Int. J. Mol. Sci. 2015, 16, 2913-2941; doi:10.3390/ijms16022913 OPEN ACCESS

International Journal of

Molecular Sciences ISSN 1422-0067 www.mdpi.com/journal/ijms Article

Systems Pharmacology Dissection of the Anti-Inflammatory Mechanism for the Medicinal Herb Folium Eriobotryae Jingxiao Zhang 1,†, Yan Li 1,†,*, Su-Shing Chen 2,3, Lilei Zhang 4, Jinghui Wang 1, Yinfeng Yang 1, Shuwei Zhang 1, Yanqiu Pan 1,*, Yonghua Wang 5,* and Ling Yang 6 1

2

3

4

5

6

†

Key laboratory of Industrial Ecology and Environmental Engineering (MOE), Faculty of Chemical, Environmental and Biological Science and Technology, Dalian University of Technology, Dalian 116024, China; E-Mails: [email protected] (J.Z.); [email protected] (J.W.); [email protected] (Y.Y.); [email protected] (S.Z.) System Biology Lab, University of Florida, Gainesville, FL 32608, USA; E-Mail: [email protected] Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL 32608, USA School of Chemistry and Environmental Engineering, Hubei University for Nationalities, Enshi 445000, China; E-Mail: [email protected] Center of Bioinformatics, College of Life Science, Northwest A&F University, Yangling 712100, China Lab of Pharmaceutical Resource Discovery, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China; E-Mail: [email protected] These authors contributed equally to this work.

* Authors to whom correspondence should be addressed; E-Mails: [email protected] (Y.L.); [email protected] (Y.P.); [email protected] (Y.W.); Tel.: +86-411-8498-6062 (Y.L.); Fax: +86-411-8498-6063 (Y.L.). Academic Editor: Vladimír Křen Received: 13 November 2014 / Accepted: 22 January 2015 / Published: 28 January 2015

Abstract: Inflammation is a hallmark of many diseases like diabetes, cancers, atherosclerosis and arthritis. Thus, lots of concerns have been raised toward developing novel anti-inflammatory agents. Many alternative herbal medicines possess excellent anti-inflammatory properties, yet their precise mechanisms of action are yet to be elucidated. Here, a novel systems pharmacology approach based on a large number of chemical, biological and pharmacological data was developed and exemplified by a probe herb

Int. J. Mol. Sci. 2015, 16

2914

Folium Eriobotryae, a widely used clinical anti-inflammatory botanic drug. The results show that 11 ingredients of this herb with favorable pharmacokinetic properties are predicted as active compounds for anti-inflammatory treatment. In addition, via systematic network analyses, their targets are identified to be 43 inflammation-associated proteins including especially COX2, ALOX5, PPARG, TNF and RELA that are mainly involved in the mitogen-activated protein kinase (MAPK) signaling pathway, the rheumatoid arthritis pathway and NF-κB signaling pathway. All these demonstrate that the integrated systems pharmacology method provides not only an effective tool to illustrate the anti-inflammatory mechanisms of herbs, but also a new systems-based approach for drug discovery from, but not limited to, herbs, especially when combined with further experimental validations. Keyword: systems pharmacology; herbal medicine; anti-inflammation; Folium Eriobotryae

1. Introduction Inflammation is a fundamental protective response that enables human survival when encountering a microbial invasion or injury, and also maintains the tissue homeostasis under various deleterious surroundings. It must be precisely fine-tuned and regulated, since insufficient or superfluous responses of inflammation may coequally lead to the morbidity as well as the shortening of lifespan [1]. Specifically, the deficiency of inflammatory responses always results in immunodeficiency phenomena such as cancer and infection, while the superfluity of inflammatory responses often contributes to higher incidence rate and mortality in several diseases including diabetes, Alzheimer’s disease, atherosclerosis, rheumatoid arthritis, Crohn’s disease, cerebral ischemia, multiple sclerosis, myocardial ischemia, etc. [1]. All these make it necessary to develop effective and safe anti-inflammatory medications. Actually, some effective medications such as the non-steroidal anti-inflammatory agents (NSAIDs) have been developed for treating various forms of inflammatory diseases. For example, as one major group of the most widely prescribed drugs in the world, NSAIDs, like aspirin and ibuprofen, are powerful anti-inflammatory medications by virtue of their abilities to suppress the biosynthesis of prostaglandins at the level of the cyclooxygenase (COX) enzymes [2]. However, though having been identified effective for treatment of various inflammations for several decades, NSAIDs have their inherent defects, i.e., severe side effects, notably the gastrointestinal hemorrhage [3], mainly due to their nonselective inhibition on two cyclooxygenase enzyme systems. In order to overcome these problems, recently, some novel selective COX2 (one isoform of COXs) inhibitors that may reduce the risk of severe gastrointestinal complications have drawn a lot of attention [4]. However, their applications are still greatly limited by the findings that selective COX2 inhibitors may also lead to a series of severe problems including hypertension, thrombosis [5], myocardial infarction and stroke [6]. Thus, developing more potent and safe anti-inflammatory drugs is still a pressing task. Alternative herbal medicines, characterized as multiple compounds and targets, might have some advantages in developing anti-inflammatory agents in two ways: (1) being natural and therefore maybe less harmful [7] and (2) containing multiple bioactive ingredients which thus may provide synergistic and/or additive pharmacodynamic effects. Fortunately, plenty of botanic medicines like

Int. J. Mol. Sci. 2015, 16

2915

Andrographis paniculata (Burm. f.) Nees (Acanthaceae), Lonicera japonica Thunb. (Caprifoliaceae) and Eriobotrya japonica (Thunb.) Lindl. (Rosaceae) have been demonstrated with potent anti-inflammatory properties. For example, Andrographis paniculata (Burm. f.) Nees has already been used as an anti-inflammatory drug for treating rheumatoid arthritis and laryngitis for several centuries [8,9], and its major constituent andrographolide exhibits anti-inflammatory effect in LPS-stimulated RAW264.7 macrophages through the suppression of STAT3-mediated inhibition of the NF-κB pathway [10]. Likewise, Lonicera japonica Thunb. is also one of the most commonly used herbs in China for treating inflammation and infectious diseases for thousands of years [11,12]. Additionally, luteolin, the bioactive ingredient of Lonicera japonica Thunb., can block the MAPK and NF-κB signaling pathways to inhibit the inflammatory mediators release, and then exhibits an anti-inflammatory effect [13,14]. Nonetheless, due to the inherent features of herbs—i.e., the complex constituents, targets and mechanisms—the exact action mechanisms of most medicinal herbs still remain unclear. Among those botanical drugs with effective anti-inflammatory properties, Eriobotrya japonica (Thunb.) Lindl. which has rich resources in China and Japan, attracts our special attention. Its dried leaves, named as Folium Eriobotryae, have been commonly used not only for clearing away the lung-heat of antitussive and anti-inflammatory agent, eliminating phlegm and relieving coughs in China [14], but also for the treatment of stomachache, ulcers, chronic bronchitis, cancer and diabetes mellitus as Japanese folk medicine [15]. Additionally, some effective anti-inflammatory ingredients were also identified from this herb like triterpenes, flavonoids, sesquiterpenes and megastigmane glycosides [16,17]. For example, triterpene ursolic acid of Folium Eriobotryae exhibits significant anti-inflammatory potency by inhibiting LPS-induced IL-8 production, iNOS mRNA expression and NF-κB activation [18]; the flavonoid ingredient kaempferol also exhibits anti-inflammatory efficacy by inhibition of iNOS protein and mRNA expression, as well as of NO production and NF-κB activation [19]. Therefore, in the present work, Folium Eriobotryae was selected as a typical herbal medicine to probe the anti-inflammatory mechanism of herbs. As a matter of fact, herbal medicines normally contain huge numbers of active pharmacological ingredients. Besides, herbal medicines also target several diverse genes or proteins to exert therapeutic effects, which makes the investigation of their pharmacological molecular mechanisms extremely difficult [20]. As an emerging field, systems pharmacology may provide a new approach to solve complex pharmacological problems [21]. In the context of systems biology, systems pharmacology utilizes an integrated method to investigate the drug actions ranging from molecular and cellular levels to tissue and organism levels [22]. Recently, as an important tool of systems pharmacology, network approach has proven to be useful for integrating abundant high-dimensional biological data [23], as well as for deciphering the relationships among drugs, targets, diseases and pathways [22–24]. Thus, it is prospected that the topological properties of the biological network may greatly enrich our knowledge about the therapeutic mechanisms underlying the multiple actions of herbs. Here, a novel systems pharmacology method was proposed with Folium Eriobotryae employed as an example to dissect the underlying anti-inflammatory mechanism of herbal medicines. In detail, first of all, the compound identification program was performed to seek the active ingredients from Folium Eriobotryae. Secondly, target identification and validation processes were implemented to verify corresponding biological targets of these active compounds. Finally, network-based analyses

Int. J. Mol. Sci. 2015, 16

2916

among compounds, targets, pathways and diseases were carried out to explore the pharmacological mechanism of Folium Eriobotryae in a systematic level as well as to unfold the main targeting proteins and signaling pathways in the process of inflammatory treatment. 2. Results and Discussion Owing to the complex components and synergistic nature of herbal medicines, using herb as therapeutic agent may serve as valuable resources in (a) extracting bioactive compounds for direct use as drugs; (b) discovering combination drugs or multi-target drugs; as well as (c) producing bioactive compounds of novel or known structures as lead compounds to produce entities with high activities and/or lower toxicities. Therefore, in the present work, we employed the systems pharmacology method to dissect the multiple mechanisms of action of herb Folium Eriobotryae, in hope that it may also provide useful information for finding novel anti-inflammatory leading compounds. 2.1. Active Compounds Identification A total of 389 compounds of Folium Eriobotryae are finally extracted from plenty literatures and TCMSP database, including 374 constituents and 15 aglycones by hydrolysis of 53 glucoside-type ingredients as displayed in Table S1. Basically, an orally administered drug must first overcome different in vivo barriers like the digestive tract, when it can further be transported into the blood circulation to arrive at the specific tissues or organs, and then exerts pharmacological effects [25]. Therefore, in the present work, in order to filter the potential bioactive compounds from Folium Eriobotryae, three crucial ADME parameters (drug-likeness, oral bioavailability and Caco-2 permeability) as typical indicators were chosen to evaluate the abilities of herbal ingredients to overcome these barriers, with the detailed screening process displayed in Figure 1.

Figure 1. Active ingredients’ screening scheme.

Int. J. Mol. Sci. 2015, 16

2917

2.1.1. Drug-likeness Evaluation As a matter of fact, druggable compounds usually have some commonalities, such as functional groups and physical properties [26]. Therefore, assuring molecules with a high drug-likeness (DL) index can increase the possibility of therapeutic success, and the higher DL value a molecule has, the larger possibility there is that it may possess certain biological properties. Here, an optimal DL model based on 6511 molecules in the DrugBank database were constructed to prescreen potential active ingredients. Among the 389 compounds, 134 ingredients pass through the threshold of DL ≥ 0.18; thus, they may exhibit favorable drug-likeness features. Actually, some of them have been experimentally validated with various biological activities. For examples, the DL value of oleanolic acid is high (0.76), indicating its large possibility to become a potential drug, which corresponds well with the fact that oleanolic acid has already achieved satisfactory effect in clinical trials in 1977 ever since its successful use as a hepatoprotective drug to treat liver disease in China [27,28]. Furthermore, this compound also exhibits anti-inflammatory and anti-hyperlipidemic properties in several animal models [27,29,30]. Moreover, maslinic acid with DL value of 0.74 possesses anti-inflammatory effect against carrageenan-induced edema and inhibitory effect on histamine-induced ileum contraction [31]; As to 2α-hydroxy ursolic acid (DL = 0.74), it shows inhibitory effects against series of cell lines including HL-60, BGC, Bel-7402 and Hela ones [32]. 2.1.2. Oral Bioavailability Prediction Compounds possessing satisfactory DL indexes have high probabilities to become promising candidate drugs, but this is not absolutely certain for the reason that the DL index is a unified descriptor. To ensure the reliability of screening model, some major specific descriptors of molecular properties including oral bioavailability (OB) and Caco-2 permeability should also be employed as screening parameters [33]. Among these two profiles, OB value is governed by the dissolution of drug in the gastrointestinal tract, the intestinal and hepatic first-pass metabolism, as well as the intestinal membrane permeation, which makes it one of the major pharmacokinetic profiles in drug development. Therefore, presently, the OB model is further utilized to prescreen active ingredients from medicinal herb. As a result, 61 out of the above DL-screened 134 constituents (about 46%) exhibit proper oral bioavailabilities (OB ≥ 30%). As a matter of fact, many compounds with high OB values, such as chlorogenic acid (OB = 63%) and pomolic acid (OB = 75.45%), have already been identified as bioactive ingredients. For examples, chlorogenic acid with dose of 50 mg/kg inhibits carrageenin-induced paw edema in the early phase of rat model, and with the same dose this compound also exhibits analgesic activities in the late phase of formalin-induced pain test [34]. Additionally, pomolic acid shows anti-inflammatory activity through inhibiting the carrageenan-induced paw oedema by 37% (1 h) at a dose of 125 mg/kg in mouse model [35]. Thus, these 61 compounds were finally extracted for further analysis. 2.1.3. Caco-2 Permeability Screening Besides oral bioavailability, Caco-2 permeability is another widely used parameter, which is employed to evaluate the absorption rate of candidate compound across the intestinal epithelial barrier. Therefore, in the present work, Caco-2 permeability model is introduced to evaluate the absorption rates

Int. J. Mol. Sci. 2015, 16

2918

of above 61 molecules. As a consequence of this screening, 26 compounds (with both high OB and DL values) are found meeting the threshold requirement of Caco-2 permeability (≥−0.4). Corresponding to this, several ingredients possessing higher Caco-2 permeability values have also been confirmed with biological activities, like β-sitosterol (Caco-2 = 1.33), methyl ursolate (0.92) and oleanolic acid (0.61). For instances, β-sitosterol exerts anti-inflammatory effect through suppressing oedema formation by 52% with dose of 0.5 mg/ear in mice model [36], while an in vitro study [37] suggests that the anti-inflammatory activity of β-sitosterol is involved in arachidonic acid cascade. In addition, methyl ursolate exhibits the strong inhibitory effect (50% inhibitory dose, ID50 = 0.43 mg/ear) on TPA-induced inflammation in mice model, and also displays inhibitory effect with IC50 value of 477 mol ratio/32 p mol TPA [38]. 2.1.4. Exceptional Molecules In fact, reality constraints of most biological models force the dependent variables to lie in some finite bounds [39], which may cause models to have some limitations, and our models are no exception. Because not only the chemo-physical properties of herbal ingredients should be weighted when building a model to explore the critical factors influencing the therapeutic effects, but also other characteristics of a herbal medicine like the ingredient contents should also be taken into consideration, due to the fact that under certain circumstances, they may also impact the pharmaceutical effects of the herb. Thus, in order to avoid possible omissions of active ingredients during our prescreening process, a large-scale text mining was also performed to acquire all existing candidate compounds of Folium Eriobotryae, which ends up with three exceptional molecules when compared with our screened results as mentioned above. The first one is ursolic acid which has been experimentally validated to be bioactive ingredient of Folium Eriobotryae exhibiting anti-inflammatory and anti-hyperlipidemic properties in animal models [27,32]. However, it only obtains a low OB predicted value (about 10%), which is not surprising due to its poor solubility [40]. Factually, the average content of ursolic acid reaches 7.57 mg/g in the herb [41] which may greatly increase its absolute OB value, and in this way compensate for and assure the acid’s pharmaceutical effects. The other two compounds are chlorogenic acid and rutin. Chlorogenic acid has anti-inflammatory effect through attenuating the activation of JNK/AP-1 and NF-κB signaling pathways [34,42,43]. However, its Caco-2 permeability is very poor (−1.07). The reason may be on account of the fact that this molecule can be further hydrolyzed into caffeic acid by esterase hydrolysis of the colonic microflora [44]. Whereas, the latter can be efficiently absorbed (OB = 51.08% and Caco-2 = 0.24). Factually, caffeic acid has been identified as having antioxidant properties to scavenge ABTS•+, DPPH• and superoxide anion radical, as well as to reduce power and metal chelating [45]. Additionally, it also exhibits anti-inflammatory property in cardiac tissue of diabetic mice [46]. Thus, caffeic acid as a metabolite should also be considered as an active ingredient, despite its low DL index (0.05) less than 0.18. Similarly, rutin has very low Caco-2 permeability as −1.91, but its metabolite quercetin (Caco-2 = 0.02) can easily pass through the colon barrier to exert biological effects [47], which we assume is the reason why rutin still possesses high biological activities in in vivo models [39]. Thus, in the end, the compounds chlorogenic acid, caffeic acid and rutin are also added into the active ingredients database. By combining the modeling results, finally, 29 compounds are selected as the

Int. J. Mol. Sci. 2015, 16

2919

potential active ingredients of Folium Eriobotryae, with their ADME parameters and biological information displayed in Table 1. Out of these chemicals, nearly two thirds (19 compounds) have been experimentally validated with various bioactivities, such as anti- inflammatory, oxidant, diabetic, cancer effects. Interestingly, among these 19 active ingredients, 15 compounds which are mostly classified as triterpene acids and flavonoids exhibit anti-inflammatory activities as shown in Table 1. All these results well demonstrate the feasibility and effectiveness of our prescreening models. Table 1. The pharmacokinetic and biological information of potential active ingredients of Folium Eriobotryae. No.

M_01

Ingredient

β-sitosterol

Structure

OB (%)

45.04

Caco-2

1.33

DL

Bioactivity (PMID)

0.75

anti-inflammation (6967611); antipyretic (6967611); anthelminthic & anti-mutagenic activities (12203259); anti-oxidative (16508147) anti-inflammation (12007601); anti-HIV (12907261); antidiabetic (11830140); antitumor-promoting (14745168)

M_02

tormentic acid

33.73

−0.20

0.71

M_03

sanguisorba, officinalis-7

35.32

−0.25

0.71

M_04

rutin

35.45

−1.91

0.28

anti-inflammation (11680812); antioxidant (10904145)

0.28

antioxidant (9414116); anti-inflammation (18549926); pro-apoptotic (14688022)

0.73

anti-inflammation (18260049); anticancer (22223345); anti-apoptotic (15694664); anti-HIV (9748372)

0.76

anti-inflammation (1359067); antioxidant (22891614); anti-HIV (9748372); antidiabetic (18066109); hepatoprotection (18066109); antibacterial (18069238)

M_05

M_06

M_07

quercetin

pomolic acid

oleanolic acid

31.36

75.45

66.91

0.02

0.26

0.61

Int. J. Mol. Sci. 2015, 16

2920 Table 1. Cont.

No

Ingredient

M_08

Structure

OB (%)

Caco-2

DL

Bioactivity (PMID)

methyl ursolate

47.63

0.92

0.74

anti-inflammation (16204964)

M_09

methyl maslinate

42.79

0.32

0.72

M_10

methyl betulinate

43.41

0.85

0.76

M_11

methyl arjunolate

56.25

−0.05

0.70

anti-inflammation (16204964) anti-inflammation (3769076); anti-HIV (8759159); antioxidant (12802735); anti-tumor (20509140); antiangiogenic (21175131)

M_12

maslinic acid

55.84

0.04

0.74

M_13

linguersinol

48.60

−0.17

0.54

OH

M_14

kaempferol

HO

O

65.70

0.27

0.24

OH OH

O

anti-inflammation (17666881); analgesic (17666881); antioxidant (17551714); anti-bacteria (15234754)

M_15

delta 7-stigmastenol

40.40

1.30

0.75

M_16

2α-hydroxy ursolic acid

39.30

0.03

0.74

anti-cancer (15922841)

M_17

arjunolic acid

43.27

−0.27

0.72

antioxidant (18273903)

M_18

3-O-trans-pcoumaroyltormentic acid

46.94

−0.17

0.34

anti-inflammation (16204964)

M_19

3-O-transcaffeoyltormentic acid

36.33

−0.07

0.32

M_20

3-O-cis-pcoumaroyltormentic acid

33.35

−0.19

0.34

anti-inflammation (16204964)

Int. J. Mol. Sci. 2015, 16

2921 Table 1. Cont.

No

Ingredient

M_21

OB (%)

Caco-2

DL

3-epiursolic acid

39.92

0.51

0.75

M_22

2α,3αdihydroxyursolic acid

37.73

0.25

0.74

M_23

2α,3α-dihydroxyurs -12-en-28-oic acid

45.16

0.21

0.74

M_24

2α,3α,23trihydroxyolean-12en-28-oic acid

41.34

−0.27

0.72

M_25

2α,3α,19αtrihydroxy-12oleanen-28-oic acid

30.03

−0.18

0.72

M_26

2α,19α-dihydroxy-3 -oxo-urs-12-en-28oic acid

46.95

−0.30

0.71

anti-HIV (12907261)

0.66

anti-inflammation (19051345); anti-HIV (8759159); hepatoprotection (8847885); antiangiogenic (21175131); anti-cancer (15922841); antibacterial (18069238)

0.33

antioxidant (12771329); anticarcinogenic (12771329); anti-inflammation (17077520); analgesic (17077520)

0.05

antioxidant (16243424); anticoagulant (19678956); anti-inflammation (19678956)

M_27

M_28

M_29

ursolic acid

chlorogenic acid

caffeic acid

Structure

9.95

63.12

51.08

0.79

−1.07

0.24

Bioactivity (PMID)

anti-inflammation (16204964)

Int. J. Mol. Sci. 2015, 16

2922

2.2. Target Identification Indisputably, herbal medicines exert their therapeutic effects through the synergistic effects of multiple compounds and targets. Thus, besides exploring active ingredients, the therapeutic targets exploration is also a necessity. In the present work, for comparison, first of all, we screened out all therapeutic proteins being targeted by FDA-approved anti-inflammatory agents from DrugBank and TTD databases. The results of target exploration show that, among the 29 active ingredients obtained as described above, 25 compounds are found interacting with 232 targets, while the other four molecules have no targets. Since our work attempts to elucidate the anti-inflammatory effect of Folium Eriobotryae, these target proteins were further sent to TTD and PharmGKB databases to identify if they have relationships with inflammatory diseases. As a result, 53 inflammation-related targets which have interactions with 13 corresponding compounds were screened out. After docking validation process, 43 targets linked with 11 active ingredients are finally considered as the candidate anti-inflammation-related targets and compounds of this herb. The detailed information of the 43 target proteins is described in Table 2. Table 2. The information of anti-inflammation-related targets of Folium Eriobotryae. No. P_01 P_02 P_03 P_04 P_05 P_06 P_07 P_08 P_09 P_10 P_11 P_12 P_13 P_14 P_15 P_16 P_17 P_18 P_19 P_20 P_21 P_22 P_23 P_24 P_25 P_26

Protein Name 3-hydroxy-3-methylglutaryl-Coenzyme A reductase Aldose reductase Arachidonate 15-lipoxygenase Arachidonate 5-lipoxygenase Baculoviral IAP repeat-containing protein 5 C-C motif chemokine 2 CD40 ligand Cell division protein kinase 2 Cholinesterase C-X-C motif chemokine 10 C-X-C motif chemokine 2 Cytochrome P450 1A2 Epidermal growth factor receptor E-selectin Glycogen synthase kinase-3 beta Heme oxygenase 1 Inhibitor of nuclear factor kappa-B kinase subunit alpha Inhibitor of nuclear factor kappa-B kinase subunit beta Intercellular adhesion molecule 1 Interleukin-1 alpha Interleukin-1 beta Interleukin-10 Interleukin-2 Interleukin-6 Interleukin-8 Mitogen-activated protein kinase 14

Gene Name HMGCR AKR1B1 ALOX15 ALOX5 BIRC5 CCL2 CD40LG CDK2 BCHE CXCL10 CXCL2 CYP1A2 EGFR SELE GSK3B HMOX1 CHUK IKBKB ICAM1 IL1A IL1B IL10 IL2 IL6 IL8 MAPK14

UniProt ID P04035 P15121 P16050 P09917 O15392 P13500 P29965 P24941 P06276 P02778 P19875 P05177 P00533 P16581 P49841 P09601 O15111 O14920 P05362 P01583 P01584 P22301 P60568 P05231 P10145 Q16539

Int. J. Mol. Sci. 2015, 16

2923 Table 2. Cont.

No. P_27 P_28 P_29 P_30 P_31 P_32 P_33 P_34 P_35 P_36 P_37 P_38 P_39 P_40 P_41 P_42 P_43

Protein Name Mitogen-activated protein kinase 8 Muscarinic acetylcholine receptor M1 NF-kappa-B inhibitor alpha Nitric-oxide synthase, endothelial Ornithine decarboxylase Osteopontin Peroxisome proliferator-activated receptor alpha Peroxisome proliferator-activated receptor delta Peroxisome proliferator-activated receptor gamma Phospholipase A2 Poly [ADP-ribose] polymerase 1 Prostaglandin G/H synthase 1 Prostaglandin G/H synthase 2 Signal transducer and activator of transcription 1-alpha/beta Transcription factor AP-1 Transcription factor p65 Tumor necrosis factor

Gene Name MAPK8 CHRM1 NFKBIA NOS3 ODC1 SPP1 PPARA PPARD PPARG PLA2G1B PARP1 COX1 COX2

UniProt ID P45983 P11229 P25963 P29474 P11926 P10451 Q07869 Q03181 P37231 Q9BS22 P09874 P23219 P35354

STAT1

P42224

JUN RELA TNF

P05412 Q96F54 P01375

Targets that shared by the herbal ingredients and FDA-approved drugs are marked in bold.

By analyzing all targets in Table 2, two interesting phenomena were observed. On one hand, seven active herbal ingredients target six therapeutic targets of FDA-approved anti-inflammatory agents—i.e., COX2, COX1, ALOX5, ICAM1, PPARG and PPARD—indicating that Folium Eriobotryae may have similar pharmacological mechanisms to Western medicines. On the other hand, two categories of inflammation closely-related proteins are found among these targets, i.e., pro-inflammatory cytokines (such as TNF, IL1A, IL1B, IL2, IL6 and IL8) and anti-inflammatory cytokines (like IL4 and IL10), which indicates that Folium Eriobotryae may also regulate these two categories of targets to exert synergistic or additive anti-inflammatory effect. In summary, compared to the usual mono-therapy for patients with Western drugs, the herb Folium Eriobotryae may have advantages for treatment of inflammation diseases in both regulating multiple inflammatory pathways and exerting additive or synergistic effects for improving the outcome. 2.3. Network Construction and Analysis In the field of systems pharmacology, network analyses provide efficient tool for enhancing the understanding of action mechanisms of multi-drugs in the context of regulatory networks [22], and it is also useful for organizing high-dimensional biological data and extracting useful information. In view of the fact that herbal medicine exerts pharmacological effects through multiple compounds, targets and mechanisms, using the biological network to dissect action mechanisms of herbal medicines is a necessity [48]. Factually, the biological network has already been successfully used in deciphering the molecular mechanism of several traditional Chinese medicines which are used in treating stroke [49] and depression [50].

Int. J. Mol. Sci. 2015, 16

2924

In this work, to better understand the therapeutic mechanism of Folium Eriobotryae, four biological networks were constructed, including D–T, I–T, T–P and T–D networks. Firstly, for comparison, the anti-inflammatory drug-target network (D–T network) for western medicines was constructed based on all FDA-approved anti-inflammatory therapeutic drugs and targets from DrugBank and TTD databases. Secondly, in order to better decipher the anti-inflammatory mechanism of Folium Eriobotryae, the anti-inflammatory ingredient-target network (I–T network) of this herb was built by employing all anti-inflammation-related active ingredients and their corresponding targets. Subsequently, in order to elucidate the major signaling pathways implicated in inflammatory treatment by Folium Eriobotryae, 16 inflammation-related biological pathways were extracted to construct the target–pathway network (T–P network) with corresponding targets of herbal compounds. Finally, target-disease network (T–D network) was achieved by connecting all active herbal ingredients to their related diseases revealing the potential therapeutic effects of Folium Eriobotryae. 2.3.1. Anti-Inflammatory D-T Network for Western Medicines To compare the therapy mechanism differences between herbs and conventional FDA-approved drugs in treating inflammatory diseases, we extracted all therapeutic targets of anti-inflammatory drugs from databases of the DrugBank and TTD, and then built a drug–target network of Western medicine. In the present work, until 30 December 2013, a total of 78 FDA-approved anti-inflammatory drugs and their corresponding 72 target proteins had been extracted, which then were employed to generate the bipartite graph of D-T network as shown in Figure 2.

Figure 2. The D–T network where fuchsia circular nodes represent the DrugBank ID of anti-inflammatory drugs and lime diamond nodes represent the therapeutic targets, respectively.

Int. J. Mol. Sci. 2015, 16

2925

As displayed in Figure 2, the D–T network embodies 150 nodes (78 drugs and 72 therapeutic targets) and 176 edges. In this network, three targets, i.e., glucocorticoid receptor (NR3C1), prostaglandin G/H synthase 2 (COX2) and prostaglandin G/H synthase 1 (COX1), which are linked with 32, 28 and 24 drugs, respectively, become the major therapeutic targets for the anti-inflammatory drugs due to their dominant positions in this network. Taking NR3C1 as an example, a kind of widely used anti-inflammatory drug, the steroids like prednisone (DB00635 as displayed in Figure 2), betamethasone (DB00443) and methylprednisolone (DB00959) exhibit anti-inflammatory properties by activating NR3C1 which regulates the transcription of some response genes containing the expression inducible COX2 and NOS2 [51]. Additionally, cyclooxygenases (COX1 and COX2) are targeted by another widely used type of anti-inflammatory agent, NSAIDs, like ibuprofen (DB01050 as shown in Figure 2), meloxicam (DB00814) and indomethacin (DB00328). Additionally, NSAIDs exhibit their anti-inflammatory effects by inhibiting cyclooxygenase activity which can diminish the prostaglandin production [52]. Therefore, we conclude that these three major therapeutic targets (NR3C1, COX1 and COX2) are closely related to the anti-inflammatory functions of Western medicine. 2.3.2. Anti-Inflammatory I-T Network for Folium Eriobotryae To decipher the action mechanism of Folium Eriobotryae against the inflammatory diseases, its I-T network was constructed based on all active compounds and their inflammation-related targets (Figure 3). As shown in Figure 3, this network is composed of 54 nodes (11 active ingredients and 43 targets) and 82 interactions.

Figure 3. I-T network for Folium Eriobotryae.

Int. J. Mol. Sci. 2015, 16

2926

From this I-T network, four characteristics of Folium Eriobotryae are observed: (1) The promiscuous properties of its ingredients. The “promiscuous property” of a compound means that a single molecule selectively targets on multiple receptors [53]. As displayed in Figure 3, almost 50% of active compounds are connected with more than two targets, which well-reflect the promiscuous property of Folium Eriobotryae; (2) The existence of highly interconnected compounds. Interconnectedness in a network means the average length of the shortest paths between any two nodes [54] and nodes with high degrees are responsible for the high interconnectedness of this network [55]. Thus, in this network, some compounds with multiple targets are observed to be responsible for the high interconnectedness of the I–T network, especially for high-degree compounds M_05 (quercetin, degree = 38), M_14 (kaempferol, degree = 15), M_04 (rutin, degree = 14) and M_29 (caffeic acid, degree = 5). Their degrees are much larger than the average value (3.04), and they thus become highly interconnected compounds of this network. Taking quercetin as an example, it is not only an inhibitor of PPAR1 [56], but also a weak partial agonist of PPARG reporter gene assay [57]; (3) The presence of highly interconnected targets. Obviously, in the I–T network, the degrees of targets COX2, ALOX5, PPARG, TNF and RELA are 6, 4, 4, 4 and 4, respectively, all higher than the average degree of target nodes (1.91), which indicates their indispensable roles in the I–T network. Thus, these five highly interconnected targets (COX2, ALOX5, PPARG, TNF and RELA) should be treated as the crucial targets of Folium Eriobotryae; (4) Synergistic or additive therapeutic actions of herbal ingredients. As mentioned in the C–T network, COX2 is the major therapeutic targets of inflammation. Additionally, protein PPARG [58], TNF [59] and MPK14 [60] are also identified as the potential therapeutic targets of inflammatory diseases. Fortunately, for the herb Folium Eriobotryae, several ingredients are involved in regulating these known or potential therapeutic targets, which may provide synergistic and/or additive therapeutic effects in patients with inflammatory diseases. Furthermore, by comparing and analyzing the D–T and I–T networks, we can observe two interesting phenomena. On one hand, six major targets—i.e., COX2, ALOX5, PPARG, ICAM1, COX1 and PPARD—exist in both networks. Among them, COX2, ALOX5 and PPARG are regarded as the crucial targets of Folium Eriobotryae as mentioned above. Therefore, we deduce that Folium Eriobotryae may show similar anti-inflammatory mechanisms as Western medicines. Secondly, mutual interactions and constraints among multiple targets may counteract or reduce the side effects of Folium Eriobotryae. For instance, as mentioned above, NSAIDs non-selectively target on cyclooxygenase enzymes (COX1 and COX2), especially COX1, and in this way increase the risk of serious gastrointestinal adverse events. Similarly, for Folium Eriobotryae, its ingredients M_05 and M_29 also interact with both COX1 and COX2, which may also produce certain side effects. However, no literature about the side effects of Folium Eriobotryae has been reported up-to-date. Why? By analyzing the I–T network, we find that the major reason may be related to the other seven targets—i.e., MAPK14, PPARG, RELA, TNF, MAPK8, HMOX1 and CXCL10—which are also implicated in gastrointestinal diseases. The interactions and constraints among these seven targets together with COX1 may offset the gastrointestinal adverse effects. Indeed, Folium Eriobotryae has no side effect on the gastrointestinal tract, and has been used for a long time in the treatment of gastroenteric disorders in Japan [61], which may be attributed to the contributions of these targets. Thus, we conclude that, exhibiting similar anti-inflammatory mechanisms as Western medicines, Folium Eriobotryae, characterized as multi-targets, may exhibit therapeutic superiority (less or no side effects) for treating inflammatory diseases in comparison to Western medicines.

Int. J. Mol. Sci. 2015, 16

2927

2.3.3. Target-Pathway Network of Folium Eriobotryae Depending on the specific inflamed tissue or organ involved, inflammatory process generates diverse diseases. However, all of the disorders have common features or conjoint cellular processes, like the activation of a stress signaling pathway and the concomitant production of inflammatory cytokines [62]. Here, to further elucidate the action mechanism of Folium Eriobotryae, the T–P network was constructed in this section. Firstly, 16 inflammation-related KEGG pathways, such as MAPK signaling pathway, NF-κB signaling pathway, rheumatoid arthritis, influenza A and TNF signaling pathway were extracted from the KEGG database. Then, our screened inflammation-related targets were mapped onto the 16 pathways. The subsequent modeling results show that, only one target, i.e., BCHE does not get involved in any related pathway. Thus, based on all other 42 targets and corresponding signaling pathways, a T–P network was erected as displayed in Figure 4.

Figure 4. T-P network. The orange nodes represent the targets, while the purple ones the pathways. Node size is proportional to its degree. Clearly, Figure 4 depicts those pathways that are intensively connected to corresponding targets. For instance, influenza A exhibits the highest number of target connections (degree = 16), as followed by the toll-like receptor signaling pathway with 14 targets, and NF-κB pathway with 12 ones, respectively. By further analyzing the T–P network, we observe an interesting phenomenon that some ingredients bind to the different targets of the network, which, however, turn out belonging to the same biological pathway, and in this way these ingredients may exert complementary actions for Folium Eriobotryae to exhibit synergistic anti-inflammatory effects. For example, seven ingredients including M_04, M_05, M_07 (oleanolic acid), M_13, M_14, M_28 and M_29 are involved in mediating the major components

Int. J. Mol. Sci. 2015, 16

2928

of NF-κB signaling pathway like COX2, TNF, RELA, ICAM1 and PARP1, which may contribute to the synergistic therapeutic effects for patients with inflammatory diseases. Additionally, five compounds of Folium Eriobotryae including M_01 (β-sitosterol), M_04, M_05, M_14 and M_29 may disturb Toll-like receptor signaling pathway through regulating its related 14 proteins like TNF, RELA and JUN, and then provide synergistic therapeutic effects to benefit patients. In addition, by observation of the T–P network, we find that NF-κB and MAPK signaling pathways should be considered as the major anti-inflammatory pathways of Folium Eriobotryae, due to their topological characteristics in the net and their core roles as cross-talk pathways with the other 15 pathways. In detail, firstly, NF-κB and MAPK signaling pathways respectively hold 12 and 10 proteins, which are responsible for the high interconnectedness of the T–P network. Additionally, the crucial targets existing in the I-T network, such as COX2, PPARG, TNF and RELA, are involved in these two pathways. Furthermore, we discover that NF-κB and MAPK signaling pathways being the major cross-talk pathways play crucial roles in several other pathways. For instance, NF-κB and toll-like receptor signaling pathways are coupled together to activate the intracellular signaling cascades, and subsequently regulate IKBKB (inhibitor of nuclear factor kappa-B kinase subunit beta) and downstream protein NF-κB. Indeed, IKBKB is essential for TNF and IL1-induced NF-κB activation and therefore becomes a potential target for strategy aimed at suppressing proinflammatory gene expression for the treatment of several inflammatory diseases [63]. For instance, the anti-inflammatory effect exerted by compound M_05 occurs, at least partially, through the binding to IKBKB, which finally interferes both NF-κB and toll-like receptor signaling pathways. In conclusion, we assume that Folium Eriobotryae may exert anti-inflammatory effect primarily through the regulation of NF-κB and MAPK signaling pathways. 2.3.4. Target-Disease Network According to the expanded human disease network, the majority of diseases have some common genetic origins with others. Furthermore, it is well known that inflammatory responses can result in several diseases, including infection, cancer, atherosclerosis, diabetes and rheumatoid arthritis [1]. Thus, we speculate that herb Folium Eriobotryae may be involved in treating other diseases besides inflammation. To prove this, here, all targets were sent to the databases of PharmGKB and TTD to find their corresponding diseases, and then a T–D network (as shown in Figure 5) was constructed based on all 43 targets and their corresponding diseases.

Int. J. Mol. Sci. 2015, 16

2929

Figure 5. T-D network. The blue nodes show the representative diseases, while the magenta ones the targets. As displayed in Figure 5, 43 proteins are connected with 13 disorders like inflammation, lung disease, arthritis, cancer and atherosclerosis. Considering the fact that all targets in this network are implicated in inflammation, we suggest that inflammation may be closely involved in the pathogenesis of other 12 disorders. Indeed, inflammation has been proved to be implicated in many diseases. For example, Balough et al. [64] showed that inflammation was present early in the course of cystic fibrosis lung disease. Moreover, it had been shown that chronic inflammation was a critical component of cancer growth and metastasis, and reducing inflammation with the appropriate mode would abrogate cancer and recuperate tissue health [65,66]. Therefore, we assume that using Folium Eriobotryae to treat inflammation may be beneficial for the improvement of other diseases which include lung disease, arthritis, cancer, atherosclerosis and another three diseases. Factually, over half of the diseases obtained from this T–D network are strongly supported by the results of the series of pharmacological function studies on Folium Eriobotryae (as displayed in Table 3). For instance, the extracts of Folium Eriobotryae can be used as anti-metastatic agent for cancer metastasis therapy [67,68]. Furthermore, triterpenes extracted from this herb also display anti-proliferative and apoptosis induction activities against cancer cells [69]. Likewise, the total sesquiterpenes [68], triterpene acids [70], flavonoids [70] and 70% ethanol extracts [71] of Folium Eriobotryae all exhibit anti-hyperglycemic effects. All these prove that this herb is a promising remedy for the treatment of diabetes mellitus and cancer. Besides, Folium Eriobotryae has also been used for a long time for the empirical treatment of gastrointestinal [71], skin [72] and lung diseases [73]. Therefore, we speculate that in addition to other eight pharmaceutical effects (including the anti-inflammation, cancer and diabetes, as well as the treatment of lung, skin, Alzheimer’s, gastrointestinal and virus diseases as listed in Table 3), Folium Eriobotryae may still have therapeutic

Int. J. Mol. Sci. 2015, 16

2930

potential for treatment of another five “not proven” diseases including cardiovascular disease, arthritis, atherosclerosis, autoimmune disease and obesity (as displayed in Table 3), which may be the subsequent research objects of pharmacological studies for this herb. Table 3. The pharmacological actions of Folium Eriobotryae. Actions Anti-inflammation

Anti-cancer Treatment of cardiovascular diseases Treating lung diseases

Anti-diabetes

Treating skin diseases Antirheumatic activity Neuroprotective effects Treatment of gastroenteric disorders Anti-atherosclerosis Treatment of autoimmune disease Antiviral effect Anti-obesity

References (PMID) Banno N, et al. [38]; Huang Y, et al. [74]; Uto T, et al. [75]; Ge JF, et al. [73]; Kim SH, et al.[76]; Shimizu M, et al. [31]; Choi YG, et al. [77]; Cha DS, et al. [78] Banno N, et al. [38]; Ito H, et al. [61]; Ito H, et al. [79]; Kim SH, et al. [80] Not proved Ge JF, et al. [73]; Lee CH, et al. [18] Li WL, et al. [71]; Lü H, et al. [70]; Chen J, et al. [81]; Qa'dan F, et al. [82]; De Tommasi N, et al. [83]; Chen J, et al. [72]; Tanaka K [84]; Noreen W, et al. [85]; Gumy C, et al. [86]; Kim SH, et al. [80]; Lü H, et al. [87]; Chen J, et al. [72] Shimizu M, et al. [72]; De Tommasi N, et al. [88] Not proved Kim MJ, et al. [89] Ito H, et al. [61] Not proved Not proved N De Tommasi, et al. [88] Not proved

3. Materials and Methods 3.1. Database Establishment In total, 374 ingredients of Folium Eriobotryae were extracted from literatures and the Traditional Chinese Medicine System Pharmacology Database and Analysis Platform (TCMSP, available online: http://tcmspnw.com), and then added into the ingredient database of Folium Eriobotryae. All structures of these compounds were saved as mol2 format, and then optimized by Sybyl 6.9 (Tripos Associates, St. Louis, MO, USA) with the same parameters as described in our previous work [90]. It is generally known that deglycosylation by colonic bacteria is a critical step in the absorption and metabolism of glycosides in humans. Therefore, based on the rule of glycosidase hydrolysis reaction, 53 ingredients with glycosyls were further deglycosylated, and the corresponding resulted 15 aglycones were added into ingredient database for further research. All information of the 389 ingredients is displayed in Table S1.

Int. J. Mol. Sci. 2015, 16

2931

3.2. Pharmacokinetic Prediction In drug development, poor pharmacokinetic properties of drugs become the major cause of costly late-stage failures. Thus, early ADME (absorption, distribution, metabolism and excretion) evaluation is extremely important [91]. Due to the disadvantages of biological experiments as time-consuming and high-cost, identification of ADME properties by in silico tools has now become an inevitable paradigm in pharmaceutical research. In this work, three ADME-related models, including the evaluation of oral bioavailability (OB), Caco-2 permeability and drug-likeness (DL), were employed to identify the potential bioactive compounds of Folium Eriobotryae. Drug-likeness evaluation. Drug-likeness is a qualitative profile used in drug design to evaluate whether a compound is chemically suitable for drug, and how drug-like a molecule is with respect to parameters affecting its pharmacodynamic and pharmacokinetic profiles which ultimately impact its ADME properties [26]. In our present work, we developed a robust in silico model to predict a compound’s DL index, which was supported by a database of 6511 structurally diverse drugs and drug-like molecules from DrugBank database. In this model, we introduced DL-related parameters as many as possible, which ends up with 1533 descriptors containing constitutional descriptors (like molecular weight), one-dimensional parameters (such as logP, number of H-donors and H-acceptors), two-dimensional factors (for example, polarity number, global topological charge index), three-dimensional profiles (for instance, average geometric distance degree and radius of gyration), as well as other parameters (like total positive and negative charges). All these parameters are calculated based on the Dragon software. Also, the Tanimoto coefficient (as displayed in Equation (1)) was engaged in constructing the DL model as described in our previous work [92].

T (A , B ) 

AB 2

A  B

2

AB

(1)

In this equation, A displays the molecular parameters of herbal ingredients, and B represents the average molecular characteristics of 6511 molecules in DrugBank (available online: http://www.drugbank.ca) database. The higher DL index the herbal ingredient has, the larger the possibility it may have certain biological effects. Finally, molecules with large DL indexes (DL ≥ 0.18) were chosen as candidate molecules for further research. This threshold is determined by consideration of the fact that the average DL index in the DrugBank is 0.18 [92]. Oral bioavailability. OB prescreening is used to determine the fraction of the oral dose of bioactive compound which reaches systemic circulation in the TCM remedy. Here, a reliable in silico model OBioavail 1.1 [93] which integrates the metabolism (P450 3A4) and transport (P-glycoprotein) information was employed to calculate the OB values of herbal ingredients. The optimal model was constructed based on a dataset of 805 structurally diverse drugs and drug-like compounds, which exhibits high predictability (with Rtraining = 0.89 and Rtest = 0.85 for the internal training and external test sets, respectively, and the standard errors of estimate SEEtraining = 0.35, SEEtest = 0.42, respectively). Finally, those molecules with proper oral bioavailability values (OB ≥ 30%) were selected for subsequent research. Caco-2 permeability. Due to the good correlations often existing between the human oral drugs’ adsorptions and their apparent permeability coefficients in human intestinal epithelial (Caco-2) cells [94],

Int. J. Mol. Sci. 2015, 16

2932

Caco-2 cell monolayers are widely applied as standard permeability-screening assay for prediction of the compound’s intestinal absorption and fraction of oral dose absorbed in humans [95]. In our present work, Caco-2 cell permeation values of all molecules were calculated by in silico model using the VolSurf approach [96]. This Caco-2 permeability model was constructed based on 100 drug compounds and displayed an optimal statistical result (R2 > 0.8) [50]. The threshold of Caco-2 permeability is set to −0.4 with regard to the fact that compound with Caco-2 ≤ −0.4 is not permeable. 3.3. Target Identification In order to identify the target proteins of herbal ingredients, we developed a combinatorial approach that integrates the text-mining and chemogenomic methods. First of all, herbal ingredients were sent to the TTD (Therapeutic Target Database, available online: http://bidd.nus.edu.sg/group/ttd/) [97], DrugBank [98] and HIT (Herb Ingredient Target, available online: http://lifecenter.sgst.cn/hit/) databases for mining their corresponding targets, resulting in compound–target interactions supported by the experiments. Secondly, the target prediction was implemented for the herbal constituents using an in-house systematic model [99] which effectively integrates the chemical, genomic and pharmacological information based on Random Forest (RF) and Support Vector Machine (SVM). The optimal model with concordance of 85.83%, sensitivity of 79.62%, and specificity of 92.76% was obtained, displaying reliable predictability. Here, the compound-target interactions with SVM score ≥ 0.8 and RF score ≥ 0.7 were extracted for further research. Finally, all targets obtained from the above two steps were sent to TTD and PharmGKB (available online: http://www.pharmgkb.org) [100] databases to eliminate the noises and overlaps. In order to increase the dependability of these predicted compound-target interactions, molecular docking was performed using GOLD5.1 software. Specific steps are as follows: firstly, the crystallographic structures of target proteins were obtained from the RCSB Protein Data Bank database (available online: http://www.pdb.org/) or were constructed using the Swiss-Model Automated Mode Serve (available online: http://swissmodel.expasy.org/) [101] if the 3D structures were not available. Secondly, before performing the docking process, crystallographic ligands were extracted and then mixed into the docking database for re-dock and polar hydrogen atoms were added. Subsequently, genetic algorithm was adopted to accomplish the ligand conformational searching for docking simulations with default parameters. Eventually, those compound-target interactions with Gold docking scores greater than 40 were selected for further research. 3.4. Network Analysis In order to further interpret the complex relationships among compounds, targets, pathways and diseases, several networks were constructed. The details are as follows: (1) Building compound–target networks which include two different nets. Firstly, drug-target network of Western medicines (D-T network) was constructed based on FDA-approved anti-inflammatory drugs and their corresponding targets which were extracted from the TTD and DrugBank databases. Secondly, in a similar way, another active ingredient-target network for

Int. J. Mol. Sci. 2015, 16

2933

Folium Eriobotryae (I-T network) was built based on the screened active ingredients and their corresponding validated targets. (2) Constructing the target-pathway network (T-P network). To further identify the major signaling pathways of Folium Eriobotryae, we extracted 16 inflammation-related pathways from the database of KEGG (Kyoto Encyclopedia of Genes and Genomes, available online: http://www.genome.jp/kegg/). Subsequently, all targets were attempted to be mapped onto these 16 pathways, resulting in a target–pathway network. (3) Building the target-disease network (T-D network). All target proteins were sent to databases of PharmGKB and TTD to reveal their relevant diseases, and then these targets and their corresponding diseases were used to build the target-disease network. All these networks were constructed by the Cytoscape 2.8.1 software [102], an open software for biological data integration and validation. 4. Conclusions Inflammation is a convoluted process involving hundreds of genes and connections with various complicated diseases which makes deciphering the inflammatory mechanisms more difficult. Folium Eriobotryae, being a potent anti-inflammatory agent, has been used in China for thousands of years. However, due to its inherent features—i.e., multiple compounds and targets—its exact mechanism of action still remains unclear. In this work, for exploring the complex therapeutic mechanisms, an integrated systems pharmacology method is employed to uncover the anti-inflammatory mechanism of this probe herb. The main results are: (1) Through ADME prescreening, 11 anti-inflammatory ingredients possessing favorable pharmacokinetic profiles, mostly flavonoids and triterpene acids, are extracted from Folium Eriobotryae. Meanwhile, 43 proteins targeted by these 11 compounds are identified as the potential anti-inflammatory targets of this herb. (2) By comparing and analyzing the I-T and D-T networks, two major conclusions are obtained: (a) Folium Eriobotryae exhibiting anti-inflammatory effect may be through similar pharmacological mechanisms as Western medicines, reflected by the fact that six proteins (i.e., COX2, ALOX5, PPARG, ICAM1, COX1 and PPARD), being the major targets, both exist in the I–T pathway of Folium Eriobotryae and the D–T pathway of Western medicines; (b) however, unlike mono-therapy of Western medicines, Folium Eriobotryae regulates dozens of anti-inflammation-related targets, and the mutual interactions and constraints among these targets may counteract or reduce the side effects that always occur with anti-inflammatory Western medicines. (3) Through building and analyzing the T-P network, we assume that NF-κB and MAPK pathways may be major signaling pathways regulated by the ingredients of Folium Eriobotryae to exert anti-inflammatory effects, mainly due to the following reasons: (a) the two pathways are highly interconnected nodes in the T-P network; (b) several targets (like COX2, PPARG, TNF and RELA) involved in NF-κB and MAPK pathways are also the crucial targets of the I-T network; (c) the two pathways as the major cross-talk pathways also play critical roles in several other pathways of the T-P network.

Int. J. Mol. Sci. 2015, 16

2934

(4) By analyzing the I-T and T-P networks, we conclude that a combinational use of herbal ingredients may offer additive or synergistic therapeutic effects for inflammation treatment. In detail, several compounds may act on the same or different targets of same signaling cascades to exhibit synergistic or additive pharmacodynamic effects. (5) The T-D network nicely interprets why the herb Folium Eriobotryae can be used not only as an anti-inflammatory agent, but also for the treatment of cancer, diabetes, lung, skin, Alzheimer’s, gastrointestinal and virus diseases. By analyzing the T-D network, we can speculate that Folium Eriobotryae may have the therapeutic potential for cardiovascular disease, arthritis, atherosclerosis, autoimmune disease and obesity. In summary, a systems pharmacology-based method involves systems biology approaches, combined with pharmacodynamics and pharmacokinetics evaluations, to study drugs, targets, pathways and effects. Applying systems pharmacology to herbal medicines not only provides a research platform for deciphering the molecular mechanism of herbs, but also introduces a novel way for developing new drugs from medicinal herbs, as well as for discovering combined therapeutic agents for complex human diseases. However, despite these significant developments, systems pharmacology for herbal medicines still has some limitations: (1) the pharmacodynamic interactions among herbal active ingredients have not been sufficiently taken into account, which can strongly influence the I–T network. However, large-scale experiments used to evaluate the interactions among herbal ingredients are very time-consuming and labor-intensive, or even ineffective in some circumstances due to the intricate their relationships; (2) Active ingredients can have different binding affinities with their corresponding targets, and depending on target engagement, active ingredients and their corresponding targets can be connected with a continuous approach. Nevertheless, the unified approach for quantitative target engagement of different compounds is yet to be further studied. Thus, to completely decipher the action mechanism of herbal medicines, huge amounts of work still need to be done. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/16/02/2913/s1. Acknowledgments Thanks for the financial support given by the National Natural Science Foundation of China (Grant No. 11201049). Author Contributions Jingxiao Zhang, and Yan Li prepared the first draft of the manuscript. Jingxiao Zhang, Su-Shing Chen, and Lilei Zhang participated in the design and analysis. Jinghui Wang and Yinfeng Yang helped to draft the manuscript. Shuwei Zhang, and Yanqiu Pan conducted the statistical analysis. Yonghua Wang, and Ling Yang participated in data interpretation and helped to draft the manuscript. Jingxiao Zhang, Su-Shing Chen, and Lilei Zhang planned and designed this research, and revised the manuscript. Yan Li, Shuwei Zhang, and Yanqiu Pan helped to revise the manuscript.

Int. J. Mol. Sci. 2015, 16

2935

Conflicts of Interest The authors declare no conflict of interest. References 1. 2. 3. 4. 5. 6. 7. 8.

9. 10.

11.

12.

13.

14.

15.

Tracey, K.J. The inflammatory reflex. Nature 2002, 420, 853–859. Crofford, L.J. Use of NSAIDs in treating patients with arthritis. Arthritis Res. Ther. 2013, 15, doi:10.1186/ar4174. Li, R.W.; Myers, S.P.; Leach, D.N.; Lin, G.D.; Leach, G. A cross-cultural study: Anti-inflammatory activity of Australian and Chinese plants. J. Ethnopharmacol. 2003, 85, 25–32. Martinez Gonzalez, J.; Badimon, L. Mechanisms underlying the cardiovascular effects of COX-inhibition: Benefits and risks. Curr. Pharm. Des. 2007, 13, 2215–2227. Gilroy, D.W.; Lawrence, T.; Perretti, M.; Rossi, A.G. Inflammatory resolution: New opportunities for drug discovery. Nat. Rev. Drug Discov. 2004, 3, 401–416. Wong, D.; Wang, M.; Cheng, Y.; FitzGerald, G.A. Cardiovascular hazard and non-steroidal anti-inflammatory drugs. Curr. Opin. Pharmacol. 2005, 5, 204–210. Ernst, M. Harmless herbs? A review of the recent literature. Am. J. Med. 1998, 104, 170–178. Shen, Y.C.; Chen, C.F.; Chiou, W.F. Andrographolide prevents oxygen radical production by human neutrophils: Possible mechanism(s) involved in its anti-inflammatory effect. Br. J. Pharmacol. 2002, 135, 399–406. Mishra, S.K.; Sangwan, N.S.; Sangwan, R.S. Andrographis paniculata (Kalmegh): A review. Pharmacog. Rev. 2007, 1, 283–298. Lee, K.C.; Chang, H.H.; Chung, Y.H.; Lee, T.Y. Andrographolide acts as an anti-inflammatory agent in LPS-stimulated RAW264.7 macrophages by inhibiting STAT3-mediated suppression of the NF-κB pathway. J. Ethnopharmacol. 2011, 135, 678–684. Tang, D.; Li, H.J.; Chen, J.; Guo, C.W.; Li, P. Rapid and simple method for screening of natural antioxidants from Chinese herb Flos Lonicerae japonicae by DPPH-HPLC-DAD-TOF/MS. J. Sep. Sci. 2008, 31, 3519–3526. Shang, X.; Pan, H.; Li, M.; Miao, X.; Ding, H. Lonicera japonica Thunb.: Ethnopharmacology, phytochemistry and pharmacology of an important traditional Chinese medicine. J. Ethnopharmacol. 2011, 138, 1–21. Kang, O.H.; Choi, J.G.; Lee, J.H.; Kwon, D.Y. Luteolin isolated from the flowers of Lonicera japonica suppresses inflammatory mediator release by blocking NF-κB and MAPKs activation pathways in HMC-1 cells. Molecules 2010, 15, 385–398. Qiusheng, Z.; Xiling, S.; Gang, L.; Meng, S.; Changhai, W. Protective effects of luteolin-7-glucoside against liver injury caused by carbon tetrachloride in rats. Pharmazie 2004, 59, 286–288. Ito, H.; Kobayashi, E.; Li, S.H.; Hatano, T.; Sugita, D.; Kubo, N.; Shimura, S.; Itoh, Y.; Yoshida, T. Megastigmane glycosides and an acylated triterpenoid from Eriobotrya japonica. J. Nat. Prod. 2001, 64, 737–740.

Int. J. Mol. Sci. 2015, 16 16.

17.

18.

19.

20. 21. 22. 23.

24.

25. 26. 27. 28.

29. 30.

2936

Yuan, X.; Fu, W.; Zheng, Q.S.; Ju, B.; Xu, B. Antioxidative and cardioprotective activities of total flavonoids from Moldavian balm (Dracocephalum moldavica L.). In Proceedings of the 2010 International Conference on Bioinformatics and Biomedical Technology (ICBBT), Chengdu, China, 16–18 April 2010. Li, Y.L.; Yang, X.W.; Li, S.M.; Shen, Y.H.; Zeng, H.W.; Liu, X.H.; Tang, J.; Zhang, W.D. Terpenoid constituents of Abies chensiensis with potential anti-inflammatory activity. J. Nat. Prod. 2009, 72, 1065–1068. Lee, C.H.; Wu, S.L.; Chen, J.C.; Li, C.C.; Lo, H.Y.; Cheng, W.Y.; Lin, J.G.; Chang, Y.H.; Hsiang, C.Y.; Ho, T.Y. Eriobotrya japonica leaf and its triterpenes inhibited lipopolysaccharide-induced cytokines and inducible enzyme production via the nuclear factor-kappaB signaling pathway in lung epithelial cells. Am. J. Chin. Med. 2008, 36, 1185–1198. Hämäläinen, M.; Nieminen, R.; Vuorela, P.; Heinonen, M.; Moilanen, E. Anti-inflammatory effects of flavonoids: Genistein, kaempferol, quercetin, and daidzein inhibit STAT-1 and NF-κB activations, whereas flavone, isorhamnetin, naringenin, and pelargonidin inhibit only NF-κB activation along with their inhibitory effect on iNOS expression and NO production in activated macrophages. Mediat. Inflamm. 2007, 2007, 45673. Yang, M.; Cheng, C.; Yang, J.; Guo, D.A. Metabolite profiling and characterization for medicinal herbal remedies. Curr. Drug Metable 2012, 13, 535–557. Lu, A.P.; Bian, Z.X.; Chen, K.J. Bridging the traditional Chinese medicine pattern classification and biomedical disease diagnosis with systems biology. Chin. J. Integr. Med. 2012, 18, 883–890. Berger, S.I.; Iyengar, R. Network analyses in systems pharmacology. Bioinformatics 2009, 25, 2466–2472. Cheng, F.; Liu, C.; Jiang, J.; Lu, W.; Li, W.; Liu, G.; Zhou, W.; Huang, J.; Tang, Y. Prediction of drug-target interactions and drug repositioning via network-based inference. PLoS Comput. Biol. 2012, 8, e1002503. Li, S.; Zhang, B.; Zhang, N. Network target for screening synergistic drug combinations with application to traditional Chinese medicine. BMC Syst. Biol. 2011, 5, doi:10.1186/1752-05095-S1-S10. Chan, L.; Lowes, S.; Hirst, B.H. The ABCs of drug transport in intestine and liver: Efflux proteins limiting drug absorption and bioavailability. Eur. J. Pharm. Sci. 2004, 21, 25–51. Walters, W.P.; Murcko, M.A. Prediction of 'drug-likeness'. Adv. Drug Deliv. Rev. 2002, 54, 255–271. Liu, J. Pharmacology of oleanolic acid and ursolic acid. J. Ethnopharmacol. 1995, 49, 57–68. Wen, X.; Sun, H.; Liu, J.; Cheng, K.; Zhang, P.; Zhang, L.; Hao, J.; Zhang, L.; Ni, P.; Zographos, S.E. Naturally occurring pentacyclic triterpenes as inhibitors of glycogen phosphorylase: Synthesis, structure-activity relationships, and X-ray crystallographic studies. J. Med. Chem. 2008, 51, 3540–3554. Price, K.; Johnson, I.; Fenwick, G.; Malinow, M. The chemistry and biological significance of saponins in foods and feedingstuffs. Crit. Rev. Food Sci. Nutr. 1987, 26, 27–135. Singh, G.B.; Singh, S.; Bani, S.; Gupta, B.D.; Banerjee, S.K. Anti-inflammatory activity of oleanolic acid in rats and mice. J. Pharm. Pharmacol. 1992, 44, 456–458.

Int. J. Mol. Sci. 2015, 16 31.

32. 33. 34.

35.

36. 37.

38.

39.

40.

41. 42.

43.

44.

45. 46.

2937

Shimizu, M.; Fukumura, H.; Tsuji, H.; Tanaami, S.; Hayashi, T.; Morita, N. Anti-inflammatory constituents of topically applied crude drugs. I. Constituents and anti-inflammatory effect of Eriobotrya japonica Lindl. Chem. Pharm. Bull. 1986, 34, 2614–2617. Ma, C.M.; Cai, S.Q.; Cui, J.R.; Wang, R.Q.; Tu, P.F.; Hattori, M.; Daneshtalab, M. The cytotoxic activity of ursolic acid derivatives. Eur. J. Med. Chem. 2005, 40, 582–589. Muegge, I. Selection criteria for drug-like compounds. Med. Res. Rev. 2003, 23, 302–321. dos Santos, M.D.; Almeida, M.C.; Lopes, N.P.; de Souza, G.E. Evaluation of the anti-inflammatory, analgesic and antipyretic activities of the natural polyphenol chlorogenic acid. Biol. Pharm. Bull. 2006, 29, 2236–2240. Schinella, G.; Aquila, S.; Dade, M.; Giner, R.; Recio Mdel, C.; Spegazzini, E.; de Buschiazzo, P.; Tournier, H.; Rios, J.L. Anti-inflammatory and apoptotic activities of pomolic acid isolated from Cecropia pachystachya. Planta Med. 2008, 74, 215–220. Navarro, A.; de Las Heras, B.; Villar, A. Anti-inflammatory and immunomodulating properties of a sterol fraction from Sideritis foetens CLEM. Biol. Pharm. Bull. 2001, 24, 470–473. Moreno, J.J. Effect of olive oil minor components on oxidative stress and arachidonic acid mobilization and metabolism by macrophages RAW 264.7. Free Radic. Biol. Med. 2003, 35, 1073–1081. Banno, N.; Akihisa, T.; Tokuda, H.; Yasukawa, K.; Taguchi, Y.; Akazawa, H.; Ukiya, M.; Kimura, Y.; Suzuki, T.; Nishino, H. Anti-inflammatory and antitumor-promoting effects of the triterpene acids from the leaves of Eriobotrya japonica. Biol. Pharm. Bull. 2005, 28, 1995–1999. Metodiewa, D.; Kochman, A.; Karolczak, S. Evidence for antiradical and antioxidant properties of four biologically active N,N-Diethylaminoethyl ethers of flavaone oximes: A comparison with natural polyphenolic flavonoid rutin action. IUBMB Life 1997, 41, 1067–1075. Yang, L.; Sun, Z.; Zu, Y.; Zhao, C.; Sun, X.; Zhang, Z.; Zhang, L. Physicochemical properties and oral bioavailability of ursolic acid nanoparticles using supercritical anti-solvent (SAS) process. Food Chem. 2012, 132, 319–325. Ju, J.H.; Zhou, L.; Zheng, R.X.; Wang, C.L.; Yang, J.S. Determination of ursolic acid in Eriobotrya japonica by HPLC analysis. China Pharm. J. 2003, 38, 657–658. Gonthier, M.P.; Verny, M.A.; Besson, C.; Remesy, C.; Scalbert, A. Chlorogenic acid bioavailability largely depends on its metabolism by the gut microflora in rats. J. Nutr. 2003, 133, 1853–1859. Shan, J.; Fu, J.; Zhao, Z.; Kong, X.; Huang, H.; Luo, L.; Yin, Z. Chlorogenic acid inhibits lipopolysaccharide-induced cyclooxygenase-2 expression in RAW264. 7 cells through suppressing NF-κB and JNK/AP-1 activation. Int. Immunopharmacol. 2009, 9, 1042–1048. Konishi, Y.; Kobayashi, S. Transepithelial transport of chlorogenic acid, caffeic acid, and their colonic metabolites in intestinal caco-2 cell monolayers. J. Agric. Food Chem. 2004, 52, 2518–2526. Gulcin, I. Antioxidant activity of caffeic acid (3,4-dihydroxycinnamic acid). Toxicology 2006, 217, 213–220. Chao, P.C.; Hsu, C.C.; Yin, M.C. Anti-inflammatory and anti-coagulatory activities of caffeic acid and ellagic acid in cardiac tissue of diabetic mice. Nutr. Metable 2009, 6, doi:10.1186/ 1743-7075-6-33.

Int. J. Mol. Sci. 2015, 16 47. 48. 49.

50. 51. 52.

53. 54. 55. 56.

57. 58. 59. 60. 61.

62. 63. 64.

2938

Scalbert, A.; Williamson, G. Dietary intake and bioavailability of polyphenols. J. Nutr. 2000, 130, 2073S-2085S. Li, P.; Qi, L.W.; Liu, E.H.; Zhou, J.L.; Wen, X.D. Analysis of Chinese herbal medicines with holistic approaches and integrated evaluation models. Trac-Trend Anal. Chem. 2008, 27, 66–77. Zhang, J.; Li, Y.; Chen, X.; Pan, Y.; Zhang, S.; Wang, Y. Systems pharmacology dissection of multi-scale mechanisms of action for herbal medicines in stroke treatment and prevention. PLoS One 2014, 9, e102506. Huang, C.; Zheng, C.; Li, Y.; Wang, Y.; Lu, A.; Yang, L. Systems pharmacology in drug discovery and therapeutic insight for herbal medicines. Brief. Bioinf. 2013, 15, 710–733. Vane, J.; Botting, R. New insights into the mode of action of anti-inflammatory drugs. Inflamm. Res. 1995, 44, 1–10. Gerstenfeld, L.C.; Thiede, M.; Seibert, K.; Mielke, C.; Phippard, D.; Svagr, B.; Cullinane, D.; Einhorn, T.A. Differential inhibition of fracture healing by non-selective and cyclooxygenase-2 selective non-steroidal anti-inflammatory drugs. J. Orthop. Res. 2003, 21, 670–675. Espinoza-Fonseca, L.M. The benefits of the multi-target approach in drug design and discovery. Bioorg. Med. Chem. 2006, 14, 896–897. Albert, R.; Jeong, H.; Barabási, A.L. Error and attack tolerance of complex networks. Nature 2000, 406, 378–382. Yıldırım, M.A.; Goh, K.I.; Cusick, M.E.; Barabási, A.L.; Vidal, M. Drug-target network. Nat. Biotechnol. 2007, 25, 1119–1126. Geraets, L.; Moonen, H.J.; Brauers, K.; Wouters, E.F.; Bast, A.; Hageman, G.J. Dietary flavones and flavonoles are inhibitors of poly(ADP-ribose)polymerase-1 in pulmonary epithelial cells. J. Nutr. 2007, 137, 2190–2195. Fang, X.K.; Gao, J.; Zhu, D.N. Kaempferol and quercetin isolated from Euonymus alatus improve glucose uptake of 3T3-L1 cells without adipogenesis activity. Life Sci. 2008, 82, 615–622. Berger, J.P.; Akiyama, T.E.; Meinke, P.T. PPARs: Therapeutic targets for metabolic disease. Trends Pharmacol. Sci. 2005, 26, 244–251. Esposito, E.; Cuzzocrea, S. TNF-α as a therapeutic target in inflammatory diseases, ischemia-reperfusion injury and trauma. Curr. Med. Chem. 2009, 16, 3152–3167. Kumar, S.; Boehm, J.; Lee, J.C. p38 MAP kinases: Key signalling molecules as therapeutic targets for inflammatory diseases. Nat. Rev. Drug Discov. 2003, 2, 717–726. Ito, H.; Kobayashi, E.; Takamatsu, Y.; Li, S.H.; Hatano, T.; Sakagami, H.; Kusama, K.; Satoh, K.; Sugita, D.; Shimura, S.; Itoh Y.; Yoshida T. Polyphenols from Eriobotrya japonica and their cytotoxicity against human oral tumor cell lines. Chem. Pharm. Bull. 2000, 48, 687–693. Haegeman, G. Inhibition of signal transduction pathways involved in inflammation. Eur. Respir. J. Suppl. 2003, 44, 16s–19s. Tak, P.P.; Firestein, G.S. NF-κB: A key role in inflammatory diseases. J. Clin. Investig. 2001, 107, 7–11. Balough, K.; McCubbin, M.; Weinberger, M.; Smits, W.; Ahrens, R.; Fick, R. The relationship between infection and inflammation in the early stages of lung disease from cystic fibrosis. Pediatr. Pulmonol. 1995, 20, 63–70.

Int. J. Mol. Sci. 2015, 16 65. 66. 67. 68.

69.

70.

71. 72.

73.

74.

75.

76.

77.

78. 79.

2939

Erdman, S.E.; Poutahidis, T. Cancer inflammation and regulatory T cells. Int. J. Cancer 2010, 127, 768–779. Coussens, L.M.; Werb, Z. Inflammation and cancer. Nature 2002, 420, 860–867. Cha, D.S.; Shin, T.Y.; Eun, J.S.; Kim, D.K.; Jeon, H. Anti-metastatic properties of the leaves of Eriobotrya japonica. Arch. Pharm. Res. 2011, 34, 425–436. Kim, M.; You, M.; Rhuy, D.; Kim, Y.; Baek, H.; Kim, H. Loquat (Eriobotrya japonica) extracts suppress the adhesion, migration and invasion of human breast cancer cell line. Nutr. Res. Pract. 2009, 3, 259–264. Uto, T.; Sakamoto, A.; Tung, N.H.; Fujiki, T.; Kishihara, K.; Oiso, S.; Kariyazono, H.; Morinaga, O.; Shoyama, Y. Anti-proliferative activities and apoptosis induction by triterpenes derived from Eriobotrya japonica in human leukemia cell lines. Int. J. Mol. Sci. 2013, 14, 4106–4120. Lu, H.; Chen, J.; Li, W.L.; Ren, B.R.; Wu, J.L.; Kang, H.Y.; Zhang, H.Q.; Adams, A.; de Kimpe, N. Hypoglycemic and hypolipidemic effects of the total triterpene acid fraction from Folium Eriobotryae. J. Ethnopharmacol. 2009, 122, 486–491. Li, W.L.; Wu, J.L.; Ren, B.R.; Chen, J.; Lu, C.G. Pharmacological studies on anti-hyperglycemic effect of Folium Eriobotryae. Am. J. Chin. Med. 2007, 35, 705–711. Chen, J.; Li, W.L.; Wu, J.L.; Ren, B.R.; Zhang, H.Q. Hypoglycemic effects of a sesquiterpene glycoside isolated from leaves of loquat (Eriobotrya japonica (Thunb.) Lindl.). Phytomedicine 2008, 15, 98–102. Ge, J.F.; Wang, T.Y.; Zhao, B.; Lv, X.W.; Jin, Y.; Peng, L.; Yu, S.C.; Li, J. Anti-inflammatory effect of triterpenoic acids of Eriobotrya japonica (Thunb.) Lindl. leaf on rat model of chronic bronchitis. Am. J. Chin. Med. 2009, 37, 309–321. Huang, Y.; Li, J.; Cao, Q.; Yu, S.C.; Lv, X.W.; Jin, Y.; Zhang, L.; Zou, Y.H.; Ge, J.F. Anti-oxidative effect of triterpene acids of Eriobotrya japonica (Thunb.) Lindl. leaf in chronic bronchitis rats. Life Sci. 2006, 78, 2749–2757. Uto, T.; Suangkaew, N.; Morinaga, O.; Kariyazono, H.; Oiso, S.; Shoyama, Y. Eriobotryae folium extract suppresses LPS-induced iNOS and COX-2 expression by inhibition of NF-kappaB and MAPK activation in murine macrophages. Am. J. Chin. Med. 2010, 38, 985–94. Kim, S.H.; Shin, T.Y. Anti-inflammatory effect of leaves of Eriobotrya japonica correlating with attenuation of p38 MAPK, ERK, and NF-kappaB activation in mast cells. Toxicol. In Vitro 2009, 23, 1215–1219. Choi, Y.G.; Seok, Y.H.; Yeo, S.; Jeong, M.Y.; Lim, S. Protective changes of inflammation-related gene expression by the leaves of Eriobotrya japonica in the LPS-stimulated human gingival fibroblast: microarray analysis. J. Ethnopharmacol. 2011, 135, 636–645. Cha, D.S.; Eun, J.S.; Jeon, H. Anti-inflammatory and antinociceptive properties of the leaves of Eriobotrya japonica. J. Ethnopharmacol. 2011, 134, 305–312. Ito, H.; Kobayashi, E.; Li, S.H.; Hatano, T.; Sugita, D.; Kubo, N.; Shimura, S.; Itoh, Y.; Tokuda, H.; Nishino, H.; Yoshida, T. Antitumor activity of compounds isolated from leaves of Eriobotrya japonica. J. Agric. Food Chem. 2002, 50, 2400–2403.

Int. J. Mol. Sci. 2015, 16 80.

81. 82.

83.

84.

85. 86.

87. 88.

89.

90.

91. 92.

93. 94.

2940

Kim, S.H.; Kwon, Y.E.; Park, W.H.; Jeon, H.; Shin, T.Y. Effect of leaves of Eriobotrya japonica on anaphylactic allergic reaction and production of tumor necrosis factor-alpha. Immunopharmacol. Immunotoxicol. 2009, 31, 314–319. Chen, J.; Li, W.L.; Wu, J.L.; Ren, B.R.; Zhang, H.Q. Euscaphic acid, a new hypoglycemic natural product from Folium Eriobotryae. Pharmazie 2008, 63, 765–767. Qa'dan, F.; Verspohl, E.J.; Nahrstedt, A.; Petereit, F.; Matalka, K.Z. Cinchonain Ib isolated from Eriobotrya japonica induces insulin secretion in vitro and in vivo. J. Ethnopharmacol. 2009, 124, 224–227. De Tommasi, N.; De Simone, F.; Cirino, G.; Cicala, C.; Pizza, C. Hypoglycemic effects of sesquiterpene glycosides and polyhydroxylated triterpenoids of Eriobotrya japonica. Planta Med. 1991, 57, 414–416. Tanaka, K.; Nishizono, S.; Makino, N.; Tamaru, S.; Terai, O.; Ikeda, I. Hypoglycemic activity of Eriobotrya japonica seeds in type 2 diabetic rats and mice. Biosci. Biotechnol. Biochem. 2008, 72, 686–693. Noreen, W.; Wadood, A.; Hidayat, H.K.; Wahid, S.A. Effect of Eriobotrya japonica on blood glucose levels of normal and alloxan-diabetic rabbits. Planta Med. 1988, 54, 196–199. Gumy, C.; Thurnbichler, C.; Aubry, E.M.; Balazs, Z.; Pfisterer, P.; Baumgartner, L.; Stuppner, H.; Odermatt, A.; Rollinger, J.M. Inhibition of 11beta-hydroxysteroid dehydrogenase type 1 by plant extracts used as traditional antidiabetic medicines. Fitoterapia 2009, 80, 200–205. Lu, H.; Chen, J.; Li, W.L.; Ren, B.R.; Wu, J.L.; Zhang, H.Q. Hypoglycemic effect of the total flavonoid fraction from folium Eriobotryae. Phytomedicine 2009, 16, 967–971. De Tommasi, N.; De Simone, F.; Pizza, C.; Mahmood, N.; Moore, P.S.; Conti, C.; Orsi, N.; Stein, M.L. Constituents of Eriobotrya japonica. A study of their antiviral properties. J. Nat. Prod. 1992, 55, 1067–1073. Kim, M.J.; Lee, J.; Seong, A.R.; Lee, Y.H.; Kim, Y.J.; Baek, H.Y.; Jun, W.J.; Yoon, H.G. Neuroprotective effects of Eriobotrya japonica against beta-amyloid-induced oxidative stress and memory impairment. Food Chem. Toxicol. 2011, 49, 780–784. Li, X.; Xu, X.; Wang, J.; Yu, H.; Wang, X.; Yang, H.; Xu, H.; Tang, S.; Li, Y.; Yang, L. A System-level investigation into the mechanisms of Chinese traditional medicine: Compound danshen formula for cardiovascular disease treatment. PLoS One 2012, 7, e43918. Van de Waterbeemd, H.; Gifford, E. ADMET in silico modelling: Towards prediction paradise? Nat. Rev. Drug Discov. 2003, 2, 192–204. Liu, H.; Wang, J.; Zhou, W.; Wang, Y.; Yang, L. Systems approaches and polypharmacology for drug discovery from herbal medicines: An example using licorice. J. Ethnopharmacol. 2013, 146, 773–793. Xu, X.; Zhang, W.; Huang, C.; Li, Y.; Yu, H.; Wang, Y.; Duan, J.; Ling, Y. A novel chemometric method for the prediction of human oral bioavailability. Int. J. Mol. Sci. 2012, 13, 6964–6982. Artursson, P.; Karlsson, J. Correlation between oral drug absorption in humans and apparent drug permeability coefficients in human intestinal epithelial (Caco-2) cells. Biochem. Biophys. Res. Commun. 1991, 175, 880–885.

Int. J. Mol. Sci. 2015, 16 95.

96. 97.

98.

99.

100.

101. 102.

2941

Ano, R.; Kimura, Y.; Shima, M.; Matsuno, R.; Ueno, T.; Akamatsu, M. Relationships between structure and high-throughput screening permeability of peptide derivatives and related compounds with artificial membranes: Application to prediction of Caco-2 cell permeability. Bioorg. Med. Chem. 2004, 12, 257–264. Hu, G.X.; Zhang, C.H.; Zhao, W.N.; Yu, Q.S. QSPR study on the permeability of drugs across Caco-2 monolayer. J. Zhejiang Univ. 2009, 3, 304–308. Zhu, F.; Shi, Z.; Qin, C.; Tao, L.; Liu, X.; Xu, F.; Zhang, L.; Song, Y.; Liu, X.; Zhang, J. Therapeutic target database update 2012: A resource for facilitating target-oriented drug discovery. Nucleic Acids Res. 2012, 40, D1128–D1136. Knox, C.; Law, V.; Jewison, T.; Liu, P.; Ly, S.; Frolkis, A.; Pon, A.; Banco, K.; Mak, C.; Neveu, V. DrugBank 3.0: A comprehensive resource for ‘omics’ research on drugs. Nucleic Acids Res. 2011, 39, D1035–D1041. Yu, H.; Chen, J.; Xu, X.; Li, Y.; Zhao, H.; Fang, Y.; Li, X.; Zhou, W.; Wang, W.; Wang, Y. A systematic prediction of multiple drug-target interactions from chemical, genomic, and pharmacological data. PLoS One 2012, 7, e37608. Whirl-Carrillo, M.; McDonagh, E.; Hebert, J.; Gong, L.; Sangkuhl, K.; Thorn, C.; Altman, R.; Klein, T. Pharmacogenomics knowledge for personalized medicine. Clin. Pharmacol. Ther. 2012, 92, 414–417. Schwede, T.; Kopp, J.; Guex, N.; Peitsch, M.C. SWISS-MODEL: An automated protein homology-modeling server. Nucleic Acids Res. 2003, 31, 3381–3385. Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N.S.; Wang, J.T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504.

© 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/)

Supplementary Information Table S1. The constituents of Folium Eriobotryae. MolID

Compound Name

CAS

OB

Caco-2

DL

001 002 003 004

(−)-epicatechin (+)-carvone (1-methylethyl)-benzene (1-methylpropyl)-benzene (2R)-naringenin 8-C-α-L-rhamnopyranosyl-(1→2) -β-D-glucopyranoside (2R)-naringenin 8-C-α-L-rhamnopyranosyl-(1→2) -β-D-glucopyranoside_qt (2S)-naringenin 8-C-α-L-rhamnopyranosyl-(1→2) -β-D-glucopyranoside (2S)-naringenin 8-C-α-L-rhamnopyranosyl-(1→2) -β-D-glucopyranoside_qt (6R,7E,9R)-9-hydroxy-4,7-megastigmadien-3-one 9-O-β-D-apiofuranosyl-(1→6)-β-D-glucopyranoside (6R,7E,9R)-9-hydroxy-4,7-megastigmadien-3-one 9-O-β-D-glucopyranoside (6R,7E,9R)-9-hydroxy-4,7-megastigmadien-3-one 9-O-β-D-xylopyranosyl-(1→6)-β-D-glucopyranoside (6R,7E,9R)-9-hydroxy-4,7-megastigmadien-3-one 9-O-β-L-arabinopyranosy-(1→6)-β-D-glucopyranoside (6R,7E,9R)-9-hydroxy-4,7-megastigmadien-3-one (6R,7E,9S)-9-hydroxy-4,7-megastigmadien-3-one 9-O-β-D-glucopyranoside (6r,9r)-3-oxo-r-ionyl-9-o-glc_qt (6S,7E,9R)-6,9-dihydroxy-4,7-megastigmadien-3-one (6s,9r)-roseoside (6S,9R)-vomifoliol-9-O-β-apiofuranosyl-(1''→6')O-βgluopyranoside (6S,9R)-vomifoliol-9-O-β-xylopyranosyl-(1''→6')O-βgluopyranoside (6S,9R)-vomifoliol-9-O-β-xylopyranosyl-(1''→6')O-βgluopyranoside_qt (cis-trans) farnesol (E)-2-hexenyl-gamma-L-arabinopyranosyl-(1-2)-beta-Dglucopyranoside (E)-2-hexenyl-gamma-L-arabinopyranosyl-(1-2)-beta-Dglucopyranoside_qt (E)-3-penten-2-one (E)-6,10-dimethyl-5,9-undecadien-2-one

490-46-0 2244-16-8 98-82-8 68411-44-9

21.97 22.23 28.73 31.13

−0.65 1.35 1.81 1.87

0.77 0.03 0.02 0.02

278614-44-1

45.22

−1.80

0.81

—

49.13

−1.23

0.71

278614-43-0

45.48

−1.59

0.78

—

18.02

−1.24

0.71

—

30.35

−1.65

0.72

—

51.53

−0.76

0.34

—

34.51

−1.70

0.72

—

45.80

−1.80

0.72

52210-15-8

19.80

0.49

0.07

—

32.01

−1.09

0.34

— — —

18.63 48.96 9.73

0.52 −0.11 −1.19

0.07 0.08 0.36

—

43.92

−2.29

0.71

—

41.86

−2.19

0.72

—

28.19

−0.16

0.08

3790-71-4

5.25

1.26

0.06

—

45.37

−1.51

0.38

—

50.63

1.08

0.01

3102-33-8 3796-70-1

48.40 5.65

1.26 1.45

0.00 0.04

005 006 * 007 008 * 009 010 011 012 013 014 015 * 016 017 018 019 020 * 021 022 023 * 024 025

S2 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044

(trans-trans)farnesol 1-(ethenyloxy)-butane 1,1,2-trichloro-ethane 1,1'-bicyclohexyl 1,2,3,4-tetramethyl-benzene 1,2,3,5-tetramethyl-benzene 1,2,4,5-tetramethyl-benzene 1,2,4-trimethyl-benzene 1,2-benzenedicarboxylic,bis(2-methylpropyl)ester 1,2-diethyl-benzene 1,2-dimethyl-naphthalene 1,3,3-trimethyl-bicyclo[2.2.1]heptan-2-ol 1,3,5-trimethyl-benzene 1,3-bis(1-methylethyl)-benzene 1,3-diethyl-benzene 1,3-dimethyl-naphthalene 1,4-bis(1-methylethyl)-benzene 1,4-dichloro-benzene 1,4-dimethoxy-2,3-dimethylbenzene 1,4-methanoazulene,decahydro-4,8,8-trimethyl-9methylene 1,6,7-trimethyl-naphthalene 1,8-dimethyl-naphthalene 10,10-dimethyl-2,6-bis(methylene)-dicyclo[7.2.0] undecane 11H-benzo[b]fluorene 1-dodecanol 1-ethyl-2-methyl-benzene 1-ethyl-3,5-dimethyl-benzene 1-ethyl-3-methyl-benzene 1-hexadecanol 1-hexadecene 1-hexyl-cyclohexen 1-methyl-2(1-methylethyl)-benzene 1-methyl-2-propyl-benzene

106-28-5 111-34-2 79-00-5 92-51-3 25619-60-7 527-53-7 — 95-63-6 — — 573-98-8 1632-73-1 108-67-8 99-62-7 25340-17-4 575-41-7 100-18-5 106-46-7 39021-83-5

27.72 30.68 13.05 24.33 34.70 54.69 37.65 32.41 20.96 7.79 76.49 32.19 14.21 21.00 11.13 12.13 11.28 1.85 10.83

1.26 1.45 1.82 1.77 1.88 1.88 1.86 1.84 0.74 1.88 1.89 1.30 1.85 1.85 1.86 1.87 1.87 1.84 1.39

0.06 0.01 0.00 0.04 0.03 0.03 0.03 0.02 0.13 0.02 0.05 0.05 0.02 0.03 0.02 0.05 0.03 0.01 0.04

475-20-7

23.49

1.79

0.11

2245-38-7 569-41-5

77.35 38.72

1.90 1.87

0.06 0.05

—

54.71

1.82

0.09

243-17-4 112-53-8 611-14-3 934-74-7 620-14-4 36653-82-4 629-73-2 — 527-84-4 1074-17-5

17.84 10.62 29.21 26.59 20.10 48.54 60.36 65.55 50.74 18.36

1.94 1.24 1.83 1.87 1.86 1.31 1.86 1.86 1.84 1.85

0.21 0.03 0.02 0.02 0.02 0.08 0.06 0.03 0.02 0.02

045 046 047 048 049 050 051 052 053 054 055 056 057 058

S3 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

059 060 061 062 063 064 065 * 066 067 068 069 070 071 072 073 074 075 076 077 078 079 080 081 082 083 084 085 086

1-methyl-3-(1-methylethyl)-benzene 1-methyl-4-(1-methylethyl)-benzene 1-methyl-naphthalene 1-methyl-phenanthrene 1-nonanal 1-O-feruloyl-β-D-glucopyranoside 1-O-feruloyl-β-D-glucopyranoside_qt 1-penten-3-ol 1-tetradecanol 1-tridecene 1-undecanol 1-verbenone 1β-hydroxyeuscaphic acid 2,2,3,3-tetramethylhexane 2,2,4,4-tetramethyl-3-pentanone 2,2-dimethylbutane 2,3,5,8-tetramethyldecane 2,3,5-trimethyl-phenol 2,3-dihydro-4,6-dimethyl-1H-indene 2,3-dihydro-5-methyl-1H-indene 2,3-dimethyl-naphthalene 2,3-diphenyl-2-cyclopropen-1-one 2,4,6-trimethyl-pyridine 2,4-bis(1,1-dimethylethyl)-phenol 2,4-heptadienal 2,6,10-trimethyl-14-pentadecanone 2,6,10-trimethyltetradecane 2,6,11-trimethyldodecane 2,6-dimethoxy-4-(2-propenyl)phenol 1-O-β-D-glucopyranoside 2,6-dimethoxy-4-(2-propenyl)phenol 2,6-dimethyloctane 2,6-dimethylpyridine 23-cis-p-coumaroyltormentic acid 23-trans-p-coumaroyltormentic acid

535-77-3 99-87-6 90-12-0 832-69-9 124-19-6 — — 616-25-1 112-72-1 2437-56-1 112-42-5 1196-01-6 — 13475-81-5 815-24-7 75-83-2 192823-15-7 697-82-5 1685-82-1 874-35-1 581-40-8 886-38-4 108-75-8 — 96-76-4 502-69-2 31295-56-4 —

17.91 9.64 40.40 12.29 46.74 47.54 17.95 3.54 28.24 72.77 65.96 5.06 24.03 31.28 32.64 1.79 62.96 35.09 31.35 30.80 33.81 47.63 31.25 30.38 32.01 41.74 33.27 27.04

1.83 1.88 1.86 1.92 1.32 −0.97 0.41 1.04 1.27 1.83 1.24 1.27 −0.38 1.80 1.40 1.77 1.82 1.60 1.89 1.89 1.90 1.42 1.61 1.67 1.33 1.44 1.82 1.83

0.02 0.02 0.04 0.11 0.02 0.32 0.06 0.00 0.05 0.03 0.03 0.06 0.69 0.02 0.02 0.01 0.04 0.03 0.04 0.03 0.05 0.11 0.02 0.06 0.01 0.10 0.06 0.04

—

17.14

−0.47

0.28

6627-88-9 2051-30-1 108-48-5 — —

17.74 34.59 30.37 43.93 20.45

1.26 1.80 1.58 −0.65 −0.64

0.05 0.01 0.01 0.32 0.32

087 088 089 090 091 092

S4 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

093 094 095 096 097 098 099 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127

2-butanone,4-(2,6,6-trimethyl-2-cyclohexen-1-ylidene) 2-butenal 2-ethyl-1-hexanol 2-ethyl-naphthalene 2-hexadecen-1-ol, 3,7,11,15-tetramethyl2-hexadecenoic acid 2-hexanoylfuran 2-hexenoic acid 2-methoxy-4-vinylphenol 2-methyl butyric acid 2-methyl-5-(1-methylethyl)-phenol 2-methyl-aziridine 2-methyl-decane 2-methyl-naphthalene 2-naphthalenamine 2-propenal 2-propenamide 2-propyn-ol 2α,19α-dihydroxy-3-oxo-urs-12-en-28-oic acid 2α,3α,19α,23-tetrahydroxy-olean-12-en-28-oic acid 2α,3α,19α-trihydroxy-12-oleanen-28-oic acid 2α,3α,23-trihydroxyolean-12-en-28-oic acid 2α,3α,24-trihydroxyolean-12-en-28-oic acid 2α,3α-dihydroxyurs-12-en-28-oic acid 2α,3α-dihydroxyursolic acid 2α,3β,13β-trihydroxyurs-11-en-28-oic acid 2α,3β,19α,23-tetrahydroxyurs-12-en-28-oic acid 3,4-dimethoxy-phenol 3,6-dimethyldecane 3,6-dimethyl-phenanthrene 3,7,11-trimethyl-[S-(Z)]-1,6,10-dodecatrien-3-ol 3,7-dimethylnonane 3,8-dimethyl-decane 3,8-dimethylundecane 3-carene

56052-61-0 4170-30-3 104-76-7 91-57-6 7541-49-3 629-56-1 14360-50-0 1191-04-4 7786-61-0 116-53-0 499-75-2 75-55-8 90622-57-4 — 91-59-8 25314-61-8 107-13-1 — — — — 102519-34-6 150821-16-2 — — — — 95-65-8 17312-53-7 1576-67-6 — 17302-32-8 17312-55-9 17301-30-3 13466-78-9

37.64 55.96 11.53 54.27 47.57 44.46 62.88 13.25 22.15 17.70 25.09 1.85 62.50 26.85 47.38 19.89 11.33 2.44 46.95 16.65 30.03 41.34 13.24 45.16 37.73 17.85 22.73 16.66 42.83 20.84 6.21 70.51 33.07 27.11 27.36

1.43 1.14 1.17 1.88 1.26 1.10 1.22 0.81 1.38 0.74 1.60 1.23 1.81 1.85 1.13 1.07 0.70 0.78 −0.30 −0.59 −0.18 −0.27 −0.20 0.21 0.25 −0.44 −0.70 1.09 1.80 1.95 1.45 1.81 1.81 1.81 1.87

0.05 0.00 0.01 0.05 0.13 0.10 0.03 0.01 0.03 0.01 0.03 0.00 0.02 0.04 0.04 0.00 0.00 0.00 0.71 0.69 0.72 0.72 0.72 0.74 0.74 0.68 0.69 0.03 0.02 0.13 0.06 0.02 0.02 0.03 0.04

S5 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

128 129 130 131 132 133 134 135 136 137 138

3-epicorosolic acid methyl ester 3-epicorosolic acid 3-epiursolic acid 3-ethyl-3-methylheptane 3-ethyloctane 3-methyl-1-butanol 3-methyl-2-cyclopenten-1-one 3-O-cis-p-coumaroyltormentic acid 3-O-trans-caffeoyltormentic acid 3-O-trans-p-coumaroylrotundic acid 3-O-trans-p-coumaroyltormentic acid 3-oxo-α-ionyl-9-O-β-D-apiofuranosyl(1→6)-β-Dglucopyranside 3-oxo-α-ionyl-9-O-β-D-glucopyranside 3-oxo-α-ionyl-9-O-β-D-glucopyranside_qt 3α-trans-feruloyloxy-2α-hydroxyurs-12-en-28-oic acid 3β,6α,19α-trihydroxyurs-12-en-28-oic acid 3β,6β,19α-trihydroxyurs-12-en-28-oic acid 4-(1-methylethyl)-phenol 4,6-dimethylundecane 4,8,12-tetradecatrienal,5,9,13-trimethyl 4-allylanisole 4-ethyl-decane 4-hexenylacetic acid 4-methoxy-1,3-benzenediamine 4-methyl-1,3-benzenediamine 4-methyl-cyclohexene 4-O-caffeoylquinic acid 5-(2-propenyl)-1,3-benzodioxole 5,6-dimethylundecane 5,9-undecadion-2-one,6,10-dimethyl 5-ethenyl-2-methyl-pyridine 6-methyl-5-hepten-2-one 6α,19α-dihydroxyursolic acid

— 52213-27-1 989-30-0 17302-01-1 5881-17-4 123-51-3 2758-18-1 121072-40-0 — — 121064-78-6

14.35 17.74 39.92 23.90 24.77 50.77 32.04 33.35 36.33 10.02 46.94

0.30 0.55 0.51 1.82 1.78 1.01 1.22 −0.19 −0.07 −0.10 −0.17

0.72 0.76 0.75 0.02 0.01 0.00 0.01 0.34 0.32 0.34 0.34

—

7.82

−1.39

0.72

— — — — — 99-89-8 17312-82-2 66408-55-7 140-67-0 1636-44-8 — 615-05-4 95-80-7 591-47-9 905-99-7 94-59-7 17615-91-7 689-67-8 20260-76-8 110-93-0 —

43.92 72.83 22.12 17.69 15.31 63.08 6.15 42.84 44.01 12.88 73.88 31.77 34.04 15.09 28.59 32.67 50.24 34.67 49.16 31.12 23.23

−0.85 0.54 0.01 −0.06 −0.37 1.56 1.82 1.55 1.74 1.80 0.92 0.03 0.20 1.78 −0.95 1.45 1.81 0.26 1.58 1.31 −0.43

0.34 0.07 0.31 0.71 0.71 0.02 0.03 0.09 0.03 0.02 0.02 0.03 0.02 0.01 0.33 0.05 0.03 0.06 0.02 0.01 0.71

139 140 141 * 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160

S6 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

161 162 163 164 165 166 167 168 169 * 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195

9,12-octadecadienoic acid (Z,Z)-,methyl ester 9-ethyl-9H-carbazol-3-amine 9-octadecenoic acid,(Z)-,methyl ester acetamide alpha,4-dimethyl-benzenmethanol alpha-bisabobol alpha-farnesene amygdalin amygdalin_qt anisic aldehyde arjunic acid arjunolic acid benzaldehyde benzene benzenemethanol benzyl benzoate benzyl salicylate beta-cadinene beta-damascone biphenyl borneol butyl benzoate butylated hydroxyanisole butylated hydroxytoluene butyl-benzene cadindiene camphene camphor caryophyllene chlorogenic acid cinchonain IIb cinchonain Ia cinchonain Ib cinchonain Ic cinchonain Id 7-O-β-glucopyranoside

112-63-0 132-32-1 112-62-9 60-35-5 536-50-5 502-61-4 — 29883-15-6 — 123-11-5 31298-06-3 465-00-9 100-52-7 71-43-2 100-51-6 120-51-4 118-58-1 523-47-7 23726-91-2 92-52-4 507-70-0 136-60-7 25013-16-5 128-37-0 104-51-8 29350-73-0 79-92-5 76-22-2 87-44-5 327-97-9 — 85081-24-9 85022-69-1 85081-22-7 —

28.07 14.68 43.98 2.81 37.51 23.78 11.55 7.77 32.59 38.50 21.80 43.27 12.65 2.87 54.08 42.89 16.13 19.09 30.85 17.74 46.99 37.92 56.06 30.93 15.55 17.05 12.50 24.25 29.67 63.12 24.81 13.92 30.62 22.82 39.36

1.43 1.24 1.43 0.56 1.21 1.38 1.92 −1.96 0.68 1.05 −0.22 −0.27 1.31 1.78 1.08 1.33 1.09 1.87 1.36 1.90 1.21 1.28 1.40 1.75 1.88 1.87 1.80 1.27 1.83 −1.07 −1.51 −0.55 −0.74 −0.55 −2.10

0.17 0.13 0.17 0.00 0.02 0.07 0.05 0.61 0.02 0.02 0.72 0.72 0.01 0.01 0.01 0.09 0.10 0.08 0.05 0.04 0.05 0.04 0.05 0.07 0.02 0.08 0.04 0.05 0.09 0.33 0.18 0.93 0.93 0.93 0.61

S7 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

196 197 198 199 200 201 202 * 203 204 205 206

cinchonain Id cis-3-hexen-1-ol cis-linalooloxide cis-verbenol colosolic acid corchoionoside C corchoionoside C_qt corosolic acid methyl ester coumaran cyclohexane cyclohexanone cyclohexene,1-methyl-4-(5-methy-1-menthylene-4hexenyl)-(s)cyclopentane daucosterol decane delta 7-stigmastenol delta-cadinol dibenz[a,h]anthracene dibenzofuran dibutyl phthalate dihydroactinidiolide diisobutyl phthalate dimethyl phthalate diphenyl ether docosane eicosane elemicin ellagic acid epicatechin-(4β→2)-phloroglucinol eriobofuran eriobotrin eriojaposide A eriojaposide B eriojaposide B_qt

85022-67-9 928-96-1 5989-33-3 1845-30-3 4547-24-4 185414-25-9 — — 496-16-2 110-82-7 108-94-1

15.22 35.69 25.90 32.58 39.30 15.22 24.51 15.48 25.01 10.45 53.50

−0.52 1.06 0.99 1.25 0.03 −1.30 −0.14 0.26 1.67 1.77 1.19

0.93 0.01 0.04 0.06 0.74 0.36 0.08 0.72 0.03 0.01 0.01

—

21.65

1.89

0.22

287-92-3 474-58-8 124-18-5 6869-99-4 36564-42-8 53-70-3 132-64-9 84-74-2 17092-92-1 84-69-5 131-11-3 101-84-8 629-97-0 112-95-8 487-11-6 476-66-4 — 97218-06-9 — — — —

1.84 23.47 52.49 40.40 14.84 9.38 9.90 78.06 38.23 41.87 66.69 8.62 12.59 46.94 45.85 7.04 43.82 36.60 43.20 29.74 47.09 18.20

1.78 −0.15 1.78 1.30 1.44 2.02 1.75 0.79 1.18 0.85 0.44 1.75 1.84 1.82 1.34 −0.49 −0.59 1.19 −0.87 −1.57 −1.45 0.51

0.00 0.63 0.01 0.75 0.09 0.62 0.08 0.13 0.07 0.13 0.06 0.05 0.18 0.13 0.06 0.43 0.64 0.17 0.62 0.71 0.73 0.07

207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 *

S8 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

230 231 232 233 * 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257

ethyl cyclohexane ethylbenzene eugenyl β-rutinoside eugenyl β-rutinoside_qt euscaphic acid farnesane farnesiferol A farnesyl acetate farnesylacetone ferulic acid furfural geranic acid hex-2-enal hexachloro-ethane hexadecane hexadecanoic acid, methyl ester hexanoic acid hexanol hydrocinnamic acid hydrocoumarin hyperoside hyptadienic acid indane indene indole isobutylamine isocyanato-methane isoeriobotrin isohumbertiol-3-O-{α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)-[α-L-(4-trans-feruloyl) rhamnopyranosyl(1→6)]}-β-D-glucopyranoside isohumbertiol-3-O-{α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)-[α-L-(4-trans-feruloyl) rhamnopyranosyl(1→6)]}-β-D-glucopyranoside_qt

— 100-41-4 — — 53155-25-2 3891-98-3 511-33-1 29548-30-9 762-29-8 1135-24-6 98-01-1 459-80-3 505-57-7 67-72-1 544-76-3 112-39-0 142-62-1 111-27-3 65-85-0 119-84-6 482-36-0 128397-09-1 496-11-7 95-13-6 120-72-9 78-81-9 624-83-9 —

9.75 9.05 17.84 7.87 17.69 43.13 8.30 37.26 16.60 67.39 5.36 21.86 31.30 24.89 21.36 11.62 31.23 67.53 42.40 52.62 34.90 32.62 19.36 50.05 33.82 32.73 51.66 15.31

1.77 1.81 −1.31 1.45 −0.22 1.91 0.72 1.40 1.46 0.61 1.09 1.04 1.26 1.80 1.82 1.38 0.83 1.06 0.90 1.19 −1.34 0.90 1.87 1.82 1.78 1.03 1.06 −0.60

0.01 0.01 0.68 0.04 0.71 0.05 0.64 0.11 0.10 0.06 0.01 0.03 0.01 0.02 0.06 0.12 0.01 0.01 0.03 0.04 0.77 0.01 0.02 0.03 0.03 0.00 0.00 0.62

—

7.02

−2.83

0.17

—

7.02

−1.29

0.47

258

259 *

S9 Table S1. Cont. MolID 260 261 262 263 264 * 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 * 284 285 286 287 288 289 290 291

Compound Name isohumbertiol-3-O-{α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)-[α-L-rhamnopyranosyl(1→6)]} -β-D-glucopyranoside isoquercitrin isorhamnetin-3-O-galactoside isorhamnetin-3-O-glucoside isorhamnetin-3-O-glucoside_qt kaempferol 3-O-α-L-(2'',4''-di-Z-p-coumaroyl)-rhamnoside kaempferol kaempferol-3-O-galactoside kaempferol-3-O-glucoside kaempferol-3-O-neohesperidoside kaempferol-3-O-rhamnoside kaempferol-3-O-sophoroside kaempferol-3-O-α-L-(2'',4''-di-E-feruloyl)-rhamnoside kaempferol-3-O-α-L-(2'',4''-di-E-p-coumaroyl)rhamnoside kaempferol-3-O-α-L-rhamnopyranosyl-(1→2)-β-Dglucopyranoside kaempferol-3-rutinoside limonene linalool linamarin linguersinol 9'-O-β-D-xylopyranoside linguersinol linolenyl alcohol loquatifolin A loquatifolin A_qt malic acid maslinic acid menthol methyl arjunolate methyl betulinate methyl chlorogenate methyl corosolate methyl maslinate

CAS

OB

Caco-2

DL

—

11.36

−2.70

0.38

482-35-9 6743-92-6 5041-82-7 —

17.72 63.56 90.59 17.25

−1.44 −1.24 −1.45 0.33

0.77 0.80 0.80 0.31

—

17.63

−0.98

0.40

520-18-3 23627-87-4 480-10-4 32602-81-6 482-39-3 19895-95-5 —

65.70 28.23 66.61 41.08 59.10 18.50 6.17

0.27 −1.14 −1.11 −2.22 −0.77 −2.57 −0.66

0.24 0.74 0.74 0.75 0.70 0.71 0.31

—

7.67

−0.70

0.40

—

42.64

−2.15

0.75

17650-84-9 138-86-3 78-70-6 554-35-8 — — 506-44-5 116174-69-7 — 6915-15-7 4373-41-5 89-78-1 22452-82-0 25493-95-2 — 4518-70-1 22425-82-7

22.50 19.72 25.24 34.59 37.15 48.60 48.38 42.19 47.60 34.60 55.84 36.33 56.25 43.41 11.49 29.88 42.79

−2.26 1.86 1.22 −0.90 −0.91 −0.17 1.39 −1.71 1.38 −1.06 0.04 1.24 −0.05 0.85 −0.94 0.30 0.32

0.74 0.02 0.02 0.09 0.83 0.54 0.12 0.59 0.06 0.02 0.74 0.03 0.70 0.76 0.36 0.72 0.72

S10 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

292 293 294 295 296 297 298 299 300 301 302 303 304

methyl mercaptan methyl salicylate methyl ursolate methyl vinyl ketone methylbenzene methyl-cyclohexyl-dimethyl oxygensilane m-tert-butyl-phenol m-xylene myrcene myristicin myrtanol nerol nerolidol nerolidol-3-O-[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-β-D-glucopyranoside nerolidol-3-O-{[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-α-L-(4-trans-feruloyl)rhamnopyranosyl(1→6)}-β-D-glucopyranoside nerolidol-3-O-{[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-α-L-rhamnopyranosyl(1→6)} -β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl(1→2)-[α-Lrhamnopyranosyl(1→6)]-β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl-(1→2)-β-Dglucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl-(1→4)-α-Lrhamnopyranosyl-(1→6)-β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl(1→6)-α-Lrhamnopyranosyl(1→2)-β-D-glucopyranoside n-heptadecane nitro-cyclohexane N-methyl-N-nitroso-benzenamine nerolidol nerolidol-3-O-[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-β-D-glucopyranoside nerolidol-3-O-{[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-α-L-(4-trans-feruloyl)rhamnopyranosyl(1→6)}-β-D-glucopyranoside

74-93-1 119-36-8 32208-45-0 78-94-4 108-88-3 — 585-34-2 108-38-3 123-35-3 607-91-0 53369-17-8 106-25-2 7212-44-4

30.64 47.71 47.63 49.96 44.09 44.03 50.18 10.49 22.06 47.76 28.48 21.95 6.43

1.40 0.83 0.92 1.16 1.80 1.48 1.56 1.85 1.87 1.34 1.17 1.17 1.45

0.00 0.03 0.74 0.00 0.01 0.03 0.03 0.01 0.02 0.07 0.06 0.02 0.06

—

34.22

−1.86

0.55

—

29.11

−2.75

0.17

—

29.62

−2.63

0.36

—

17.83

−1.84

0.59

—

44.57

−1.06

0.68

—

38.33

−2.01

0.55

—

8.84

−1.88

0.59

629-78-7 — 614-00-6 7212-44-4

37.15 9.81 27.05 6.43

1.84 1.35 0.95 1.45

0.07 0.02 0.02 0.06

—

34.22

−1.86

0.55

—

29.11

−2.75

0.17

305

306

307

308 309 310 311 312 313 314 304 305

306

S11 Table S1. Cont. MolID 307

308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336

Compound Name nerolidol-3-O-{[α-L-rhamnopyranosyl(1→4)-α-Lrhamnopyranosyl(1→2)]-α-L-rhamnopyranosyl(1→6)} -β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl(1→2)-[α-Lrhamnopyranosyl(1→6)]-β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl-(1→2)-β-Dglucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl-(1→4)-α-Lrhamnopyranosyl-(1→6)-β-D-glucopyranoside nerolidol-3-O-α-L-rhamnopyranosyl(1→6)-α-Lrhamnopyranosyl(1→2)-β-D-glucopyranoside n-heptadecane nitro-cyclohexane N-methyl-N-nitroso-benzenamine N-nitrosodimethylamine nonanoic acid octanoic acid oleanolic acid O-xylene palmitic acid ethyl ester palmitic acid p-aminotoluene p-cymene pentadecane phenanthrene phenethyl alcohol phenylacetaldehyde pinocarvone piperidine pomolic acid procyanidin B procyanidin B1 procyanidin B2 procyanidin B4 procyanidin C1 procyanidin C2

CAS

OB

Caco-2

DL

—

29.62

−2.63

0.36

—

17.83

−1.84

0.59

—

44.57

−1.06

0.68

—

38.33

−2.01

0.55

—

8.84

−1.88

0.59

629-78-7 — 614-00-6 62-75-9 112-05-0 124-07-2 508-02-1 95-47-6 628-97-7 57-10-3 106-49-0 — 629-62-9 85-01-8 60-12-8 122-78-1 16812-40-1 110-89-4 13849-91-7 — 20315-25-7 29106-49-8 29106-51-2 37064-30-5 —

37.15 9.81 27.05 32.76 16.49 9.93 66.91 47.04 43.92 39.33 13.70 40.91 6.56 55.31 54.12 22.17 55.27 51.47 75.44 16.08 32.34 15.72 11.72 35.78 25.35

1.84 1.35 0.95 1.20 0.94 0.90 0.61 1.82 1.38 1.09 1.08 1.84 1.81 1.89 1.12 1.31 1.32 1.46 0.26 1.02 −1.06 −1.05 −0.85 −2.02 −2.05

0.07 0.02 0.02 0.00 0.02 0.02 0.76 0.01 0.14 0.10 0.01 0.02 0.05 0.10 0.02 0.02 0.06 0.01 0.73 0.74 0.66 0.66 0.66 0.10 0.10

S12 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

337 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369

propanoic acid, anhydride procyanidin c2 propanoic acid,anhydride p-xylene pyrene quercetin quercetin-3-O-sophoroside quercetin-3-rhamnoside quercetin-3-sambubioside quercetin-4'-O-β-D-galactoside quercetin-7-α-L-rhamnoside quinoline rhamnocitrin roseoside roxburic acid rutin safranal sanguisorba,officinalis-7 sanguisorba,officinalis-8 tartaric acid terpineol tert-butyl-benzene tetradecanoic acid thymol tormentic acid methyl ester tormentic acid trans-2-hexenol trans-cinnamaldehyde trans-farnesol trans-linalool oxide tricosane tricosanoic acid tridecane trimethylamine undecane

123-62-6 — 123-62-6 106-42-3 129-00-0 117-39-5 18609-17-1 522-12-3 83048-35-5 — — 91-22-5 569-92-6 54835-70-0 108657-25-6 153-18-4 116-26-7 — — 526-83-0 8000-41-7 98-06-6 544-63-8 89-83-8 13850-15-2 13850-16-3 928-95-0 14371-10-9 — 34995-77-2 638-67-5 2433-96-7 629-50-5 75-50-3 1120-21-4

17.20 25.35 17.20 42.90 29.04 31.36 48.08 11.61 40.37 7.22 17.74 39.70 17.41 9.71 19.59 62.82 14.61 35.32 17.76 57.03 15.69 45.26 59.01 27.15 17.82 33.73 27.85 29.96 44.45 17.60 5.02 23.50 18.23 4.57 17.71

0.78 −2.05 0.78 1.84 1.93 0.04 −2.67 −0.85 −2.46 −1.26 −1.14 1.56 0.39 −1.23 −0.57 −2.50 1.42 −0.25 −0.52 −1.85 1.33 1.85 1.06 1.65 0.10 −0.20 1.10 1.37 1.26 0.99 1.88 1.21 1.84 1.87 1.83

0.01 0.10 0.01 0.01 0.19 0.28 0.67 0.74 0.74 0.81 0.76 0.03 0.27 0.36 0.69 0.68 0.04 0.71 0.69 0.02 0.03 0.02 0.07 0.03 0.69 0.71 0.01 0.02 0.06 0.07 0.21 0.30 0.03 0.00 0.02

S13 Table S1. Cont. MolID

Compound Name

CAS

OB

Caco-2

DL

370 371 372 373

ursolic acid ursolic acid lactone uvaol valeric acid

77-52-1 — 545-46-0 109-52-4

9.95 15.61 17.56 29.97

0.79 0.54 0.87 0.75

0.65 0.76 0.76 0.01

vomifoliol

23526-45-6

22.38

−0.05

0.08

—

39.33

−1.65

0.71

481-34-5 127-41-3 80-56-8 14912-44-8 18794-84-8 79-77-6 127-91-3 20307-83-9 83-46-5 695-06-7 — 472-15-1 6713-26-4 331-39-5

36.26 20.14 25.08 18.56 17.73 36.33 37.62 22.33 45.04 52.44 3.45 17.10 12.51 51.08

1.32 1.38 1.82 1.83 1.90 1.41 1.82 1.89 1.33 1.11 −3.03 0.55 0.58 0.24

0.09 0.05 0.05 0.12 0.05 0.05 0.05 0.06 0.75 0.01 0.64 0.78 0.76 0.05

374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 *

vomifoliol-9-o-β-D-apiofuranosyl(1→6)-β-Dglucopyranside α-cadinol α-ionone α-pinene α-ylangene β-farnesene β-ionone β-pinene β-sesquiphellandrene β-sitosterol γ-hexanolide quercetin-3-O-galactosyl-(1→6)-glucoside betulinic acid δ-oleanolic acid caffeic acid

* deglycosylated molecule.

Copyright of International Journal of Molecular Sciences is the property of MDPI Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.