1521-009X/43/8/1181–1189$25.00 http://dx.doi.org/10.1124/dmd.115.063800DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 43:1181–1189, August 2015Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics

Ginsenosides Regulate PXR/NF-kB Signaling and Attenuate DextranSulfate Sodium–Induced Colitis s

Jun Zhang, Lijuan Cao, Hong Wang, Xuefang Cheng, Lin Wang, Lin Zhu, Tingting Yan, Yang Xie,Yuzheng Wu, Min Zhao, Sijing Ma, Mengqiu Wu, Guangji Wang, and Haiping Hao

State Key Laboratory of Natural Medicines, Key Laboratory of Drug Metabolism and Pharmacokinetics, China PharmaceuticalUniversity, Nanjing, China (J.Z., L.C., H.W., X.C., L.W., L.Z., T.Y., Y.X., Y.W., M.Z., S.M., M.W., G.W., H.H.); and School of Pharmacy,

Nanjing Medical University, Nanjing, China (J.Z.)

Received May 15, 2015; accepted May 15, 2015

ABSTRACT

Pregnane X receptor (PXR) activation exhibits anti-inflammatoryeffects via repressing nuclear factor-kB (NF-kB); however, itsoveractivation may disrupt homeostasis of various enzymes andtransporters. Here we found that ginsenosides restore PXR/NF-kB signaling in inflamed conditions without disrupting PXRfunction in normal conditions. The effects and mechanisms ofginsenosides in regulating PXR/NF-kB signals were determinedboth in vitro and in vivo. Ginsenosides significantly inhibited NF-kB activation and restored the expression of PXR target genes intumor necrosis factor-a–stimulated LS174T cells. Despite notbeing PXR agonists, ginsenosides repressed NF-kB activation ina PXR-dependent manner. Ginsenosides significantly increased

the physical association between PXR and the NF-kB p65 subunitand thereby decreased the nuclear translocation of p65. GinsenosideRb1 and compound K (CK) were major bioactive compounds inthe regulating PXR/NF-kB signaling. Consistently, ginsenosidessignificantly attenuated dextran sulfate sodium–induced experi-mental colitis, which was associated with restored PXR/NF-kBsignaling. This study indicates that ginsenosides may elicit anti-inflammatory effects via targeting PXR/NF-kB interaction withoutdisrupting PXR function in healthy conditions. Ginsenoside Rb1and CK may serve as leading compounds in the discovery of newdrugs that target PXR/NF-kB interaction in therapy for inflamma-tory bowel disease.

Introduction

Pregnane X receptor (PXR) is a ligand-activated transcription factorand a master regulator of drug transport and metabolism (Cheng et al.,2012). PXR is expressed predominantly in the liver and intestine ofmammals. It is activated by a range of structurally diverse xenobiotics,such as the antibiotic rifampicin (RIF), the herbal antidepressant St.John’s wort, endogenous hormones, and bile acids (Ma et al., 2008).On activation, PXR forms a heterodimer with retinoid X receptor thatbinds to specific PXR response elements on the promoters of targetgenes, such as cytochrome P450 3A (CYP3A), multidrug resistance 1(MDR1), and multidrug resistance protein 2. In addition to the wellvalidated functions in regulating transporters and drug-metabolizingenzymes, PXR plays important roles in the pathologic development ofa panel of diseases, including cancer, diabetes, and inflammatorybowel diseases (IBD). Of particular interest, reciprocal repression

between PXR and nuclear factor-kB (NF-kB) has been described(Zhou et al., 2006). The expression of PXR and several of its targetgenes has been shown to be suppressed in the colon of patients withIBD (Langmann et al., 2004; Blokzijl et al., 2007), suggesting thatPXR may play a critical role in IBD.NF-kB is a key transcription factor that can modulate several steps

in the inflammatory cascade by inducing the expression ofproinflammatory genes (Hayden and Ghosh, 2004; Perkins andGilmore, 2006; Karrasch and Jobin, 2008). Excessive production ofproinflammatory cytokines, such as tumor necrosis factor a (TNF-a),interleukin-1b (IL-1b), IL-6, and interferons, was detected in IBD(Sartor, 1994; Ishiguro, 1999.) The activated NF-kB is a heterodimer,which consists predominantly consists of p65 and p50 (Hayden andGhosh, 2004; Sarkar et al., 2008). In the absence of activating signals,NF-kB proteins are normally sequestered in the cytoplasm by inhibitorof kBa (IkBa). Activation of NF-kB requires the phosphorylation ofIkBa by the IkB kinase (IKK) complex and subsequent degradationby 26S proteasome, which induces the translocation of NF-kB fromthe cytosol to the nucleus and facilitates the transcription of specifictarget genes. Several investigations have revealed that the inhibition ofNF-kB activity ameliorated the severity of intestinal inflammation(Neurath et al., 1996; Atreya et al., 2008). In consideration of the wellvalidated important roles of PXR and NF-kB in IBD and their mutual

This study was financially supported by Jiangsu provincial promotionfoundation for the key laboratory of drug metabolism and pharmacokinetics[Grant BM2012012]; Natural Science Foundation of Jiangsu Province foroutstanding youth scholar [Grant BK2012026]; and National Natural ScienceFoundation of China [Grant 81430091, 81325025, and 81273586].

dx.doi.org/10.1124/dmd.115.063800.s This article has supplemental material available at dmd.aspetjournals.org.

ABBREVIATIONS: CK, compound K; COX, cyclooxygenase; DAI, disease activity index; DSS, dextran sulfate sodium; GAPDH, gluceraldelyde-3-phosphate dehydrogenase; GSE, ginseng saponin extract; IBD, inflammatory bowel diseases; IL-1b, interleukin-1b; IkB, inhibitor of kB; IKK, IkBkinase; iNOS, inducible nitric oxide synthase; MDR, multidrug resistance; MPO, myeloperoxidase; NF-kB, nuclear factor-kB; PBS, phosphate-buffered saline; PVDF, polyvinylidene fluoride; PXR, pregnane X receptor; RIF, rifampicin; RT-PCR, reverse-transcription quantitative real-timepolymerase chain reaction; siRNA, small interfering RNA; TBST, Tris-buffered saline containing 0.5% Tween-20; TNBS, 2,4,6-trinitrobenzenesulfonic acid; TNF-a, tumor necrosis factor a.

1181

http://dmd.aspetjournals.org/content/suppl/2015/05/18/dmd.115.063800.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

repressive relationship, it is reasonable to expect that restoration of thebalance between PXR and NF-kB would be a promising strategy forthe therapy of IBD, a disease with complicated causes and poortherapeutic means (Zhou et al., 2014).Ginseng, the root of Panax ginseng, has been widely used as a food

product and medicinal ingredient in Asian countries for more than2000 years (Jia and Zhao, 2009; Kang et al., 2011). Ginsenosides areconsidered the major active ingredients of ginseng (Ouyang et al., 2014;Zhao et al., 2014). Ginseng saponins have various pharmacologiceffects, such as antioxidation, antitumor, anti-inflammatory activities,and neuroprotective effects (Attele et al., 1999; Radad et al., 2006; Kimet al., 2010). A recent investigation suggested that ginseng suppressedkey inflammatory players such as cyclooxygenase (COX)-2, induciblenitric oxide synthase (iNOS), and NF-kB (Park et al., 2009; Kim et al.,2012). Ginsenoside compound K (CK) was shown to promote recoveryof DSS-induced colitis by suppressing NF-kB activation (Li et al.,2014). Ginsenoside Rd can attenuate 2,4,6-trinitrobenzene sulfonic acid(TNBS)-induced colitis via modulating mitogen-activated protein kinasesignaling pathways (Yang et al., 2012b). Ginseng has also beensuggested as an anti-inflammatory agent or adjuvant treatment of thepatients of IBD (Hofseth and Wargovich, 2007).Although the anti-inflammatory effects of ginsenosides and its

remedy effects against IBD have been well reported, the molecularmechanisms remain established (Yang et al., 2012a; Li et al., 2014; Yeet al., 2014). Here we hypothesized that ginsenosides may elicit itsanti-inflammatory effects in IBD via restoring PXR/NF-kB signalingin the inflamed colon. Unexpectedly, we found that ginsenosides,without the influence of PXR activity in the normal conditions,significantly combat NF-kB signaling in a PXR-dependent manner.

Materials and Methods

Chemicals and Reagents. Ginseng saponin extract (GSE); ginsenoside Rb1,Rb2, Rc, Rd, Re, Rg1, Rg3, Rh1, Rh2, CK, protopanaxadiol (PPD); andprotopanaxatriol (PPT) were purchased from Hongjiu Biotech Co. Ltd. (Jilin,China). The contents of ginsenosides (Rb1 11.2%, Rb2 11.1%, Rc 10.5%, Rd7.7%, Re 8.9%, Rf 0.9%, Rg1 3.3%, Rg2 1.4%, and Rh1 0.2%) weredetermined using a validated liquid chromatography-mass spectrometry methodin our laboratory (Hao et al., 2010). Rifampicin (RIF) was purchased fromSigma-Aldrich (Shanghai, China). Dextran sulfate sodium (DSS) was providedby MP Biomedicals (Solon, OH) (mol. wt. 36–50 kDa). TNF-a (human,recombinant) was purchased from Promega Corporation (Madison, WI). PXRsmall interfering RNA (siRNA) and control siRNA were purchased fromInvitrogen Trading Shanghai Co. (Invitrogen Life Technologies, Shanghai,China).

Animals and Treatment. Male C57BL/6 mice (8 weeks old, 20–22g) wereobtained from Academy of Military Medical Sciences, China. They weremaintained in standard cages in a controlled room (temperature 24 to 25�C,humidity 70%–75%, lighting regimen of 12-hour light-dark cycle), with freeaccess to food and water. All animal studies were approved by the AnimalEthics Committee of China Pharmaceutical University. Mice were divided intofour groups: normal, GSE, DSS, and DSS + GSE. Acute colitis was induced byadministration of 3% (w/v) DSS in drinking water for 5 days. GSE at a dose of400 mg/kg of body weight was administered via oral gavage 3 days before theDSS treatment of a consecutive 9 days. The dose for GSE treatment wasdesigned on the basis of previous reports (Wong et al., 2010; Jang et al., 2012)and the results from a preliminary experiment.

Disease activity index (DAI) was used for evaluating the grade and extent ofintestinal inflammation (Azuma et al., 2010; Han et al., 2011). Body weight,stool consistency, and blood in the stool were monitored daily for DAIdetermination. The score for each index was assigned as previously described:body weight loss (0, none; 1,,5%; 2, 5%–10%; 3, 10%–20%, 4,.20%), fecalblood (0, negative; 2, slight bleeding; 4, gross bleeding), stool consistency(0, normal; 2, loose stools; 4, diarrhea). The DAI score, ranging from 0 to 12(total score), was summarized from these three indexes.

Sample Collection and Histologic Evaluation. After the mice were killedunder anesthesia, the colons were immediately removed and fixed in 10%buffered formalin, paraffin-embedded, sectioned, and stained with H&E.Histologic score of H&E-stained specimens of the colon was determined bytwo pathologists in a blinded fashion. Histologic sections were scored usinga validated scoring system as described previously (Williams et al., 2001; Wenet al., 2014).

Cell Culture. The human colon cancer cell line LS174T was obtained fromShanghai Cell Resource Center (Shanghai Institutes for Biologic Sciences,China) and cultivated at 37�C in an atmosphere of 95% air and 5% CO2 withDulbecco’s modified Eagle’s medium supplemented with 10% fetal bovineserum (Hyclone, Logan, Utah), 100 U/ml penicillin, 100 mg/ml streptomycin,and 20 mM L-glutamine. The cell medium was changed every other day. Cellswere passaged on reaching ;80% confluence.

Real-Time Quantitative Reverse Transcriptase PCR. Total RNA wasextracted by using Trizol reagent (TaKaRa Biotechnology Co., Ltd., Dalian,China). The content of RNA was estimated by measuring the absorbance at260 nm. Purified total RNA was reverse-transcribed using the PrimerScript RTReagent Kit (TaKaRa Biotechnology Co., Ltd.) according to the manufacturer’sprotocol. Primer sequences are as follows (forward 59-39, reverse 59-39): mouseIL-1b (CCT CGT GCT GTC GGA CCC AT, TCC AGC TGC AGG GTGGGT GT), mouse IL-6 (ATC CAG TTG CCT TCT TGG GAC TGA, TAAGCC TCC GAC TTG TGA AGT GGT), mouse Cox-2 (TGA GCA ACT ATTCCA AAC CAG C, GCA CGT AGT CTT CGA TCA CTA TC), mouse iNOS(CGC TGT GCT TTG AGA ACT GT, AGG TCC TTG CCT ACT TGC TG),mouse Pxr (GTT CAA GGG CGT CAT CAA CT, TTC TGG AAG CCA CCATTA GG), mouse Cyp3a11 (CCG AGT GGA TTT TCT TCA GC, GAG CCTCAT CGA TCT CAT CC), mouse Mdr1a (TGT GAT TGC GTT TGG AGGAC, CCA TAC CAG AAT GCC AGA GC), mouse gluceraldelyde-3-phosphate dehydrogenase (Gapdh) (CGA CTT CAA CAG CAA CTC, GTAGCC GTA TTC ATT GTC AT), human IL-1b (TAT AGC CTG GAC TTTCCT GTT GTC, GCT GAC TGT CCT GGC TGA TG), human TNF-a (AACCTC CTC TCT GCC ATC AA, GGA AGA CCC CTC CCA GAT AG), humanCOX-2 (TGA GCA TCT ACG GTT TGC TG, TGC TTG TCT GGA ACAACT GC), human iNOS (TCA TTC TGC TGC TTG CTG AG, CAA CAATGT GGA GAA AGC CC), human PXR (TCT CAC CTC CAG GTT TGCTT, ATT GTT CAA AGT GGA CCC CA), human CYP3A4 (CAG GAT CTGTGA TAG CCA GC, CCC ACA CCT CTG CCT TTT T), human MDR1(AGC CCA TCC TGT TTG ACT GC, TGT ATG TTG GCC TCC TTT GC),and human GAPDH (GGC CTC CAA GGA GTA AGA CC, AGG GGA GATTCA GTG TGG TG). Real-time PCR was performed by using an RT-PCRsystem (Bio-Rad, Carlsbad, CA; C1000TM Thermal cycler, CFX96TM) andSYBR Green reagent kit (TaKaRa Biotechnology Co., Ltd.). Changes of geneexpression were calculated by the d-d Ct method, and the values werenormalized to endogenous reference GAPDH.

Western Blot Analysis. Colon tissues were disrupted by homogenization onice and centrifuged at 4�C (12,000g, 15 minutes). The supernatants werecollected, and the protein concentrations were determined by BCA assay.Protein lysates were loaded onto 10% SDS-polyacrylamide gels (BeyotimeInstitute of Biotechnology, Jiangsu, China), subjected to electrophoresis, andtransferred to Immun-Blot polyvinylidenefluoride (PVDF) membranes (0.2mm; Bio-Rad Laboratories). PVDF membranes were blocked in 5% skim milkand incubated in primary antibody overnight at 4�C, after which they werewashed three times for 15 minutes in Tris-buffered saline containing 0.5%Tween-20 (TBST) and incubated with the appropriate secondary antibodies for1 hour. Blots were again washed with TBST and then developed by enhancedchemiluminescence detection regents (GE Healthcare BioSciences KK, Piscataway,NJ). The protein bands were quantified using image processing and analyzedby digitalized software (Bio-Rad) and normalized to the reference bandb-actin. The antibodies used were as follows: anti-NF-kB p65 (no. 8242,Cell Signaling Technology, Danvers, MA), anti-NF-kB phospho-p65 (no.3033, Cell Signaling Technology), anti-PXR (sc-48403, Santa CruzBiotechnology, Dallas, TX), and anti-b-actin (AP0060, Bioworld Technology,St. Louis Park, MN). The horseradish peroxidase-conjugated goat anti-mouse/rabbit IgG secondary antibodies were purchased from Cell SignalingTechnology.

RNA Interference Assay. LS174T cells were seeded into six-well plates,incubated at 37�C, and transfected 24 hours after reaching 70% confluence.

1182 Zhang et al.

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

Cells were transiently transfected with PXR siRNA targeting the human PXRmRNA. Control siRNA, a nontargeting siRNA, was used as a negative control.After transfection, cells were incubated with GSE (100 mg/ml), Rb1 (10 mM),CK (10 mM), or RIF (10 mM) for 24 hours, followed by an additionalincubation with or without TNF-a (20 ng/ml) for 6 hours. At the end of theincubation, cells were rinsed, scraped, and used in RT-PCR or Western blotanalysis as described herein.

Immunohistochemistry. Tissue specimens were fixed in 10% formalin for24 hours, dehydrated, and paraffin-embedded. Standard immunohistochemicalprocedures were performed. Tissue sections were incubated with the primaryantibody (rabbit monoclonal anti-NF-kB p65, rabbit polyclonal anti-PXR) for 2hours at room temperature, followed by incubation with the predilutedhorseradish peroxidase-conjugated secondary antibody for 30 minutes. Fornegative controls, 1% of nonimmune serum in phosphate-buffered saline (PBS)replaced the primary antibodies. The immunohistochemistry staining of NF-kBp65, and PXR was scored by measuring the integrated optical density of at leastthree visions of each slice using Image Pro-plus 6.0 software (Media Cybernetics,Inc., Rockville, VA). Data are expressed as mean 6 S.E.M. of six mice.

Immunofluorescence Staining. LS174T cells were seeded in cell imagingdish (Eppendorf, Germany). After overnight incubation, cells were treated withGSE (100 mg/ml), Rb1 (10 mM), or CK (10 mM) for 3 hours, followed by anadditional incubation with or without TNF-a (20 ng/ml) for 6 hours. At the end ofthe incubation, cells were harvested and fixed with 4% paraformaldehyde solutionat 20�C for 20 minutes. After washing in PBS, cells were permeabilized with0.2% Triton X-100 in PBS at room temperature for 5 minutes. After incubation inblocking buffer containing 0.1% Triton X-100 and 5% bovine serum albumin,cells were incubated with rabbit NF-kB p65 antibody (no. 8242, Cell SignalingTechnology) at 4�C overnight and then with Alexa Fluor 488-conjugated anti-rabbit IgG antibody (A-21206, Invitrogen) at room temperature for 30 minutes in1% bovine serum albumin in PBS. Fluorescence photographs were obtainedusing a Zeiss 710 confocal microscope (Zeiss, Germany).

Coimmunoprecipitation. A coimmunoprecipitation assay was conducted toexamine the physical association between PXR and NF-kB p65 subunits.Briefly, cells were collected and lysed in NP-40 buffer and sonicated. Aftercentrifuging the cell lysate for 10 minutes at 15,000g, the clear lysate wasimmunoprecipitated with PXR (sc-48403) antibody or an irrelevant rabbit IgGas a negative control, together with protein A-agarose (Invitrogen, Carlsbad,CA).The immunoprecipitates were washed three times with washing buffer,resuspended in 4� sample buffer (Bio-Rad), and heated for 5 minutes at 100�C.After a spin-down, equal amounts of protein from the supernatants wereelectrophoresed on 10% SDS-PAGE gels, transferred onto PVDF membranes,and immunoblotted with an antibody against PXR and NF-kB p65.

Statistical Analysis. Data are expressed as mean 6 S.E.M. Prism 5.0statistical software was used for the analysis. Differences between groups werecompared using one way analysis of variance and two-tailed Student’s t test.P , 0.05 was considered significant.

Results

GSE Regulates PXR/NF-kB Signaling in Colonic Cell Lines. Todetermine the effect of GSE in regulating PXR/NF-kB signaling, themRNA levels of multiple PXR and NF-kB target genes weredetermined in LS174T cells in both basal and TNF-a-stimulatedconditions. In the basal condition, the typical human PXR agonist RIFsignificantly increased mRNA expression of PXR target genes CYP3A4and MDR1 in LS174T cells, whereas GSE had little effect; however,both GSE and +RIF significantly counteracted TNFa-induced reductionof the mRNA levels of PXR, CYP3A4, and MDR1 in LS174T cells(Supplemental Fig. 1, A–C). Likewise, both GSE and RIF significantlyreduced TNFa-induced upregulation of IL-1b, TNF-a, and iNOS(Supplemental Fig. 1, D–F). TNF-a treatment to LS174T cells for 36hours caused a significant reduction of PXR protein levels and increasein the ratio of phosphorylated to total NF-kB p65, both of which wereabolished by GSE (Supplemental Fig. 1G). These results suggest thatGSE is able to recover the imbalanced PXR/NF-kB signaling on TNF-astimulation, although GSE is not a PXR agonist.

PXR Silencing Abrogates the Anti-inflammatory Activity ofGSE. Our results suggest that GSE can restore PXR/NF-kB signalingwithout the activation of PXR. Previous studies indicated that PXRand NF-kB are mutually repressed. We thus asked whether the effectsof GSE in attenuating NF-kB activation were PXR dependent. To thisend, we silenced the PXR gene using siRNA in human coloncarcinoma LS174T cells and determined the effects of GSE in NF-kBactivation. With scrambled siRNA treatment, GSE and RIF signifi-cantly attenuated TNF-a-induced activation of NF-kB; the mRNAlevels of CYP3A4, MDR1, IL-1b, and iNOS were largely recovered;however, the PXR siRNA treatment resulted in a significant reductionof the protein levels of PXR (Supplemental Fig. 2A) and largelyabrogated the effect of GSE and RIF in attenuating TNF-a-inducedactivation of NF-kB (Supplemental Fig. 2, B–F). In agreement, GSEsignificantly repressed p-p65 accumulation without the influence ofthe total p65 levels, which was markedly abrogated by PXR silencing.These results indicate that the effects of GSE in combating TNF-a-induced activation of NF-kB are largely mediated by PXR.Screening for Anti-inflammatory Compounds in GSE. GSE is

a mixture containing dozens of ginseng saponin compounds. We thussought to screen for the dominant compounds contributing to the anti-inflammatory activity of GSE. Using the mRNA level of iNOS as anindex, we screened a panel of 12 compounds and found thatginsenoside Rb1 and CK are the compounds with the most potentanti-inflammatory activity in GSE (Supplemental Fig. 3). Thereafter,we focused on evaluating the effects of ginsenoside Rb1 and CK onPXR/NF-kB signaling. The concentration for ginsenoside Rb1 (10mM) and CK (10 mM) was optimized from a preliminary study toensure sufficient anti-inflammatory activity and without apparentcytotoxicity. Ginsenoside Rb1 and CK significantly reduced TNF-a–induced upregulation of IL-1b and iNOS mRNA levels (Fig. 1, A, B,E, and F), and restored the mRNA levels of PXR and CYP3A4 inLS174T cells (Fig. 1, C, D, G, and H). TNF-a caused a significantreduction in PXR protein levels and increase in the ratio ofphosphorylated to total NF-kB p65, both of which were significantlyabrogated by ginsenoside Rb1 and CK (Fig. 1I).Anti-inflammatory Activity of Ginsenoside Rb1 and CK Is PXR

Dependent. Because GSE elicits its anti-inflammatory activity ina PXR-dependent manner, we asked whether this also holds true forginsenoside Rb1 and CK. To this end, the anti-inflammatory activityof ginsenoside Rb1 and CK was determined in LS174T cellstransfected with negative control or PXR siRNA. As expected, theknockdown of PXR protein levels by PXR siRNA almost completelyabrogated the anti-inflammatory activity of ginsenoside Rb1 and CK.In PXR silenced LS174T cells, the treatment of ginsenoside Rb1 andCK could not counteract TNF-a-induced upregulation of IL-1b andiNOS, and downregulation of CYP3A4 (Fig. 2, A–F). Furthermore,PXR silencing obviously abolished the effect of ginsenoside Rb1 andCK on counteracting TNFa-induced increase in the ratio ofphosphorylated to total NF-kB p65 (Fig. 2, G and H).To further validate the role of PXR in the anti-inflammatory activity

of ginsenosides, we performed a confocal microscopy analysis of p65nuclear translocation in LS174T cells transfected with negative controlor PXR siRNA (Fig. 2J). The nuclear levels of p65 translocation wereincreased in the LS174T cells treated with TNF-a, which was largelyabolished by GSE, ginsenoside Rb1, or CK. Silence of PXR almostcompletely abrogated the effect of GSE, ginsenoside Rb1, and CK oncounteracting TNF-a–stimulated p65 nuclear translocation. All theseresults support that the effects of ginsenosides in combating TNF-a–triggered inflammation are largely PXR dependent.Ginsenosides Promote PXR and NF-kB p65 Interaction.

Ginsenosides have little effect in PXR function in the basal condition;

Ginsenosides Regulate PXR/NF-kB Signaling 1183

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

Fig. 1. Effects of ginsenoside Rb1 and CK on PXR and NF-kB signaling in vitro. The mRNA levels of (A, E) PXR, (B, F) CYP3A4, (C, G) IL-1b, and (D, H) iNOS weredetermined using RT-PCR in LS174T cells. (I) Phospho-p65, p65, and PXR protein levels were determined by Western blot analysis. (J) The relative density of the proteinlevels was performed by densitometric analysis of the blots. After overnight incubation, LS174T cells were treated with ginsenoside Rb1 (10 mM), CK (10 mM), or RIF (10mM) for 24 hours, followed by an additional incubation with or without TNF-a (20 ng/ml) for 6 hours (Fig. 1A-H) or 36 hours (I and J). At the end of the incubation, cells

1184 Zhang et al.

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

however, their anti-inflammatory activities are largely PXR de-pendent. These results prompted us to hypothesize that ginsenosidesmight influence the physical association between PXR and NF-kB p65

subunit (Mencarelli et al., 2011). To this end, we performed acoimmunoprecipitation assay using anti-PXR antibody. Becausethe long-duration treatment (36 hours) of TNF-a significantly reduced

were harvested and analyzed by RT-PCR or Western blot. Data are expressed as mean 6 S.E.M. of triplicates of two independent experiments. *P , 0.05, **P , 0.01versus vehicle-treated wells; #P , 0.05, ##P , 0.01 versus TNF-a treated wells.

Fig. 2. PXR is an important determinant inrepressing NF-kB signaling by ginsenoside Rb1and CK. The mRNA levels of (A, D) CYP3A4, (B,E) IL-1b, and (C, F) iNOS were determined usingRT-PCR in LS174T cells. (G, H) Phospho-p65 andp65 protein levels were determined by Western blotanalysis. (I) The relative density of the protein levelswas performed by densitometric analysis of theblots. (J) NF-kB nuclear translocation was detectedby immunofluorescence analysis using p65 anti-body. LS174T cells were transfected with scrambledsiRNA or PXR siRNA. After incubation for 48hours, cells were treated with ginsenoside Rb1 (10mM), CK (10 mM), or RIF (10 mM) for 24 hours,followed by an additional incubation with or withoutTNF-a (20 ng/ml). At the end of the incubation,cells were harvested and analyzed by RT-PCR orWestern blot. Data are expressed as mean 6 S.E.M.of triplicates of two independent experiments. *P ,0.05, **P , 0.01 versus vehicle-treated wells; #P ,0.05, ##P , 0.01 versus TNF-a-treated wells.

Ginsenosides Regulate PXR/NF-kB Signaling 1185

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

the protein levels of PXR that might make it difficult to evaluate thephysical interaction between PXR and NF-kB, a protocol of shorterduration of treatment (6 hours) was applied in the coimmunoprecipitationassay. As shown in Fig. 3, GSE, ginsenoside Rb1, and CK, without theinfluence of the total PXR levels, increased the degree of the physicalassociation between PXR and NF-kB p65 in TNF-a–stimulatedLS174T cells. This result indicates that ginsenosides may combatTNF-a–stimulated inflammatory responses via promoting the physicalinteraction between PXR and NF-kB p65 subunit.GSE Ameliorates DSS-Induced Colitis. Several previous reports

indicated that several isolated ginsenoside compounds includingginsenoside Rd and Rh2 (Yang et al., 2012a,b; Ye et al., 2014) areeffective against experimental colitis. It remains unclear, however,whether GSE, which has been widely used by human beings as a foodsupplement, is also effective against the pathologic development ofcolitis. To answer this question, we extended this study to validate theeffect of GSE against DSS-induced colitis. DSS induced a significantloss of body weight, whereas GSE treatment dramatically amelioratedDSS-induced body weight loss (Fig. 4A). GSE treatment alsosignificantly combated DSS-induced colon shortening (Fig. 4B).Diarrhea symptoms appeared after day 5; the mice with GSE treatmentshowed less diarrhea and bloody diarrhea than did the DSS group (Fig.4C). Histologic studies showed that GSE administration to DSS miceresulted in significant protection of the colon crypt structures and lesssevere histologic inflammation (Fig. 4, D and E). Additionally, GSEadministration significantly attenuated DSS-triggered activation ofmyeloperoxidase, a biochemical marker of lipid peroxidation (Fig.4F).GSE Recovers PXR/NF-kB Signaling in Inflamed Colon. We

extended to determine whether the anti-inflammatory effect of GSEwas associated with the recovery of PXR/NF-kB balance. The resultsshowed that the mRNA levels of IL-1b, IL-6, Cox-2, and iNOS, all ofwhich are typical NF-kB target genes, were remarkably induced in thecolonic tissue of DSS mice. In contrast, the increase in theseinflammatory mediators after DSS administration was obviouslyretarded in the mice with GSE treatment (Fig. 5, A–D). Moreover,Western blotting analysis showed that DSS-induced phosphorylationof NF-kB p65 was markedly decreased by GSE (Fig. 5H). GSE hadlittle effect on PXR activation in healthy mice, which is in line with

the results from in vitro study; however, GSE significantly attenuatedDSS-induced reduction in the mRNA levels of PXR and its targetgenes (Fig. 5, E–G). Immunohistochemical analysis showed that NF-kB p65 was significantly reduced, whereas PXR increased by GSEtreatment compared with DSS group (Fig. 5, I and K).

Discussion

PXR and NF-kB were mutually repressed. Thereafter, the activationof PXR by several agonists has been found effective in combatingagainst IBD (Shah et al., 2007; Dou et al., 2012). In this study, wefound that the compounds of ginsenosides, which are not PXRagonists themselves, repressed NF-kB-activated inflammatoryresponses in a PXR-dependent manner via enforcing the interactionbetween PXR and p65. In cultured colon cancer cell lines, thetreatment with ginsenosides markedly inhibited TNF-a-induced NF-kB activation and restored the expression of PXR and its target genesCYP3A4 and MDR1 in a PXR-dependent manner. Ginsenosides havelittle effect, however, on the activation of PXR in normal cells. Thetreatment of ginsenosides significantly attenuated DSS-induced colitis,which was associated with the recovery of the balance between PXRand NF-kB. Our study suggests that ginsenosides may selectivelyregulate the function of PXR in inflamed tissues via targeting PXR/NF-kB interaction without the influence of PXR activation in healthyconditions.The anti-inflammatory activities of ginseng extract and its isolated

compounds of ginsenosides have been widely confirmed in various invitro and in vivo conditions; however, the molecular mechanismsremain largely unclear. Ginseng extract could block NF-kB activity invarious cell types, including microglial cells and human breast cancercells (Park et al., 2009; Peralta et al., 2009). In line with these findings,we found that GSE markedly repressed TNF-a–stimulated NF-kBactivation in LS174T cells. An interesting finding was that theexpression of PXR and its target genes CYP3A4 and MDR1 wassignificantly recovered with GSE treatment; however, unlike thetypical PXR agonist RIF, GSE has little effect in PXR function inthe basal condition of LS174T cells. A previous report also indi-cated that ginseng had marginal effect in the activation of PXR(Mooiman et al., 2013). These results suggest that the restored expres-sion and activation of PXR on GSE treatment is probably via a directinhibition of NF-kB signaling; however, the silencing of PXR largelyabrogated the anti-inflammatory activity of GSE in LS174T cells,supporting a PXR-dependent effect in inhibiting NF-kB signalingby GSE.To further validate this finding, we screened the isolated single

compounds of ginsenosides in GSE with anti-inflammatory activity.The screening results showed that ginsenoside Rb1, a dominantcompound in GSE, possesses the strongest anti-inflammatory activityamong the ginsenosides screened. Additionally, we found that CK,a major metabolite of ginsenosides in the gastrointestinal tract, ondegradation by gut microbiota also possesses strong anti-inflammatoryactivity. In accordance with the finding of GSE, both ginsenoside Rb1and CK inhibit NF-kB signaling in a PXR-dependent manner,although both compounds affect little of the basal PXR function.Previous studies indicated that NF-kB p65 directly interacted with theDNA-binding domain of retinoid X receptor a, an important partnerfor PXR transactivation, providing a mechanistic explanation to themutual repression between PXR and NF-kB (Gu et al., 2006). Thismechanism might not explain the PXR-dependent effect of ginsenosidesin repressing NF-kB signaling, however, because ginsenosides are notPXR agonists. To this end, we tested whether PXR could directlyinteract with NF-kB p65. The coimmunoprecipitation test supports that

Fig. 3. A coimmunoprecipitation assay of the physical association between PXRand NF-kB p65 subunit. The cell lysates were immunoprecipitated with PXRantibody and an irrelevant rabbit IgG as a negative control, together with proteinA-agarose. The immunoprecipitates were then electrophoresed on SDS-PAGEgels, transferred onto PVDF membranes, and immunoblotted with an antibodyagainst PXR and NF-kB p65. Similar results were observed in three independentexperiments.

1186 Zhang et al.

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

PXR directly interacts with p65 and that ginsenosides can markedlypromote this interaction. In addition, the confocal analysis of p65nuclear translocation confirmed that ginsenosides could decreasep65 nuclear translocation in a PXR-dependent manner. All theseresults suggest that PXR might physically associate with p65 in thecytoplasm, thereby preventing p65 nuclear translocation, which is a basicstep for NF-kB transactivation of its downstream proinflammatory genes.It seems that ginsenosides can strengthen the physical interaction betweenPXR and p65, thus showing PXR-dependent anti-inflammatory effects.In addition, it is important to note that TNF-a treatment significantlyreduces the mRNA and protein levels of PXR, which was recovered onGSE, as well as ginsenoside Rb1 and CK treatment. Similar results wereobserved in the colon of DSS-treated mice. It is reasonable to assumethat the restored expression of PXR may contribute to repress NF-kB

activation. These results suggest that ginsenosides may inhibit NF-kBactivation and downstream inflammation via the restoration of PXRexpression and the enforcement of the physical interaction betweenPXR and NF-kB.PXR is well known for its central role in the regulation of drug

metabolism and is a major regulator of CYP3A gene expression inboth human and rodents (Cheng et al., 2012). CYP3A is a member ofthe cytochrome P450 monooxygenase superfamily, which is responsiblefor the oxidative metabolism of numerous drugs and toxicants(Goodwin et al., 2002). PXR expression and several of its target geneshave been shown to be suppressed in the colon of patients with IBD(Langmann et al., 2004; Blokzijl et al., 2007). Humans with geneticvariation in the PXR encoding gene were associated with a decrease inPXR activity and an increase in susceptibility to IBD (Dring et al.,

Fig. 4. GSE attenuates DSS-induced colitis in mice. (A) Body weight changes after DSS induction of colitis. Data are plotted as percentage of basal body weight. (B) Colonlength. (C) Representative H&E-stained colon sections (magnification 200�). (D) Histology score. (E) Disease activity index. (F) myeloperoxidase (MPO) activity (U/gtissue). Values are expressed as mean 6 S.E.M. of six to eight mice in each group.*P , 0.05, **P , 0.01 versus normal; #P , 0.05, ##P , 0.01 versus DSS model.

Ginsenosides Regulate PXR/NF-kB Signaling 1187

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

2006). In the current study, we showed that GSE significantly attenuatedDSS-induced colitis, which is associated with repressed NF-kB activityand restored PXR and CYP3A mRNA expression in the inflamed colon.Although it has previously been reported that Panax ginseng influenceCYP3A activity(Malati et al., 2012), we confirmed that ginsenosides had

no effect on Pxr and Cyp3a11 mRNA expression in the healthy mice,which is consistent with the findings from in vitro LS174T cells and theresults from other groups (Gurley et al., 2002; Anderson et al., 2003).Ginseng and the ginsenosides extracts are widely available as food

supplements. The current study suggests that the consumption of GSE

Fig. 5. GSE restores PXR/NF-kB signaling in theinflamed colon. The mRNA levels of (A) IL-1b, (B) IL-6, (C) COX-2, (D) iNOS, (E) Pxr, (F) Cyp3a11, and (G)Mdr1a were determined using RT-PCR. Colon sampleswere isolated from mice (six to eight per group) treatedwith vehicle, GSE, DSS, or GSE, and DSS. Data areexpressed as mean 6 S.E.M. (H) Western blot wasperformed with phospho-p65 antibody. Representativeimages and scores of immunohistochemical staining of(I, J) NF-kB p65 and (K, L) PXR in colon sections. *P, 0.05, **P , 0.01 versus normal; #P , 0.05, ##P ,0.01 versus DSS model.

1188 Zhang et al.

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from

may be of benefit to IBD patients. Moreover, we showed thatginsenoside Rb1 and its metabolite CK may serve as useful leadingcompounds for the development of drugs targeting PXR/NF-kBinteraction against IBD. Because the strong activation of PXR mayarouse undesirable effects, such as the disrupted expression of variouskinds of drug-metabolizing enzymes and transporters, the strategy ofdeveloping PXR agonists to anti-inflammatory drugs remains to beestablished. In contrast, the current study indicates that targeting PXR/NF-kB interaction may be an alternative strategy for the developmentof anti-inflammatory drugs without disrupting PXR function in normaltissues.

Authorship ContributionsParticipated in research design: Hao, G. Wang, Zhang.Conducted experiments: Zhang, Yan, Xie, Y. Wu, Zhao, Ma, M. Wu.Contributed new reagents or analytic tools: Cao, Cheng, L. Wang, Zhu.Performed data analysis: Zhang, Hao, H. Wang.Wrote or contributed to the writing of the manuscript: Zhang, Hao,

G. Wang.

References

Anderson GD, Rosito G, Mohustsy MA, and Elmer GW (2003) Drug interaction potential of soyextract and Panax ginseng. J Clin Pharmacol 43:643–648.

Atreya I, Atreya R, and Neurath MF (2008) NF-kappaB in inflammatory bowel disease. J InternMed 263:591–596.

Attele AS, Wu JA, and Yuan CS (1999) Ginseng pharmacology: multiple constituents andmultiple actions. Biochem Pharmacol 58:1685–1693.

Azuma YT, Nishiyama K, Matsuo Y, Kuwamura M, Morioka A, Nakajima H, and Takeuchi T(2010) PPARa contributes to colonic protection in mice with DSS-induced colitis. IntImmunopharmacol 10:1261–1267.

Blokzijl H, Vander Borght S, Bok LI, Libbrecht L, Geuken M, van den Heuvel FA, Dijkstra G,Roskams TA, Moshage H, and Jansen PL, et al. (2007) Decreased P-glycoprotein (P-gp/MDR1) expression in inflamed human intestinal epithelium is independent of PXR proteinlevels. Inflamm Bowel Dis 13:710–720.

Cheng J, Shah YM, and Gonzalez FJ (2012) Pregnane X receptor as a target for treatment ofinflammatory bowel disorders. Trends Pharmacol Sci 33:323–330.

Dou W, Mukherjee S, Li H, Venkatesh M, Wang H, Kortagere S, Peleg A, Chilimuri SS, WangZT, and Feng Y, et al. (2012) Alleviation of gut inflammation by Cdx2/Pxr pathway in a mousemodel of chemical colitis. PLoS ONE 7:e36075.

Dring MM, Goulding CA, Trimble VI, Keegan D, Ryan AW, Brophy KM, Smyth CM, KeelingPW, O’Donoghue D, and O’Sullivan M, et al. (2006) The pregnane X receptor locus isassociated with susceptibility to inflammatory bowel disease. Gastroenterology 130:341–348,quiz 592.

Goodwin B, Redinbo MR, and Kliewer SA (2002) Regulation of cyp3a gene transcription by thepregnane x receptor. Annu Rev Pharmacol Toxicol 42:1–23.

Gu X, Ke S, Liu D, Sheng T, Thomas PE, Rabson AB, Gallo MA, Xie W, and Tian Y (2006)Role of NF-kappaB in regulation of PXR-mediated gene expression: a mechanism for thesuppression of cytochrome P-450 3A4 by proinflammatory agents. J Biol Chem 281:17882–17889.

Gurley BJ, Gardner SF, Hubbard MA, Williams DK, Gentry WB, Cui Y, and Ang CY (2002)Cytochrome P450 phenotypic ratios for predicting herb-drug interactions in humans. ClinPharmacol Ther 72:276–287.

Han ES, Oh JY, and Park HJ (2011) Cordyceps militaris extract suppresses dextran sodiumsulfate-induced acute colitis in mice and production of inflammatory mediators from macro-phages and mast cells. J Ethnopharmacol 134:703–710.

Hao H, Lai L, Zheng C, Wang Q, Yu G, Zhou X, Wu L, Gong P, and Wang G (2010) Microsomalcytochrome p450-mediated metabolism of protopanaxatriol ginsenosides: metabolite profile,reaction phenotyping, and structure-metabolism relationship. Drug Metab Dispos 38:1731–1739.

Hayden MS and Ghosh S (2004) Signaling to NF-kappaB. Genes Dev 18:2195–2224.Hofseth LJ and Wargovich MJ (2007) Inflammation, cancer, and targets of ginseng. J Nutr 137(1,Suppl)183S–185S.

Ishiguro Y (1999) Mucosal proinflammatory cytokine production correlates with endoscopicactivity of ulcerative colitis. J Gastroenterol 34:66–74.

Jang S, Lim Y, Valacchi G, Sorn S, Park H, Park NY, and Lee M (2012) Preventive effects ofprotopanaxadiol and protopanaxatriol ginsenosides on liver inflammation and apoptosis inhyperlipidemic apoE KO mice. Genes Nutr 7:319–329.

Jia L and Zhao Y (2009) Current evaluation of the millennium phytomedicine—ginseng (I):etymology, pharmacognosy, phytochemistry, market and regulations. Curr Med Chem 16:2475–2484.

Kang A, Hao H, Zheng X, Liang Y, Xie Y, Xie T, Dai C, Zhao Q, Wu X, and Xie L, et al. (2011)Peripheral anti-inflammatory effects explain the ginsenosides paradox between poor braindistribution and anti-depression efficacy. J Neuroinflammation 8:100–113.

Karrasch T and Jobin C (2008) NF-kappaB and the intestine: friend or foe? Inflamm Bowel Dis14:114–124.

Kim JA, Son JH, Yang SY, Song SB, Song GY, and Kim YH (2012) A new lupane-type triterpenefrom the seeds of Panax ginseng with its inhibition of NF-kB. Arch Pharm Res 35:647–651.

Kim KR, Chung TY, Shin H, Son SH, Park KK, Choi JH, and Chung WY (2010) Red ginsengsaponin extract attenuates murine collagen-induced arthritis by reducing pro-inflammatoryresponses and matrix metalloproteinase-3 expression. Biol Pharm Bull 33:604–610.

Langmann T, Moehle C, Mauerer R, Scharl M, Liebisch G, Zahn A, Stremmel W, and Schmitz G(2004) Loss of detoxification in inflammatory bowel disease: dysregulation of pregnane Xreceptor target genes. Gastroenterology 127:26–40.

Li J, Zhong W, Wang W, Hu S, Yuan J, Zhang B, Hu T, and Song G (2014) Ginsenosidemetabolite compound K promotes recovery of dextran sulfate sodium-induced colitis andinhibits inflammatory responses by suppressing NF-kB activation. PLoS ONE 9:e87810.

Ma X, Idle JR, and Gonzalez FJ (2008) The pregnane X receptor: from bench to bedside. ExpertOpin Drug Metab Toxicol 4:895–908.

Malati CY, Robertson SM, Hunt JD, Chairez C, Alfaro RM, Kovacs JA, and Penzak SR (2012)Influence of Panax ginseng on cytochrome P450 (CYP)3A and P-glycoprotein (P-gp) activityin healthy participants. J Clin Pharmacol 52:932–939.

Mencarelli A, Renga B, Palladino G, Claudio D, Ricci P, Distrutti E, Barbanti M, Baldelli F,and Fiorucci S (2011) Inhibition of NF-kB by a PXR-dependent pathway mediates counter-regulatory activities of rifaximin on innate immunity in intestinal epithelial cells. Eur JPharmacol 668:317–324.

Mooiman KD, Maas-Bakker RF, Moret EE, Beijnen JH, Schellens JH, and Meijerman I (2013)Milk thistle’s active components silybin and isosilybin: novel inhibitors of PXR-mediatedCYP3A4 induction. Drug Metab Dispos 41:1494–1504.

Neurath MF, Pettersson S, Meyer zum Büschenfelde KH, and Strober W (1996) Local admin-istration of antisense phosphorothioate oligonucleotides to the p65 subunit of NF-kappa Babrogates established experimental colitis in mice. Nat Med 2:998–1004.

Ouyang LF, Wang ZL, Dai JG, Chen L, and Zhao YN (2014) Determination of total ginsenosidesin ginseng extracts using charged aerosol detection with post-column compensation of thegradient. Chin J Nat Med 12:857–868.

Park JS, Park EM, Kim DH, Jung K, Jung JS, Lee EJ, Hyun JW, Kang JL, and Kim HS (2009)Anti-inflammatory mechanism of ginseng saponins in activated microglia. J Neuroimmunol209:40–49.

Peralta EA, Murphy LL, Minnis J, Louis S, and Dunnington GL (2009) American Ginsenginhibits induced COX-2 and NFKB activation in breast cancer cells. J Surg Res 157:261–267.

Perkins ND and Gilmore TD (2006) Good cop, bad cop: the different faces of NF-kappaB. CellDeath Differ 13:759–772.

Radad K, Gille G, Liu L, and Rausch WD (2006) Use of ginseng in medicine with emphasis onneurodegenerative disorders. J Pharmacol Sci 100:175–186.

Sarkar FH, Li Y, Wang Z, and Kong D (2008) NF-kappaB signaling pathway and its therapeuticimplications in human diseases. Int Rev Immunol 27:293–319.

Sartor RB (1994) Cytokines in intestinal inflammation: pathophysiological and clinical consid-erations. Gastroenterology 106:533–539.

Shah YM, Ma X, Morimura K, Kim I, and Gonzalez FJ (2007) Pregnane X receptor activationameliorates DSS-induced inflammatory bowel disease via inhibition of NF-kappaB target geneexpression. Am J Physiol Gastrointest Liver Physiol 292:G1114–G1122.

Wen XD, Wang CZ, Yu C, Zhao L, Zhang Z, Matin A, Wang Y, Li P, Xiao SY, and Du W, et al.(2014) Panax notoginseng attenuates experimental colitis in the azoxymethane/dextran sulfatesodium mouse model. Phytother Res 28:892–898.

Williams KL, Fuller CR, Dieleman LA, DaCosta CM, Haldeman KM, Sartor RB, and Lund PK(2001) Enhanced survival and mucosal repair after dextran sodium sulfate-induced colitis intransgenic mice that overexpress growth hormone. Gastroenterology 120:925–937.

Wong VK, Cheung SS, Li T, Jiang ZH, Wang JR, Dong H, Yi XQ, Zhou H, and Liu L (2010)Asian ginseng extract inhibits in vitro and in vivo growth of mouse lewis lung carcinoma viamodulation of ERK-p53 and NF-kB signaling. J Cell Biochem 111:899–910.

Yang XL, Guo TK, Wang YH, Gao MT, Qin H, and Wu YJ (2012a) Therapeutic effect ofginsenoside Rd in rats with TNBS-induced recurrent ulcerative colitis. Arch Pharm Res 35:1231–1239.

Yang XL, Guo TK, Wang YH, Huang YH, Liu X, Wang XX, Li W, Zhao X, Wang LP, and YanS, et al. (2012b) Ginsenoside Rd attenuates the inflammatory response via modulating p38 andJNK signaling pathways in rats with TNBS-induced relapsing colitis. Int Immunopharmacol12:408–414.

Ye H, Wu Q, Zhu Y, Guo C, and Zheng X (2014) Ginsenoside Rh2 alleviates dextran sulfatesodium-induced colitis via augmenting TGFb signaling. Mol Biol Rep 41:5485–5490.

Zhao YN, Wang ZL, Dai JG, Chen L, and Huang YF (2014) Preparation and quality assessmentof high-purity ginseng total saponins by ion exchange resin combined with macroporousadsorption resin separation. Chin J Nat Med 12:382–392.

Zhou C, Tabb MM, Nelson EL, Grün F, Verma S, Sadatrafiei A, Lin M, Mallick S, Forman BM,and Thummel KE, et al. (2006) Mutual repression between steroid and xenobiotic receptor andNF-kappaB signaling pathways links xenobiotic metabolism and inflammation. J Clin Invest116:2280–2289.

Zhou X, Cao L, Jiang C, Xie Y, Cheng X, Krausz KW, Qi Y, Sun L, Shah YM, and Gonzalez FJ,et al. (2014) PPARa-UGT axis activation represses intestinal FXR-FGF15 feedback signallingand exacerbates experimental colitis. Nat Commun 5:4573–4587.

Address correspondence to: Dr. Guangji Wang, State Key Laboratory of NaturalMedicines, Key Laboratory of Drug Metabolism and Pharmacokinetics, ChinaPharmaceutical University, No. 24, Tongjiaxiang, Nanjing 210009, China. E-mail:guangjiwang@hotmail.com; or Dr. Haiping Hao, State Key Laboratory of NaturalMedicines, Key Laboratory of Drug Metabolism and Pharmacokinetics, ChinaPharmaceutical University, No. 24, Tongjiaxiang, Nanjing 210009, China. E-mail:hhp_770505@hotmail.com.

Ginsenosides Regulate PXR/NF-kB Signaling 1189

at ASPE

T Journals on February 15, 2020

dmd.aspetjournals.org

Dow

nloaded from