revealing synergistic mechanism of multiple components in

Research ArticleRevealing Synergistic Mechanism of MultipleComponents in Gandi Capsule for Diabetic NephropathyTherapeutics by Network Pharmacology

Jian Zhang12 Qiqiang Zhang1 Xiaofei Chen2 Yan Liu1 Jiyang Xue3 Arik Dahan4

Hai Zhang 3 and Yifeng Chai 2

1Department of Pharmacy Xinhua Hospital Shanghai Jiao Tong University School of Medicine Shanghai 200092 China2School of Pharmacy Second Military Medical University Shanghai 200433 China3Department of Pharmacy Shanghai First Maternity and Infant Hospital Tongji University School of MedicineShanghai 201204 China4Department of Clinical Biochemistry and Pharmacology School of Pharmacy Ben-Gurion University of the Negev84105 Beer-Sheva Israel

Correspondence should be addressed to Hai Zhang zhxdks2005126com and Yifeng Chai yfchaismmueducn

Received 23 December 2017 Revised 4 March 2018 Accepted 18 March 2018 Published 26 April 2018

Academic Editor I-Min Liu

Copyright copy 2018 Jian Zhang et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Gandi capsule a traditional Chinese herbal medicinal formulation that consists of eight herbs is used as a clinical therapy fordiabetic nephropathy To clarify the potential synergisticmechanism this study adopted a network pharmacology strategy to screenthe action targets that corresponded to the active components in the Gandi capsule We first constructed a compound databaseof 315 components in the Gandi capsule and a target database of diabetic nephropathy which included 155 target proteins Sixrepresentative compounds were selected to dock with 99 proteins found in the UniProtKB database with their PDB code andinteraction networks between the active ingredients of the Gandi capsule and their targets were mapped out Results revealed 47proteins with a high affinity with at least one compoundmolecule in the Gandi capsuleThemain action pathways closely related tothe development of diabetic nephropathy were the TGF-1205731 AMPK insulin TNF-120572 and lipid metabolism pathways as per networkpharmacology analysis In the interaction network ACC1 SOD2 COX2 PKC-B IR and ROCK1 proteins had the most frequentinteractions with the six compounds We performed visual molecular docking in silico and experimentally confirmed competitivecomponent-protein binding by SPR and an enzyme activity test which highlighted the relationships of wogonin to COX2 andSOD2 astragaloside IV to ACC1 and morroniside to ACC1 We concluded that the potential synergistic mechanism of the Gandicapsule resulted fromhigh affinitieswithmultiple proteins and intervention inmultiple pathways in combination therapy of diabeticnephropathy

1 Introduction

Diabetic nephropathy is a severe complication of diabetesmellitus especially type 2 diabetes One-third of patientswith type 2 diabetes will develop diabetic nephropathywhich ultimately leads to end-stage renal disease (ESRD) [1]Moreover diabetic nephropathy is also the major cause ofESRD in China and behind glomerulonephritis After dia-betic nephropathy has developed into ESRD kidney failureis unavoidable and irreversible and can only be treated with

renal replacement [2]Meanwhile although some therapeuticmethods can relieve the symptoms of diabetic nephropathyincluding strict control of blood pressure the administrationof angiotensin-converting enzyme inhibitors and the phar-macological suppression of the renin-angiotensin systemthere are few methods currently available to reverse thedevelopment of diabetic nephropathy [3]

Gandi capsule (GDC) is a traditional Chinese medicinalformulation created by the Pharmacy Department of XinhuaHospital Shanghai Jiao TongUniversity and used for diabetic

HindawiEvidence-Based Complementary and Alternative MedicineVolume 2018 Article ID 6503126 11 pageshttpsdoiorg10115520186503126

2 Evidence-Based Complementary and Alternative Medicine

nephropathy It consists mainly of eight traditional Chineseherbs Astragali Radix Corni Fructus Radix RehmanniaeLeonuri Herba Sophorae Flos Scutellariae Radix BombyxBatryticatus and Phyllanthi Fruit [4] Thus it is clear thatGDC is a broad-spectrum formulation with many com-pounds Many herbs in GDC have been confirmed to havepositive effects on diabetic nephropathy such as AstragaliRadix a monarch medicine in GDC which was demon-strated to reduce levels of fasting blood glucose and albumin-uria and ameliorate the pathological changes of early diabeticnephropathy in animals [5] Moreover many componentsfrom GDC have been shown to have a positive influencein diabetic nephropathy For example some studies haveshown that loganin exerts a renal protective role in diabeticnephropathy through the reduction of renal connective tissuegrowth factor (CTGF) expression through the extracellular-signal-regulated kinase (ERK) signaling pathway [6] More-over calycosin and calycosin-7-O-120573-D-glucoside inhibitedhigh glucose-induced mesangial cell proliferation and AGE-induced endothelial cell apoptosis in diabetic nephropathy[7] Meanwhile some studies have indicated that quercetinmay ameliorate the epithelial-mesenchymal transition indiabetic nephropathy through the inhibition of the twotranscriptional factors (Snail and Twist) and the activationof mTORC1p70S6K [8] A previous study has described thecharacteristics of GDC from several perspectives such asthe analysis of ingredients in GDC [4] However it is stillunclear if the synergistic mechanism of multiple componentsin theGDC acts onmultiple targets for the therapy of diabeticnephropathy It is known that the pharmacological mech-anisms of traditional Chinese medicine (TCM) arise fromthemultiple compounds withmultiple targets and synergisticinteractions so it is difficult to clarify the comprehensivepharmacological mechanisms [9]

To reveal the main pharmacological action mechanismsof GDC we employed a network pharmacology strategyNetwork pharmacology which is based on the interactionnetwork of diseases genes target proteins and drugs is asystematic analytical method [10] This method emphasizesthe translation from the opinion of one target to one drugto a novel strategy of multicomponent to multitarget [11]It can reveal the action mechanism of the drug throughthe combination of computational biology systems biologyand ldquoomicsrdquo related to target drug [12] Recent reports haveemphasized the broad application of network pharmacologyto determine the mechanism of action of TCM formulationFor example the network pharmacology approach was usedto define the active components and potential targets inCurculigo orchioides for the treatment of osteoporosis [13]In this method multiple target synchronous regulation canbe verified which supports the complexity of TCMs In thisstudy two other methods molecular docking and networkanalysis were involved Molecular docking is a commonapproach for the identification of the interactions amongthe receptor protein and molecules with high accuracy [14]Therefore we can screen for potential molecules with a highbinding affinity Network analysis is based on the features ofthe diagram drawn based on the interactions between theprotein and protein or molecule

Therefore based on the advantages of network phar-macological research for the study of TCM the validationof the binding affinity between the proteins and moleculeswas undertaken Furthermore special proteins with specificnetwork pharmacological mechanisms in diabetic nephropa-thy can be identified The results could provide a betterunderstanding of the synergistic effect of compounds inGDCfor the treatment of diabetic nephropathy Moreover thenetwork pharmacology strategy could assist researchers inthe simplification of the study of complex TCM formulas andprovide a novel strategy to research the mechanism of actionof other complex TCM formulae

2 Materials and Methods

21 Materials Standard substances including astragalosideIV morroniside and wogonin were purchased from Stan-dard (Shanghai) acetyl-CoA carboxylase 1 was purchasedfrom Proteintech (Shanghai) and cyclooxygenase-2 andsuperoxide dismutase were purchased from Sino Biological(Shanghai) Other kinase proteins were purchased fromProteintech (Shanghai)

22 Construction of Compound and Protein Target DatabaseWe collected the compounds from the original herbs inGDC from the Chemistry Database developed by Shang-hai Organic Institute A proportion of the representativecompounds were selected out to proceed with the dock-ing These compounds were imported into PubMed tofind their structure canonical SMILES and any relevantother features Meanwhile we compiled the protein databaseabout diabetes diabetic complication diabetic nephropathyand microvascular disease collected from Therapeutic Tar-get Database (httpbiddnusedusggroupcjttd) andDrug-Bank (httpwwwdrugbankca) To identify the completefeatures of GDC no prescreening filters were applied If aPDB code was available in UniProtKB (httpwwwuniprotorg) of these proteins with lower resolution their crystalstructures were derived from the Research Collaboratoryfor Structural Bioinformatics (RCSB) Protein Data Bank(httpwwwpdborg) with their own ligands All proteinstructures used were from humans

23 Molecular Docking In this step the canonical SMILESof the compounds selected were imported into DiscoveryStudio 30 to define the structures of these molecules andthe structures of proteins were also entered At first thePreparation Module was used to clean the protein structureby using the CHARMM force field The Preparation Modulewas used to optimize the structures so that they couldbe adapted to docking Molecular docking was used toprove the potential interaction between the proteins and thedrug molecules The receptor-ligand interaction module ofDiscovery Studio 30 was used in our study at the earliestSpecific binding sites in the target protein were defined basedon the crystal structure of the protein-ligand compounds andthe docking procedure was used to dock the ligand into aspecific binding site To generate all possible conformations

Evidence-Based Complementary and Alternative Medicine 3

Table 1 The information of drug molecules

ID Compounds Short name Molecular formula Molecular weight(gsdotmolminus1) Origin Classification

1 Astragaloside IV AS C41H68O14

7849702 Astragali Radix Glucoside2 Morroniside MO C

17H26O11

4063840 Corni Fructus Glucoside3 Ferulic acid FA C

10H10O4

1941840 Corni Fructus Phenolic acid4 Rutin RU C

27H30O16

6105175 Sophorae FlosLeonuri Herba Flavonoid5 Wogonin WO C

22H20O11

4603876 Scutellariae Radix Glucosides6 Kaempferide KA C

16H12O6

3002629 Astragali RadixPhyllanthi Fruit Sugar acids

of molecules a conformational search was applied Thegenerated conformations were matched with a binding siteto select the optimal conformations which could interactwith the binding site This generated a LibDock score if themolecule was successfully docked with the target proteinsand the LibDock score of the ligand provided the baselinevalue to define the affinity between the protein andmoleculeThe whole docking procedure was conducted under theCHARMM force field The docking tolerance was set to025 the docking performance was set to be high qualitythe conformation method was set to be FAST the energythreshold was 200 and the minimization algorithm was setto smartminimizerThe default values were selected for otherparameters

24 Construction of Compound-Target Interaction NetworkTo analyze the interaction of the compound-protein-pathwayin diabetic nephropathy we used analytical software Cyto-scape 351 to construct a network diagram The compound-protein network was founded by the interaction of thecandidate molecule with the corresponding proteins Theprotein-pathway network was built through the connectionof the proteins and their own pathways respectively Inthe network the nodes represent compounds proteins andpathways the edges represent the interactions between themolecule protein and pathway The information was col-lected from TTD (httpbiddnusedusggroupcjttd) andKEGG (httpwwwgenomejpkeggpathwayhtml)The net-work parameters were calculated by the network analyzermodule in Cytoscape 351 Only results with a high bindingaffinity in molecular docking could be used for networkanalysis

25 SPR Assay In order to prove the validity of the networkpharmacological method different protein-molecule inter-actions were selected to participate in this surface plasmonresonance (SPR) test In the SPR test the target proteins werefirst combined with the CM5 chip After correction throughstandard solutions a series of different concentrations ofmolecules (0 2 4 8 16 32 64 and 128 120583moll) were testedfor their binding to the protein Finally the affinity betweenthe protein and molecule was determined by an array ofparameters such as the119870119889 values Rmax offset and 1205942

26 EnzymeActivity Test As there were some kinase proteinsin our protein database with different characteristics to other

proteins we used an enzyme activity test to validate the realinteractions between the protein and molecule

Selected pairs of protein and molecule participated inenzyme activity test while electrophoretic mobility shirtassay (EMSA) and Lance ultra were conducted in an experi-ment with ten different concentrations (100 50 25 125 625312 156 078 and 039 120583moll) Finally the IC

50(nM) value

was used to assess the ability of a molecule to act or inhibitthe protein

3 Results

31 Construction of Compound and Target Protein DatabaseFrom the constituents of GDC we acquired 315 compoundsfor our compound database Among these compounds 72were from Astragali Radix 71 were from Corni Fructus33 were from Radix Rehmanniae 23 were from SophoraeFlos 60 were from Scutellariae Radix 29 were from BombyxBatryticatus 13 were from Leonuri Herba and 14 were fromPhyllanthi Fruit Finally six representative compounds fromthis compound database were selected astragaloside IVmorroniside ferulic acid rutin wogonin and kaempferoltheir features are listed in Table 1 and the structures fromPubMed are shown in Figure 1

In the protein database the proteins were classified bydisease There were 155 proteins in this database whichwere divided into eight categories diabetes mellitus (DM)special diabetes mellitus (special DM) diabetes mellitustype 1 (DM1) diabetes mellitus type 2 (DM2) nonrenalcomplications of diabetes mellitus (DM complication) dia-betic nephropathy (DN) microvascular disease (MD) anddiabetic microangiopathy The principle of classification isbased on the keyword of retrieval in Therapeutic TargetDatabase Finally there were 49 proteins associated with DM13 proteins associated with DM1 33 proteins associated withDM2 two proteins associated with special DM four proteinsassociatedwithDMcomplication 28 proteins associatedwithdiabetic nephropathy 13 proteins associated with MD and12 proteins associated with diabetic microangiopathy Theproteins are listed in Appendix 1 with their characteristicfeatures After the construction of the target protein databasemore specific proteins parameters were sought If there werepharmacological parameters of the RCSB Protein Data Bankavailable for the PDB code from UniProtKB such as 119870119889 ICand the action diagram with small molecules they were usedin subsequent procedures


O

O

OHOH

OH

OH

OH

OH

H

H

HO

HO

HO

O

O

O

(a) Astragaloside IV

OH

OH

OH

O O

OO

O

H

H

H

HO

HO

O

(b) Morroniside

OHO

O

HO(c) Ferulic acid

OH

OH

OH

OH

OH

OH

HO

HO

HO

HO

O

O

O

OO O

(d) RutinOH

OHO

O

O(e) Wogonin

OH

OH

HO

HO

O

O

(f) Kaempferol

Figure 1 The structure of the six representative compounds

32 Target Prediction In the docking study the bindingaffinity between the molecule and protein was verified bycomparing the LibDock score with the ligandThemoleculesthat had a higher LibDock score than the ligand wereregarded to have a high affinity with protein otherwise theywere considered to have a low affinity In total 99 proteinswere examined for docking and all results that could beexported to LibDock score are shown in Appendix Finally47 target proteins had a higher LibDock score than the ligandwhich meant that there was a high binding affinity betweenthe molecules and proteins The virtual docking of part ofrepresentative compound-protein pairs is shown in Figure 2

From the analysis of the docking results we identified thatferulic acid had the most target molecules with 75 differentvalid results followed by kaempferide with 69 morronisideandwogoninwith 67 and rutin with 44 Astragaloside IV hadthe fewest targets with only 34 However morroniside wasthe most affiliative drug molecule of the six molecules with37 high affinity results whereas ferulic acid had only sevenFor the other compounds wogonin had 35 results rutin had25 kaempferide had 20 and astragaloside IV had 16

Moreover every molecule could dock with proteins witha high affinity for at least one protein in the different

diseases except astragaloside IV and rutin in DM1 Mean-while only RXRa could dock with high affinity to all sixmolecules 25 proteins could dock with all six moleculeswith valid affinity such as angiotensin-converting enzyme(ACE) integrin beta-7 and interstitial collagenase (MMP-1) 47 proteins could dock with at least one drug moleculewith high affinity such as acetyl-CoA carboxylase 1 (ACC1)glucagon receptor protein kinase C beta type (PKC-B)purinoceptor tumor necrosis factor-alpha (TNF-120572) trans-forming growth factor-1205731 (TGF-1205731) vasopressin receptor(V2R) aldosterone receptor (ALD) rho-associated pro-tein kinase 1 (ROCK1) and superoxide dismutase [Mn](SOD2)

33 Network Analysis Thenetworks based on the interactionbetween the protein molecule and pathway were shown inFigures 3 and 4

There were 91 nodes and 250 edges in the protein-molecule-pathway network and the proteins and drugmolecules were independent terms Meanwhile according tothe protein analysis the interaction between these proteinsand molecules that had high affinity could be classified into


(a) ACC1 with AS (b) ACC1 with MO (c) COX2 with WO (d) SOD2 with WO

(e) ROCK1 with MO (f) PKC-B with MO (g) mTOR with AS

Figure 2The docking results between the compounds and the proteins respectively Only the compound-protein pairs with valid results areshown

Pathway related to multi-pathway

Pathway related to cardiovascular system

Pathway related to immune system

Target protein

Pathway related to apoptosis

Drug molecule

Pathway related to metabolism

Figure 3 The compound-protein-pathway network diagram including six compounds and 47 target proteins and their pathways

five different systems immune system cardiovascular sys-tem apoptosis metabolism andmultiple pathways Differentclassifications were drawn in different colors and shapes Thecentralization and heterogeneity of the network were 0359and 0913 respectively

There were 42 proteins related to metabolism 33 relatedto the immune system 32 related to apoptosis 21 related tomultiple pathways and 10 related to cardiovascular systemMoreover the TGF-related pathway (degree = 9) and theadenosine 51015840-monophosphate- (AMP-) activated protein


ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR

SIRT1 AMPK signaling pathway

(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K

Insulin related signaling pathway

(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1

TGF-120573 related pathway

(d)

CT2 SOD2

PKC-B

TNF-120572COX2

TGFR-1 TNF related pathway

(e)

Figure 4 Five main pathways with their target proteins

kinase (AMPK) signaling pathway (degree = 9) were linkedwith nine proteins and the insulin-related signaling pathway(degree = 8) and lipid metabolism pathway (degree = 8)were linked with eight proteins and the TNF-related pathway(degree = 7) was linked with seven proteins other pathwayscorresponded to at least one protein Meanwhile for theproteins PKC-B was involved in the most pathways (15) IRwas involved in 12 COX2 was involved in 10 and ROCK1transforming growth factor receptor 1 (TGFR-1) and mTORwere involved in 9 each Other proteins had a lower numberof related pathways for example ACC1 was related to fivepathways and SOD2 was related to four pathways

34 Validity Assay of Target Protein Seven protein-moleculepairs were selected to take part in the SPR test ACC1 withastragaloside IV and morroniside COX2 with astragalosideIV morroniside and wogonin and SOD2 with morronisideand wogonin The results showed that astragaloside IV andmorroniside had a high response and affinity with ACC1with 119870119889 values of 925 120583M and 1363 120583M respectively Incontrast wogonin has a broader affinity with COX2 andSOD2 with119870119889 values of 4467 120583Mand 508120583M respectivelyThree protein-molecule pairs PKC-B with morronisideROCK1 with morroniside and mTOR with astragalosideIV were examined in the enzyme activity test Howeverall compound-protein interactions were negative The IC

50

values of these protein-molecule pairs were more than10000 120583M

The information and results of proteins and their respec-tive molecules are listed in Table 2 The results of SPR areshown in Figure 5

4 Discussion

TCM formulations are composed of multiple componentswith mechanisms of action that are always connected withmultiple target proteins and pathways GDC a TCM formu-lation administered only in hospital has been used to ame-liorate microvascular complications in DM such as diabeticnephropathy and diabetic retinopathy (DR) for some yearsIn the pharmaceutical analysis of GDC with HPLC-ESI-TOFMS 52 different components were identified includ-ing morroniside rutin astragaloside IV and wogonin Thebiopharmaceutical analysis of drug molecules from GDCshowed that 14 components including morroniside rutinastragaloside IV and wogonin could be absorbed in bloodafter the oral administration of GDC which means that theymight act through blood circulation [12] Given increasedrequirements for specific pharmacological mechanisms weexamined GDC in this study The complete procedure of thisresearch is presented in Figure 6

The construction of the compound and target proteindatabases was the first step in our study For the compounddatabase we collected compounds according to the herbs ofGDC formula obtained fromChemistry Database developedby Shanghai Organic Institute To identify the main sub-stances with pharmacological action inGDCwe used the fol-lowing three rules for selection (1) the rules of compositiontheory of the prescription in traditional Chinese medicine(2) the quantity of substance in GDC and (3) the hypo-glycemic effect reported about diabetic nephropathy Finallythe six representative compounds astragaloside IV morroni-side ferulic acid rutin wogonin and kaempferol were high-lighted To extend the database as much as possible we chose


Table 2 Information and results of the target protein in the validity assay

Number Full name ofprotein

Short name ofprotein

Respectivemolecule Test method

Results of experiment119870119889(120583M)

119877max(RU)

Offset(RU) 1205942 (RU2)

1Acetyl-CoAcarboxylase 1 ACC1 Astragaloside IV SPR 925 7856 02424 00243

Morroniside SPR 1363 2204 minus02651 00962

2 Cyclooxygenase-2 COX2Astragaloside IV SPRMorroniside SPR 4467 3954 06314 0799Wogonin SPR

3Superoxide

dismutase [Mn] SOD2 Morroniside SPRWogonin SPR 508 143 minus3611 211

IC50 (120583M)

4Protein kinase C

beta type PKC-B Morroniside EMSA gt10000

5Rho-associatedprotein kinase 1 ROCK1 Morroniside EMSA gt10000

6 Mammalian targetof rapamycin mTOR Astragaloside IV Lance ultra gt10000

Kd = 925 120583M

Cycle 13 AS 0002 mM

Cycle 15 AS 0008 mM Cycle 18 AS 0064 mMCycle 14 AS 0004 mM Cycle 17 AS 0032 mM

Cycle 16 AS 0016 mM

minus50 0 50 100 150 200 250 300 350minus100Time (s)

Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM

Cycle 25 MO 0008 mM Cycle 28 MO 0064 mMCycle 24 MO 0004 mM Cycle 27 MO 0032 mM

Cycle 26 MO 0016 mM

minus100 minus50 0 50Time (s)

100 150 200 250 300 350minus10

minus8minus6minus4minus2

02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM

Cycle 35 WO 0008 mM Cycle 38 WO 0064 mMCycle 34 WO 0004 mM Cycle 37 WO 0032 mM

Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2

Figure 5 The results of SPR Only compound-protein pairs with valid results are shown


Highlight the pathway (5) and protein

Select out compound-protein pairs

Network construction and analysis

Validity assay

Docking

Pre-treatment of structureCanonical Smile of compound

from Pubmed

Import into Discovery studio 30

Representative compounds (6)

Acquire protein structure (99)

Start

Protein databaseTarget protein collection (155)Compound information

collection (315)TCM database

Figure 6 Flow chart of entire research

the classification of diseases related to diabetic nephropathyas a criterion of acquiring target proteins there were eightkeywords for searching target proteins The exact classifica-tions were DM DM1 DM2 special DM DM complicationDN MD and diabetic microangiopathy Finally there are155 proteins in our target protein database Some studieshave shown that many target proteins such as angiotensin-converting enzyme (ACE) protein kinase C (PKC) and120572-glucosidase might be a therapeutic target for diabeticnephropathy [3] These proteins were also identified in ourtarget protein database As can be seen the protein databasewas shown to contain proteins covering different diseases ofdiabetic nephropathy which was characteristic of extensivediversity that is supported by the complexity of TCM

After the construction of the target protein database 155proteins were searched in UniProtKB if they were identifiedin UniProtKB the bias of the protein structure was comparedwith reality After this procedure 99 individual proteins werealso found in the RCSBProteinData BankThepharmacolog-ical parameters in RCSB Protein Data Bank such as the 119870119889IC and action mechanism with small molecules supportedtheir reliable crystal structure for docking For moleculedocking a high-throughput virtual screening approach was

used Then there were 47 proteins within our networkanalysis and the degree of centrality was chosen to be thestandard for influence of proteins or pathways in our studysystem Finally five representative pathways were definedand 10 compound-protein pairs were examined in the validityassay The methods above displayed a powerful capacity forthe acquisition processing and integration of informationon compounds in a TCM formulation furthermore it wasspecific to the complexity of the GDC capsule

In this study 99 proteins were evaluated for themoleculardocking of six representative compoundsThe results showedthat morroniside was the most affiliative drug of the sixmolecules with a high binding affinity with numerous pro-teins furthermore the action of morroniside may be clearerthan that for other components Ferulic acid could dock themost target molecules but it also had the lowest number ofhigh affinity results indicating that the action of ferulic acidin the treatment of diabetic nephropathy occurs only becauseof the sum of various weak interactions Other compoundsalso show affinities to different proteins which may be thebasis of their action as diabetic nephropathy therapeutics

In this study according to the result of molecular dock-ing and network analysis all protein-pathway pairs were


distributed among metabolism immune system apoptosiscardiovascular system and multiple pathways Moreoverthe most frequent pathways were the TGF-related pathwayand AMPK signaling pathway which were linked with nineproteins followed by the insulin-related signaling pathwaylipid metabolism and the TNF-related pathway

The TGF-related pathway mainly includes TGF pathwayand TGFR pathway in which TGF-120573 is a key mediator TGF-120573 is active in human mesangial cells and causes oxidativestress and apoptosis in DN [15] It also can regulate the Smadsignaling pathway which induces type I collagen synthesisin tubular epithelial cells and mesangial cells ultimatelyleading to fibronectin in mesangial cells [16] AMPK is aserinethreonine protein kinase that influences the processof DN through the regulation of energy metabolism prolif-eration inflammation and apoptosis in which the AMPKsignaling pathway can also intervene [17] Some proteins inthe signaling pathway above such as p38 MAPK and TGF-120573 can regulate the AGE-RAGE pathway which activatesmultiple signaling pathways implicated in inflammationcancer and other diseases [18] That is the AMPK signalingpathway and the TGF-related pathwaymay have potential forsynergistic action in the development of DN

The insulin-related signaling pathway which has an arrayof mediators including IR and GLPR-1 is active in theprocess of DNThe pathway of lipid metabolism covers eightproteins such as ACC1 ACC2 IR and mTOR which canregulate the generation decomposition or actions of lipidA study indicated that dyslipidemia was a risk factor in theprocess of kidney disease in type 2 diabetes mellitus and thatthe accumulation of lipid causes podocyte dysfunction andapoptosis in DN [19]

TNF-120572 is one of the most active proteins in the TNF-related pathway which may be activated in inflammationdevelopmentThe research byGuo et al showed that the pres-ence of infiltrating macrophages in DN promotes podocyteapoptosis via TNF-120572-related pathway [20] Presently DNhas been defined as the occurrence of a series of abnormalchanges in pathophysiology including metabolic disorderoxidative stress inflammatory processes apoptosis andhemodynamic changes [3 21 22] In addition based on theinformation fromKEGG themain pathways of DN are AGE-RAGE signaling pathway renin-angiotensin system (RAS)FoxO signaling pathway and longevity regulation Somecommon pathways also exist in this pathway such as theAMPK pathway TNF-120572 signaling pathway and TGF signal-ing pathway

In addition to these five main pathways above somepathways related to apoptosis were also found in our investi-gation such as the Ras signaling pathway the Rap1 signalingpathway and theWnt signaling pathway It appears thatmanypairs of protein and molecules can influence these pathwayssuch as PKC-B with morroniside FGF with astragaloside IVand morroniside and GSK-3120573 with rutin Moreover proteinmetabolism may be another approach for the treatment ofGDC However only a small number of proteins are involvedin this way such as DPP-IV and NEP Meanwhile as forRAS some proteins are involved with compounds such asV2R ALD and NEP ALD has been defined as a regulatory

activator of sodium and water which is unbalanced in DN[3]

As for proteins the centrality degree of proteins in thenetwork analysis and their evidence of influence for the pro-cess of DN have been considered Ultimately PKC-B COX2ROCK1 mTOR SOD2 and ACC1 were chosen to be the rep-resentative compounds In combinationwith the informationextracted from the five pathways above we found that ACC1might influence the AMPK signaling pathway the insulin-related signaling pathway and lipidmetabolismmTORcouldinfluence the AMPK signaling pathway the insulin-relatedsignaling pathway and lipid metabolism COX2 could alterlipidmetabolism and the TNF-related pathway PKC-B couldparticipate in the insulin-related signaling pathway and theTNF-related pathway SOD2 influenced the TGF-related andTNF-related pathways and ROCK1 influenced the TGF-related pathway According to the pathway information theaction of ACC1 andmTOR focused on themetabolism of glu-cose and lipids inflammation and apoptosis in DN whereasPKC-B and COX2 were concentrated on metabolism andinflammation SOD2 altered oxidative stress and apoptosisand ROCK1 mainly influenced apoptosis Collectively theactions of these proteins are concentrated on metabolisminflammation apoptosis and oxidative stress in DN

Moreover the results of the validity assay indicated thatthe action of ACC1 can be influenced by astragaloside IVandmorroniside whereas COX2 and SOD2 can be altered bywogonin

Astragaloside IV can regulate the AMPK signalingpathway the insulin-related signaling pathway and themetabolism of lipids through an influence on the activity ofACC1 The research of Zhang showed that astragaloside IValtered the NF-kappa B pathway to counteract the expressionof E-selectin andVCAM-1 in response to inflammatorymedi-ators which meant that it has credible anti-inflammatoryability [23] Combined with our study the anti-inflammatoryaction of astragaloside IV may occur through the changes inACC1

While morroniside can take part in the TGF-relatedpathway the insulin-related signaling pathway and the TNF-related pathway may be altered by ROCK1 and PKC-B Anearly study pointed out that morroniside alleviated oxida-tive stress under diabetic pathological conditions througha reduction in ROS level and the downregulation of theexpression of iNOS and COX2 followed by the inhibitionof NF-120581B transcription [24] Some research has shown thatmorroniside can alter the AGE-RAGE pathway to suppressthe high expression of p38 MAPK and the synthesis andrelease of NF-kB ICAM-1 MCP-1 and TGF-120573 [18] Thusmorroniside may have an effect through the alteration ofROCK1 and PKC-B in the TGF-related pathway insulin-related signaling pathway and the TNF-related pathway

Wogonin can affect the TGF-related pathway lipidmetabolism and TNF-related pathway through the regula-tion of COX2 and SOD2 Clearly anti-inflammationmay be amethod of action of wogonin and a study demonstrated thatwogonin inhibited high-glucose-induced vascular inflamma-tion which was attributed to antioxidant ability and theinhibition of LPS-induced nitric oxide (NO) production and


the expression of the nitric oxide synthase (iNOS) gene [25]According to our research wogonin affected the progress ofinflammation in DN through the regulating of COX2 andSOD2 so that it can regulate the TGF-related pathway lipidmetabolism and TNF-related pathway

In conclusion our research demonstrated that these sixrepresentative compounds inGDC exert a potential influenceon the pathways that are disordered in diabetic nephropathythrough their constituent proteins However the results areonly based on virtual methods based on the results of SPRand the enzyme activity test further research is required toreveal the actual action of GDC in diabetic nephropathy

5 Conclusions

In this study six representative compounds inGDCwere cho-sen to dock with 155 proteins related to diabetic nephropathyAs a result 47 proteins were identified with a high affinitywith at least one compound molecule These 47 proteinswere analyzed in the protein-molecule-pathway networkIt was found that the TGF-1205731 pathway AMPK signalingpathway insulin-related signaling pathway lipidmetabolismand TNF-120572 pathway were the major pathways of treatmentof GDC for diabetic nephropathy These five pathways maybe the primary aspects of the treatment for GDC in diabeticnephropathy Moreover astragaloside IV and morronisidewere shown to alter ACC1 directly (119870119889 = 925 120583M and1363 120583M) whereas wogonin could bind with COX2 andSOD2 (119870119889 = 4467 120583M and 508 120583M)

In summary the action of GDC in the treatment ofdiabetic nephropathy was concentrated on metabolism reg-ulation anti-inflammation the inhibition of apoptosis andexerted pharmaceutical actions based on the regulation of themedial proteins in these pathways

Data Availability

The data of our research can be acquired from the Supple-mentary Materials uploaded with this article

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the submission of this manuscript

Authorsrsquo Contributions

Jian Zhang and Qiqiang Zhang equally contributed to thiswork All the authors have approved the manuscript that isenclosed

Acknowledgments

This research was supported by the National Natural ScienceFoundation of China (no 81773797) Shanghai MunicipalMedical and Health Discipline Construction Projects (no2017ZZ02015) and the Fundamental Research Funds for theCentral Universities (no 1515219051)

Supplementary Materials

Six files have been uploaded as supplementary materials forthis research The first file is the WORD file of the appendixwhich has two forms One is the target protein database andthe other is the form of valid results between the compoundsand proteins in docking The second file is an EXCEL fileof the compound database including 315 compounds Thethird file is the graphical abstract in PDF which shows themain process of the research with picturesThe fourth file is aformwith three sheets in EXCEL format which includes totalresults of docking valid results of docking and high affinityresults of docking Meanwhile the fifth file and the sixth fileare EXCEL files of the network diagram of representativetarget-compound networks and target-compound-pathwaynetworks respectively (Supplementary Materials)

References

[1] J Collins Allan N Foley Robert H Charles et al ldquoUS RenalData System 2012 Annual Data Reportrdquo American Journal ofKidney Diseases vol 61 Suppl 1 no 1 pp e1ndashe476 2013

[2] Z-H Liu ldquoNephrology in Chinardquo Nature Reviews Nephrologyvol 9 no 9 pp 523ndash528 2013

[3] Y B Tripathi and D Yadav ldquoDiabetic Nephropathy Causesand Managementsrdquo Recent Patents on Endocrine Metabolic ampImmune Drug Discovery vol 7 no 1 pp 57ndash64 2012

[4] Z J Lin Z Y Tang Y N Huang et al ldquoSimultaneous deter-mination of seven compounds in Gandi capsules by UHPLC-MSMS methodrdquo Chinese Journal of Pharmaceutical Analysisvol 34 no 4 pp 622ndash627 2014

[5] J Zhang X Xie C Li and P Fu ldquoSystematic review of the renalprotective effect of Astragalusmembranaceus (root) on diabeticnephropathy in animal modelsrdquo Journal of Ethnopharmacologyvol 126 no 2 pp 189ndash196 2009

[6] W-L Jiang S-P Zhang J Hou and H-B Zhu ldquoEffect ofloganin on experimental diabetic nephropathyrdquo Phytomedicinevol 19 no 3-4 pp 217ndash222 2012

[7] D Tang B He Z-G Zheng et al ldquoInhibitory effects of twomajor isoflavonoids in Radix Astragali on high glucose-inducedmesangial cells proliferation and AGEs-induced endothelialcells apoptosisrdquo Planta Medica vol 77 no 7 pp 729ndash732 2011

[8] Q Lu X-J Ji Y-X Zhou et al ldquoQuercetin inhibits themTORC1p70S6K signaling-mediated renal tubular epithelial-mesenchymal transition and renal fibrosis in diabetic nephro-pathyrdquo Pharmacological Research vol 99 article no 2848 pp237ndash247 2015

[9] J Xu Z Song Q Guo and J Li ldquoSynergistic effect andmolecular mechanisms of traditional Chinese medicine onregulating tumor microenvironment and cancer cellsrdquo BioMedResearch International vol 2016 Article ID 1490738 14 pages2016

[10] A L Hopkins ldquoNetwork pharmacologyrdquoNature Biotechnologyvol 25 no 10 pp 1110-1111 2007

[11] S Li and B Zhang ldquoTraditional Chinese medicine networkpharmacology theory methodology and applicationrdquo ChineseJournal of Natural Medicines vol 11 no 2 pp 110ndash120 2013

[12] A L Hopkins ldquoNetwork pharmacology the next paradigm indrug discoveryrdquoNature Chemical Biology vol 4 no 11 pp 682ndash690 2008


[13] N Wang G Zhao Y Zhang et al ldquoA network pharmacologyapproach to determine the active components and potential tar-gets of Curculigo orchioides in the treatment of osteoporosisrdquoMedical Science Monitor vol 23 pp 5113ndash5122 2017

[14] D Plewczynski A Philips M von Grotthuss L Rychlewskiand K Ginalski ldquoHarmonyDOCK the structural analysis ofposes in protein-ligand dockingrdquo Journal of ComputationalBiology vol 21 no 3 pp 247ndash256 2014

[15] M Jiang H Zhang L Zhai B Ye Y Cheng and C ZhaildquoALALA ameliorates glucose toxicity on HK-2 cells by attenu-ating oxidative stress and apoptosis through the ROSp38TGF-1205731 pathwayrdquo Lipids in Health and Disease vol 16 no 1 articleno 216 2017

[16] J Zhao S Miyamoto Y-H You and K Sharma ldquoAMP-activated protein kinase (AMPK) activation inhibits nucleartranslocation of Smad4 inmesangial cells and diabetic kidneysrdquoAmerican Journal of Physiology-Renal Physiology vol 308 no10 pp F1167ndashF1177 2015

[17] M Lee D Feliers M M Mariappan et al ldquoA role for AMP-activated protein kinase in diabetes-induced renal hypertro-phyrdquo American Journal of Physiology-Renal Physiology vol 292no 2 pp F617ndashF627 2007

[18] G Lv X Lv Y Tao and H Xu ldquoEffect of morroniside onglomerular mesangial cells through AGEndashRAGE pathwayrdquoHuman Cell vol 29 no 4 pp 148ndash154 2016

[19] M Herman-Edelstein P Scherzer A Tobar M Levi andU Gafter ldquoAltered renal lipid metabolism and renal lipidaccumulation in human diabetic nephropathyrdquo Journal of LipidResearch vol 55 no 3 pp 561ndash572 2014

[20] Y Guo Z Song M Zhou et al ldquoInfiltrating macrophages indiabetic nephropathy promote podocytes apoptosis via TNF-120572-ROS-p38MAPK pathwayrdquoOncotarget vol 8 no 32 pp 53276ndash53287 2017

[21] Y Lytvyn B A Perkins and D Z I Cherney ldquoUric acid asa biomarker and a therapeutic target in diabetesrdquo CanadianJournal of Diabetes vol 39 no 3 pp 239ndash246 2015

[22] Y Maezawa M Takemoto and K Yokote ldquoCell biology of dia-betic nephropathy roles of endothelial cells tubulointerstitialcells and podocytesrdquo Journal of Diabetes Investigation vol 6 no1 pp 3ndash15 2015

[23] WZhang PHufnagl B R Binder and JWojta ldquoAntiinflamma-tory activity of astragaloside IV ismediated by inhibition of NF-120581B activation and adhesion molecule expressionrdquo Thrombosisand Haemostasis vol 90 no 5 pp 904ndash914 2003

[24] C H Park J S Noh T Tanaka and T Yokozawa ldquoEffects ofmorroniside isolated from Corni Fructus on renal lipids andinflammation in type 2 diabetic micerdquo Journal of Pharmacy andPharmacology vol 62 no 3 pp 374ndash380 2010

[25] S-K Ku and J-S Bae ldquoBaicalin baicalein and wogonin inhibitshigh glucose-induced vascular inflammation in vitro and invivordquo BMB Reports vol 48 no 9 pp 519ndash524 2015

Stem Cells International

Hindawiwwwhindawicom Volume 2018


MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwwwhindawicom Volume 2018

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine


Behavioural Neurology

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwwwhindawicom

Submit your manuscripts atwwwhindawicom


nephropathy It consists mainly of eight traditional Chineseherbs Astragali Radix Corni Fructus Radix RehmanniaeLeonuri Herba Sophorae Flos Scutellariae Radix BombyxBatryticatus and Phyllanthi Fruit [4] Thus it is clear thatGDC is a broad-spectrum formulation with many com-pounds Many herbs in GDC have been confirmed to havepositive effects on diabetic nephropathy such as AstragaliRadix a monarch medicine in GDC which was demon-strated to reduce levels of fasting blood glucose and albumin-uria and ameliorate the pathological changes of early diabeticnephropathy in animals [5] Moreover many componentsfrom GDC have been shown to have a positive influencein diabetic nephropathy For example some studies haveshown that loganin exerts a renal protective role in diabeticnephropathy through the reduction of renal connective tissuegrowth factor (CTGF) expression through the extracellular-signal-regulated kinase (ERK) signaling pathway [6] More-over calycosin and calycosin-7-O-120573-D-glucoside inhibitedhigh glucose-induced mesangial cell proliferation and AGE-induced endothelial cell apoptosis in diabetic nephropathy[7] Meanwhile some studies have indicated that quercetinmay ameliorate the epithelial-mesenchymal transition indiabetic nephropathy through the inhibition of the twotranscriptional factors (Snail and Twist) and the activationof mTORC1p70S6K [8] A previous study has described thecharacteristics of GDC from several perspectives such asthe analysis of ingredients in GDC [4] However it is stillunclear if the synergistic mechanism of multiple componentsin theGDC acts onmultiple targets for the therapy of diabeticnephropathy It is known that the pharmacological mech-anisms of traditional Chinese medicine (TCM) arise fromthemultiple compounds withmultiple targets and synergisticinteractions so it is difficult to clarify the comprehensivepharmacological mechanisms [9]

To reveal the main pharmacological action mechanismsof GDC we employed a network pharmacology strategyNetwork pharmacology which is based on the interactionnetwork of diseases genes target proteins and drugs is asystematic analytical method [10] This method emphasizesthe translation from the opinion of one target to one drugto a novel strategy of multicomponent to multitarget [11]It can reveal the action mechanism of the drug throughthe combination of computational biology systems biologyand ldquoomicsrdquo related to target drug [12] Recent reports haveemphasized the broad application of network pharmacologyto determine the mechanism of action of TCM formulationFor example the network pharmacology approach was usedto define the active components and potential targets inCurculigo orchioides for the treatment of osteoporosis [13]In this method multiple target synchronous regulation canbe verified which supports the complexity of TCMs In thisstudy two other methods molecular docking and networkanalysis were involved Molecular docking is a commonapproach for the identification of the interactions amongthe receptor protein and molecules with high accuracy [14]Therefore we can screen for potential molecules with a highbinding affinity Network analysis is based on the features ofthe diagram drawn based on the interactions between theprotein and protein or molecule

Therefore based on the advantages of network phar-macological research for the study of TCM the validationof the binding affinity between the proteins and moleculeswas undertaken Furthermore special proteins with specificnetwork pharmacological mechanisms in diabetic nephropa-thy can be identified The results could provide a betterunderstanding of the synergistic effect of compounds inGDCfor the treatment of diabetic nephropathy Moreover thenetwork pharmacology strategy could assist researchers inthe simplification of the study of complex TCM formulas andprovide a novel strategy to research the mechanism of actionof other complex TCM formulae

2 Materials and Methods

21 Materials Standard substances including astragalosideIV morroniside and wogonin were purchased from Stan-dard (Shanghai) acetyl-CoA carboxylase 1 was purchasedfrom Proteintech (Shanghai) and cyclooxygenase-2 andsuperoxide dismutase were purchased from Sino Biological(Shanghai) Other kinase proteins were purchased fromProteintech (Shanghai)

22 Construction of Compound and Protein Target DatabaseWe collected the compounds from the original herbs inGDC from the Chemistry Database developed by Shang-hai Organic Institute A proportion of the representativecompounds were selected out to proceed with the dock-ing These compounds were imported into PubMed tofind their structure canonical SMILES and any relevantother features Meanwhile we compiled the protein databaseabout diabetes diabetic complication diabetic nephropathyand microvascular disease collected from Therapeutic Tar-get Database (httpbiddnusedusggroupcjttd) andDrug-Bank (httpwwwdrugbankca) To identify the completefeatures of GDC no prescreening filters were applied If aPDB code was available in UniProtKB (httpwwwuniprotorg) of these proteins with lower resolution their crystalstructures were derived from the Research Collaboratoryfor Structural Bioinformatics (RCSB) Protein Data Bank(httpwwwpdborg) with their own ligands All proteinstructures used were from humans

23 Molecular Docking In this step the canonical SMILESof the compounds selected were imported into DiscoveryStudio 30 to define the structures of these molecules andthe structures of proteins were also entered At first thePreparation Module was used to clean the protein structureby using the CHARMM force field The Preparation Modulewas used to optimize the structures so that they couldbe adapted to docking Molecular docking was used toprove the potential interaction between the proteins and thedrug molecules The receptor-ligand interaction module ofDiscovery Studio 30 was used in our study at the earliestSpecific binding sites in the target protein were defined basedon the crystal structure of the protein-ligand compounds andthe docking procedure was used to dock the ligand into aspecific binding site To generate all possible conformations






17H26O11


10H10O4


27H30O16


22H20O11


16H12O6








50(nM) value


3 Results




O

O

OHOH

OH

OH

OH

OH

H

H

HO

HO

HO

O

O

O


OH

OH

OH

O O

OO

O

H

H

H

HO

HO

O

(b) Morroniside

OHO

O

HO(c) Ferulic acid

OH

OH

OH

OH

OH

OH

HO

HO

HO

HO

O

O

O

OO O

(d) RutinOH

OHO

O

O(e) Wogonin

OH

OH

HO

HO

O

O

(f) Kaempferol















Target protein


Drug molecule






ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR


(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K


(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1


(d)

CT2 SOD2

PKC-B

TNF-120572COX2


(e)




50



4 Discussion









119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of
























17H26O11


10H10O4


27H30O16


22H20O11


16H12O6








50(nM) value


3 Results




O

O

OHOH

OH

OH

OH

OH

H

H

HO

HO

HO

O

O

O


OH

OH

OH

O O

OO

O

H

H

H

HO

HO

O

(b) Morroniside

OHO

O

HO(c) Ferulic acid

OH

OH

OH

OH

OH

OH

HO

HO

HO

HO

O

O

O

OO O

(d) RutinOH

OHO

O

O(e) Wogonin

OH

OH

HO

HO

O

O

(f) Kaempferol















Target protein


Drug molecule






ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR


(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K


(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1


(d)

CT2 SOD2

PKC-B

TNF-120572COX2


(e)




50



4 Discussion









119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of




















O

O

OHOH

OH

OH

OH

OH

H

H

HO

HO

HO

O

O

O


OH

OH

OH

O O

OO

O

H

H

H

HO

HO

O

(b) Morroniside

OHO

O

HO(c) Ferulic acid

OH

OH

OH

OH

OH

OH

HO

HO

HO

HO

O

O

O

OO O

(d) RutinOH

OHO

O

O(e) Wogonin

OH

OH

HO

HO

O

O

(f) Kaempferol















Target protein


Drug molecule






ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR


(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K


(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1


(d)

CT2 SOD2

PKC-B

TNF-120572COX2


(e)




50



4 Discussion









119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of


























Target protein


Drug molecule






ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR


(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K


(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1


(d)

CT2 SOD2

PKC-B

TNF-120572COX2


(e)




50



4 Discussion









119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of




















ACC1

ACC2

PI3K

GCGR

MLYCD

mTOR

IR


(a)

GSK-3 betamTOR

ACC1

GLPR-1

IR

PKC-B

PI3K


(b)

JAK2

mTOR

ACC1

ACC2

CRAT

IR

RXRa

COX2

metabolism of lipid

(c)

TGFR-1CT2

SCC2

ACC2

DPP-IV

TNF-120572CGB-PDE

FGF

ROCK1


(d)

CT2 SOD2

PKC-B

TNF-120572COX2


(e)




50



4 Discussion









119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of

























119877max(RU)





3Superoxide


IC50 (120583M)

4Protein kinase C




Kd = 925 120583M

Cycle 13 AS 0002 mM


Cycle 16 AS 0016 mM


Resp

onse

(RU

) 6420

810

minus4minus6

minus2

minus8minus10

(a) AS with ACC1

Kd = 1363 120583M

Cycle 23 MO 0002 mM


Cycle 26 MO 0016 mM


100 150 200 250 300 350minus10


02468

10

Resp

onse

(RU

)

(b) MO with ACC1

Kd = 4467 120583M

Cycle 33 WO 0002 mM


Cycle 36 WO 0016 mM


minus10

minus5

0

5

10

15

20

25

30

35

Resp

onse

(RU

)

(c) WO with COX2

Kd = 508 120583M

Cycle 23 WO 0002 mM


Cycle 26 WO 0016 mM


minus40

minus30

minus20

minus10

0

10

20

30

40

Resp

onse

(RU

)

(d) WO with SOD2






Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of























Validity assay

Docking


from Pubmed




Start
























5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of


































5 Conclusions



Data Availability






Acknowledgments




References































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of





































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of



















revealing synergistic mechanism of multiple components in

Documents