Research Article dTGS: Method for Effective Components ...
9
Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2013, Article ID 840427, 9 pages http://dx.doi.org/10.1155/2013/840427 Research Article dTGS: Method for Effective Components Identification from Traditional Chinese Medicine Formula and Mechanism Analysis Ji Luo, 1 Yinglong Ren, 1 Hao Gu, 1 Yi Wu, 2 and Yun Wang 1 1 Beijing University of Chinese Medicine, School of Chinese Pharmacy No. 6, Zhonghuan South Road, Wangjing, Chaoyang, Beijing 100102, China 2 Pennsylvania State University Erie, e Behrend College, School of Engineering, 5101 Jordan Road, Erie, PA 16563, USA Correspondence should be addressed to Yun Wang; [email protected] Received 24 July 2013; Revised 27 September 2013; Accepted 27 September 2013 Academic Editor: Weidong Zhang Copyright © 2013 Ji Luo 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. Because of the complexity of the components in Traditional Chinese Medicine formula (TCM formula), it is still a challenge to identify its effective components, to elucidate the mechanism of the components, and to discover the relationship between components and therapy objectives. In this paper, a method called directed TCM grammar systems (dTGS) for effective component identification was proposed using entity grammar systems (EGS) as the theoretical framework. e component-disease relationship of a TCM formula (i.e., Bai-Hu decoction plus Wasting-irsting formula, BHDWT) and one disease (i.e., type 2 diabetes mellitus) treated with it was studied, and the effective component groups (ECGs) were identified. 19 compounds were found acting on 20 proteins in type 2 diabetes mellitus (T2D) disease network, and 15 compounds were determined as the candidate effective components. Results indicated that this method can be used to identify the effective components and provide an innovative way to elucidate the molecular mechanism of TCM formulas. 1. Introduction e components in Traditional Chinese Medicine formula (TCM formula) were very complex and their molecular mechanism was unclear. For the treatment of one disease, some components may be favorable and others may not. Identification of favorable components and analysis of their action mechanisms will benefit the optimization of culti- vation condition, processing technology, extraction process, and new drug development. At present, experiment screening was still the main method for the identification of effective components and effective component groups (ECGs). For example, high-performance liquid chromatography-mass spectrometry (HPLC-MS) was used to analyze the active con- stituents of Xiao-Xu-Ming decoction [1]; drosophila trans- genic models were used to identify combinatorial drug, such as suberoylanilide hydroxamic acid (SAHA) and gel- danamycin, for the treatment of Huntington’s disease [2]; cell- based assays technology was used to screen two-component combinations for the treatment of cancer, infectious dis- eases, and CNS disorders [3]. However, the results identified through experimental screening were limited, due to the complexity of components and the high cost associated with experiments. Recently, computational systems biology was used to study TCM because of the technical advantages of study- ing large and complex systems and the relative lower cost compared to experimental screening. e applications of complex network analysis techniques, in particular, led to many new findings. For instance, microarray technology and connectivity maps were integrated into the research of molecular mechanisms of Si-Wu decoction (composed of four herbs: Radix Rehmanniae preparata, Radix Angelicae Sinensis, Rhizoma Ligustici Chuanxiong, and Radix Paeoniae Alba) [4]. Multilayer map of “Phenotype network-Biological network-Herb network” was applied to uncover the underly- ing network systems of TCM syndromes and herb formulas [5]; the drug-target network was implemented to elucidate the mechanism of one TCM formula for the treatment of T2D [6]. Although those applications of computational systems biology in the study of TCM formulas are still in the exploratory stages, they demonstrate the feasibility of
-
Upload
khangminh22 -
Category
Documents
-
view
3 -
download
0
Transcript of Research Article dTGS: Method for Effective Components ...
Hindawi Publishing CorporationEvidence-Based Complementary and Alternative MedicineVolume 2013 Article ID 840427 9 pageshttpdxdoiorg1011552013840427
Research ArticledTGS Method for Effective Components Identification fromTraditional Chinese Medicine Formula and Mechanism Analysis
Ji Luo1 Yinglong Ren1 Hao Gu1 Yi Wu2 and Yun Wang1
1 Beijing University of Chinese Medicine School of Chinese Pharmacy No 6 Zhonghuan South Road WangjingChaoyang Beijing 100102 China
2 Pennsylvania State University Erie The Behrend College School of Engineering 5101 Jordan Road Erie PA 16563 USA
Correspondence should be addressed to Yun Wang wangyunbucmeducn
Received 24 July 2013 Revised 27 September 2013 Accepted 27 September 2013
Academic Editor Weidong Zhang
Copyright copy 2013 Ji Luo et al This is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
Because of the complexity of the components in Traditional Chinese Medicine formula (TCM formula) it is still a challengeto identify its effective components to elucidate the mechanism of the components and to discover the relationship betweencomponents and therapy objectives In this paper amethod called directed TCMgrammar systems (dTGS) for effective componentidentificationwas proposed using entity grammar systems (EGS) as the theoretical frameworkThe component-disease relationshipof a TCM formula (ie Bai-Hu decoction plusWasting-Thirsting formula BHDWT) and one disease (ie type 2 diabetes mellitus)treated with it was studied and the effective component groups (ECGs) were identified 19 compounds were found acting on20 proteins in type 2 diabetes mellitus (T2D) disease network and 15 compounds were determined as the candidate effectivecomponents Results indicated that this method can be used to identify the effective components and provide an innovative way toelucidate the molecular mechanism of TCM formulas
1 Introduction
The components in Traditional Chinese Medicine formula(TCM formula) were very complex and their molecularmechanism was unclear For the treatment of one diseasesome components may be favorable and others may notIdentification of favorable components and analysis of theiraction mechanisms will benefit the optimization of culti-vation condition processing technology extraction processand newdrug development At present experiment screeningwas still the main method for the identification of effectivecomponents and effective component groups (ECGs) Forexample high-performance liquid chromatography-massspectrometry (HPLC-MS)was used to analyze the active con-stituents of Xiao-Xu-Ming decoction [1] drosophila trans-genic models were used to identify combinatorial drugsuch as suberoylanilide hydroxamic acid (SAHA) and gel-danamycin for the treatment ofHuntingtonrsquos disease [2] cell-based assays technology was used to screen two-componentcombinations for the treatment of cancer infectious dis-eases and CNS disorders [3] However the results identified
through experimental screening were limited due to thecomplexity of components and the high cost associated withexperiments
Recently computational systems biology was used tostudy TCM because of the technical advantages of study-ing large and complex systems and the relative lower costcompared to experimental screening The applications ofcomplex network analysis techniques in particular led tomany new findings For instance microarray technologyand connectivity maps were integrated into the research ofmolecular mechanisms of Si-Wu decoction (composed offour herbs Radix Rehmanniae preparata Radix AngelicaeSinensis RhizomaLigustici Chuanxiong andRadix PaeoniaeAlba) [4] Multilayer map of ldquoPhenotype network-Biologicalnetwork-Herb networkrdquo was applied to uncover the underly-ing network systems of TCM syndromes and herb formulas[5] the drug-target network was implemented to elucidatethe mechanism of one TCM formula for the treatmentof T2D [6] Although those applications of computationalsystems biology in the study of TCM formulas are still inthe exploratory stages they demonstrate the feasibility of
2 Evidence-Based Complementary and Alternative Medicine
integrating the biological network and the experiences intraditional medicines for the analysis of TCM formulas
To date graph theory is a primary approach for networkresearch [7] It is viable to study a network graph composedof dozens or hundreds of nodes through visual inspectionHowever it is not practical to analyze a network containingmassive nodes or complex relations between them even withthe help of three-dimensional display techniquesMostmeth-ods developed for complex networks such as the path-lengthmethod and the nodes-distribution method [8] focus onthe topological structure instead of the specific relationshipbetween nodes Therefore it is still challenging to study acomplex disease network with intercrossing pathways andto understand the final effects of the components of TCMformulas based on the biological signal pathways In additionthose biological effects may be ambiguous (positive throughone pathway while negative through the other one) plus someproteins affected by the components of TCM which have notbeen identified as disease targets In order to solve these issuesand to discover components with positive effect from TCMformulas we proposed a new approach called directed TCMgrammar systems (dTGS) to identify effective componentsfrom TCM formula based on an entity grammar system(EGS) In dTGS the TCM component-protein network andthe disease network are viewed as grammar systems andthe ECGs can be identified through syntax rules Bai-Hudecoction plus Wasting-Thirsting formula (BHDWT) wasselected as an example in this paper to illustrate the basic ideaof the method
2 Materials and Methods
21 Definition of dTGS Model in the Framework of EGS EGSis a formal grammar system that aims at complex biologicalsystem modeling [9] Because of its scalable feature EGShas already been used to establish the flow graph models ofchemical processes [10] and to illustrate the mechanisms ofTCM [11] The details for establishing a specific EGS weredescribed in [9] and are briefly summarized here
An entity grammar system 119866 is a quintuple 119866 = (119881119873
119881119879 119865 119875 119878) whereas 119881
119873cup 119881119879= 119881 119881
119873is a finite set of
nonterminal symbols 119881119879is a finite set of terminal symbols
and 119881119873cap 119881119879= Φ 119865 is a finite set of relations for 119881 119875 is a set
of rules to deduce relationships between entities and 119878 is thestarting entity
dTGS has the same structure as EGS set 119881 containsdifferent types of nodes (compounds proteins T2D apop-tosis inflammation etc) set 119865 contains different types ofrelationships between adjacent nodes set 119875 defines the rulesto derive the relationship of nodes as described by thefollowing
119881 = 1198811cup 1198812cup 1198813cup 1198814 (1)
1198811is the set of the compounds in TCM formulas 119881
2
is the set of proteins in the disease network on which thecompounds in TCM formulas act directly 119881
3is the set
of the rest proteins in the disease network and 1198814is the
i
m
k
T
b
h
f
e
c
A
g
d
Figure 1 Network used in dTGS for deduction Red triangle nodechemical compound blue circle node protein ldquo997888rarrrdquo one chemicalcomponent or one protein enables the expression of the next proteinraises the expression or enhances the activity of the next proteinldquo⊣rdquo one chemical component or one protein inhibits the expressionof the next protein lowers the expression or weakens the activity ofthe next protein
set of nonprotein nodes in the disease network (eg T2Dapoptosis inflammation etc) Consider
119865 = cp (119883 119884 119885) pp (119883 119884 119885) draw (119883 119884 119885) tag (119883) (2)
In cp(119883 119884 119885) 119883 isin 1198811 119884 isin 119881
2 119885 isin pos neg in
pp(119883 119884 119885)119883 119884 isin 1198813cup1198814 119885 isin pos neg in draw(119883 119884 119885)
119883 119884 isin 119881 119885 isin pos neg in tag(119883) 119883 isin 119881cp(119883 119884 119885) defines that119883 (compound in TCM formulas)
acts on119884 (protein directly reacts to compound) with an effectdescribed in119885 pp(119883 119884 119885) defines that119883 (protein) acts on 119884(protein) with an effect described in 119885 tag(119883) labels nodesof interest draw(119883 119884 119885) extracts tagged nodes and theirrelationships defined in cp(119883 119884 119885) or pp(119883 119884 119885) in dTGS119864(119881 119865) refers to all entities composed by the elements in 119881with the structure cp(119883 119884 119885) pp(119883 119884 119885) draw(119883 119884 119885) ortag(119883) Consider
119875 = 1198751cup 1198752cup 1198753cup 1198754cup 1198755cup 1198756
1198751= pp (119883 119884 pos) pp (119884119872 pos) 997904rArr pp (119883119872 pos)
Evidence-Based Complementary and Alternative Medicine 3
q9y617
p30566p30086
q3lxa3
q7l266
q9ui42
q99489q6pi48o60391q9p2j5
p78330
p14868q16773q96ab6p45381o15399q9nsb8
p48664
q9y5z0
p23515q5tdp6
p48549o43555
p08243
p36957
p15153
p05166q9bzw2
q13488
p21579
q8wz71
q96i99q02338p48200p02812
o14925q6pce3
q9y5j9
p10515p05108 q9bxa5
p23378q9ujm8q86wu2q96s66p12236
q9h3h5o43837
q9ubx3q9y276
q8n907p0cg04
p31937
p16112p12235
q01813
p56524
q6ibs0
p50213q06124
p48039
q96hd9
q92542p22234
p47897
q13155q8n142
q00577q15046q7z429
q9bvj7o15239
q969s0
q8tdd2p03905
q53gq0p17858
q9h9p8
p04839
q8nbi2q9nz42
p31944
q9ui32
q96rs0
p41252p51793
o75306q9nvh6
q9y2z2
q0vdg4o43181
p52564p48764p19634q9ugb7
p13725q9btd8q49a26
p32856
p05113
q8n0w3
p48165
p02750
p47929
q01433
p09972p04075q4u2r8
o00462
p08574
o14638
ts00140
q8tct9
o95139
q8ne62
rs00003
p22830 p14927
q9uig4
p49447
p02766p04062
q9y215
p20336
q07837
q96t21
q0vdf9
o94916
q16623
p14550
q6ub98
q96h96q9y436o15460
q8ivl5
q5t6j7o75251q7z4n8p34896
q96ta2
p06576
q9p2r7
q96q83
p39060
o60568
c9j9i1
q8nhu3q8tct1
q9gzz6
p32320
q9y227p56539
q9ukg4p01730
q13045
q9hbg6
p30411
o75563p25208
o14556q8nci6
p53597
p25705
p35080q07912q9uhl4
p48163p49286
q9nti2p23368p56134
q9p2j9
q6uwu2
p02708
q92478
p36544o43405
q9hbu6q14498
q5m775q13510
o00754
q96cw1q13905
q9hat2
p19235
q9ntj4q9y2e5
o00220
o94850p51688
o95471
q96ri9
p78356q5vx52p01008
p09693q9y616q92636p26715
q9y6x0q01459
q8n5d6
q86wv6p19438q9bzp6
q96hz4
p08727
q13601
q16552
p10809
p25942o60759
p29033
p13236
o15455
as00275
o94864
q9y223q8n4e4
p14210p50591
q13158q9np84 p08571q8tdq7
p43403p20963
q9uhj3o75382p08912
p16152q9uha7
q9y5z7q9ubh0
q9nzh7
q5t280q9nvg8q15677
p42330p48730p34949p05452
q8wwz1
p49674p49221
p08913
q9nzh6q8iyr6q9nzh8p01583
q9bzz5
p11233o60870p21802
p17676
q5tba9
q9h4b6
p23771
q5vzb9p43405
p30273q9bzx2
p23769
q8wxg6
q12792
q13287
p27930
p04234p07766q9uj70
p49590q9ha47
q02223
p18428
p15586
o94808
q9p0l0p35232
q9ui33
q9y573p42701p43251
o95259o00590
q2tb90p49354p53675
p01266
q00610o75600p01857
q7rtx1p78316p07355
p01106p20382o00442q15825
p21439q53gd3q9nvf9q96rl7
q13822q07001
q9ui17q9bvg9
p01344
q86uu9
q00994q8iwa5p63027q04844
q05315
q6p1j6p17787
q15822 q9ul12q05901
p07510
p15421
q9nyc9q7rtx0
p07948o00244p29762
p02787
q6uwr7
p43681p04080
p12830q9ugi8
q16134
o75955
q14254
p52788p11473
p02452p49356
q9hcg7
q96pd5p31513q8wz75o15244
q8te23
q9h8t0
q9y3i0q9h3z4
q9ui36o75197
p10606p13987p01834o75792q13515
p27824p22413
q9ubm1
p81605
o15245o60271
q9npp4
p54760
p62244q9brg1
q14410q16878q99471p85298 p38606
q96t52
q13952
q6uwm7o96004p13942
q9haz2 q9uqp3p24390q9npg3q92922
q14409
q9uhv8
q9ugm1
q00325q9ug56p11230q9gzv3
p32297
q7kz85q9y6m1q96bw1
q969g3
q8n5y2
q16595q15049q7z353
q96fz7
q96s36q9hb07q8tb68
o60260
q9p1t7
q9ubm8p10109
o75884
p16442
p08922q96cf2
q9by43
q9y3e7o43633
p35520
q13163p01591
q9hci7q68dk7q96lu5q9nzz3
q5t870
p11717
q9ns23
q9y6k0o75751
p49789
q96h20
q969l2q9h444
q8ni17
p01574p01563q8ncc3q12794p01570
p61244p05014
q86wn2p01571p19099q86wc4p32881
p09038
q96rq9
ls00113q8wxi4
gs00298q99616o95810
q8tdu6
q9y4g2
o15554q92597
q9p232
p21926
q7rts9p51787o60674
p03923p10914q09472o60443
p41212q5eg05
q6zna5q14151q15424o95718o14990
q16637q9npd5
p15538q9hau4p28676
q14249
o15120p17927q6zmt1
q8wua8q15878
ks00036
gs00302
q92753q00975
ls00110p00441
q9ubh6p10827
p68400
q12762p19784
q13522
q9ujy1bs00293q96t58
p80365p40313
as00350
q9y6q6as00353
p11171
o60499
q8n7p1q96c36
q13336
q8izw2
p31645
q16587
q8wyr5q00613q9y4k3
p84090 q14790xs00013cs00499
o95206 p20333
ls00125
cs00498
p14735q9h173
p06401p04150
q9nrz3
q86uw1q5sw79
q9bxc9q9y2g4
p38398
o15427q96n87q9bxs1
q02318
q9bxc0
q8nej9
q16719
p31751
p48552cs00875
p53985q96ru7
p08134q16342q03001
o00626
p47992p05787
p30519
q8n0z6
q13490p04196
q13794o95140
o00399
q16621
q14721
p56470
p48147
p07949
q8iwl1
o95342q5t5p0p37840
p55957 p49675
o75444
p69905
p04049
q99884p32942
q9nyl5p51659o43866o95750
p37268
q01581q6tdu7
q9upn6q12770o75881
q92496q99487
o95445p43034q6ux71
ls00187ls00188
o00353ts00144
o60906fs00061
q6rui8
Glycyrrhiza uralensis
p03372
coptidis rhizomaq9nq94
is00086
q92887
ls00105
is00187
p51671
p21397
se00911
q53rt3
q9y4x3p17516
o43236
gs00307
p52895
p02679
hs00046
anemarrhenae
ls00122
q14314
o60725gs00305q14432
p42345p10242
q03591
q13370q15493p51530
q9h2c0q96ap7
p42126
cs00415
ls00127p42679
p19623
ms00103
is00211
p06493p20248p78396
q16678
ls00183p11511
ms00035
os00089
q96du3
q14315q8iy17
p10745
q9uhj9o00501js00007
q75t13
q5t8e0
is00181
q14258
q96d42
q8n695
os00007
ls00106
p11182
ps00017
p80098
q01415p31040
rs00138
p01033
ns00060
p35557
ms00098
ss00414
q16625
q07157p57739
q93088
p14410
p04745 p24298q8td30
p16278o00443p17540
p06732o14756
p12532
p08519
p23416
a0jlq5
gs00279
q96c90
q06278
q99572
o14649
p04818
q9nws1q9hc52
us00016
q9y2c4q96eb6
p43490p38936q5syt8
o15294p16581
q969s2
q9y6l6p30307
o95644
q9gzz1
p04035
p09874
p08697
p11166ss00388
ns00013
q04828
p16284
p11802q92934
p46531
o14791
q9uq10
p51452
p19174
p17096
q8nec7
cs00212p78381
p55316q86ug4
q16762
p52565
ns00227
p14635
p04085
q92731p25963
a4fu51
p15408
o14763q8n8n7
se00354is00186
q86xx4p11245p09429q5t7c0ss00115ss00290
os00009
q9nzc2
ss00165
gs00240p28799
p18847
p02771q07817
p23942
p54259
q5fya8
p22460p19525
q14164q13489
p39877
q02156
q10588
q9y5q0
q8wxh0
q86vf5
q9npg2
p35637p50225
q9bww8p32929
q8ius7p54257
q6q788
p34982q8wwm9
o95866q96s37
q9y697q14686p07498q5hyw3q5snw6p49019q8n6m5q96n76
q92765p09105p05090
p14616
q9bzm1
q9nyp7
q9uf11
q03181
as00429es00030
p35568
q8tdv5
q9y2u9
p02647
p04271
q16517
q86uk5
p02656
q16850q9h3y6
q9y226
o96017
p30279p30281
q9unu6q14746
q9bxh1q9una4
p57679
p43351
q14693q9y239p54868
q2tac6q9nvs9
p20062p02652q9uj68
p05129
p09960
p34913
q9byj1p28300
p05121
sd00054
q9uhc9
q9y6a2
q9ubu8
o15218p48775
o60239a0pjb0p07327
q9unh5o75469o14683q9ukr8
o75293 p19320
q9nxr7
q9upn3cs00903
o43464q92859
o94972
p34969p78423
q5sxq6
q92838p30044
q96c86
p62491
q8ixh7
p78545
q9hab8q9y2i1
q9nq25q9y6a5p62508
p30043p18510
q14671
q14457q9ukg1
q96ir7
q92583o15162q9nrp7
p84098p50613
q13950p59665
q5t7p3o14920q14145
p98170o75112
p11474o95831
q9y6k9p47944
q9nqx6
q9bq39
q16769o75365q8iwl2
p05455
p00450
q702n8
gs00067
q8tcu4o14972q02161o15068p57735
q9uja3
o15553q15596
q8tb72
q14773
q96hc4p01127 p04921
p0c0l5
rehmannia dride rhizome
p00505
q6uw88o60733
q16549p41743
ws00006
q9haw9rs00107bs00006
p40261
gs00192
p14679
q93038
p07478 p07477
bs00002
q8ner1
p06213
p07585
p16435
p09601
p08246
q5tka1
q5vv67
o95864sd00083
q92911
p04054q9p1z2
o15296p28907p10632
o60282
p33527
p35610
p02753
o75908
gs00150
p16662
bs00188
p33316p18054
p09917
q13427
p05177
q96rn5
q6zwj8
fs00063
q99735
p02649
p47712
p60568
p35638
q05655
p06400p00395
q9umx9
p16050
p48506
q04206
p35228
q00653
p17252q9uj83
p60321
p25440
p07741p50991
p02788p02458p52789o43463
o14793q15582p29459
fs00041
p45984
p27815
q96pn7
q9bs16
q9nr28
ms00083p08172
gs00239
p11229 p15121p17174
p69892
p01579
q9unq0
p00390
p13674
se01327
q15800
p35225
cs00312p27361
p22301
q15759q99707
p80370
o43852
p53778
p12931
ms00089
p26447vs00026
o00142
q15546cs00977
p32754
p01189
q53fz2q15758p28482
p00747
q8tet4
q9bq90
p35542
p21912
p10600p01562
q14116p29460
p08253
o00292gs00238
q06330
p17302
q01201
q16585
p02768
p05771q05513
p29474
ls00115
p42574
p08183
q32p28p29372
p01023p46926
as00349
is00241
o15440q8ng68
o14798o75886
q76g19q9ul42
p57730
p51816
q8n5j4
q6zut3p43307
q96pd4q9npj8
q92772p01566p05000
q9p0w0p60508p60981 q9ubc7
p01569
p11678q9bqq7q9nrr2
p46976p52926
q16659q86vu5
p01215q9uqf0p30041 q9nzq3
se00987p08235
q13231
cs00298
o43765
q96ba8q14c87
p28845o95251
as00351
q9y6h8o43395
q04864
o00327
q8td08
p31152q9bpy8
p23560q02153p36897
o94760 q9gzv4p51841p14902
p52569o43508p33764
p63241b6zdq1
q9y619
o75343p17342p25106 q9bxl6
q9nwm0p54108p33402o43272
o60684q09013o00505q12979
o00629o15131p52292p52294
q8wwu5
q674r7q9np91b6zde8q9y2j8
p14652
p20930
p51648
q3c1w8p06733q8wy07q8neu8
q4w4y0p82251q96br1o60828
q7rtu3q02108
q9hbh5q15077
p13521
q8nbx0
q15672
q9y5l3p30793
q7rtt9
p05186
p16562 q5t6x5q02846
o14836p49683p0cg37q9nyz2p07900
q9bxi2p20594
q9bxl7p24158p16066p31327q92886q8izt6 p59046
q9um07q9h497
p31321
p22694
o43708
p11310
p28330p49748o43772
p98077q969e2q9ud71p11309
p49642
q9by32
q9ubf8
o75533
q9nun7
p33765
p20839
q9ubl9
q05639
p54107
p01350
p23526o00555
p51575
o14965
p51161q13085o00763p08247
p22676q05193q05469
q6u949
p00558
q05cv2
o14544q92847
q9h2w2
p54315
cs00064q8nfu7
a6nk58p55084
q13237
p56192
q16854
q9ul49
p12268
o43175
cs00334
q9hau5
p27338
p31644q08397
p01374
o75874 p11498
q96fz2
p09543
p21918
q5hyk3
p42356p21781
p26678q14654
p17480p41231
p14416
q13126
q71rs6p49639q8n8r3
q9y234
q86yv9
q93052p54289o43865 q13573 q9hb75b3kn93
q9h477q9udr5
q13247
q9y4k1q04671
q8tbp6
q7rtu1q15637q9hcu5
q9ulj3q8iyp9
q96s52 p14222p26885
q9y5k5
q7l5y6q7z7k0p50120
o14493
o43193
q8tcc3o00232p55769o95965
q99590q99732p51665q15021q96cm8q06889p09958
q504t8
q15287q9upz3
p21953q86vi4
p51810p17643
o95059q5h8a4
q8nfz5
p40967
q13509p57729q969f9
p48304
q6zsm3p00790
p42765o94788
o43766
o60641
q16186p11308
q16740
q6p1a2
q9bpx3q9nx24o14625
q03014o00233q9p0k8
p28066q8iwt6p98153q9nqe7o95461
p62633 q8wwy8
q13765
q9bqr3p49682q01970o00231
p55036p62333q9unm6 q15003
p62988o95841
o75800
o75832p19883 q9ns56
p26436p31153 q01081q86sq9q96s55
q9uff9
q9nrb3p09172
q00266gypsump15018p52747
p00540
o00571
p35222
o00116
p60484
q9upi3
p84077
q08050
p21730
o95249
q14201
p62191
o60664
p10768
p07360
q9y5x9
q86z14
p20936 q92506
q8n6m6a2myd5p18859
q6p3u7
p10275
p22894
p06280
q9y259
p40926
p11686p01185
p05112
q13164
p49961
o15173
q9gzt9
p29972
vs00017
p01236
p14138p53396
q15165p78542
p13501
p49585
p11926q9byz2
q02388
q9y5k3
p35790
p04629
q14749
q4le39q13423p07550
p01138
p07998
ms00068p01178
p61073
p00480
p49915
p40126p11177
p55263
p00966
p04424
p39086p14920
q02127
q99798
ls00054
p28222
o75746
q03154
p28221
o75936
q8n4m1
p07339
p00338p07195q6zmr3p04181q5t2r2
q9ujs0
p07864
p30990p22061
p09622
p22033
p30520p49189
p98174
q96e11
p54803q01432p23109
ss00413
p22303
a8k5i5
q14739p40925
q14957p63252
q86u10q12904
q9h2b2
p49768p10144
q86yt5q13183q9uld0
q9h6q4q16572
q9y600q16445
p30838q8ndx2
q9ubs5o75791q9h2x9
q14416
q8tdu9q9h992
q99928p05937
o60296
q05586q14833q13948
q13620
p30531p30536p18505
q9p2u7q12879q9p2u8q16478
q13003p43005p01148
p41594
q03518p43004
o60220
p54886
q13224
p51649
q86xe5q9y5l4
p06737
p80404
p11216
q9ula0q8nhs2
o75027
q12797
b4del5p07101q53ga4
p05165
p43007
q16099
q99259
p20472q13255p43003
q9nsd5o75311o95166p49747
p48058p48167
q9h237
q8n808q96g03
p13498
o00757o00217f5gxe5
o95394
q02978
p09110
q6iq20p80108
p01275
p62158
q6gpi1q99895
q9y5k2
p60893
o95467
o95749
p14324o76082
p27169
q9bv23q8wy41p60323
p35558
o14939
p13569
p34059o14578
p01024
ns00183o15360p35030
p10620
q16698p01308
p05305
q96s42
p01584
p47710
o95347
q16822 q7z2e3p29466
q13867
p24752
p06132
p06681
p08174q05066
q8wwi5
q16798
p19440
o43451
q13326
q13093p60174
radix trichosanthisp05814
p07814
p00734
p13726o14495o43688
p03891
q8n726p09848
o95503
q6ns38
p33897q7z5q5
p00709q9bxw9p50336
q02809q15735
o14832
q9nui1
p55199p52209
q96bz4q8iv08
p23141
q13393as00075
p43155q695t7
q9nrd0
p11766
q96rq3
q6pf09
q9h400
ps00173
o75390
p14618q16778
p14136p28838
q8wxs4p20396
p06307
q9um01
p07954
o15146
q03519
p48735q92973
cs00392p01100
p30613p06753
q9y5q6
p49448p00367
q7z569 q6uwp7
p15104
p30542p13798
q08ah1
q9byc2p55809p09683
o00330
p31939
q6ibw4p42695q86xi2
p10636
p21399
p78504
q9ntj3
p21728
p00439
q09428
p38571p10599
q93099
cs00067q04609
q13519q9h2s9p01303o94855p60520q7l0j3
p08908
p78334q401n2q8n1c3q14832
p11086q8wvc0
p56178p07108 p05091o75899
p17752
q7z2h8
p18507q8te85
p78509
p28476q13619 o00222p04217
q9brk5
q05940p50553
p42263
q9umx0o14764
q9upv9p23381
p01213p48169p41250
p28472
p49238q9h598p19801p47869p34903
p78310
p48066q6ib77
p14867
p24046p47870
q9nqx3
p23415p56177
p07320
p42261p47895
o00522
p30038q96nt0
p04207q6tgc4 p02144q07283
p11831
p10826 q9bze9
q96a70
p58215p54920
p53602
q5vyx0
p11274
p35611
p25054q96hn2
p29274p49902
p30837
p02686sd00044
q15842
p09923
q96jp5
q9npb1
p48047p01588
p29275p43119
q8n7h5p21549
p01282 p42262
p05451p05089
o43612p78540
q8n159
q96g91q7z6v5q9bzd2p27707
p55265q9bub4
q9h0p0
p54851
p05187
q16620
q13976
p51580
p47900 q99571
p78563
q9y3b8q9bse5q9ulw8
o00487p55735
o95865
q13505
p21589
o43868q9bxi3
p35251q8tcs8q9h244
q96sz5p16109
p09603
o00337
q9ns39p39905o43318
p16220
p22079
p13866
p00746
q00796
p25445
p01009
p48065
p55011
q9c000
p04746
p58012
p19961
o43280
p05062
ps00098
q14397
q03393
p51570
ns00189
q99643
p01241
p43304
ss00357
p21796
q9y3z3
p17735
p11217
p35670
q05329
q9br01q8wwt9
p28329
q9uby0
q7z721
p05019
q8net4
q16851
p45985
q8n699q99759
p15382
q9uj14
q9nr45
p36269
p09919
q9y5p6 p36871
o43488p20800p04732
p02760
cs00201p16591p09467
q16538
p21695
p05067p32189
p00156o00591
q9ulc6
p07288
p06276q9nph2
q9y4x5
p02675hs00118
p57773
p61812
p16070
hs00119
p46098
p02671
o75610
o60701
q9nz01q96hy7
p08237
p35575
p09455
p08559q8iww8
p08100
p43630
o60565
q12772q08257
p99999
q96ha7
p55210
p02735
p07947
p00533ls00097p22466
p11168
p61626
q9h227
p10253q53h76
q32zl2
q16539
p55212
p00492
gs00029q8ivi9p54136
q9h300p28288p10909q9h115p01210q16836
q13145q9y314p27708
p29323
q96jb6q08629
q8iuf8q9y4k0
p50440q9bvc4
q9nr19p36551o95389
q8tas1p01286
q15166
p11488q9nra2p41217
p07306
p28715p43699
p08729
q8nbp7
q9nrc8
q9p2w7
q3kpi0
q8tat5p22674p35900
q9uif7
p08779q5jvs0
o43542
q9uk23
p23946
p48061
q15468
q9byv7
p10586
q3ybr2
q9h2h9
p22614
o15121q15399
o14678q99447
q12873
p62805q9ubi9
q14191p06731o95948
q16531
q96t60q13569p17661
o60673q99607q12809p48751p32971 q13315
q9ubz9
q9nti5
q9y6m5p09327
o75794
p49281
p07307
q9c056q06430
o43524
p13611q96q15
q92630q01959q16891
q9y272o60936
p34972p41143
q05516
q9y5s1
q86yb8
q9nrm7p19021
q13258
q9c026p09238o95727
p51812
p57059
o15297p23025
p09488
q9ntg7q7l1v2
o15350q00839
o75528
q15181p33993
q9hba0
p67809
o14735
o00519
ps00182
p49757
q13470
p13073
p20719o15409
q13547
q8wx93q96q91
q9nrg4
p61567
o75928
p49959o15392
q12834
p48651
q00534
p08263
p09211
cs00649
p16104
p02790q9un88
q9y6f1q13156p39748p07992
p06702o15354p16455
p18887
q92889q96fi4p08670
p18074
q9nwh7
q01831
q7z7h5
q6p1n9
e5kn95
p54278q92698
p15088p19013o15527p13051q96se4
o14757
q9ugn5p49917q96qu6q59ek9
q9ukk3q86vb7
p16422p28906q9ugp5
q9nuw8
q86wg3
p49916p04233
q8ivs2
p22083
p46939q96ja1
p40692
p18858
p16150q86yl7
q13421o95271
q8iun9
p35408p54277q9ns91
p23396
p28336o60907p23975p12956
q6gmr7
q93009q9uhb4
q53lp3q9h2x6
q96he7
q96a56q9nrn9
ps00183os00096
p10721
q16611
q96qu1
q5t4w7
q9gzp4
o00764
ms00241
q16584
o15231
q9h3d4o95997
q9h7z6q15717p37023 q9bzq4
o14980q86wx3
q08431o75083
q63hm1q14186
p00738q969k7
q9h8v3o15229
p30466p68402
p02655q16609
q8wwy7q9nww6q96i15
q8tcv5
q9gzn6
q8tb61p49247
q9bw04p18405
p46952
q3t906
q9ubm7q15102
q96t66q8ixj6q8tds4
q9nxa8
p48023
q9han9
q9h221
ophiopogonis tuber
q6xqn6p29401
q9h222
p43220
q9y5s8
p06727p08138
p11021q13426
o43819p21980
q9h3s4
q9y6e7p05546
q08945p26358
o95155
q9buj2
q8nb91
q9bzz2b4dln7
p33121
sd00085
p53634
p07148
o00767
p01130
p55851ps00024
p15692
q15303
q9hb55
q9bx93
p33261
p41159
q92993
q9nz20
q9ubk2
q9y618
p49792
q7z6w7p07359
rs00133p05412
p07099
as00063
ss00282
p20813
p04637
o76074
is00262ns00109
q8izm9
q99217cs00029p24385
q07820
o95477
p11712p00813q15848
p00491
p05181
q59f19
p10415
q13133
q86v15
o75342
q03013
p08684
q96kp6
o75477
p04798q13829
q6p589q07812
p21860
p11215
p02749
q9nrf9
p08758
p11387
ws00005
o14788
p37837
q03169p54098
p35354
p45983 p18440
scutellariap06746
o95379q5gj75q8wvp5
p15559
p01375
p21964
p34810
ls00171p05231
p19838
q8tdx5
p35869
q9h7z7
p23219
p05362
p20815
o95622
q8tdf5
p14555
p48357p06858
q8nf37
q6p9f0
p20933
p82932
q86ue4
q9h2a7
p08603
p53779
q92186
p00739
q14289
p55087p54252
q14376
o60218
p50053
q9nxz1
o43914
o95968
p15260
q04446
gs00157
q9uj99
q08499o75355q9ugm6o95500q7z2y5p36382p35212q96kn9 q5t442p50747
a6nfd8se01038o15195
q13546
q15628
q9y285
p06850
q9hc78
p83916p08034
q5nv69q9bx67
p57087q92959
o43474q9nqz7
o95169
q06210
o75054p01589
p07315
q8n1n0
q8wwv6
p09525
o75712p61421
q92851
o15439p37059
p52790
q96gd0p07902q9y6a1
q96qt6
p15311
p22732
p46734
q9bqt8
p42226p04083o95452p23582o43396
p43652p78562
p29376
q92616q9nr09
o14521
js00028 p01137
q15004
p04406
o14494q03135
o15264
p01160
p00568
p04040
p01270
p04179
p06729
p61278
q16665
q9hbf4
o14727
p11413
ss00360
p36383p13500
gs00300
p14780o95479 p08236
ps00477
p00797
p05164
q96ap0
p49419
o43306
p06744q06609
p55211
q15392
es00154
q14994
p20702
o94806
p33260
q8n5b7
p22392
p10163q8wtv0
p78380
p01133q5hya0
q9y2z9
q96gr2p21583p04053
o60346
q13526
p53801p00750
ns00107q14534
p18031
q15274
q96ri1q8izn7
p55055
q8n104
q9y2r9
p49888q9ubb5
cs00128
us00005
p22680
o15197
o95544
p02741
p24941
se00229q96pd7
p11509
o95905
q969g5
bs00084p02008
p27797
o95411
q9y5b9
q9nqa5
q9nyg8
p60510
p78524
ls00212
q05823
o75310q92839
o75795
p36537q9y4x1
q16411p30405
q9npz5p54855
p10619p08833q9np50
o14503
q15031 q9hcn2
q6uwm9
q8iwy9
os00093 q16696q9h5v8
p26440
q03468q12778q6ia69
q15047o94776
q13464q00987
q12888
q86vz5
p06133q9by64
p24903
q92819o94910p12955o15169
o94766
p48449
q14677o00219o00560
p61086p78549
q9h2k2
q96kb5
p36639
q96lt4
p19447
p19544
p52701
q9bzj4
q8tdg4q12980
q96rr4p28908q96fl8
p40198
q14683p07093
p08575p09936
p15941
p09210
o60760p35754q9np99q04760o15217p23921
q9h4y5p21266
o95881
p30711q8ted1
q16772
q7lg56p48637
p31350o14880
p0cg29q16775
q7rtv2p78417
q8tdv2q92600q15904
p30953q9npf7
q99685q9y316
p35913q5vwk5
p22891
q14690
q5sci4q9byb4
q16236
p28340p10646
p32780q9uq84
p27695
p07204
q9y385
p12004q96la8
q5xg87
q13352p09884 q86sq4
o95243p09104
q8tdm6q9ubx5
p41219q9y4e5
p56180
p02545q676u5
o95069
p09382
q92820
q9nzj6
q8net8 q8nhy0
p32418
q8wux1
q9np80p51572
q99807
q9up65
p06127
p0c869
p34932
p41161o43242
q9hay6
p53370q8ndx5q9gzr5
p59796
q12974
p10243p20700
q07108
p18283
o00180
q99218
q7z449
q8ner5
o75665
p98155
p53041p22105
p55268
p51582
o15496
q8nff5
q9hcm9
q86sj6
q86tx2p33176 q9up95
q5fve4p15144
q9unk4
p25874
o15254
o14842
q8wui4
q96cc6p24462
q12791q9bqi3
p31431 q8n6t7
q86yn6
q5nul3
q99873p25092
q99808
q99698
p49753
p49591
p08709
ls00058p68871 o15540
a0mnp2
p55916
p46527
o95573
p02751
q9nud9
q16655
p36969
q04759
cs00116o15151
q13790
q8n1l4
q9uid3
p58743
q02928
cs00989
cs00880o43808
p17900
p20142
q9npf0
q9uku0
ms00393
p22736q9p296
q6p5z2q15139
q13508p22307q96l15
p24723q9bzl6
p52961
q9y5v3o60603q99729
q7z408q9hcc0
q99445p31946
p13584
p16401
d3dvq6q75wm6q02539q8iv61p11142
p61204
q8td17o60547q573b4p84243
q99570
p07305
p19878
p46094
p41235
o14645
q9btm1p0c5z0
o43399q9nrw4q06265
q3syg4
q8wvc6
q63hq0q16853
p30301q9uql6p19484
a5pkw4q9by66
q9nr22q9ukl6
p17677p15151p10412
o60551p50135
p43080o00159
p62330
q96a08
p14770
q96pb1p49006
q9y6e0
a8k1m2q8ivk1o00458p78556
p0c5y9
q9h4x1
p14598
q8nff2
p63098q9ui40
o75367p01112
p57723
p16402q6qhc5
p16403
q9h305
q12884
q9hc58
q13702
q9ubs4
o00622q01101q99836
q96pz7o95429p55198
q6nt55
o43781
p41238q86tg7
o75715p22352
p02808
p07203q96sl4
q7z407
q9p0m6
q8wwk9p62745
q16829p29966o95267
q7z2w7
p18085p08567
q9h1r2p30968p35243q9bxl5
p23527p22492
q71ui9
q99418q8nb12
p22570
p42357
q9nz09p98194
q9h211p13497
q9umx3
q9ul01
q8nbh2
q9y397
q9nv23 q7z5g4p54840p38117
q96k12
q9y5r8
q8iuh5
o75460q8n6n7
p47974p60880
p0c7u2
q9y296
q8wvx9
p06703q14541
q14542q9nqg7p28039
p0c0s5
p56704o60488
o43918
o95237
q9nzj5
p35556q9nyx4
q9h009q99767
p12259
p40879
p24386
o00472
o43617
p06241
q9nr71o95822
q7l5a8
q99784q15434q9hd26o76015q96rj0
q05084q96mw7
q13371q96ga7
a4fu01
p24666
q01469p61769p13807
q5t215q06203
p17861o76054
b3kp71
p13804
p05198
o15228q96c24q16654
p09466
a2arm6
o00115p07988
o75781
p10145q96ph1
p51589
p37231
q9hcl2
q9y6r1
p04114
p40145
q8nfm4
p16233
o60503
o95969 q8n9l9
p14672
hs00094
p12104
p19835 q8tcg5q08462
p45844
p32322
q9nzd2
p23786
q96dy3
q86x55
p03956
p40939
p55157
q6p1m0
o75159
o14561
cs00250
q9has3p22612
p10644
p30825p31323
p13861
p17612
p61586
q9nwu1
q08426
p36956
p11597
p20292
p15090
p41222
as00249
q86sk9
cs00285
es00010
q9h015
q08828p00167
q9y6y9
q53hv7
q9nzk7se00285
p31749q9uhd2
p04275o75907
o95255 q9bzm2
q15067p49327
p51828
p08311
p29083
q9ubu3
o00628
q92523
q07869
p50416
os00061
p19971o60266
p16671
q9p0j1
o00206
p07205
q6jqn1
q8tcd5
cs00644
p05413
p11150
o00154
q96a54
o43581q14108
q9ulc5
p32238
p07098
p50542
p00414
o95278
q16401
q99541 q86x29
ps00111
p30084
o75381
p30419
q96t53
p31025
p50897
q96iy1
q9y2p5
o95158
q14653
p02774
q96ad5cs00020
q8wts1
p54317q6piu2
o75688
p12643
q9gzx6o43315
q9nst1
o00548q5k4l6q9hd89p10082
q9uhr6
p24557
p55061p06748
q16647
q9uet6
q9npa8p01298
q9y2p4
o60240
q08477
q99988
p20132
p46063
q86y26
q16626q93070o75478
q00978
q9gzx7
p0c1z6
q9ha82
q6ub99
q9bvm2
p37275
p63096
p17081
q9hb03
q6uwv6
as00574
q13123
p02654
Figure 2 Compound-protein network of BHDWT Red triangle node compound in BHDWT blue circle node protein
1198752= pp (119883 119884 neg) pp (119884119872 neg) 997904rArr pp (119883119872 pos)
1198753= pp (119883 119884 pos) pp (119884119872 neg) 997904rArr pp (119883119872 neg)
1198754= pp (119883 119884 neg) pp (119884119872 pos) 997904rArr pp (119883119872 neg)
1198755= pp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
1198756= cp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
(3)
1198751cup 1198752cup 1198753cup 1198754is the set of rules to deduce the eventual
effects of chemical compounds on the disease 1198751indicates
that if the effect of119883 to 119884 is positive and the effect of 119884 to119872is positive too then the effect of 119883 to119872 is positive Similarderivations are defined in 119875
2 1198753 and 119875
4 They may be used
as many times as necessary to the final disease node 1198755and
1198756are the rules to extract the nodes for the network They
can also be used to draw the subnetwork when the originalnetworks are too complicated to analyze 119875
5indicates that if
119883 (protein) acts on 119884 (protein) with the effect of 119885 and 119883 istagged then this relation is extracted and 119884 is also labeledfor further derivation 119875
6is similar to 119875
5except that the
starting entity is chemical compound 119883 In the construction
4 Evidence-Based Complementary and Alternative Medicine
PI3K
AIFENDO-G
CASP12
TRADD
Cn
stresssignals
MYD88
calpain
JNK
LEP
IL-1R
IL-1
GHRP
G6PT1 GIPCPTI
Beta3AR
IRS
FFASOCS
AA1R
NGF
FATCD36 TRKA
FACS
SGLT2P K C
TNFR2 TNFR1
erstress
GLR IAPPmNCETNFA
INSR
SHC
INS(hyperinsulinism) PTP1B
LAR
APS
ATMIL-3R
D-GlucoseDNAdamage
D-GlucosehyperglycemiaIL-3 AMPK TNFa
GLUT2
BAX
RosFAS-L
PDX-1
FLIPTRAIL-R
APAF-1
MafA
FAS
BAD
CASP10
P53 BID
CASP8
GK
FADD
TRAIL
DFF40
DFF45
CASP3
CASP6
CASP9
CASP7
cytc
cleavagesubstrate
DNAFragmentation
impaired insulin secretion
Pyruvate
apoptosis
PYK
INS
ACC
lipogenesis
ATP
N F B
I B IKBA
IKK
BCL-2IAP
PDK12
SREBP-1c
GLP1R
IRAK
11HSD1 UCP1
P K C
UCP3
T2D
glucolysis
SOCS3
P G C 1 -
mTOR
4EBP1
STAT3
lipidhomeostasis
S6
- M S H
P70S6K
NIK
RIP1TRAF2
JNK
Kir62SUR2ASUR1SUR2
VDCC
cb1
CRKII
Obesity
ADIPO
ADIPOR TC10
GRF2
EX070
MEK12
Insulin resistance
RAF
DPPIV
LEPR
Retinoic acid
EIF4E
P K C
PKA
HSL
antilipolysis
GRB2
SOS
GLUT4
RAS
PDE3TSC2FOXO1PP1
RHEBGYSFBPPHK
AKT
cAMP
GSK3B
PEPCK
AMPK
G6PC
AGRP
SHP-2
MNK
CPT-1
Ceramide
RXR
ERK
AMPKK
P P A R
Long-chain fatty acid
CIP42
PYGPPARG
glycogen
glycogenesis IRS
glucosehomeostasis JAK
Figure 3 Biological network of T2D Blue circle node protein purple circle node nonprotein node green octagon nodes type 2 diabetesldquo997888rarrrdquo one node enables the expression of the next node raises the expression or enhances the activity of the next node ldquo⊣rdquo one node inhibitsthe expression of the next node lowers the expression or weakens the activity of the next node
of network 1198756will be used once and 119875
5may be used as many
times as necessary to the targetThe modes of action between nodes include the positive
(pos) and the negative (neg) effects If we define the positiveeffect as ldquo1rdquo and the negative effect as ldquominus1rdquo the ultimateinfluence of intervention will depend on the product of eachstep in the whole signal pathway For example compound Ain Figure 1 influences protein T through 3 pathsThe effect ofA to T is negative through the path ldquoA-c-d-e-f-Trdquo and the pathldquoA-b-h-k-i-m-Trdquo This effect is uncertain through the thirdpath ldquoA-g-h-k-i-m-Trdquo taking into account the negative effectfrom the feedback path m to g In this paper we neglectedthe effects produced through feedback because the role offeedback is expected to regulate the magnitude of effect butnot to alter the overall mode of action of effect So theeffect is positive through the third path After the effectsof one compound on the ultimate node (ie T in Figure 1)through all signal pathways are determined we can selectthe effective compounds based on the desired effect If thedesired effect on the ultimate node is positive the compounds
with positive effects through all pathways will be selectedas active components The compounds with negative effectsthrough all pathways will be ruled out The components withboth effects through different pathways need further analysison molecular mechanism If their undesired effects can becountered by other compounds they may also be selected aseffective components to be used together with the counteringcompounds The opposite analysis will be done if the desiredeffect on the ultimate node is negative Consider
119878 = 1198781cup 1198782 (4)
1198781is the set of entities with structure cp(119883 119884 119885) or
pp(119883 119884 119885) in biological network of disease which are thebackground for deduction 119878
2is the set of labeled compounds
or proteins expressed by tag(119883) 1198782is the initial conditions
for deduction
22 Data for Construction of Component-Protein Networkof BHDWT Formula For decades BHDWT has been usedto treat T2D at the Beijing Guang-An-Men Hospital [12]
Evidence-Based Complementary and Alternative Medicine 5
beta-Sitosterol
Baicalein
Adenosine
T2D
linoleic acidIsoliquiritigenin
Acteoside
Chrysin
Ferulic acid
Nicotinic acid
Lauric acid
CatalpolMangiferin
Berberine
Mannitol
Phenylacetic acid
Scutellarin
Rutin
Succinic acid
Wogonin
Figure 4The effect of compound of BHDWT on T2D Red triangle node chemical compound green octagon nodes type 2 diabetes ldquo997888rarrrdquoone compound has the positive effect on T2D ldquo⊣rdquo one compound has the negative effect on T2D
BHDWT has a positive effect on blood glucose control andsymptom control for some patients in the early stages ofT2D The BHDWT formula consists of eight herbs includ-ing gypsum Anemarrhena asphodeloides Bunge rehmanniadried rhizome radix trichosanthisOphiopogon japonicusKerGawl Coptis chinensis Franch Scutellaria baicalensis Georgiand Glycyrrhiza uralensis
The components of the BHDWT formula came fromthe Traditional Chinese Medicines Database (TCMD) [13]the State Administration of Traditional Chinese MedicineBasic Information Database (httpdbsharecintcmcomZhongYaoJiChu) and A Handbook on the Analysis of theActive Composition in Traditional Chinese Medicine [14]
The compound-targeted proteins were derived from theSTITCH system (httpstitchemblde) [15] By entering thenames or identifiers of compounds or proteins of interestSTITCH provides the list of proteins withmatching or higherconfidence score that the user specified up to the numberthe user specified The required confidence score representsthe possibility of interaction between the entities In orderto obtain more general results the parameter of the requiredconfidence score was set higher than 0 and the interactingentities number was set to be 500 The interacting entitieswith clear mode of action (positive or negative) were chosenfor further analysis
23 Data for the Construction of T2D Biological Network Toconstruct the T2D network we used the data collected fromthe Kyoto Encyclopedia of Genes and Genomes (KEGG)and therapeutic targets database (TTD) KEGG lists thesignal pathways related to T2D and TTD lists the chemicalcomponents used to treat T2D The biological network ofT2D (Figure 3) was constructed using the signal pathwaysfrom KEGG the chemical components from TTD andthe positive or negative relationship between targets andT2D from STITCH The networks were visualized with thesoftware Cytoscape [16]
3 Results and Discussion
31 Compound-Protein Network of BHDWT Formula andthe Biological Network of T2D We first constructed thecomponent-protein network of BHDWT using the datafrom Section 22 (Figure 2 additional file 1 in Supplemen-tary Material available online at httpdxdoiorg1011552013840427) In this network there are 144 compoundsand 2865 proteins The red triangle nodes represent thecompound in BHDWT and the blue circle nodes representthe proteins The biological network of T2D (Figure 3) wasderived from the data in Section 23 using 119875
5and 119875
6defined
in Section 21 with T2D as the ultimate node Biologicalnetwork of T2D contains 146 proteins The effects of each
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
2 Evidence-Based Complementary and Alternative Medicine
integrating the biological network and the experiences intraditional medicines for the analysis of TCM formulas
To date graph theory is a primary approach for networkresearch [7] It is viable to study a network graph composedof dozens or hundreds of nodes through visual inspectionHowever it is not practical to analyze a network containingmassive nodes or complex relations between them even withthe help of three-dimensional display techniquesMostmeth-ods developed for complex networks such as the path-lengthmethod and the nodes-distribution method [8] focus onthe topological structure instead of the specific relationshipbetween nodes Therefore it is still challenging to study acomplex disease network with intercrossing pathways andto understand the final effects of the components of TCMformulas based on the biological signal pathways In additionthose biological effects may be ambiguous (positive throughone pathway while negative through the other one) plus someproteins affected by the components of TCM which have notbeen identified as disease targets In order to solve these issuesand to discover components with positive effect from TCMformulas we proposed a new approach called directed TCMgrammar systems (dTGS) to identify effective componentsfrom TCM formula based on an entity grammar system(EGS) In dTGS the TCM component-protein network andthe disease network are viewed as grammar systems andthe ECGs can be identified through syntax rules Bai-Hudecoction plus Wasting-Thirsting formula (BHDWT) wasselected as an example in this paper to illustrate the basic ideaof the method
2 Materials and Methods
21 Definition of dTGS Model in the Framework of EGS EGSis a formal grammar system that aims at complex biologicalsystem modeling [9] Because of its scalable feature EGShas already been used to establish the flow graph models ofchemical processes [10] and to illustrate the mechanisms ofTCM [11] The details for establishing a specific EGS weredescribed in [9] and are briefly summarized here
An entity grammar system 119866 is a quintuple 119866 = (119881119873
119881119879 119865 119875 119878) whereas 119881
119873cup 119881119879= 119881 119881
119873is a finite set of
nonterminal symbols 119881119879is a finite set of terminal symbols
and 119881119873cap 119881119879= Φ 119865 is a finite set of relations for 119881 119875 is a set
of rules to deduce relationships between entities and 119878 is thestarting entity
dTGS has the same structure as EGS set 119881 containsdifferent types of nodes (compounds proteins T2D apop-tosis inflammation etc) set 119865 contains different types ofrelationships between adjacent nodes set 119875 defines the rulesto derive the relationship of nodes as described by thefollowing
119881 = 1198811cup 1198812cup 1198813cup 1198814 (1)
1198811is the set of the compounds in TCM formulas 119881
2
is the set of proteins in the disease network on which thecompounds in TCM formulas act directly 119881
3is the set
of the rest proteins in the disease network and 1198814is the
i
m
k
T
b
h
f
e
c
A
g
d
Figure 1 Network used in dTGS for deduction Red triangle nodechemical compound blue circle node protein ldquo997888rarrrdquo one chemicalcomponent or one protein enables the expression of the next proteinraises the expression or enhances the activity of the next proteinldquo⊣rdquo one chemical component or one protein inhibits the expressionof the next protein lowers the expression or weakens the activity ofthe next protein
set of nonprotein nodes in the disease network (eg T2Dapoptosis inflammation etc) Consider
119865 = cp (119883 119884 119885) pp (119883 119884 119885) draw (119883 119884 119885) tag (119883) (2)
In cp(119883 119884 119885) 119883 isin 1198811 119884 isin 119881
2 119885 isin pos neg in
pp(119883 119884 119885)119883 119884 isin 1198813cup1198814 119885 isin pos neg in draw(119883 119884 119885)
119883 119884 isin 119881 119885 isin pos neg in tag(119883) 119883 isin 119881cp(119883 119884 119885) defines that119883 (compound in TCM formulas)
acts on119884 (protein directly reacts to compound) with an effectdescribed in119885 pp(119883 119884 119885) defines that119883 (protein) acts on 119884(protein) with an effect described in 119885 tag(119883) labels nodesof interest draw(119883 119884 119885) extracts tagged nodes and theirrelationships defined in cp(119883 119884 119885) or pp(119883 119884 119885) in dTGS119864(119881 119865) refers to all entities composed by the elements in 119881with the structure cp(119883 119884 119885) pp(119883 119884 119885) draw(119883 119884 119885) ortag(119883) Consider
119875 = 1198751cup 1198752cup 1198753cup 1198754cup 1198755cup 1198756
1198751= pp (119883 119884 pos) pp (119884119872 pos) 997904rArr pp (119883119872 pos)
Evidence-Based Complementary and Alternative Medicine 3
q9y617
p30566p30086
q3lxa3
q7l266
q9ui42
q99489q6pi48o60391q9p2j5
p78330
p14868q16773q96ab6p45381o15399q9nsb8
p48664
q9y5z0
p23515q5tdp6
p48549o43555
p08243
p36957
p15153
p05166q9bzw2
q13488
p21579
q8wz71
q96i99q02338p48200p02812
o14925q6pce3
q9y5j9
p10515p05108 q9bxa5
p23378q9ujm8q86wu2q96s66p12236
q9h3h5o43837
q9ubx3q9y276
q8n907p0cg04
p31937
p16112p12235
q01813
p56524
q6ibs0
p50213q06124
p48039
q96hd9
q92542p22234
p47897
q13155q8n142
q00577q15046q7z429
q9bvj7o15239
q969s0
q8tdd2p03905
q53gq0p17858
q9h9p8
p04839
q8nbi2q9nz42
p31944
q9ui32
q96rs0
p41252p51793
o75306q9nvh6
q9y2z2
q0vdg4o43181
p52564p48764p19634q9ugb7
p13725q9btd8q49a26
p32856
p05113
q8n0w3
p48165
p02750
p47929
q01433
p09972p04075q4u2r8
o00462
p08574
o14638
ts00140
q8tct9
o95139
q8ne62
rs00003
p22830 p14927
q9uig4
p49447
p02766p04062
q9y215
p20336
q07837
q96t21
q0vdf9
o94916
q16623
p14550
q6ub98
q96h96q9y436o15460
q8ivl5
q5t6j7o75251q7z4n8p34896
q96ta2
p06576
q9p2r7
q96q83
p39060
o60568
c9j9i1
q8nhu3q8tct1
q9gzz6
p32320
q9y227p56539
q9ukg4p01730
q13045
q9hbg6
p30411
o75563p25208
o14556q8nci6
p53597
p25705
p35080q07912q9uhl4
p48163p49286
q9nti2p23368p56134
q9p2j9
q6uwu2
p02708
q92478
p36544o43405
q9hbu6q14498
q5m775q13510
o00754
q96cw1q13905
q9hat2
p19235
q9ntj4q9y2e5
o00220
o94850p51688
o95471
q96ri9
p78356q5vx52p01008
p09693q9y616q92636p26715
q9y6x0q01459
q8n5d6
q86wv6p19438q9bzp6
q96hz4
p08727
q13601
q16552
p10809
p25942o60759
p29033
p13236
o15455
as00275
o94864
q9y223q8n4e4
p14210p50591
q13158q9np84 p08571q8tdq7
p43403p20963
q9uhj3o75382p08912
p16152q9uha7
q9y5z7q9ubh0
q9nzh7
q5t280q9nvg8q15677
p42330p48730p34949p05452
q8wwz1
p49674p49221
p08913
q9nzh6q8iyr6q9nzh8p01583
q9bzz5
p11233o60870p21802
p17676
q5tba9
q9h4b6
p23771
q5vzb9p43405
p30273q9bzx2
p23769
q8wxg6
q12792
q13287
p27930
p04234p07766q9uj70
p49590q9ha47
q02223
p18428
p15586
o94808
q9p0l0p35232
q9ui33
q9y573p42701p43251
o95259o00590
q2tb90p49354p53675
p01266
q00610o75600p01857
q7rtx1p78316p07355
p01106p20382o00442q15825
p21439q53gd3q9nvf9q96rl7
q13822q07001
q9ui17q9bvg9
p01344
q86uu9
q00994q8iwa5p63027q04844
q05315
q6p1j6p17787
q15822 q9ul12q05901
p07510
p15421
q9nyc9q7rtx0
p07948o00244p29762
p02787
q6uwr7
p43681p04080
p12830q9ugi8
q16134
o75955
q14254
p52788p11473
p02452p49356
q9hcg7
q96pd5p31513q8wz75o15244
q8te23
q9h8t0
q9y3i0q9h3z4
q9ui36o75197
p10606p13987p01834o75792q13515
p27824p22413
q9ubm1
p81605
o15245o60271
q9npp4
p54760
p62244q9brg1
q14410q16878q99471p85298 p38606
q96t52
q13952
q6uwm7o96004p13942
q9haz2 q9uqp3p24390q9npg3q92922
q14409
q9uhv8
q9ugm1
q00325q9ug56p11230q9gzv3
p32297
q7kz85q9y6m1q96bw1
q969g3
q8n5y2
q16595q15049q7z353
q96fz7
q96s36q9hb07q8tb68
o60260
q9p1t7
q9ubm8p10109
o75884
p16442
p08922q96cf2
q9by43
q9y3e7o43633
p35520
q13163p01591
q9hci7q68dk7q96lu5q9nzz3
q5t870
p11717
q9ns23
q9y6k0o75751
p49789
q96h20
q969l2q9h444
q8ni17
p01574p01563q8ncc3q12794p01570
p61244p05014
q86wn2p01571p19099q86wc4p32881
p09038
q96rq9
ls00113q8wxi4
gs00298q99616o95810
q8tdu6
q9y4g2
o15554q92597
q9p232
p21926
q7rts9p51787o60674
p03923p10914q09472o60443
p41212q5eg05
q6zna5q14151q15424o95718o14990
q16637q9npd5
p15538q9hau4p28676
q14249
o15120p17927q6zmt1
q8wua8q15878
ks00036
gs00302
q92753q00975
ls00110p00441
q9ubh6p10827
p68400
q12762p19784
q13522
q9ujy1bs00293q96t58
p80365p40313
as00350
q9y6q6as00353
p11171
o60499
q8n7p1q96c36
q13336
q8izw2
p31645
q16587
q8wyr5q00613q9y4k3
p84090 q14790xs00013cs00499
o95206 p20333
ls00125
cs00498
p14735q9h173
p06401p04150
q9nrz3
q86uw1q5sw79
q9bxc9q9y2g4
p38398
o15427q96n87q9bxs1
q02318
q9bxc0
q8nej9
q16719
p31751
p48552cs00875
p53985q96ru7
p08134q16342q03001
o00626
p47992p05787
p30519
q8n0z6
q13490p04196
q13794o95140
o00399
q16621
q14721
p56470
p48147
p07949
q8iwl1
o95342q5t5p0p37840
p55957 p49675
o75444
p69905
p04049
q99884p32942
q9nyl5p51659o43866o95750
p37268
q01581q6tdu7
q9upn6q12770o75881
q92496q99487
o95445p43034q6ux71
ls00187ls00188
o00353ts00144
o60906fs00061
q6rui8
Glycyrrhiza uralensis
p03372
coptidis rhizomaq9nq94
is00086
q92887
ls00105
is00187
p51671
p21397
se00911
q53rt3
q9y4x3p17516
o43236
gs00307
p52895
p02679
hs00046
anemarrhenae
ls00122
q14314
o60725gs00305q14432
p42345p10242
q03591
q13370q15493p51530
q9h2c0q96ap7
p42126
cs00415
ls00127p42679
p19623
ms00103
is00211
p06493p20248p78396
q16678
ls00183p11511
ms00035
os00089
q96du3
q14315q8iy17
p10745
q9uhj9o00501js00007
q75t13
q5t8e0
is00181
q14258
q96d42
q8n695
os00007
ls00106
p11182
ps00017
p80098
q01415p31040
rs00138
p01033
ns00060
p35557
ms00098
ss00414
q16625
q07157p57739
q93088
p14410
p04745 p24298q8td30
p16278o00443p17540
p06732o14756
p12532
p08519
p23416
a0jlq5
gs00279
q96c90
q06278
q99572
o14649
p04818
q9nws1q9hc52
us00016
q9y2c4q96eb6
p43490p38936q5syt8
o15294p16581
q969s2
q9y6l6p30307
o95644
q9gzz1
p04035
p09874
p08697
p11166ss00388
ns00013
q04828
p16284
p11802q92934
p46531
o14791
q9uq10
p51452
p19174
p17096
q8nec7
cs00212p78381
p55316q86ug4
q16762
p52565
ns00227
p14635
p04085
q92731p25963
a4fu51
p15408
o14763q8n8n7
se00354is00186
q86xx4p11245p09429q5t7c0ss00115ss00290
os00009
q9nzc2
ss00165
gs00240p28799
p18847
p02771q07817
p23942
p54259
q5fya8
p22460p19525
q14164q13489
p39877
q02156
q10588
q9y5q0
q8wxh0
q86vf5
q9npg2
p35637p50225
q9bww8p32929
q8ius7p54257
q6q788
p34982q8wwm9
o95866q96s37
q9y697q14686p07498q5hyw3q5snw6p49019q8n6m5q96n76
q92765p09105p05090
p14616
q9bzm1
q9nyp7
q9uf11
q03181
as00429es00030
p35568
q8tdv5
q9y2u9
p02647
p04271
q16517
q86uk5
p02656
q16850q9h3y6
q9y226
o96017
p30279p30281
q9unu6q14746
q9bxh1q9una4
p57679
p43351
q14693q9y239p54868
q2tac6q9nvs9
p20062p02652q9uj68
p05129
p09960
p34913
q9byj1p28300
p05121
sd00054
q9uhc9
q9y6a2
q9ubu8
o15218p48775
o60239a0pjb0p07327
q9unh5o75469o14683q9ukr8
o75293 p19320
q9nxr7
q9upn3cs00903
o43464q92859
o94972
p34969p78423
q5sxq6
q92838p30044
q96c86
p62491
q8ixh7
p78545
q9hab8q9y2i1
q9nq25q9y6a5p62508
p30043p18510
q14671
q14457q9ukg1
q96ir7
q92583o15162q9nrp7
p84098p50613
q13950p59665
q5t7p3o14920q14145
p98170o75112
p11474o95831
q9y6k9p47944
q9nqx6
q9bq39
q16769o75365q8iwl2
p05455
p00450
q702n8
gs00067
q8tcu4o14972q02161o15068p57735
q9uja3
o15553q15596
q8tb72
q14773
q96hc4p01127 p04921
p0c0l5
rehmannia dride rhizome
p00505
q6uw88o60733
q16549p41743
ws00006
q9haw9rs00107bs00006
p40261
gs00192
p14679
q93038
p07478 p07477
bs00002
q8ner1
p06213
p07585
p16435
p09601
p08246
q5tka1
q5vv67
o95864sd00083
q92911
p04054q9p1z2
o15296p28907p10632
o60282
p33527
p35610
p02753
o75908
gs00150
p16662
bs00188
p33316p18054
p09917
q13427
p05177
q96rn5
q6zwj8
fs00063
q99735
p02649
p47712
p60568
p35638
q05655
p06400p00395
q9umx9
p16050
p48506
q04206
p35228
q00653
p17252q9uj83
p60321
p25440
p07741p50991
p02788p02458p52789o43463
o14793q15582p29459
fs00041
p45984
p27815
q96pn7
q9bs16
q9nr28
ms00083p08172
gs00239
p11229 p15121p17174
p69892
p01579
q9unq0
p00390
p13674
se01327
q15800
p35225
cs00312p27361
p22301
q15759q99707
p80370
o43852
p53778
p12931
ms00089
p26447vs00026
o00142
q15546cs00977
p32754
p01189
q53fz2q15758p28482
p00747
q8tet4
q9bq90
p35542
p21912
p10600p01562
q14116p29460
p08253
o00292gs00238
q06330
p17302
q01201
q16585
p02768
p05771q05513
p29474
ls00115
p42574
p08183
q32p28p29372
p01023p46926
as00349
is00241
o15440q8ng68
o14798o75886
q76g19q9ul42
p57730
p51816
q8n5j4
q6zut3p43307
q96pd4q9npj8
q92772p01566p05000
q9p0w0p60508p60981 q9ubc7
p01569
p11678q9bqq7q9nrr2
p46976p52926
q16659q86vu5
p01215q9uqf0p30041 q9nzq3
se00987p08235
q13231
cs00298
o43765
q96ba8q14c87
p28845o95251
as00351
q9y6h8o43395
q04864
o00327
q8td08
p31152q9bpy8
p23560q02153p36897
o94760 q9gzv4p51841p14902
p52569o43508p33764
p63241b6zdq1
q9y619
o75343p17342p25106 q9bxl6
q9nwm0p54108p33402o43272
o60684q09013o00505q12979
o00629o15131p52292p52294
q8wwu5
q674r7q9np91b6zde8q9y2j8
p14652
p20930
p51648
q3c1w8p06733q8wy07q8neu8
q4w4y0p82251q96br1o60828
q7rtu3q02108
q9hbh5q15077
p13521
q8nbx0
q15672
q9y5l3p30793
q7rtt9
p05186
p16562 q5t6x5q02846
o14836p49683p0cg37q9nyz2p07900
q9bxi2p20594
q9bxl7p24158p16066p31327q92886q8izt6 p59046
q9um07q9h497
p31321
p22694
o43708
p11310
p28330p49748o43772
p98077q969e2q9ud71p11309
p49642
q9by32
q9ubf8
o75533
q9nun7
p33765
p20839
q9ubl9
q05639
p54107
p01350
p23526o00555
p51575
o14965
p51161q13085o00763p08247
p22676q05193q05469
q6u949
p00558
q05cv2
o14544q92847
q9h2w2
p54315
cs00064q8nfu7
a6nk58p55084
q13237
p56192
q16854
q9ul49
p12268
o43175
cs00334
q9hau5
p27338
p31644q08397
p01374
o75874 p11498
q96fz2
p09543
p21918
q5hyk3
p42356p21781
p26678q14654
p17480p41231
p14416
q13126
q71rs6p49639q8n8r3
q9y234
q86yv9
q93052p54289o43865 q13573 q9hb75b3kn93
q9h477q9udr5
q13247
q9y4k1q04671
q8tbp6
q7rtu1q15637q9hcu5
q9ulj3q8iyp9
q96s52 p14222p26885
q9y5k5
q7l5y6q7z7k0p50120
o14493
o43193
q8tcc3o00232p55769o95965
q99590q99732p51665q15021q96cm8q06889p09958
q504t8
q15287q9upz3
p21953q86vi4
p51810p17643
o95059q5h8a4
q8nfz5
p40967
q13509p57729q969f9
p48304
q6zsm3p00790
p42765o94788
o43766
o60641
q16186p11308
q16740
q6p1a2
q9bpx3q9nx24o14625
q03014o00233q9p0k8
p28066q8iwt6p98153q9nqe7o95461
p62633 q8wwy8
q13765
q9bqr3p49682q01970o00231
p55036p62333q9unm6 q15003
p62988o95841
o75800
o75832p19883 q9ns56
p26436p31153 q01081q86sq9q96s55
q9uff9
q9nrb3p09172
q00266gypsump15018p52747
p00540
o00571
p35222
o00116
p60484
q9upi3
p84077
q08050
p21730
o95249
q14201
p62191
o60664
p10768
p07360
q9y5x9
q86z14
p20936 q92506
q8n6m6a2myd5p18859
q6p3u7
p10275
p22894
p06280
q9y259
p40926
p11686p01185
p05112
q13164
p49961
o15173
q9gzt9
p29972
vs00017
p01236
p14138p53396
q15165p78542
p13501
p49585
p11926q9byz2
q02388
q9y5k3
p35790
p04629
q14749
q4le39q13423p07550
p01138
p07998
ms00068p01178
p61073
p00480
p49915
p40126p11177
p55263
p00966
p04424
p39086p14920
q02127
q99798
ls00054
p28222
o75746
q03154
p28221
o75936
q8n4m1
p07339
p00338p07195q6zmr3p04181q5t2r2
q9ujs0
p07864
p30990p22061
p09622
p22033
p30520p49189
p98174
q96e11
p54803q01432p23109
ss00413
p22303
a8k5i5
q14739p40925
q14957p63252
q86u10q12904
q9h2b2
p49768p10144
q86yt5q13183q9uld0
q9h6q4q16572
q9y600q16445
p30838q8ndx2
q9ubs5o75791q9h2x9
q14416
q8tdu9q9h992
q99928p05937
o60296
q05586q14833q13948
q13620
p30531p30536p18505
q9p2u7q12879q9p2u8q16478
q13003p43005p01148
p41594
q03518p43004
o60220
p54886
q13224
p51649
q86xe5q9y5l4
p06737
p80404
p11216
q9ula0q8nhs2
o75027
q12797
b4del5p07101q53ga4
p05165
p43007
q16099
q99259
p20472q13255p43003
q9nsd5o75311o95166p49747
p48058p48167
q9h237
q8n808q96g03
p13498
o00757o00217f5gxe5
o95394
q02978
p09110
q6iq20p80108
p01275
p62158
q6gpi1q99895
q9y5k2
p60893
o95467
o95749
p14324o76082
p27169
q9bv23q8wy41p60323
p35558
o14939
p13569
p34059o14578
p01024
ns00183o15360p35030
p10620
q16698p01308
p05305
q96s42
p01584
p47710
o95347
q16822 q7z2e3p29466
q13867
p24752
p06132
p06681
p08174q05066
q8wwi5
q16798
p19440
o43451
q13326
q13093p60174
radix trichosanthisp05814
p07814
p00734
p13726o14495o43688
p03891
q8n726p09848
o95503
q6ns38
p33897q7z5q5
p00709q9bxw9p50336
q02809q15735
o14832
q9nui1
p55199p52209
q96bz4q8iv08
p23141
q13393as00075
p43155q695t7
q9nrd0
p11766
q96rq3
q6pf09
q9h400
ps00173
o75390
p14618q16778
p14136p28838
q8wxs4p20396
p06307
q9um01
p07954
o15146
q03519
p48735q92973
cs00392p01100
p30613p06753
q9y5q6
p49448p00367
q7z569 q6uwp7
p15104
p30542p13798
q08ah1
q9byc2p55809p09683
o00330
p31939
q6ibw4p42695q86xi2
p10636
p21399
p78504
q9ntj3
p21728
p00439
q09428
p38571p10599
q93099
cs00067q04609
q13519q9h2s9p01303o94855p60520q7l0j3
p08908
p78334q401n2q8n1c3q14832
p11086q8wvc0
p56178p07108 p05091o75899
p17752
q7z2h8
p18507q8te85
p78509
p28476q13619 o00222p04217
q9brk5
q05940p50553
p42263
q9umx0o14764
q9upv9p23381
p01213p48169p41250
p28472
p49238q9h598p19801p47869p34903
p78310
p48066q6ib77
p14867
p24046p47870
q9nqx3
p23415p56177
p07320
p42261p47895
o00522
p30038q96nt0
p04207q6tgc4 p02144q07283
p11831
p10826 q9bze9
q96a70
p58215p54920
p53602
q5vyx0
p11274
p35611
p25054q96hn2
p29274p49902
p30837
p02686sd00044
q15842
p09923
q96jp5
q9npb1
p48047p01588
p29275p43119
q8n7h5p21549
p01282 p42262
p05451p05089
o43612p78540
q8n159
q96g91q7z6v5q9bzd2p27707
p55265q9bub4
q9h0p0
p54851
p05187
q16620
q13976
p51580
p47900 q99571
p78563
q9y3b8q9bse5q9ulw8
o00487p55735
o95865
q13505
p21589
o43868q9bxi3
p35251q8tcs8q9h244
q96sz5p16109
p09603
o00337
q9ns39p39905o43318
p16220
p22079
p13866
p00746
q00796
p25445
p01009
p48065
p55011
q9c000
p04746
p58012
p19961
o43280
p05062
ps00098
q14397
q03393
p51570
ns00189
q99643
p01241
p43304
ss00357
p21796
q9y3z3
p17735
p11217
p35670
q05329
q9br01q8wwt9
p28329
q9uby0
q7z721
p05019
q8net4
q16851
p45985
q8n699q99759
p15382
q9uj14
q9nr45
p36269
p09919
q9y5p6 p36871
o43488p20800p04732
p02760
cs00201p16591p09467
q16538
p21695
p05067p32189
p00156o00591
q9ulc6
p07288
p06276q9nph2
q9y4x5
p02675hs00118
p57773
p61812
p16070
hs00119
p46098
p02671
o75610
o60701
q9nz01q96hy7
p08237
p35575
p09455
p08559q8iww8
p08100
p43630
o60565
q12772q08257
p99999
q96ha7
p55210
p02735
p07947
p00533ls00097p22466
p11168
p61626
q9h227
p10253q53h76
q32zl2
q16539
p55212
p00492
gs00029q8ivi9p54136
q9h300p28288p10909q9h115p01210q16836
q13145q9y314p27708
p29323
q96jb6q08629
q8iuf8q9y4k0
p50440q9bvc4
q9nr19p36551o95389
q8tas1p01286
q15166
p11488q9nra2p41217
p07306
p28715p43699
p08729
q8nbp7
q9nrc8
q9p2w7
q3kpi0
q8tat5p22674p35900
q9uif7
p08779q5jvs0
o43542
q9uk23
p23946
p48061
q15468
q9byv7
p10586
q3ybr2
q9h2h9
p22614
o15121q15399
o14678q99447
q12873
p62805q9ubi9
q14191p06731o95948
q16531
q96t60q13569p17661
o60673q99607q12809p48751p32971 q13315
q9ubz9
q9nti5
q9y6m5p09327
o75794
p49281
p07307
q9c056q06430
o43524
p13611q96q15
q92630q01959q16891
q9y272o60936
p34972p41143
q05516
q9y5s1
q86yb8
q9nrm7p19021
q13258
q9c026p09238o95727
p51812
p57059
o15297p23025
p09488
q9ntg7q7l1v2
o15350q00839
o75528
q15181p33993
q9hba0
p67809
o14735
o00519
ps00182
p49757
q13470
p13073
p20719o15409
q13547
q8wx93q96q91
q9nrg4
p61567
o75928
p49959o15392
q12834
p48651
q00534
p08263
p09211
cs00649
p16104
p02790q9un88
q9y6f1q13156p39748p07992
p06702o15354p16455
p18887
q92889q96fi4p08670
p18074
q9nwh7
q01831
q7z7h5
q6p1n9
e5kn95
p54278q92698
p15088p19013o15527p13051q96se4
o14757
q9ugn5p49917q96qu6q59ek9
q9ukk3q86vb7
p16422p28906q9ugp5
q9nuw8
q86wg3
p49916p04233
q8ivs2
p22083
p46939q96ja1
p40692
p18858
p16150q86yl7
q13421o95271
q8iun9
p35408p54277q9ns91
p23396
p28336o60907p23975p12956
q6gmr7
q93009q9uhb4
q53lp3q9h2x6
q96he7
q96a56q9nrn9
ps00183os00096
p10721
q16611
q96qu1
q5t4w7
q9gzp4
o00764
ms00241
q16584
o15231
q9h3d4o95997
q9h7z6q15717p37023 q9bzq4
o14980q86wx3
q08431o75083
q63hm1q14186
p00738q969k7
q9h8v3o15229
p30466p68402
p02655q16609
q8wwy7q9nww6q96i15
q8tcv5
q9gzn6
q8tb61p49247
q9bw04p18405
p46952
q3t906
q9ubm7q15102
q96t66q8ixj6q8tds4
q9nxa8
p48023
q9han9
q9h221
ophiopogonis tuber
q6xqn6p29401
q9h222
p43220
q9y5s8
p06727p08138
p11021q13426
o43819p21980
q9h3s4
q9y6e7p05546
q08945p26358
o95155
q9buj2
q8nb91
q9bzz2b4dln7
p33121
sd00085
p53634
p07148
o00767
p01130
p55851ps00024
p15692
q15303
q9hb55
q9bx93
p33261
p41159
q92993
q9nz20
q9ubk2
q9y618
p49792
q7z6w7p07359
rs00133p05412
p07099
as00063
ss00282
p20813
p04637
o76074
is00262ns00109
q8izm9
q99217cs00029p24385
q07820
o95477
p11712p00813q15848
p00491
p05181
q59f19
p10415
q13133
q86v15
o75342
q03013
p08684
q96kp6
o75477
p04798q13829
q6p589q07812
p21860
p11215
p02749
q9nrf9
p08758
p11387
ws00005
o14788
p37837
q03169p54098
p35354
p45983 p18440
scutellariap06746
o95379q5gj75q8wvp5
p15559
p01375
p21964
p34810
ls00171p05231
p19838
q8tdx5
p35869
q9h7z7
p23219
p05362
p20815
o95622
q8tdf5
p14555
p48357p06858
q8nf37
q6p9f0
p20933
p82932
q86ue4
q9h2a7
p08603
p53779
q92186
p00739
q14289
p55087p54252
q14376
o60218
p50053
q9nxz1
o43914
o95968
p15260
q04446
gs00157
q9uj99
q08499o75355q9ugm6o95500q7z2y5p36382p35212q96kn9 q5t442p50747
a6nfd8se01038o15195
q13546
q15628
q9y285
p06850
q9hc78
p83916p08034
q5nv69q9bx67
p57087q92959
o43474q9nqz7
o95169
q06210
o75054p01589
p07315
q8n1n0
q8wwv6
p09525
o75712p61421
q92851
o15439p37059
p52790
q96gd0p07902q9y6a1
q96qt6
p15311
p22732
p46734
q9bqt8
p42226p04083o95452p23582o43396
p43652p78562
p29376
q92616q9nr09
o14521
js00028 p01137
q15004
p04406
o14494q03135
o15264
p01160
p00568
p04040
p01270
p04179
p06729
p61278
q16665
q9hbf4
o14727
p11413
ss00360
p36383p13500
gs00300
p14780o95479 p08236
ps00477
p00797
p05164
q96ap0
p49419
o43306
p06744q06609
p55211
q15392
es00154
q14994
p20702
o94806
p33260
q8n5b7
p22392
p10163q8wtv0
p78380
p01133q5hya0
q9y2z9
q96gr2p21583p04053
o60346
q13526
p53801p00750
ns00107q14534
p18031
q15274
q96ri1q8izn7
p55055
q8n104
q9y2r9
p49888q9ubb5
cs00128
us00005
p22680
o15197
o95544
p02741
p24941
se00229q96pd7
p11509
o95905
q969g5
bs00084p02008
p27797
o95411
q9y5b9
q9nqa5
q9nyg8
p60510
p78524
ls00212
q05823
o75310q92839
o75795
p36537q9y4x1
q16411p30405
q9npz5p54855
p10619p08833q9np50
o14503
q15031 q9hcn2
q6uwm9
q8iwy9
os00093 q16696q9h5v8
p26440
q03468q12778q6ia69
q15047o94776
q13464q00987
q12888
q86vz5
p06133q9by64
p24903
q92819o94910p12955o15169
o94766
p48449
q14677o00219o00560
p61086p78549
q9h2k2
q96kb5
p36639
q96lt4
p19447
p19544
p52701
q9bzj4
q8tdg4q12980
q96rr4p28908q96fl8
p40198
q14683p07093
p08575p09936
p15941
p09210
o60760p35754q9np99q04760o15217p23921
q9h4y5p21266
o95881
p30711q8ted1
q16772
q7lg56p48637
p31350o14880
p0cg29q16775
q7rtv2p78417
q8tdv2q92600q15904
p30953q9npf7
q99685q9y316
p35913q5vwk5
p22891
q14690
q5sci4q9byb4
q16236
p28340p10646
p32780q9uq84
p27695
p07204
q9y385
p12004q96la8
q5xg87
q13352p09884 q86sq4
o95243p09104
q8tdm6q9ubx5
p41219q9y4e5
p56180
p02545q676u5
o95069
p09382
q92820
q9nzj6
q8net8 q8nhy0
p32418
q8wux1
q9np80p51572
q99807
q9up65
p06127
p0c869
p34932
p41161o43242
q9hay6
p53370q8ndx5q9gzr5
p59796
q12974
p10243p20700
q07108
p18283
o00180
q99218
q7z449
q8ner5
o75665
p98155
p53041p22105
p55268
p51582
o15496
q8nff5
q9hcm9
q86sj6
q86tx2p33176 q9up95
q5fve4p15144
q9unk4
p25874
o15254
o14842
q8wui4
q96cc6p24462
q12791q9bqi3
p31431 q8n6t7
q86yn6
q5nul3
q99873p25092
q99808
q99698
p49753
p49591
p08709
ls00058p68871 o15540
a0mnp2
p55916
p46527
o95573
p02751
q9nud9
q16655
p36969
q04759
cs00116o15151
q13790
q8n1l4
q9uid3
p58743
q02928
cs00989
cs00880o43808
p17900
p20142
q9npf0
q9uku0
ms00393
p22736q9p296
q6p5z2q15139
q13508p22307q96l15
p24723q9bzl6
p52961
q9y5v3o60603q99729
q7z408q9hcc0
q99445p31946
p13584
p16401
d3dvq6q75wm6q02539q8iv61p11142
p61204
q8td17o60547q573b4p84243
q99570
p07305
p19878
p46094
p41235
o14645
q9btm1p0c5z0
o43399q9nrw4q06265
q3syg4
q8wvc6
q63hq0q16853
p30301q9uql6p19484
a5pkw4q9by66
q9nr22q9ukl6
p17677p15151p10412
o60551p50135
p43080o00159
p62330
q96a08
p14770
q96pb1p49006
q9y6e0
a8k1m2q8ivk1o00458p78556
p0c5y9
q9h4x1
p14598
q8nff2
p63098q9ui40
o75367p01112
p57723
p16402q6qhc5
p16403
q9h305
q12884
q9hc58
q13702
q9ubs4
o00622q01101q99836
q96pz7o95429p55198
q6nt55
o43781
p41238q86tg7
o75715p22352
p02808
p07203q96sl4
q7z407
q9p0m6
q8wwk9p62745
q16829p29966o95267
q7z2w7
p18085p08567
q9h1r2p30968p35243q9bxl5
p23527p22492
q71ui9
q99418q8nb12
p22570
p42357
q9nz09p98194
q9h211p13497
q9umx3
q9ul01
q8nbh2
q9y397
q9nv23 q7z5g4p54840p38117
q96k12
q9y5r8
q8iuh5
o75460q8n6n7
p47974p60880
p0c7u2
q9y296
q8wvx9
p06703q14541
q14542q9nqg7p28039
p0c0s5
p56704o60488
o43918
o95237
q9nzj5
p35556q9nyx4
q9h009q99767
p12259
p40879
p24386
o00472
o43617
p06241
q9nr71o95822
q7l5a8
q99784q15434q9hd26o76015q96rj0
q05084q96mw7
q13371q96ga7
a4fu01
p24666
q01469p61769p13807
q5t215q06203
p17861o76054
b3kp71
p13804
p05198
o15228q96c24q16654
p09466
a2arm6
o00115p07988
o75781
p10145q96ph1
p51589
p37231
q9hcl2
q9y6r1
p04114
p40145
q8nfm4
p16233
o60503
o95969 q8n9l9
p14672
hs00094
p12104
p19835 q8tcg5q08462
p45844
p32322
q9nzd2
p23786
q96dy3
q86x55
p03956
p40939
p55157
q6p1m0
o75159
o14561
cs00250
q9has3p22612
p10644
p30825p31323
p13861
p17612
p61586
q9nwu1
q08426
p36956
p11597
p20292
p15090
p41222
as00249
q86sk9
cs00285
es00010
q9h015
q08828p00167
q9y6y9
q53hv7
q9nzk7se00285
p31749q9uhd2
p04275o75907
o95255 q9bzm2
q15067p49327
p51828
p08311
p29083
q9ubu3
o00628
q92523
q07869
p50416
os00061
p19971o60266
p16671
q9p0j1
o00206
p07205
q6jqn1
q8tcd5
cs00644
p05413
p11150
o00154
q96a54
o43581q14108
q9ulc5
p32238
p07098
p50542
p00414
o95278
q16401
q99541 q86x29
ps00111
p30084
o75381
p30419
q96t53
p31025
p50897
q96iy1
q9y2p5
o95158
q14653
p02774
q96ad5cs00020
q8wts1
p54317q6piu2
o75688
p12643
q9gzx6o43315
q9nst1
o00548q5k4l6q9hd89p10082
q9uhr6
p24557
p55061p06748
q16647
q9uet6
q9npa8p01298
q9y2p4
o60240
q08477
q99988
p20132
p46063
q86y26
q16626q93070o75478
q00978
q9gzx7
p0c1z6
q9ha82
q6ub99
q9bvm2
p37275
p63096
p17081
q9hb03
q6uwv6
as00574
q13123
p02654
Figure 2 Compound-protein network of BHDWT Red triangle node compound in BHDWT blue circle node protein
1198752= pp (119883 119884 neg) pp (119884119872 neg) 997904rArr pp (119883119872 pos)
1198753= pp (119883 119884 pos) pp (119884119872 neg) 997904rArr pp (119883119872 neg)
1198754= pp (119883 119884 neg) pp (119884119872 pos) 997904rArr pp (119883119872 neg)
1198755= pp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
1198756= cp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
(3)
1198751cup 1198752cup 1198753cup 1198754is the set of rules to deduce the eventual
effects of chemical compounds on the disease 1198751indicates
that if the effect of119883 to 119884 is positive and the effect of 119884 to119872is positive too then the effect of 119883 to119872 is positive Similarderivations are defined in 119875
2 1198753 and 119875
4 They may be used
as many times as necessary to the final disease node 1198755and
1198756are the rules to extract the nodes for the network They
can also be used to draw the subnetwork when the originalnetworks are too complicated to analyze 119875
5indicates that if
119883 (protein) acts on 119884 (protein) with the effect of 119885 and 119883 istagged then this relation is extracted and 119884 is also labeledfor further derivation 119875
6is similar to 119875
5except that the
starting entity is chemical compound 119883 In the construction
4 Evidence-Based Complementary and Alternative Medicine
PI3K
AIFENDO-G
CASP12
TRADD
Cn
stresssignals
MYD88
calpain
JNK
LEP
IL-1R
IL-1
GHRP
G6PT1 GIPCPTI
Beta3AR
IRS
FFASOCS
AA1R
NGF
FATCD36 TRKA
FACS
SGLT2P K C
TNFR2 TNFR1
erstress
GLR IAPPmNCETNFA
INSR
SHC
INS(hyperinsulinism) PTP1B
LAR
APS
ATMIL-3R
D-GlucoseDNAdamage
D-GlucosehyperglycemiaIL-3 AMPK TNFa
GLUT2
BAX
RosFAS-L
PDX-1
FLIPTRAIL-R
APAF-1
MafA
FAS
BAD
CASP10
P53 BID
CASP8
GK
FADD
TRAIL
DFF40
DFF45
CASP3
CASP6
CASP9
CASP7
cytc
cleavagesubstrate
DNAFragmentation
impaired insulin secretion
Pyruvate
apoptosis
PYK
INS
ACC
lipogenesis
ATP
N F B
I B IKBA
IKK
BCL-2IAP
PDK12
SREBP-1c
GLP1R
IRAK
11HSD1 UCP1
P K C
UCP3
T2D
glucolysis
SOCS3
P G C 1 -
mTOR
4EBP1
STAT3
lipidhomeostasis
S6
- M S H
P70S6K
NIK
RIP1TRAF2
JNK
Kir62SUR2ASUR1SUR2
VDCC
cb1
CRKII
Obesity
ADIPO
ADIPOR TC10
GRF2
EX070
MEK12
Insulin resistance
RAF
DPPIV
LEPR
Retinoic acid
EIF4E
P K C
PKA
HSL
antilipolysis
GRB2
SOS
GLUT4
RAS
PDE3TSC2FOXO1PP1
RHEBGYSFBPPHK
AKT
cAMP
GSK3B
PEPCK
AMPK
G6PC
AGRP
SHP-2
MNK
CPT-1
Ceramide
RXR
ERK
AMPKK
P P A R
Long-chain fatty acid
CIP42
PYGPPARG
glycogen
glycogenesis IRS
glucosehomeostasis JAK
Figure 3 Biological network of T2D Blue circle node protein purple circle node nonprotein node green octagon nodes type 2 diabetesldquo997888rarrrdquo one node enables the expression of the next node raises the expression or enhances the activity of the next node ldquo⊣rdquo one node inhibitsthe expression of the next node lowers the expression or weakens the activity of the next node
of network 1198756will be used once and 119875
5may be used as many
times as necessary to the targetThe modes of action between nodes include the positive
(pos) and the negative (neg) effects If we define the positiveeffect as ldquo1rdquo and the negative effect as ldquominus1rdquo the ultimateinfluence of intervention will depend on the product of eachstep in the whole signal pathway For example compound Ain Figure 1 influences protein T through 3 pathsThe effect ofA to T is negative through the path ldquoA-c-d-e-f-Trdquo and the pathldquoA-b-h-k-i-m-Trdquo This effect is uncertain through the thirdpath ldquoA-g-h-k-i-m-Trdquo taking into account the negative effectfrom the feedback path m to g In this paper we neglectedthe effects produced through feedback because the role offeedback is expected to regulate the magnitude of effect butnot to alter the overall mode of action of effect So theeffect is positive through the third path After the effectsof one compound on the ultimate node (ie T in Figure 1)through all signal pathways are determined we can selectthe effective compounds based on the desired effect If thedesired effect on the ultimate node is positive the compounds
with positive effects through all pathways will be selectedas active components The compounds with negative effectsthrough all pathways will be ruled out The components withboth effects through different pathways need further analysison molecular mechanism If their undesired effects can becountered by other compounds they may also be selected aseffective components to be used together with the counteringcompounds The opposite analysis will be done if the desiredeffect on the ultimate node is negative Consider
119878 = 1198781cup 1198782 (4)
1198781is the set of entities with structure cp(119883 119884 119885) or
pp(119883 119884 119885) in biological network of disease which are thebackground for deduction 119878
2is the set of labeled compounds
or proteins expressed by tag(119883) 1198782is the initial conditions
for deduction
22 Data for Construction of Component-Protein Networkof BHDWT Formula For decades BHDWT has been usedto treat T2D at the Beijing Guang-An-Men Hospital [12]
Evidence-Based Complementary and Alternative Medicine 5
beta-Sitosterol
Baicalein
Adenosine
T2D
linoleic acidIsoliquiritigenin
Acteoside
Chrysin
Ferulic acid
Nicotinic acid
Lauric acid
CatalpolMangiferin
Berberine
Mannitol
Phenylacetic acid
Scutellarin
Rutin
Succinic acid
Wogonin
Figure 4The effect of compound of BHDWT on T2D Red triangle node chemical compound green octagon nodes type 2 diabetes ldquo997888rarrrdquoone compound has the positive effect on T2D ldquo⊣rdquo one compound has the negative effect on T2D
BHDWT has a positive effect on blood glucose control andsymptom control for some patients in the early stages ofT2D The BHDWT formula consists of eight herbs includ-ing gypsum Anemarrhena asphodeloides Bunge rehmanniadried rhizome radix trichosanthisOphiopogon japonicusKerGawl Coptis chinensis Franch Scutellaria baicalensis Georgiand Glycyrrhiza uralensis
The components of the BHDWT formula came fromthe Traditional Chinese Medicines Database (TCMD) [13]the State Administration of Traditional Chinese MedicineBasic Information Database (httpdbsharecintcmcomZhongYaoJiChu) and A Handbook on the Analysis of theActive Composition in Traditional Chinese Medicine [14]
The compound-targeted proteins were derived from theSTITCH system (httpstitchemblde) [15] By entering thenames or identifiers of compounds or proteins of interestSTITCH provides the list of proteins withmatching or higherconfidence score that the user specified up to the numberthe user specified The required confidence score representsthe possibility of interaction between the entities In orderto obtain more general results the parameter of the requiredconfidence score was set higher than 0 and the interactingentities number was set to be 500 The interacting entitieswith clear mode of action (positive or negative) were chosenfor further analysis
23 Data for the Construction of T2D Biological Network Toconstruct the T2D network we used the data collected fromthe Kyoto Encyclopedia of Genes and Genomes (KEGG)and therapeutic targets database (TTD) KEGG lists thesignal pathways related to T2D and TTD lists the chemicalcomponents used to treat T2D The biological network ofT2D (Figure 3) was constructed using the signal pathwaysfrom KEGG the chemical components from TTD andthe positive or negative relationship between targets andT2D from STITCH The networks were visualized with thesoftware Cytoscape [16]
3 Results and Discussion
31 Compound-Protein Network of BHDWT Formula andthe Biological Network of T2D We first constructed thecomponent-protein network of BHDWT using the datafrom Section 22 (Figure 2 additional file 1 in Supplemen-tary Material available online at httpdxdoiorg1011552013840427) In this network there are 144 compoundsand 2865 proteins The red triangle nodes represent thecompound in BHDWT and the blue circle nodes representthe proteins The biological network of T2D (Figure 3) wasderived from the data in Section 23 using 119875
5and 119875
6defined
in Section 21 with T2D as the ultimate node Biologicalnetwork of T2D contains 146 proteins The effects of each
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
Evidence-Based Complementary and Alternative Medicine 3
q9y617
p30566p30086
q3lxa3
q7l266
q9ui42
q99489q6pi48o60391q9p2j5
p78330
p14868q16773q96ab6p45381o15399q9nsb8
p48664
q9y5z0
p23515q5tdp6
p48549o43555
p08243
p36957
p15153
p05166q9bzw2
q13488
p21579
q8wz71
q96i99q02338p48200p02812
o14925q6pce3
q9y5j9
p10515p05108 q9bxa5
p23378q9ujm8q86wu2q96s66p12236
q9h3h5o43837
q9ubx3q9y276
q8n907p0cg04
p31937
p16112p12235
q01813
p56524
q6ibs0
p50213q06124
p48039
q96hd9
q92542p22234
p47897
q13155q8n142
q00577q15046q7z429
q9bvj7o15239
q969s0
q8tdd2p03905
q53gq0p17858
q9h9p8
p04839
q8nbi2q9nz42
p31944
q9ui32
q96rs0
p41252p51793
o75306q9nvh6
q9y2z2
q0vdg4o43181
p52564p48764p19634q9ugb7
p13725q9btd8q49a26
p32856
p05113
q8n0w3
p48165
p02750
p47929
q01433
p09972p04075q4u2r8
o00462
p08574
o14638
ts00140
q8tct9
o95139
q8ne62
rs00003
p22830 p14927
q9uig4
p49447
p02766p04062
q9y215
p20336
q07837
q96t21
q0vdf9
o94916
q16623
p14550
q6ub98
q96h96q9y436o15460
q8ivl5
q5t6j7o75251q7z4n8p34896
q96ta2
p06576
q9p2r7
q96q83
p39060
o60568
c9j9i1
q8nhu3q8tct1
q9gzz6
p32320
q9y227p56539
q9ukg4p01730
q13045
q9hbg6
p30411
o75563p25208
o14556q8nci6
p53597
p25705
p35080q07912q9uhl4
p48163p49286
q9nti2p23368p56134
q9p2j9
q6uwu2
p02708
q92478
p36544o43405
q9hbu6q14498
q5m775q13510
o00754
q96cw1q13905
q9hat2
p19235
q9ntj4q9y2e5
o00220
o94850p51688
o95471
q96ri9
p78356q5vx52p01008
p09693q9y616q92636p26715
q9y6x0q01459
q8n5d6
q86wv6p19438q9bzp6
q96hz4
p08727
q13601
q16552
p10809
p25942o60759
p29033
p13236
o15455
as00275
o94864
q9y223q8n4e4
p14210p50591
q13158q9np84 p08571q8tdq7
p43403p20963
q9uhj3o75382p08912
p16152q9uha7
q9y5z7q9ubh0
q9nzh7
q5t280q9nvg8q15677
p42330p48730p34949p05452
q8wwz1
p49674p49221
p08913
q9nzh6q8iyr6q9nzh8p01583
q9bzz5
p11233o60870p21802
p17676
q5tba9
q9h4b6
p23771
q5vzb9p43405
p30273q9bzx2
p23769
q8wxg6
q12792
q13287
p27930
p04234p07766q9uj70
p49590q9ha47
q02223
p18428
p15586
o94808
q9p0l0p35232
q9ui33
q9y573p42701p43251
o95259o00590
q2tb90p49354p53675
p01266
q00610o75600p01857
q7rtx1p78316p07355
p01106p20382o00442q15825
p21439q53gd3q9nvf9q96rl7
q13822q07001
q9ui17q9bvg9
p01344
q86uu9
q00994q8iwa5p63027q04844
q05315
q6p1j6p17787
q15822 q9ul12q05901
p07510
p15421
q9nyc9q7rtx0
p07948o00244p29762
p02787
q6uwr7
p43681p04080
p12830q9ugi8
q16134
o75955
q14254
p52788p11473
p02452p49356
q9hcg7
q96pd5p31513q8wz75o15244
q8te23
q9h8t0
q9y3i0q9h3z4
q9ui36o75197
p10606p13987p01834o75792q13515
p27824p22413
q9ubm1
p81605
o15245o60271
q9npp4
p54760
p62244q9brg1
q14410q16878q99471p85298 p38606
q96t52
q13952
q6uwm7o96004p13942
q9haz2 q9uqp3p24390q9npg3q92922
q14409
q9uhv8
q9ugm1
q00325q9ug56p11230q9gzv3
p32297
q7kz85q9y6m1q96bw1
q969g3
q8n5y2
q16595q15049q7z353
q96fz7
q96s36q9hb07q8tb68
o60260
q9p1t7
q9ubm8p10109
o75884
p16442
p08922q96cf2
q9by43
q9y3e7o43633
p35520
q13163p01591
q9hci7q68dk7q96lu5q9nzz3
q5t870
p11717
q9ns23
q9y6k0o75751
p49789
q96h20
q969l2q9h444
q8ni17
p01574p01563q8ncc3q12794p01570
p61244p05014
q86wn2p01571p19099q86wc4p32881
p09038
q96rq9
ls00113q8wxi4
gs00298q99616o95810
q8tdu6
q9y4g2
o15554q92597
q9p232
p21926
q7rts9p51787o60674
p03923p10914q09472o60443
p41212q5eg05
q6zna5q14151q15424o95718o14990
q16637q9npd5
p15538q9hau4p28676
q14249
o15120p17927q6zmt1
q8wua8q15878
ks00036
gs00302
q92753q00975
ls00110p00441
q9ubh6p10827
p68400
q12762p19784
q13522
q9ujy1bs00293q96t58
p80365p40313
as00350
q9y6q6as00353
p11171
o60499
q8n7p1q96c36
q13336
q8izw2
p31645
q16587
q8wyr5q00613q9y4k3
p84090 q14790xs00013cs00499
o95206 p20333
ls00125
cs00498
p14735q9h173
p06401p04150
q9nrz3
q86uw1q5sw79
q9bxc9q9y2g4
p38398
o15427q96n87q9bxs1
q02318
q9bxc0
q8nej9
q16719
p31751
p48552cs00875
p53985q96ru7
p08134q16342q03001
o00626
p47992p05787
p30519
q8n0z6
q13490p04196
q13794o95140
o00399
q16621
q14721
p56470
p48147
p07949
q8iwl1
o95342q5t5p0p37840
p55957 p49675
o75444
p69905
p04049
q99884p32942
q9nyl5p51659o43866o95750
p37268
q01581q6tdu7
q9upn6q12770o75881
q92496q99487
o95445p43034q6ux71
ls00187ls00188
o00353ts00144
o60906fs00061
q6rui8
Glycyrrhiza uralensis
p03372
coptidis rhizomaq9nq94
is00086
q92887
ls00105
is00187
p51671
p21397
se00911
q53rt3
q9y4x3p17516
o43236
gs00307
p52895
p02679
hs00046
anemarrhenae
ls00122
q14314
o60725gs00305q14432
p42345p10242
q03591
q13370q15493p51530
q9h2c0q96ap7
p42126
cs00415
ls00127p42679
p19623
ms00103
is00211
p06493p20248p78396
q16678
ls00183p11511
ms00035
os00089
q96du3
q14315q8iy17
p10745
q9uhj9o00501js00007
q75t13
q5t8e0
is00181
q14258
q96d42
q8n695
os00007
ls00106
p11182
ps00017
p80098
q01415p31040
rs00138
p01033
ns00060
p35557
ms00098
ss00414
q16625
q07157p57739
q93088
p14410
p04745 p24298q8td30
p16278o00443p17540
p06732o14756
p12532
p08519
p23416
a0jlq5
gs00279
q96c90
q06278
q99572
o14649
p04818
q9nws1q9hc52
us00016
q9y2c4q96eb6
p43490p38936q5syt8
o15294p16581
q969s2
q9y6l6p30307
o95644
q9gzz1
p04035
p09874
p08697
p11166ss00388
ns00013
q04828
p16284
p11802q92934
p46531
o14791
q9uq10
p51452
p19174
p17096
q8nec7
cs00212p78381
p55316q86ug4
q16762
p52565
ns00227
p14635
p04085
q92731p25963
a4fu51
p15408
o14763q8n8n7
se00354is00186
q86xx4p11245p09429q5t7c0ss00115ss00290
os00009
q9nzc2
ss00165
gs00240p28799
p18847
p02771q07817
p23942
p54259
q5fya8
p22460p19525
q14164q13489
p39877
q02156
q10588
q9y5q0
q8wxh0
q86vf5
q9npg2
p35637p50225
q9bww8p32929
q8ius7p54257
q6q788
p34982q8wwm9
o95866q96s37
q9y697q14686p07498q5hyw3q5snw6p49019q8n6m5q96n76
q92765p09105p05090
p14616
q9bzm1
q9nyp7
q9uf11
q03181
as00429es00030
p35568
q8tdv5
q9y2u9
p02647
p04271
q16517
q86uk5
p02656
q16850q9h3y6
q9y226
o96017
p30279p30281
q9unu6q14746
q9bxh1q9una4
p57679
p43351
q14693q9y239p54868
q2tac6q9nvs9
p20062p02652q9uj68
p05129
p09960
p34913
q9byj1p28300
p05121
sd00054
q9uhc9
q9y6a2
q9ubu8
o15218p48775
o60239a0pjb0p07327
q9unh5o75469o14683q9ukr8
o75293 p19320
q9nxr7
q9upn3cs00903
o43464q92859
o94972
p34969p78423
q5sxq6
q92838p30044
q96c86
p62491
q8ixh7
p78545
q9hab8q9y2i1
q9nq25q9y6a5p62508
p30043p18510
q14671
q14457q9ukg1
q96ir7
q92583o15162q9nrp7
p84098p50613
q13950p59665
q5t7p3o14920q14145
p98170o75112
p11474o95831
q9y6k9p47944
q9nqx6
q9bq39
q16769o75365q8iwl2
p05455
p00450
q702n8
gs00067
q8tcu4o14972q02161o15068p57735
q9uja3
o15553q15596
q8tb72
q14773
q96hc4p01127 p04921
p0c0l5
rehmannia dride rhizome
p00505
q6uw88o60733
q16549p41743
ws00006
q9haw9rs00107bs00006
p40261
gs00192
p14679
q93038
p07478 p07477
bs00002
q8ner1
p06213
p07585
p16435
p09601
p08246
q5tka1
q5vv67
o95864sd00083
q92911
p04054q9p1z2
o15296p28907p10632
o60282
p33527
p35610
p02753
o75908
gs00150
p16662
bs00188
p33316p18054
p09917
q13427
p05177
q96rn5
q6zwj8
fs00063
q99735
p02649
p47712
p60568
p35638
q05655
p06400p00395
q9umx9
p16050
p48506
q04206
p35228
q00653
p17252q9uj83
p60321
p25440
p07741p50991
p02788p02458p52789o43463
o14793q15582p29459
fs00041
p45984
p27815
q96pn7
q9bs16
q9nr28
ms00083p08172
gs00239
p11229 p15121p17174
p69892
p01579
q9unq0
p00390
p13674
se01327
q15800
p35225
cs00312p27361
p22301
q15759q99707
p80370
o43852
p53778
p12931
ms00089
p26447vs00026
o00142
q15546cs00977
p32754
p01189
q53fz2q15758p28482
p00747
q8tet4
q9bq90
p35542
p21912
p10600p01562
q14116p29460
p08253
o00292gs00238
q06330
p17302
q01201
q16585
p02768
p05771q05513
p29474
ls00115
p42574
p08183
q32p28p29372
p01023p46926
as00349
is00241
o15440q8ng68
o14798o75886
q76g19q9ul42
p57730
p51816
q8n5j4
q6zut3p43307
q96pd4q9npj8
q92772p01566p05000
q9p0w0p60508p60981 q9ubc7
p01569
p11678q9bqq7q9nrr2
p46976p52926
q16659q86vu5
p01215q9uqf0p30041 q9nzq3
se00987p08235
q13231
cs00298
o43765
q96ba8q14c87
p28845o95251
as00351
q9y6h8o43395
q04864
o00327
q8td08
p31152q9bpy8
p23560q02153p36897
o94760 q9gzv4p51841p14902
p52569o43508p33764
p63241b6zdq1
q9y619
o75343p17342p25106 q9bxl6
q9nwm0p54108p33402o43272
o60684q09013o00505q12979
o00629o15131p52292p52294
q8wwu5
q674r7q9np91b6zde8q9y2j8
p14652
p20930
p51648
q3c1w8p06733q8wy07q8neu8
q4w4y0p82251q96br1o60828
q7rtu3q02108
q9hbh5q15077
p13521
q8nbx0
q15672
q9y5l3p30793
q7rtt9
p05186
p16562 q5t6x5q02846
o14836p49683p0cg37q9nyz2p07900
q9bxi2p20594
q9bxl7p24158p16066p31327q92886q8izt6 p59046
q9um07q9h497
p31321
p22694
o43708
p11310
p28330p49748o43772
p98077q969e2q9ud71p11309
p49642
q9by32
q9ubf8
o75533
q9nun7
p33765
p20839
q9ubl9
q05639
p54107
p01350
p23526o00555
p51575
o14965
p51161q13085o00763p08247
p22676q05193q05469
q6u949
p00558
q05cv2
o14544q92847
q9h2w2
p54315
cs00064q8nfu7
a6nk58p55084
q13237
p56192
q16854
q9ul49
p12268
o43175
cs00334
q9hau5
p27338
p31644q08397
p01374
o75874 p11498
q96fz2
p09543
p21918
q5hyk3
p42356p21781
p26678q14654
p17480p41231
p14416
q13126
q71rs6p49639q8n8r3
q9y234
q86yv9
q93052p54289o43865 q13573 q9hb75b3kn93
q9h477q9udr5
q13247
q9y4k1q04671
q8tbp6
q7rtu1q15637q9hcu5
q9ulj3q8iyp9
q96s52 p14222p26885
q9y5k5
q7l5y6q7z7k0p50120
o14493
o43193
q8tcc3o00232p55769o95965
q99590q99732p51665q15021q96cm8q06889p09958
q504t8
q15287q9upz3
p21953q86vi4
p51810p17643
o95059q5h8a4
q8nfz5
p40967
q13509p57729q969f9
p48304
q6zsm3p00790
p42765o94788
o43766
o60641
q16186p11308
q16740
q6p1a2
q9bpx3q9nx24o14625
q03014o00233q9p0k8
p28066q8iwt6p98153q9nqe7o95461
p62633 q8wwy8
q13765
q9bqr3p49682q01970o00231
p55036p62333q9unm6 q15003
p62988o95841
o75800
o75832p19883 q9ns56
p26436p31153 q01081q86sq9q96s55
q9uff9
q9nrb3p09172
q00266gypsump15018p52747
p00540
o00571
p35222
o00116
p60484
q9upi3
p84077
q08050
p21730
o95249
q14201
p62191
o60664
p10768
p07360
q9y5x9
q86z14
p20936 q92506
q8n6m6a2myd5p18859
q6p3u7
p10275
p22894
p06280
q9y259
p40926
p11686p01185
p05112
q13164
p49961
o15173
q9gzt9
p29972
vs00017
p01236
p14138p53396
q15165p78542
p13501
p49585
p11926q9byz2
q02388
q9y5k3
p35790
p04629
q14749
q4le39q13423p07550
p01138
p07998
ms00068p01178
p61073
p00480
p49915
p40126p11177
p55263
p00966
p04424
p39086p14920
q02127
q99798
ls00054
p28222
o75746
q03154
p28221
o75936
q8n4m1
p07339
p00338p07195q6zmr3p04181q5t2r2
q9ujs0
p07864
p30990p22061
p09622
p22033
p30520p49189
p98174
q96e11
p54803q01432p23109
ss00413
p22303
a8k5i5
q14739p40925
q14957p63252
q86u10q12904
q9h2b2
p49768p10144
q86yt5q13183q9uld0
q9h6q4q16572
q9y600q16445
p30838q8ndx2
q9ubs5o75791q9h2x9
q14416
q8tdu9q9h992
q99928p05937
o60296
q05586q14833q13948
q13620
p30531p30536p18505
q9p2u7q12879q9p2u8q16478
q13003p43005p01148
p41594
q03518p43004
o60220
p54886
q13224
p51649
q86xe5q9y5l4
p06737
p80404
p11216
q9ula0q8nhs2
o75027
q12797
b4del5p07101q53ga4
p05165
p43007
q16099
q99259
p20472q13255p43003
q9nsd5o75311o95166p49747
p48058p48167
q9h237
q8n808q96g03
p13498
o00757o00217f5gxe5
o95394
q02978
p09110
q6iq20p80108
p01275
p62158
q6gpi1q99895
q9y5k2
p60893
o95467
o95749
p14324o76082
p27169
q9bv23q8wy41p60323
p35558
o14939
p13569
p34059o14578
p01024
ns00183o15360p35030
p10620
q16698p01308
p05305
q96s42
p01584
p47710
o95347
q16822 q7z2e3p29466
q13867
p24752
p06132
p06681
p08174q05066
q8wwi5
q16798
p19440
o43451
q13326
q13093p60174
radix trichosanthisp05814
p07814
p00734
p13726o14495o43688
p03891
q8n726p09848
o95503
q6ns38
p33897q7z5q5
p00709q9bxw9p50336
q02809q15735
o14832
q9nui1
p55199p52209
q96bz4q8iv08
p23141
q13393as00075
p43155q695t7
q9nrd0
p11766
q96rq3
q6pf09
q9h400
ps00173
o75390
p14618q16778
p14136p28838
q8wxs4p20396
p06307
q9um01
p07954
o15146
q03519
p48735q92973
cs00392p01100
p30613p06753
q9y5q6
p49448p00367
q7z569 q6uwp7
p15104
p30542p13798
q08ah1
q9byc2p55809p09683
o00330
p31939
q6ibw4p42695q86xi2
p10636
p21399
p78504
q9ntj3
p21728
p00439
q09428
p38571p10599
q93099
cs00067q04609
q13519q9h2s9p01303o94855p60520q7l0j3
p08908
p78334q401n2q8n1c3q14832
p11086q8wvc0
p56178p07108 p05091o75899
p17752
q7z2h8
p18507q8te85
p78509
p28476q13619 o00222p04217
q9brk5
q05940p50553
p42263
q9umx0o14764
q9upv9p23381
p01213p48169p41250
p28472
p49238q9h598p19801p47869p34903
p78310
p48066q6ib77
p14867
p24046p47870
q9nqx3
p23415p56177
p07320
p42261p47895
o00522
p30038q96nt0
p04207q6tgc4 p02144q07283
p11831
p10826 q9bze9
q96a70
p58215p54920
p53602
q5vyx0
p11274
p35611
p25054q96hn2
p29274p49902
p30837
p02686sd00044
q15842
p09923
q96jp5
q9npb1
p48047p01588
p29275p43119
q8n7h5p21549
p01282 p42262
p05451p05089
o43612p78540
q8n159
q96g91q7z6v5q9bzd2p27707
p55265q9bub4
q9h0p0
p54851
p05187
q16620
q13976
p51580
p47900 q99571
p78563
q9y3b8q9bse5q9ulw8
o00487p55735
o95865
q13505
p21589
o43868q9bxi3
p35251q8tcs8q9h244
q96sz5p16109
p09603
o00337
q9ns39p39905o43318
p16220
p22079
p13866
p00746
q00796
p25445
p01009
p48065
p55011
q9c000
p04746
p58012
p19961
o43280
p05062
ps00098
q14397
q03393
p51570
ns00189
q99643
p01241
p43304
ss00357
p21796
q9y3z3
p17735
p11217
p35670
q05329
q9br01q8wwt9
p28329
q9uby0
q7z721
p05019
q8net4
q16851
p45985
q8n699q99759
p15382
q9uj14
q9nr45
p36269
p09919
q9y5p6 p36871
o43488p20800p04732
p02760
cs00201p16591p09467
q16538
p21695
p05067p32189
p00156o00591
q9ulc6
p07288
p06276q9nph2
q9y4x5
p02675hs00118
p57773
p61812
p16070
hs00119
p46098
p02671
o75610
o60701
q9nz01q96hy7
p08237
p35575
p09455
p08559q8iww8
p08100
p43630
o60565
q12772q08257
p99999
q96ha7
p55210
p02735
p07947
p00533ls00097p22466
p11168
p61626
q9h227
p10253q53h76
q32zl2
q16539
p55212
p00492
gs00029q8ivi9p54136
q9h300p28288p10909q9h115p01210q16836
q13145q9y314p27708
p29323
q96jb6q08629
q8iuf8q9y4k0
p50440q9bvc4
q9nr19p36551o95389
q8tas1p01286
q15166
p11488q9nra2p41217
p07306
p28715p43699
p08729
q8nbp7
q9nrc8
q9p2w7
q3kpi0
q8tat5p22674p35900
q9uif7
p08779q5jvs0
o43542
q9uk23
p23946
p48061
q15468
q9byv7
p10586
q3ybr2
q9h2h9
p22614
o15121q15399
o14678q99447
q12873
p62805q9ubi9
q14191p06731o95948
q16531
q96t60q13569p17661
o60673q99607q12809p48751p32971 q13315
q9ubz9
q9nti5
q9y6m5p09327
o75794
p49281
p07307
q9c056q06430
o43524
p13611q96q15
q92630q01959q16891
q9y272o60936
p34972p41143
q05516
q9y5s1
q86yb8
q9nrm7p19021
q13258
q9c026p09238o95727
p51812
p57059
o15297p23025
p09488
q9ntg7q7l1v2
o15350q00839
o75528
q15181p33993
q9hba0
p67809
o14735
o00519
ps00182
p49757
q13470
p13073
p20719o15409
q13547
q8wx93q96q91
q9nrg4
p61567
o75928
p49959o15392
q12834
p48651
q00534
p08263
p09211
cs00649
p16104
p02790q9un88
q9y6f1q13156p39748p07992
p06702o15354p16455
p18887
q92889q96fi4p08670
p18074
q9nwh7
q01831
q7z7h5
q6p1n9
e5kn95
p54278q92698
p15088p19013o15527p13051q96se4
o14757
q9ugn5p49917q96qu6q59ek9
q9ukk3q86vb7
p16422p28906q9ugp5
q9nuw8
q86wg3
p49916p04233
q8ivs2
p22083
p46939q96ja1
p40692
p18858
p16150q86yl7
q13421o95271
q8iun9
p35408p54277q9ns91
p23396
p28336o60907p23975p12956
q6gmr7
q93009q9uhb4
q53lp3q9h2x6
q96he7
q96a56q9nrn9
ps00183os00096
p10721
q16611
q96qu1
q5t4w7
q9gzp4
o00764
ms00241
q16584
o15231
q9h3d4o95997
q9h7z6q15717p37023 q9bzq4
o14980q86wx3
q08431o75083
q63hm1q14186
p00738q969k7
q9h8v3o15229
p30466p68402
p02655q16609
q8wwy7q9nww6q96i15
q8tcv5
q9gzn6
q8tb61p49247
q9bw04p18405
p46952
q3t906
q9ubm7q15102
q96t66q8ixj6q8tds4
q9nxa8
p48023
q9han9
q9h221
ophiopogonis tuber
q6xqn6p29401
q9h222
p43220
q9y5s8
p06727p08138
p11021q13426
o43819p21980
q9h3s4
q9y6e7p05546
q08945p26358
o95155
q9buj2
q8nb91
q9bzz2b4dln7
p33121
sd00085
p53634
p07148
o00767
p01130
p55851ps00024
p15692
q15303
q9hb55
q9bx93
p33261
p41159
q92993
q9nz20
q9ubk2
q9y618
p49792
q7z6w7p07359
rs00133p05412
p07099
as00063
ss00282
p20813
p04637
o76074
is00262ns00109
q8izm9
q99217cs00029p24385
q07820
o95477
p11712p00813q15848
p00491
p05181
q59f19
p10415
q13133
q86v15
o75342
q03013
p08684
q96kp6
o75477
p04798q13829
q6p589q07812
p21860
p11215
p02749
q9nrf9
p08758
p11387
ws00005
o14788
p37837
q03169p54098
p35354
p45983 p18440
scutellariap06746
o95379q5gj75q8wvp5
p15559
p01375
p21964
p34810
ls00171p05231
p19838
q8tdx5
p35869
q9h7z7
p23219
p05362
p20815
o95622
q8tdf5
p14555
p48357p06858
q8nf37
q6p9f0
p20933
p82932
q86ue4
q9h2a7
p08603
p53779
q92186
p00739
q14289
p55087p54252
q14376
o60218
p50053
q9nxz1
o43914
o95968
p15260
q04446
gs00157
q9uj99
q08499o75355q9ugm6o95500q7z2y5p36382p35212q96kn9 q5t442p50747
a6nfd8se01038o15195
q13546
q15628
q9y285
p06850
q9hc78
p83916p08034
q5nv69q9bx67
p57087q92959
o43474q9nqz7
o95169
q06210
o75054p01589
p07315
q8n1n0
q8wwv6
p09525
o75712p61421
q92851
o15439p37059
p52790
q96gd0p07902q9y6a1
q96qt6
p15311
p22732
p46734
q9bqt8
p42226p04083o95452p23582o43396
p43652p78562
p29376
q92616q9nr09
o14521
js00028 p01137
q15004
p04406
o14494q03135
o15264
p01160
p00568
p04040
p01270
p04179
p06729
p61278
q16665
q9hbf4
o14727
p11413
ss00360
p36383p13500
gs00300
p14780o95479 p08236
ps00477
p00797
p05164
q96ap0
p49419
o43306
p06744q06609
p55211
q15392
es00154
q14994
p20702
o94806
p33260
q8n5b7
p22392
p10163q8wtv0
p78380
p01133q5hya0
q9y2z9
q96gr2p21583p04053
o60346
q13526
p53801p00750
ns00107q14534
p18031
q15274
q96ri1q8izn7
p55055
q8n104
q9y2r9
p49888q9ubb5
cs00128
us00005
p22680
o15197
o95544
p02741
p24941
se00229q96pd7
p11509
o95905
q969g5
bs00084p02008
p27797
o95411
q9y5b9
q9nqa5
q9nyg8
p60510
p78524
ls00212
q05823
o75310q92839
o75795
p36537q9y4x1
q16411p30405
q9npz5p54855
p10619p08833q9np50
o14503
q15031 q9hcn2
q6uwm9
q8iwy9
os00093 q16696q9h5v8
p26440
q03468q12778q6ia69
q15047o94776
q13464q00987
q12888
q86vz5
p06133q9by64
p24903
q92819o94910p12955o15169
o94766
p48449
q14677o00219o00560
p61086p78549
q9h2k2
q96kb5
p36639
q96lt4
p19447
p19544
p52701
q9bzj4
q8tdg4q12980
q96rr4p28908q96fl8
p40198
q14683p07093
p08575p09936
p15941
p09210
o60760p35754q9np99q04760o15217p23921
q9h4y5p21266
o95881
p30711q8ted1
q16772
q7lg56p48637
p31350o14880
p0cg29q16775
q7rtv2p78417
q8tdv2q92600q15904
p30953q9npf7
q99685q9y316
p35913q5vwk5
p22891
q14690
q5sci4q9byb4
q16236
p28340p10646
p32780q9uq84
p27695
p07204
q9y385
p12004q96la8
q5xg87
q13352p09884 q86sq4
o95243p09104
q8tdm6q9ubx5
p41219q9y4e5
p56180
p02545q676u5
o95069
p09382
q92820
q9nzj6
q8net8 q8nhy0
p32418
q8wux1
q9np80p51572
q99807
q9up65
p06127
p0c869
p34932
p41161o43242
q9hay6
p53370q8ndx5q9gzr5
p59796
q12974
p10243p20700
q07108
p18283
o00180
q99218
q7z449
q8ner5
o75665
p98155
p53041p22105
p55268
p51582
o15496
q8nff5
q9hcm9
q86sj6
q86tx2p33176 q9up95
q5fve4p15144
q9unk4
p25874
o15254
o14842
q8wui4
q96cc6p24462
q12791q9bqi3
p31431 q8n6t7
q86yn6
q5nul3
q99873p25092
q99808
q99698
p49753
p49591
p08709
ls00058p68871 o15540
a0mnp2
p55916
p46527
o95573
p02751
q9nud9
q16655
p36969
q04759
cs00116o15151
q13790
q8n1l4
q9uid3
p58743
q02928
cs00989
cs00880o43808
p17900
p20142
q9npf0
q9uku0
ms00393
p22736q9p296
q6p5z2q15139
q13508p22307q96l15
p24723q9bzl6
p52961
q9y5v3o60603q99729
q7z408q9hcc0
q99445p31946
p13584
p16401
d3dvq6q75wm6q02539q8iv61p11142
p61204
q8td17o60547q573b4p84243
q99570
p07305
p19878
p46094
p41235
o14645
q9btm1p0c5z0
o43399q9nrw4q06265
q3syg4
q8wvc6
q63hq0q16853
p30301q9uql6p19484
a5pkw4q9by66
q9nr22q9ukl6
p17677p15151p10412
o60551p50135
p43080o00159
p62330
q96a08
p14770
q96pb1p49006
q9y6e0
a8k1m2q8ivk1o00458p78556
p0c5y9
q9h4x1
p14598
q8nff2
p63098q9ui40
o75367p01112
p57723
p16402q6qhc5
p16403
q9h305
q12884
q9hc58
q13702
q9ubs4
o00622q01101q99836
q96pz7o95429p55198
q6nt55
o43781
p41238q86tg7
o75715p22352
p02808
p07203q96sl4
q7z407
q9p0m6
q8wwk9p62745
q16829p29966o95267
q7z2w7
p18085p08567
q9h1r2p30968p35243q9bxl5
p23527p22492
q71ui9
q99418q8nb12
p22570
p42357
q9nz09p98194
q9h211p13497
q9umx3
q9ul01
q8nbh2
q9y397
q9nv23 q7z5g4p54840p38117
q96k12
q9y5r8
q8iuh5
o75460q8n6n7
p47974p60880
p0c7u2
q9y296
q8wvx9
p06703q14541
q14542q9nqg7p28039
p0c0s5
p56704o60488
o43918
o95237
q9nzj5
p35556q9nyx4
q9h009q99767
p12259
p40879
p24386
o00472
o43617
p06241
q9nr71o95822
q7l5a8
q99784q15434q9hd26o76015q96rj0
q05084q96mw7
q13371q96ga7
a4fu01
p24666
q01469p61769p13807
q5t215q06203
p17861o76054
b3kp71
p13804
p05198
o15228q96c24q16654
p09466
a2arm6
o00115p07988
o75781
p10145q96ph1
p51589
p37231
q9hcl2
q9y6r1
p04114
p40145
q8nfm4
p16233
o60503
o95969 q8n9l9
p14672
hs00094
p12104
p19835 q8tcg5q08462
p45844
p32322
q9nzd2
p23786
q96dy3
q86x55
p03956
p40939
p55157
q6p1m0
o75159
o14561
cs00250
q9has3p22612
p10644
p30825p31323
p13861
p17612
p61586
q9nwu1
q08426
p36956
p11597
p20292
p15090
p41222
as00249
q86sk9
cs00285
es00010
q9h015
q08828p00167
q9y6y9
q53hv7
q9nzk7se00285
p31749q9uhd2
p04275o75907
o95255 q9bzm2
q15067p49327
p51828
p08311
p29083
q9ubu3
o00628
q92523
q07869
p50416
os00061
p19971o60266
p16671
q9p0j1
o00206
p07205
q6jqn1
q8tcd5
cs00644
p05413
p11150
o00154
q96a54
o43581q14108
q9ulc5
p32238
p07098
p50542
p00414
o95278
q16401
q99541 q86x29
ps00111
p30084
o75381
p30419
q96t53
p31025
p50897
q96iy1
q9y2p5
o95158
q14653
p02774
q96ad5cs00020
q8wts1
p54317q6piu2
o75688
p12643
q9gzx6o43315
q9nst1
o00548q5k4l6q9hd89p10082
q9uhr6
p24557
p55061p06748
q16647
q9uet6
q9npa8p01298
q9y2p4
o60240
q08477
q99988
p20132
p46063
q86y26
q16626q93070o75478
q00978
q9gzx7
p0c1z6
q9ha82
q6ub99
q9bvm2
p37275
p63096
p17081
q9hb03
q6uwv6
as00574
q13123
p02654
Figure 2 Compound-protein network of BHDWT Red triangle node compound in BHDWT blue circle node protein
1198752= pp (119883 119884 neg) pp (119884119872 neg) 997904rArr pp (119883119872 pos)
1198753= pp (119883 119884 pos) pp (119884119872 neg) 997904rArr pp (119883119872 neg)
1198754= pp (119883 119884 neg) pp (119884119872 pos) 997904rArr pp (119883119872 neg)
1198755= pp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
1198756= cp (119883 119884 119885) tag (119883) 997904rArr draw (119883 119884 119885) tag (119884)
(3)
1198751cup 1198752cup 1198753cup 1198754is the set of rules to deduce the eventual
effects of chemical compounds on the disease 1198751indicates
that if the effect of119883 to 119884 is positive and the effect of 119884 to119872is positive too then the effect of 119883 to119872 is positive Similarderivations are defined in 119875
2 1198753 and 119875
4 They may be used
as many times as necessary to the final disease node 1198755and
1198756are the rules to extract the nodes for the network They
can also be used to draw the subnetwork when the originalnetworks are too complicated to analyze 119875
5indicates that if
119883 (protein) acts on 119884 (protein) with the effect of 119885 and 119883 istagged then this relation is extracted and 119884 is also labeledfor further derivation 119875
6is similar to 119875
5except that the
starting entity is chemical compound 119883 In the construction
4 Evidence-Based Complementary and Alternative Medicine
PI3K
AIFENDO-G
CASP12
TRADD
Cn
stresssignals
MYD88
calpain
JNK
LEP
IL-1R
IL-1
GHRP
G6PT1 GIPCPTI
Beta3AR
IRS
FFASOCS
AA1R
NGF
FATCD36 TRKA
FACS
SGLT2P K C
TNFR2 TNFR1
erstress
GLR IAPPmNCETNFA
INSR
SHC
INS(hyperinsulinism) PTP1B
LAR
APS
ATMIL-3R
D-GlucoseDNAdamage
D-GlucosehyperglycemiaIL-3 AMPK TNFa
GLUT2
BAX
RosFAS-L
PDX-1
FLIPTRAIL-R
APAF-1
MafA
FAS
BAD
CASP10
P53 BID
CASP8
GK
FADD
TRAIL
DFF40
DFF45
CASP3
CASP6
CASP9
CASP7
cytc
cleavagesubstrate
DNAFragmentation
impaired insulin secretion
Pyruvate
apoptosis
PYK
INS
ACC
lipogenesis
ATP
N F B
I B IKBA
IKK
BCL-2IAP
PDK12
SREBP-1c
GLP1R
IRAK
11HSD1 UCP1
P K C
UCP3
T2D
glucolysis
SOCS3
P G C 1 -
mTOR
4EBP1
STAT3
lipidhomeostasis
S6
- M S H
P70S6K
NIK
RIP1TRAF2
JNK
Kir62SUR2ASUR1SUR2
VDCC
cb1
CRKII
Obesity
ADIPO
ADIPOR TC10
GRF2
EX070
MEK12
Insulin resistance
RAF
DPPIV
LEPR
Retinoic acid
EIF4E
P K C
PKA
HSL
antilipolysis
GRB2
SOS
GLUT4
RAS
PDE3TSC2FOXO1PP1
RHEBGYSFBPPHK
AKT
cAMP
GSK3B
PEPCK
AMPK
G6PC
AGRP
SHP-2
MNK
CPT-1
Ceramide
RXR
ERK
AMPKK
P P A R
Long-chain fatty acid
CIP42
PYGPPARG
glycogen
glycogenesis IRS
glucosehomeostasis JAK
Figure 3 Biological network of T2D Blue circle node protein purple circle node nonprotein node green octagon nodes type 2 diabetesldquo997888rarrrdquo one node enables the expression of the next node raises the expression or enhances the activity of the next node ldquo⊣rdquo one node inhibitsthe expression of the next node lowers the expression or weakens the activity of the next node
of network 1198756will be used once and 119875
5may be used as many
times as necessary to the targetThe modes of action between nodes include the positive
(pos) and the negative (neg) effects If we define the positiveeffect as ldquo1rdquo and the negative effect as ldquominus1rdquo the ultimateinfluence of intervention will depend on the product of eachstep in the whole signal pathway For example compound Ain Figure 1 influences protein T through 3 pathsThe effect ofA to T is negative through the path ldquoA-c-d-e-f-Trdquo and the pathldquoA-b-h-k-i-m-Trdquo This effect is uncertain through the thirdpath ldquoA-g-h-k-i-m-Trdquo taking into account the negative effectfrom the feedback path m to g In this paper we neglectedthe effects produced through feedback because the role offeedback is expected to regulate the magnitude of effect butnot to alter the overall mode of action of effect So theeffect is positive through the third path After the effectsof one compound on the ultimate node (ie T in Figure 1)through all signal pathways are determined we can selectthe effective compounds based on the desired effect If thedesired effect on the ultimate node is positive the compounds
with positive effects through all pathways will be selectedas active components The compounds with negative effectsthrough all pathways will be ruled out The components withboth effects through different pathways need further analysison molecular mechanism If their undesired effects can becountered by other compounds they may also be selected aseffective components to be used together with the counteringcompounds The opposite analysis will be done if the desiredeffect on the ultimate node is negative Consider
119878 = 1198781cup 1198782 (4)
1198781is the set of entities with structure cp(119883 119884 119885) or
pp(119883 119884 119885) in biological network of disease which are thebackground for deduction 119878
2is the set of labeled compounds
or proteins expressed by tag(119883) 1198782is the initial conditions
for deduction
22 Data for Construction of Component-Protein Networkof BHDWT Formula For decades BHDWT has been usedto treat T2D at the Beijing Guang-An-Men Hospital [12]
Evidence-Based Complementary and Alternative Medicine 5
beta-Sitosterol
Baicalein
Adenosine
T2D
linoleic acidIsoliquiritigenin
Acteoside
Chrysin
Ferulic acid
Nicotinic acid
Lauric acid
CatalpolMangiferin
Berberine
Mannitol
Phenylacetic acid
Scutellarin
Rutin
Succinic acid
Wogonin
Figure 4The effect of compound of BHDWT on T2D Red triangle node chemical compound green octagon nodes type 2 diabetes ldquo997888rarrrdquoone compound has the positive effect on T2D ldquo⊣rdquo one compound has the negative effect on T2D
BHDWT has a positive effect on blood glucose control andsymptom control for some patients in the early stages ofT2D The BHDWT formula consists of eight herbs includ-ing gypsum Anemarrhena asphodeloides Bunge rehmanniadried rhizome radix trichosanthisOphiopogon japonicusKerGawl Coptis chinensis Franch Scutellaria baicalensis Georgiand Glycyrrhiza uralensis
The components of the BHDWT formula came fromthe Traditional Chinese Medicines Database (TCMD) [13]the State Administration of Traditional Chinese MedicineBasic Information Database (httpdbsharecintcmcomZhongYaoJiChu) and A Handbook on the Analysis of theActive Composition in Traditional Chinese Medicine [14]
The compound-targeted proteins were derived from theSTITCH system (httpstitchemblde) [15] By entering thenames or identifiers of compounds or proteins of interestSTITCH provides the list of proteins withmatching or higherconfidence score that the user specified up to the numberthe user specified The required confidence score representsthe possibility of interaction between the entities In orderto obtain more general results the parameter of the requiredconfidence score was set higher than 0 and the interactingentities number was set to be 500 The interacting entitieswith clear mode of action (positive or negative) were chosenfor further analysis
23 Data for the Construction of T2D Biological Network Toconstruct the T2D network we used the data collected fromthe Kyoto Encyclopedia of Genes and Genomes (KEGG)and therapeutic targets database (TTD) KEGG lists thesignal pathways related to T2D and TTD lists the chemicalcomponents used to treat T2D The biological network ofT2D (Figure 3) was constructed using the signal pathwaysfrom KEGG the chemical components from TTD andthe positive or negative relationship between targets andT2D from STITCH The networks were visualized with thesoftware Cytoscape [16]
3 Results and Discussion
31 Compound-Protein Network of BHDWT Formula andthe Biological Network of T2D We first constructed thecomponent-protein network of BHDWT using the datafrom Section 22 (Figure 2 additional file 1 in Supplemen-tary Material available online at httpdxdoiorg1011552013840427) In this network there are 144 compoundsand 2865 proteins The red triangle nodes represent thecompound in BHDWT and the blue circle nodes representthe proteins The biological network of T2D (Figure 3) wasderived from the data in Section 23 using 119875
5and 119875
6defined
in Section 21 with T2D as the ultimate node Biologicalnetwork of T2D contains 146 proteins The effects of each
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
4 Evidence-Based Complementary and Alternative Medicine
PI3K
AIFENDO-G
CASP12
TRADD
Cn
stresssignals
MYD88
calpain
JNK
LEP
IL-1R
IL-1
GHRP
G6PT1 GIPCPTI
Beta3AR
IRS
FFASOCS
AA1R
NGF
FATCD36 TRKA
FACS
SGLT2P K C
TNFR2 TNFR1
erstress
GLR IAPPmNCETNFA
INSR
SHC
INS(hyperinsulinism) PTP1B
LAR
APS
ATMIL-3R
D-GlucoseDNAdamage
D-GlucosehyperglycemiaIL-3 AMPK TNFa
GLUT2
BAX
RosFAS-L
PDX-1
FLIPTRAIL-R
APAF-1
MafA
FAS
BAD
CASP10
P53 BID
CASP8
GK
FADD
TRAIL
DFF40
DFF45
CASP3
CASP6
CASP9
CASP7
cytc
cleavagesubstrate
DNAFragmentation
impaired insulin secretion
Pyruvate
apoptosis
PYK
INS
ACC
lipogenesis
ATP
N F B
I B IKBA
IKK
BCL-2IAP
PDK12
SREBP-1c
GLP1R
IRAK
11HSD1 UCP1
P K C
UCP3
T2D
glucolysis
SOCS3
P G C 1 -
mTOR
4EBP1
STAT3
lipidhomeostasis
S6
- M S H
P70S6K
NIK
RIP1TRAF2
JNK
Kir62SUR2ASUR1SUR2
VDCC
cb1
CRKII
Obesity
ADIPO
ADIPOR TC10
GRF2
EX070
MEK12
Insulin resistance
RAF
DPPIV
LEPR
Retinoic acid
EIF4E
P K C
PKA
HSL
antilipolysis
GRB2
SOS
GLUT4
RAS
PDE3TSC2FOXO1PP1
RHEBGYSFBPPHK
AKT
cAMP
GSK3B
PEPCK
AMPK
G6PC
AGRP
SHP-2
MNK
CPT-1
Ceramide
RXR
ERK
AMPKK
P P A R
Long-chain fatty acid
CIP42
PYGPPARG
glycogen
glycogenesis IRS
glucosehomeostasis JAK
Figure 3 Biological network of T2D Blue circle node protein purple circle node nonprotein node green octagon nodes type 2 diabetesldquo997888rarrrdquo one node enables the expression of the next node raises the expression or enhances the activity of the next node ldquo⊣rdquo one node inhibitsthe expression of the next node lowers the expression or weakens the activity of the next node
of network 1198756will be used once and 119875
5may be used as many
times as necessary to the targetThe modes of action between nodes include the positive
(pos) and the negative (neg) effects If we define the positiveeffect as ldquo1rdquo and the negative effect as ldquominus1rdquo the ultimateinfluence of intervention will depend on the product of eachstep in the whole signal pathway For example compound Ain Figure 1 influences protein T through 3 pathsThe effect ofA to T is negative through the path ldquoA-c-d-e-f-Trdquo and the pathldquoA-b-h-k-i-m-Trdquo This effect is uncertain through the thirdpath ldquoA-g-h-k-i-m-Trdquo taking into account the negative effectfrom the feedback path m to g In this paper we neglectedthe effects produced through feedback because the role offeedback is expected to regulate the magnitude of effect butnot to alter the overall mode of action of effect So theeffect is positive through the third path After the effectsof one compound on the ultimate node (ie T in Figure 1)through all signal pathways are determined we can selectthe effective compounds based on the desired effect If thedesired effect on the ultimate node is positive the compounds
with positive effects through all pathways will be selectedas active components The compounds with negative effectsthrough all pathways will be ruled out The components withboth effects through different pathways need further analysison molecular mechanism If their undesired effects can becountered by other compounds they may also be selected aseffective components to be used together with the counteringcompounds The opposite analysis will be done if the desiredeffect on the ultimate node is negative Consider
119878 = 1198781cup 1198782 (4)
1198781is the set of entities with structure cp(119883 119884 119885) or
pp(119883 119884 119885) in biological network of disease which are thebackground for deduction 119878
2is the set of labeled compounds
or proteins expressed by tag(119883) 1198782is the initial conditions
for deduction
22 Data for Construction of Component-Protein Networkof BHDWT Formula For decades BHDWT has been usedto treat T2D at the Beijing Guang-An-Men Hospital [12]
Evidence-Based Complementary and Alternative Medicine 5
beta-Sitosterol
Baicalein
Adenosine
T2D
linoleic acidIsoliquiritigenin
Acteoside
Chrysin
Ferulic acid
Nicotinic acid
Lauric acid
CatalpolMangiferin
Berberine
Mannitol
Phenylacetic acid
Scutellarin
Rutin
Succinic acid
Wogonin
Figure 4The effect of compound of BHDWT on T2D Red triangle node chemical compound green octagon nodes type 2 diabetes ldquo997888rarrrdquoone compound has the positive effect on T2D ldquo⊣rdquo one compound has the negative effect on T2D
BHDWT has a positive effect on blood glucose control andsymptom control for some patients in the early stages ofT2D The BHDWT formula consists of eight herbs includ-ing gypsum Anemarrhena asphodeloides Bunge rehmanniadried rhizome radix trichosanthisOphiopogon japonicusKerGawl Coptis chinensis Franch Scutellaria baicalensis Georgiand Glycyrrhiza uralensis
The components of the BHDWT formula came fromthe Traditional Chinese Medicines Database (TCMD) [13]the State Administration of Traditional Chinese MedicineBasic Information Database (httpdbsharecintcmcomZhongYaoJiChu) and A Handbook on the Analysis of theActive Composition in Traditional Chinese Medicine [14]
The compound-targeted proteins were derived from theSTITCH system (httpstitchemblde) [15] By entering thenames or identifiers of compounds or proteins of interestSTITCH provides the list of proteins withmatching or higherconfidence score that the user specified up to the numberthe user specified The required confidence score representsthe possibility of interaction between the entities In orderto obtain more general results the parameter of the requiredconfidence score was set higher than 0 and the interactingentities number was set to be 500 The interacting entitieswith clear mode of action (positive or negative) were chosenfor further analysis
23 Data for the Construction of T2D Biological Network Toconstruct the T2D network we used the data collected fromthe Kyoto Encyclopedia of Genes and Genomes (KEGG)and therapeutic targets database (TTD) KEGG lists thesignal pathways related to T2D and TTD lists the chemicalcomponents used to treat T2D The biological network ofT2D (Figure 3) was constructed using the signal pathwaysfrom KEGG the chemical components from TTD andthe positive or negative relationship between targets andT2D from STITCH The networks were visualized with thesoftware Cytoscape [16]
3 Results and Discussion
31 Compound-Protein Network of BHDWT Formula andthe Biological Network of T2D We first constructed thecomponent-protein network of BHDWT using the datafrom Section 22 (Figure 2 additional file 1 in Supplemen-tary Material available online at httpdxdoiorg1011552013840427) In this network there are 144 compoundsand 2865 proteins The red triangle nodes represent thecompound in BHDWT and the blue circle nodes representthe proteins The biological network of T2D (Figure 3) wasderived from the data in Section 23 using 119875
5and 119875
6defined
in Section 21 with T2D as the ultimate node Biologicalnetwork of T2D contains 146 proteins The effects of each
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
Evidence-Based Complementary and Alternative Medicine 5
beta-Sitosterol
Baicalein
Adenosine
T2D
linoleic acidIsoliquiritigenin
Acteoside
Chrysin
Ferulic acid
Nicotinic acid
Lauric acid
CatalpolMangiferin
Berberine
Mannitol
Phenylacetic acid
Scutellarin
Rutin
Succinic acid
Wogonin
Figure 4The effect of compound of BHDWT on T2D Red triangle node chemical compound green octagon nodes type 2 diabetes ldquo997888rarrrdquoone compound has the positive effect on T2D ldquo⊣rdquo one compound has the negative effect on T2D
BHDWT has a positive effect on blood glucose control andsymptom control for some patients in the early stages ofT2D The BHDWT formula consists of eight herbs includ-ing gypsum Anemarrhena asphodeloides Bunge rehmanniadried rhizome radix trichosanthisOphiopogon japonicusKerGawl Coptis chinensis Franch Scutellaria baicalensis Georgiand Glycyrrhiza uralensis
The components of the BHDWT formula came fromthe Traditional Chinese Medicines Database (TCMD) [13]the State Administration of Traditional Chinese MedicineBasic Information Database (httpdbsharecintcmcomZhongYaoJiChu) and A Handbook on the Analysis of theActive Composition in Traditional Chinese Medicine [14]
The compound-targeted proteins were derived from theSTITCH system (httpstitchemblde) [15] By entering thenames or identifiers of compounds or proteins of interestSTITCH provides the list of proteins withmatching or higherconfidence score that the user specified up to the numberthe user specified The required confidence score representsthe possibility of interaction between the entities In orderto obtain more general results the parameter of the requiredconfidence score was set higher than 0 and the interactingentities number was set to be 500 The interacting entitieswith clear mode of action (positive or negative) were chosenfor further analysis
23 Data for the Construction of T2D Biological Network Toconstruct the T2D network we used the data collected fromthe Kyoto Encyclopedia of Genes and Genomes (KEGG)and therapeutic targets database (TTD) KEGG lists thesignal pathways related to T2D and TTD lists the chemicalcomponents used to treat T2D The biological network ofT2D (Figure 3) was constructed using the signal pathwaysfrom KEGG the chemical components from TTD andthe positive or negative relationship between targets andT2D from STITCH The networks were visualized with thesoftware Cytoscape [16]
3 Results and Discussion
31 Compound-Protein Network of BHDWT Formula andthe Biological Network of T2D We first constructed thecomponent-protein network of BHDWT using the datafrom Section 22 (Figure 2 additional file 1 in Supplemen-tary Material available online at httpdxdoiorg1011552013840427) In this network there are 144 compoundsand 2865 proteins The red triangle nodes represent thecompound in BHDWT and the blue circle nodes representthe proteins The biological network of T2D (Figure 3) wasderived from the data in Section 23 using 119875
5and 119875
6defined
in Section 21 with T2D as the ultimate node Biologicalnetwork of T2D contains 146 proteins The effects of each
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
6 Evidence-Based Complementary and Alternative Medicine
impaired insulin secretion
SOCS3
CIP42
GLUT4
N F B
EX070
- M S H
P G C 1 -
TC10
STAT3
LEPR
AMPKK
ADIPOR
Insulin resistance
CPT-1
BAD
AKT
T2DFOXO1
PI3K
P K CPyruvate
proteinsynthesisatp
RHEBMNK
P70S6K
lipidhomeostasisG6PCAGRP
glucolysis
FBP GK
NPY
PDK12glucosehomeostasis
P K C
fasn
SREBP-1c
PHK
GYS
TSC2
PP1
glycogenesis
PDE3glycogen
S6cAMP
EIF4E
GSK3B
PYK
HSL
antilipolysis
PKA
PYG
lipogenesis
ACC
P P A R
AMPK
ADIPO
JAK
PEPCK
GRF2
INSR
cb1
CRKII
IAP
APS
DFF40
CASP6
CASP9
CASP3
DFF45
cytcCASP7
cleavagesubstrate
BID
FADD
TRADD
CASP8
CASP10
P53
TNFR2 RIP1
NIK
BCL-2
TNFR1
DNAFragmentation
apoptosis
TRAF2
I B
SHP-2
growthr
mTOR
Berberine
IKBA
4EBP1
INS
IRSIKK
JNK
proliferationdifferentiation
ERK
TNFa
SHC
RAS
SOS
RAF
MEK12
ELK1
GRB2
(a)
AMPKK
berberine
ADIPO
AMPK
ADIPOR
P P A R
Insulin resistance
AGRP
glucolysis
G6PC
T2D
PEPCK GLUT4
CPT-1
ACC
lipogenesis
(b)
DFF45 CASP6
P53CASP3
CASP9
Berberine
cytc
CASP7
BCL-2
impaired insulin secretion
apoptosis
T2D
DNAFragmentation
DFF40cleavagesubstrate
(c)
Figure 5 Sub-network affected by berberine (a) Sub-network of berberine acting on six proteins (b) Sub-network of berberine acting onADIPO (c) Sub-network of berberine acting on BCL-2 and caspase 3
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
Evidence-Based Complementary and Alternative Medicine 7
AA1R
Mannitol
NGF
Adenosine
Nicotinic acid
Berberine
- M S H
Mangiferin
JNK
P P A R
Scutellarin
Acteoside
GLUT4
N F B
CASP3
P K C
P K C
Wogonin
PYG
Chrysin
Lauric acid
Succinic acid
Ferulic acid
INSR
ERK
TNFa
Catalpol
BAX
Baicalein
P53
BCL-2
INS
ADIPO
LEP
linoleic acid
SUR2A
Figure 6 Relationship between effective components of BHDWT and reactive proteins Red triangle node active chemical componentGreen nodes proteins in the first category (related to insulin resistance) Purple nodes proteins in the second category (related to apoptosis)Blue nodes proteins in the third category (related to both insulin resistance and apoptosis) Yellow nodes proteins in the fourth category(related to insulin secretion)
compound in BHDWT on T2Dwere derived using 1198751 1198752 1198753
and 1198754rules defined in Section 21
32The Effect of Chemical Components on T2D Theultimateeffects of each compound of BHDWT on T2D can be foundby combining Figures 2 and 3 through dTGS We appliedrule 119875
6for each compound node in Figure 3 as 119883 with
the linked protein with 119883 to be 119884 in 1198756rule As a result
all the compounds in Figure 2 that have clear modes ofaction on linked protein were labeledThe proteins presentedin both Figures 2 and 3 were also labeled by applying 119875
6
With the labeled protein as 119883 and the connected node as119884 the pathway describing the relationship can be extractedby applying 119875
5 Those compoundsrsquo ultimate effects on T2D
(Figure 4) can be derived through 1198751 1198752 1198753 and 119875
4 Totally
45 compounds in 7 TCMs (except gypsum) showed effectson 61 proteins in the T2D biological network Among 45compounds 19 (additional file 2) have a clear mode of action(positive or negative) recorded in STITCH Three kinds ofeffect were found positive negative and bidirectional effectsThe desired effect on T2D is negative Among these 19compounds 120573-sitosterol isoliquiritigenin played a positiveeffect on T2D (ie negative effect on the treatment of T2D)and four compounds (ie scutellarin catalpol mangiferinand acteoside) have negative effects on T2D (ie positiveeffect on the treatment of T2D)The rest of the 13 compounds
show bidirectional effects and will be studied further in thenext section
33 Extraction of the Subnet and Effective Components Foreach of the 13 bidirectional compounds we extracted thesubnetworks affected by each of them to study their effectsin more detail The method of extracting the subnetworkhas been explained in Section 21 We found through sub-networks that the negative effect of rutin and phenylaceticacid on T2D originates from the feedback pathways beingnot expected to override the positive effect through directpathways So these two compounds were not consideredas candidate effective components Some other bidirectionalcomponents have a complex sub-network For exampleberberine acts on six proteins The derived sub-networkincluding all six proteins is too complex for further analysis(Figure 5) Therefore we derived six sub-networks includingone or two proteins reacting with berberine Two of thosesub-networks were shown in Figures 5(b) and 5(c) Throughanalyzing these six sub-networks we found that berberinersquosnegative effect on T2D arises from the direct pathway Finallyall 13 bidirectional components except rutin and phenylaceticacid were selected as candidate effective components Theytogether with the other four negative compounds (ie scutel-larin catalpol mangiferin and acteoside) form 15 candidate
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
8 Evidence-Based Complementary and Alternative Medicine
effective components combinations and will be screenedfurther in the next section
34 Combination of Candidate Effective Components In thissection we try to find out which proteins each of the 15effective components affects This will help us figure out howto combine those components to achieve optimal resultsAccording to the literature insulin resistance and impairedinsulin secretion are two major etiological factors of T2D[17] and 120573-cell apoptosis was considered as one reason forthe impaired insulin secretion [18] Therefore we dividedthe proteins into four categories The first category is onlyrelated to insulin resistance the second category is onlyrelated to apoptosis the third category is related to bothinsulin resistance and apoptosis the fourth category is onlyrelated to insulin secretion Then the effects of each activecomponent were screened according to the category ofproteins (Figure 6) For instance mangiferin acts on insulinresistance related proteins (PPARa) and catalpol acts onapoptosis related proteins (BCL-2) hence the combinationof mangiferin and catalpol was predicted to treat T2D byameliorating insulin resistance and inhibiting apoptosis It isworth noting that Figure 6 missed one protein in additionalfile 2 that is CASP9 This is due to the fact that beta-sitosterol the only compound which enables CASP9 wasignored because of beta-sitosterolrsquos positive effect on T2DThe rest of the compounds in Figure 6 did not act on CASP9
Some of the findings revealed in Figure 6 are consistentwith numerous studies on the treatment of T2D Ferulicacid showed antidiabetic effects in experiments on diabeticmice [19] Mangiferin exhibited the potential to improveblood lipids in T2D [20] Baicalein was demonstrated toprotect pancreatic beta-cells from apoptosis and amelio-rates hyperglycemia in a mouse model of T2D [21] Theexperiment in vitro indicated that berberine can improveglucose consumption (GC) over 30when the concentrationis above 5 times 10minus6molL at the same time berberine alsodepressed cell growth remarkably at the same concentration[22] This finding was consistent with our analysis thatberberine promotes cell apoptosis by promoting caspase 3and inhibiting BCL-2
The compounds that have multiple and counterpartpathways in Figure 6 were still selected as candidate effectivecomponents to treat T2D because their unfavorable effectsmay not be dominant or counteracted by other compoundsas demonstrated in clinical practice For example somephysicians have used berberine to treat hyperglycemia agentin China for many years [23]
Figure 6 also discloses some information useful fordesigning or analyzing component combination Althoughthe hepatotoxicity or pancreotoxicity (typically resultingfrom enhanced cell apoptosis) induced by berberine hasnever been observed in clinics Figure 6 indicated that itstoxicity may be counteracted while constructing drug com-binations with the components that can inhibit cell apop-tosis such as catalpol scutellarin acteoside and baicaleinIn clinical practice BHDWT was used to treat early stages
of T2D when the main disease factor is insulin resistance [8]This can be explained by several effective components actingon green nodes (ie proteins related to insulin resistance)in Figure 6 Similar effects to suppress the insulin resistancecan also be achieved by the combinations of some of theseeffective components according to Figure 6 such as (i) thecombinations of berberine and mangiferin (ii) the combi-nations of berberine and catalpol (or scutellarin or acteosideor baicalein) and (iii) of berberine mangiferin compinationand catalpol (or scutellarin or acteoside or baicalein) Someof these combinations have been validated by the workfrom other researchers the combination of berberine andmangiferin was granted a patent [24] and the combinationof berberine and catalpol [25] has been filed for a patent Allof the results indicate that TCM formula plays its role throughsynergistic effects of multiple components
4 Conclusions
This paper proposed dTGS as an innovative method tostudy TCM formulas It integrates the research achievementsfrom three fields TCM chemistry drug discovery and thenetwork biology The findings include the action trends ofchemical components against one disease (T2D) and theactive component combinations from BHDWT formula Itcan also be applied on other TCM formulas to benefit theresearch on the mechanism of TCM formulas
In addition our work would benefit the developmentof fixed-dose combinations Nowadays drug combinationsor fixed-dose combinations (FDCs) are widely used in thetreatment of complex diseases because of the low cost and theclinical efficiency TCM formulas due to their characteristicsof multicomponents multitargets andmultipath effects mayembody some component combinations or combinationprinciples beneficial to the design of drug combinationOur method provides a systemic approach to reveal thoseprinciples
Last but not least our method provided a novel idea fornetwork analysis Our method is different from the primaryapproach in network research (eg graph theory) in thatwe proposed a series of inference rules derived from therelationship of the nodes and provided a new theoreticalframework for analyzing the complex networkThe feasibilityof this theoretical framework was proved by its successto identify the effective component combinations in TCMformulas
Acknowledgments
This work is supported and sponsored by the Natural ScienceFoundation of China (NSFC 81373985 NSFC 81173568)Program for New Century Excellent Talents in University(NCET-11-0605) and Key Projects in the National Scienceamp Technology Pillar Program during the Eleventh Five-YearPlan Period (2008BAI51B01)
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
Evidence-Based Complementary and Alternative Medicine 9
References
[1] Y Wang C Ding C Wu et al ldquoHPLC-MS and HPLC-MSMS analysis of seven active constituents of Xiao-Xu-Mingdecoction and application to a pharmacokinetic study after oraladministration to ratrdquo Acta Pharmaceutica Sinica B vol 2 no2 pp 188ndash197 2012
[2] N Agrawal J Pallos N Slepko et al ldquoIdentification of com-binatorial drug regimens for treatment of Huntingtonrsquos diseaseusing Drosophilardquo Proceedings of the National Academy ofSciences of theUnited States of America vol 102 no 10 pp 3777ndash3781 2005
[3] A A Borisy P J Elliott NWHurst et al ldquoSystematic discoveryof multicomponent therapeuticsrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 100 no13 pp 7977ndash7982 2003
[4] Z Wen Z Wang S Wang et al ldquoDiscovery of molecularmechanisms of traditional Chinese medicinal formula Si-Wu-Tang using gene expression microarray and connectivity maprdquoPLoS One vol 6 no 3 Article ID e18278 2011
[5] S Li ldquoNetwork systems underlying traditional Chinesemedicine syndrome and herb formulardquo Current Bioinformaticsvol 4 no 3 pp 188ndash196 2009
[6] J Gu H Zhang L Chen S Xu G Yuan and X Xu ldquoDrug-target network and polypharmacology studies of a traditionalChinese medicine for type II diabetes mellitusrdquo ComputationalBiology and Chemistry vol 35 no 5 pp 293ndash297 2011
[7] M E J Newman ldquoThe structure and function of complex net-worksrdquo Society for Industrial and Applied Mathematics Reviewvol 45 no 2 pp 167ndash256 2003
[8] R Albert and A-L Barabasi ldquoStatistical mechanics of complexnetworksrdquo Reviews of Modern Physics vol 74 no 1 pp 47ndash972002
[9] Y Wang ldquoEntity grammar systems a grammatical tool forstudying the hierarchal structures of biological systemsrdquo Bul-letin of Mathematical Biology vol 66 no 3 pp 447ndash471 2004
[10] R Zheng K S Wang and Y Wang ldquoRegulating flow graphmodel of chemical processes in framework of entity grammarsystemsrdquo Computer Engineering and Applications vol 46 pp245ndash248 2010
[11] J Yan Y Wang S-J Luo and Y-J Qiao ldquoTCM grammarsystems an approach to aid the interpretation of the molecularinteractions in Chinese herbal medicinerdquo Journal of Ethnophar-macology vol 137 no 1 pp 77ndash84 2011
[12] L Lin Contemporary Diabetes Mellitus in Traditional ChineseMedicine Peoplersquos Medical Publishing House Beijing China2008
[13] H B Liu Y X Qiao B Liu and J J Zhou ldquoTransformationof traditional Chinese medicine database into data warehouserdquoChinese Pharmaceutical Journal vol 9 pp 645ndash648 2006
[14] X Q Chang and L X Ding A Handbook on Analysis of theActive Composition in Traditional Chinese Medicine AcademyPress Beijing China 2002
[15] MKuhn C vonMeringMCampillos L J Jensen and P BorkldquoSTITCH interaction networks of chemicals and proteinsrdquoNucleic Acids Research vol 36 supplement 1 pp D684ndashD6882008
[16] P Shannon A Markiel O Ozier et al ldquoCytoscape a softwareEnvironment for integratedmodels of biomolecular interactionnetworksrdquo Genome Research vol 13 no 11 pp 2498ndash25042003
[17] F Halperin X Lopez R Manning C R Kahn R N Kulka-rni and A B Goldfine ldquoInsulin augmentation of glucose-stimulated insulin secretion is impaired in insulin-resistanthumansrdquo Diabetes vol 61 no 2 pp 301ndash309 2012
[18] M Y Donath J A Ehses K Maedler et al ldquoMechanisms of 120573-cell death in type 2 diabetesrdquo Diabetes vol 54 supplement 2pp S108ndashS113 2005
[19] M S Balasubashini R Rukkumani and V P Menon ldquoProtec-tive effects of ferulic acid on hyperlipidemic diabetic ratsrdquo ActaDiabetologica vol 40 no 3 pp 118ndash122 2003
[20] T Miura N Iwamoto M Kato et al ldquoThe suppressive effectmangiferin with exercise on blood lipids in type 2 diabetesrdquoBiological and Pharmaceutical Bulletin vol 24 no 9 pp 1091ndash1092 2001
[21] C-H Chen L L H Huang C-C Huang C-C Lin Y Lee andF-J Lu ldquoBaicalein a novel apoptotic agent for hepatoma celllines a potential medicine for hepatomardquoNutrition and Cancervol 38 no 2 pp 287ndash295 2000
[22] J Yin R Hu M Chen et al ldquoEffects of berberine on glucosemetabolism in vitrordquoMetabolism vol 51 no 11 pp 1439ndash14432002
[23] J Yin H Xing and J Ye ldquoEfficacy of berberine in patients withtype 2 diabetes mellitusrdquoMetabolism vol 57 no 5 pp 712ndash7172008
[24] G A Xu ldquoDrug composition for treating 2 type diabetesand diabetic chronicity complicationsrdquo US publication no20090176816 A1
[25] J Young ldquoMethod of treating non-insulin dependent dia-betes mellitus and related complicationsrdquo US publication no20050019435 A1
