroactive ligand-receptor interaction

Figure 6: e pharmacological network (yellow diamonds, compounds; red ellipse, targets; blue triangle, pathway).

prostaglandin E2, which inhibits T regulatory cell differentiation to induce AR-related inflammation [34]. We also discovered compounds that acted together on the targets P01375 and P48736 to regulate the Fc epsilon RI signaling pathway. Gallic acid was observed to interact with PIK3CG (P48736), a key protein that activates the B-cell receptor, T-cell receptor, and Toll-like receptor signal pathways. Finally, targets such as P48736, P01375, P48736, P01375, and P49841, which may provide protection against AR, interacted with most of the compounds.

Unlike Western medicines, TCMs are commonly prescribed as herbal formulas that contain a mixture of herbs. Each herb may contain many active ingredients that have single or multiple targets; thus, it is difficult to pinpoint the mechanisms of action of TCM. Network pharmacology shares the core concepts of the holistic philosophy of TCM and meets the requirements to treat complex diseases systematically [35]. However, in the traditional research of network pharmacology, the compounds are mostly collected from databases. Some compounds cannot be detected in the decoction, which may yield false positive results.

In this study, LC-MS/MS identification of compounds in the decoction followed by network pharmacology analyses provided insights into the mechanism of MFXD in the treatment of AR. In total, 24 bioactive compounds were identified in MFXD and 37 common targets were obtained

and analyzed. e results indicated that MFXD was effective in the treatment of AR by regulating key pathways, including the Fc epsilon RI, B-cell receptor, Toll-like receptor, NF-κB, T-cell receptor, PI3K-AKT, cAMP, and AMPK signaling pathways. Ten compounds in the decoction prepared from MFXD were identified as candidates that could target AR.

ese results reduce the prediction range, increase the accuracy of the prediction results, and provide important information for further pharmacological investigations on MFXD. is method, LC-MS/MS for bioactive compound identification followed by network pharmacology analyses, can increase the understanding of the mechanisms of Chinese herbal formulas and promote drug research and development.