Activation of Nur77 by Selected 1,1-Bis(3�-indolyl)-1-(p-substituted phenyl)methanes Induces Apoptosis throughNuclear Pathways*

Received for publication, January 4, 2005, and in revised form, April 26, 2005Published, JBC Papers in Press, May 3, 2005, DOI 10.1074/jbc.M500107200

Sudhakar Chintharlapalli‡, Robert Burghardt§, Sabitha Papineni¶, Shashi Ramaiah�,Kyungsil Yoon**, and Stephen Safe‡¶**‡‡

From the ‡Department of Biochemistry and Biophysics, §Department of Veterinary Anatomy and Public Health,¶Department of Veterinary Physiology and Pharmacology, and �Department of Veterinary Pathobiology, Texas A&MUniversity, College Station, Texas 77843 and **Institute of Biosciences and Technology, Texas A&M University SystemHealth Science Center, Houston, Texas 77030

Nur77 is an orphan receptor and a member of thenerve growth factor-I-B subfamily of the nuclear recep-tor family of transcription factors. Based on the resultsof transactivation assays in pancreatic and other cancercell lines, we have now identified for the first time Nur77agonists typified by 1,1-bis(3-indolyl)-1-(p-anisyl)meth-ane that activate GAL4-Nur77 chimeras expressing wild-type and the ligand binding domain (E/F) of Nur77. InPanc-28 pancreatic cancer cells, Nur77 agonists activatethe nuclear receptor, and downstream responses in-clude decreased cell survival and induction of cell deathpathways, including tumor necrosis factor-related apo-ptosis-inducing ligand (TRAIL) and poly(ADP-ribose)polymerase (PARP) cleavage. Moreover, the transacti-vation and apoptotic responses are also induced inother pancreatic, prostate, and breast cancer cells thatexpress Nur77. In Panc-28 cells, small inhibitory RNAfor Nur77 reverses ligand-dependent transactivationand induction of TRAIL and PARP cleavage. Nur77 ago-nists also inhibit tumor growth in vivo in athymic micebearing Panc-28 cell xenografts. These results identifycompounds that activate Nur77 through the ligand bind-ing domain and show that ligand-dependent activationof Nur77 through nuclear pathways in cancer cells in-duces cell death and these compounds are a novel classof anticancer agents.

The nuclear receptor superfamily of eukaryotic transcriptionfactors encompasses steroid hormone and other nuclear recep-tors for which ligands have been identified and orphan recep-tors with no known ligands (1–7). Nuclear receptors sharecommon structural features that include an N-terminal A/Bdomain, containing activation function-1 (AF-1),1 and a C-ter-

minal E domain, which contains AF-2 and the ligand bindingdomain (LBD). Nuclear receptors also have a DNA bindingdomain (C domain), a variable hinge (D domain), and C-termi-nal F regions. Ligand activation of class 1 steroid hormonereceptors induces homo- or heterodimer formations, which in-teract with consensus or nonconsensus palindromic responseelements. In contrast, class 2 receptors form heterodimers withthe retinoic X receptor as a common partner, whereas class 3and 4 orphan receptors act as homodimers or monomers andbind to direct response element repeats or single sites, respec-tively. The DNA binding domains of nuclear receptors all con-tain two zinc finger motifs that interact with similar half-sitemotifs; however, these interactions vary with the number ofhalf-sites (1 or 2), their orientation, and spacing. Differences innuclear receptor action are also determined by their otherdomains, which dictate differences in ligand binding, receptordimerization, and interaction with other nuclear cofactors.

Most orphan receptors were initially cloned and identified asmembers of the nuclear receptor family based on their domainstructure and endogenous or exogenous ligands have subse-quently been identified for many of these proteins (5–7). Thenerve growth factor I-B (NGFI-B) family of orphan receptorswere initially characterized as immediate early genes inducedby nerve growth factor in PC12 cells, and the three members ofthis family include NGFI-B� (Nur77), NGFI-B� (Nurr1), andNGFI-B� (Nor1) (8–10).

Nur77 plays an important role in thymocyte-negative selec-tion and in T-cell receptor-mediated apoptosis in thymocytes(11, 12), and overexpression of Nur77 in transgenic mice re-sulted in high levels of apoptosis in thymocytes (13, 14). Incancer cells, several mechanisms for Nur77-mediated apoptosishave been described, and differences between studies may bedue to the apoptosis-inducing agent or cell line (15–21). Forexample, the retinoid 6-[3-(1-adamantyl)-4-hydroxyphenyl]-2-naphthalene carboxylic acid (CD437) and 12-O-tetradecanoyl-phorbol-13-acetate (TPA) induce translocation of Nur77 fromthe nucleus to the mitochondria where Nur77 binds Bcl-2 toform a pro-apoptotic complex (15, 16). In contrast, it has beensuggested that TPA-induced Nur77 in LNCaP prostate cancercells activates transcription of E2F1, which is also pro-apo-ptotic (21). These studies are examples of ligand-independent

* This work was supported by National Institutes of Health GrantsES09106 and CA108718, M. D. Anderson Cancer Center PancreaticCancer Spore Grant P20CA10193, and the Texas Agricultural Experi-ment Station. The costs of publication of this article were defrayed inpart by the payment of page charges. This article must therefore behereby marked “advertisement” in accordance with 18 U.S.C. Section1734 solely to indicate this fact.

‡‡ To whom correspondence should be addressed: Dept. of VeterinaryPhysiology and Pharmacology, Texas A&M University, 4466 TAMU,Vet. Res. Bldg. 409, College Station, TX 77843-4466. Tel.: 979-845-5988;Fax: 979-862-4929; E-mail: ssafe@cvm.tamu.edu.

1 The abbreviations used are: AF, activation function; LBD, ligandbinding domain; NGFI-B, nerve growth factor I-B; CD437, 6-[3-(1-ada-mantyl)-4-hydroxyphenyl]-2-naphthalene carboxylic acid; TPA, 12-O-tetradecanoylphorbol-13-acetate; PARP, poly(ADP-ribose) polymerase;TRAIL, tumor necrosis factor-related apoptosis-inducing ligand; LMB,

leptomycin B; DMEM, Dulbecco’s modified Eagle’s medium; PI, pro-pidium iodide; TUNEL, terminal deoxynucleotidyl transferase-medi-ated dUTP nick end labeling; DIM, 3,3�-diindolylmethane; C, methyl-ene; PPAR, peroxisome proliferator-activated receptor; NurRE, Nur77response element; NBRE, Nur77 binding response element; z, benzyl-oxycarbonyl; fmk, fluoromethyl ketone; H&E, hematoxylin & eosin.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 26, Issue of July 1, pp. 24903–24914, 2005© 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 24903

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

pathways where Nur77 expression is induced and/or Nur77protein undergoes intracellular translocation, because ligandsfor this receptor have hitherto not been reported. This reportshows that 1,1-bis(3�-indolyl)-1-(p-substitutedphenyl)meth-anes containing trifluoromethyl, hydrogen, and methoxy sub-stituents induce Nur77-dependent transactivation in Panc-28pancreatic and other cancer cell lines. Nur77 agonists alsoinduce typical cellular signatures of apoptosis, including PARPcleavage and induction of TRAIL, and both ligand-dependenttransactivation and induction of apoptosis were associatedwith the action of nuclear Nur77. This study shows for the firsttime that ligand-dependent activation of the orphan receptorNur77 induces apoptosis in cancer cells, suggesting that Nur77agonists represent a new class of anticancer drugs.

MATERIALS AND METHODS

Cell Lines and Reagents—Panc-28, Panc-1, MiaPaCa-2, LNCaP,MCF-7, HT-29, and HCT-15 cancer cell lines were obtained from theAmerican Type Culture Collection (Manassas, VA). RKO, DLD-1, andSW-480 colon cancer cells were provided by Dr. S. Hamilton, and KU7and 253-JB-V-33 bladder cells were provided by Dr. A. Kamat (M. D.Anderson Cancer Center, Houston, TX). The C-substituted DIMs weresynthesized in this laboratory as previously described (22). Antibodiesfor PARP (sc8007), Sp1 (sc-59), and TRAIL (sc7877) were purchasedfrom Santa Cruz Biotechnology (Santa Cruz, CA) and Nur77 (IMG-528)from Imgenex (San Diego, CA). The GAL4 reporter containing fiveGAL4 response elements (pGAL4) was provided by Dr. Marty Mayo(University of North Carolina, Chapel Hill, NC). The GAL4-Nur77(full-length) and GAL4-Nur77 (E/F) chimeras were provided by Dr. JaeW. Lee (Baylor College of Medicine, Houston, TX) and Dr. T. Perlmann(Ludwig Institute for Cancer Research, Stockholm, Sweden), respec-tively, and Dr. Lee also provided the Nur77 response element-luciferase(NurRE-Luc) reporter construct. The GAL-4-coactivator fusion plas-mids pMSRC1, pMSRC2, pMSRC3, pMDRIP205, and pMCARM-1 werekindly provided by Dr. Shigeaki Kato (University of Tokyo, Tokyo,Japan). For RNA interference assays, we used a nonspecific scrambled(iScr) oligonucleotide as described (23). The small inhibitory RNA forNur77 (iNur77) was identical to the reported oligonucleotide (16), andthese were purchased from Dharmacon Research (Lafayette, CO). Lep-tomycin B (LMB) was obtained from Sigma, and caspase inhibitorswere purchased from BD Pharmingen. The following oligonucleotideswere prepared by IDT (Coralville, IA) and were used in gel mobilityshift assays; NBRE, 5�-GAT CCT CGT GCG AAA AGG TCA AGC GCTA-3�; NurRE, 5�-GAT CCT AGT GAT ATT TAC CTC CAA ATG CCAGGA-3�.

Transfection Assays—Transfection assays were essentially carriedout as previously described using Lipofectamine Plus reagent (Invitro-gen), and luciferase activities were normalized to �-galactosidase activ-ity. For RNA interference studies, cells were transfected with smallinhibitor RNAs for 36 h to ensure protein knockdown prior to thestandard transfection and treatment protocols (22, 23). Results areexpressed as means � S.E. for at least three replicate determinationsfor each treatment group.

Mammalian Two-hybrid Assay—Panc-28 cells were plated in 12-wellplates at 1 � 105 cells/well in DMEM/F-12 media supplemented with2.5% charcoal-stripped fetal bovine serum. After growth for 16 h, var-ious amounts of DNA, i.e. Gal4Luc (0.4 �g), �-gal (0.04 �g), VP-Nur77(E/F) (0.04 �g), pMSRC1 (0.04 �g), pMSRC2 (0.04 �g), pMSRC3(0.04 �g), pMDRIP205 (0.04 �g), and pMCARM-1 (0.04 �g) were trans-fected by Lipofectamine (Invitrogen) according to the manufacturer’sprotocol. After 5 h of transfection, the transfection mix was replacedwith complete media containing either vehicle (Me2SO) or the indicatedligand for 20–22 h. Cells were then lysed with 100 ml of 1� reporterlysis buffer, and 30 �l of cell extract was used for luciferase and�-galactosidase assays. Lumicount was used to quantitate luciferaseand �-galactosidase activities, and the luciferase activities were nor-malized to �-galactosidase activity.

Cell Growth and Apoptosis Assays—The different cancer cell lineswere cultured under standardized conditions. Panc-28 cells were grownin DMEM/Ham’s F-12 media containing 2.5% charcoal-stripped fetalbovine serum, and cells were treated with Me2SO and different concen-trations of test compounds as indicated. For longer term cell survivalstudies, the media was changed every second day, and values werepresented for a 4-day experiment. For all other assays, cytosolic, nu-

clear fractions, or whole cell lysates were obtained at various timepoints, analyzed by Western blot analysis, and bands were quantitatedas previously described (22, 23). Immunocytochemical analysis wasdetermined using Nur77 antibodies as previously reported (23).

Gel Shift Assay—Cells were seeded in DMEM/F-12 medium supple-mented with 2.5% charcoal-stripped serum and treated with 10 �M

DIM-C-pPhOCH3 for 30 min. Nuclear extracts were obtained usingNE-PER nuclear and cytoplasmic extraction reagents (Pierce ChemicalCo.). Oligonucleotides were synthesized, purified, and annealed, and 5pmol of specific oligonucleotides was 32P-labeled at the 5�-end using T4

polynucleotide kinase and [�-32P]ATP. Nuclear extracts were incubatedin HEPES with ZnCl2 and 1 �g of polydeoxyinosine-deoxycytidine for 5min; 100-fold excess of unlabeled wild-type or mutant oligonucleotideswere added for competition experiments and incubated for 5 min. Themixture was incubated with labeled DNA probe for 15 min on ice. Thereaction mixture was loaded onto a 5% polyacrylamide gel and ran at150 V for 2 h. The gel was dried, and protein-DNA complexes werevisualized by autoradiography using a Storm 860 PhosphorImager(Amersham Biosciences).

Annexin-V Staining—Detection of phosphatidylserine on the outsideof the cell membrane, a unique and early marker for apoptosis, wasperformed using a commercial kit (Vybrant Apoptosis Assay Kit #2,Molecular Probes, Eugene, OR). Panc-28 cells were cultured as de-scribed above, and treated with 10 �M DIM-C-pPhOCH3 or camptoth-ecin for 6, 12, and 24 h. Binding of annexin V-Alexa-488 conjugate andpropidium iodide (PI) was performed according to the manufacturer’sinstructions. After binding and washing, cells were observed underphase contrast and epifluorescent illumination using a 495 nm excita-tion filter and a 520 nm absorption filter for annexin V-Alexa 488 anda 546 nm excitation filter and a 590 nm absorption filter for PI. Healthycells were unstained by either dye; cells in early stages of apoptosiswere stained only by annexin V, whereas dead cells were stained byannexin V and PI. The assay was repeated on three separate Panc-28cell preparations.

Quantitative Real-time PCR—cDNA was prepared from the Panc-28cell line using a combination of oligodeoxythymidylic acid (Oligo-d(T)16),and dNTP mix (Applied Biosystems) and Superscript II (Invitrogen).Each PCR was carried out in triplicate in a 20-�l volume using SybrGreen Mastermix (Applied Biosystems) for 15 min at 95 °C for initialdenaturing, followed by 40 cycles of 95 °C for 30 s and 60 °C for 1 minin the ABI Prism 7700 Sequence Detection System. The ABI Dissocia-tion Curves software was used following a brief thermal protocol (95 °C15 s and 60 °C 20 s, followed by a slow ramp to 95 °C) to control formultiple species in each PCR amplification. Values for each gene werenormalized to expression levels of TATA-binding protein. The se-quences of the primers used for reverse transcription-PCR were asfollows: TRAIL forward, 5�-CGT GTA CTT TAC CAA CGA GCT GA-3�,reverse, 5�-ACG GAG TTG CCA CTT GAC TTG-3�; and TATA-bindingprotein forward, 5�-TGC ACA GGA GCC AAG AGT GAA-3�, reverse,5�-CAC ATC ACA GCT CCC CAC CA-3�.

Xenograft Experiment—Male athymic nude mice (BALB/c, ages 8–12weeks) were purchased from Harlan (Indianapolis, IN). The mice werehoused and maintained in laminar flow cabinets under specific patho-gen-free conditions. Panc-28 cells were harvested from subconfluentcultures by trypsinization and washed. Panc-28 cells (2 � 106) wereinjected subcutaneously into each mouse on both flanks using a 30-gauge needle. The tumors were allowed to grow for 11 days until tumorswere palpable. Mice were then randomized into two groups of sevenmice per group and dosed by oral gavage with either corn oil or DIM-C-pPhOCH3 every second day. The volume of corn oil was 75 �l, and thedose of DIM-C-pPhOCH3 was 25 mg/kg/day. The mice were weighed,and tumor areas were also measured every other day. Final body andtumor weights were determined at the end of the dosing regiment,and selected tissues were further examined by routine H & E stainingand immunohistochemical analysis for apoptosis using the TUNELassay.

RESULTS

Nur77 Expression and Structure-dependent Activation by C-substituted DIMs—Studies in this laboratory have been inves-tigating the anticarcinogenic activities of a series of ring-substituted 3,3�-diindolylmethanes (DIMs) and methylene(C)-substituted DIMs, and many of these compounds were ac-tive in vivo and in cell culture assays (22, 24–26). Some mem-bers of a series of C-substituted DIMs activated peroxisome

Nuclear Nur77 Induces Transactivation and Apoptosis24904

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

proliferator-activated receptor � (PPAR�) but not PPAR�, ret-inoic acid receptor, retinoic X receptor, estrogen receptor �, orthe aryl hydrocarbon receptor. Previous studies have linkedNur77 to decreased cell survival and activation of cell death

pathways by apoptosis-inducing agents in some cancer celllines (15–21), and we therefore investigated expression ofNur77 in cancer cell lines and the effects of a series of elevenC-substituted DIMs on Nur77 activation/translocation. Fig. 1A

FIG. 1. Nur77 expression and activation in cancer cell lines. A, Nur77 protein expression. Whole cell lysates from 12 different cancer celllines were analyzed for Nur77, Nurr1, and Nor1 by Western blot analysis as described under “Materials and Methods.” Nor1 protein was notdetected in these cell lines. Activation of Gal4-Nur77 (B) and NuRE (C) Panc-28 cells were treated with 10 or 20 �M of the various compounds,transfected with GAL4-Nur77/pGAL4 or NuRE, and luciferase activity was determined as described under “Materials and Methods.” D, Nur77activation by isomeric DIM-C-PhOCH3 compounds. Panc-28 cells were treated with 10 or 20 �M of the DIM-C-PhOCH3 isomers, transfected withGAL4-Nur77/pGAL4 and luciferase activity determined as described under “Materials and Methods.” Results are expressed as means � S.E. forat least three separate determinations for each treatment group, and significant (p � 0.05) induction is indicated by an asterisk. The compoundsin B and C were 1,1-bis(3�-indolyl)-1-(p-substitutedphenyl)methanes and the p-substituent X is shown directly in the figure. D compares theactivity of the p-substituted methoxy derivative (CIM-pPhOCH3) with the meta (DIM-C-mPhOCH3) and ortho (DIM-C-oPhOCH3) isomers.

Nuclear Nur77 Induces Transactivation and Apoptosis 24905

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

summarizes Western blot analysis of Nur77 in whole cell ly-sates from 12 different cancer cell lines derived from pancre-atic, prostate, breast, colon, and bladder tumors. Only the 253JB-V-33 bladder cancer cell line exhibited relatively low ex-pression of Nur77, and the antibodies and electrophoretic con-ditions gave two immunostained bands as previously reportedin other studies. Western blot analysis of the other NGFI-Bproteins showed variable expression of Nurr1, and Nor1 wasnot detectable in these cancer cell lines (data not shown).Similar results were also obtained in Jurkat T-cell leukemiacells (data not shown). Structure-dependent activation ofNur77 by a series of eleven C-substituted DIMs was investi-gated in Panc-28 cells transfected with a GAL4-Nur77 (full-length) chimera and a reporter construct containing five GAL4response elements linked to a luciferase reporter gene(pGAL4). The results (Fig. 1B) showed that three compoundscontaining p-trifluoromethyl (DIM-C-pPhCF3) and methoxy(DIM-C-pPhOCH3) substituents or the unsubstituted phenylgroup (DIM-C-Ph) activated luciferase activity. Similar resultswere also obtained in Panc-28 cells transfected with a constructcontaining a Nur response element (NurRE) (Fig. 1C), andthese same compounds also activated GAL4-Nur77/pGAL4 andNurRE in MiaPaCa-1 pancreatic, HCT-15 colon, and MCF-7breast cancer cells (data not shown). The structure-dependentactivation of Nur77 was also investigated using DIM-C-pPhOCH3 as a model, and the position of the methoxyl groupwas changed to the meta (DIM-C-mPhOCH3) and ortho (DIM-C-oPhOCH3) positions (Fig. 1D). Only the para-substitutedcompound was active. We also investigated N-methyl and2-methyl indole ring-substituted analogs of DIM-C-pPhOCH3,DIM-C-Ph, and DIM-C-pPhCF3, and these compounds did notactivate Nur77 (data not shown). These results demonstratethat activation of Nur77 by C-DIMs was structure-dependentand sensitive to substitution on the phenyl and indole rings.Thus, at least three C-substituted DIMs activate Nur77; one ofthese compounds (DIM-C-pPhCF3) also activates PPAR� (22, 26),whereas DIM-C-pPhOCH3 and DIM-C-Ph are PPAR�-inactive(22). DIM-C-pPhOH was inactive in both transactivation assaysand, at higher concentrations, decreased activity lower than ob-served in solvent (Me2SO) control.

Characterization and Interactions of C-DIMs That Activateand Inhibit Nur77-mediated Transactivation—The role of theLBD or E/F region in ligand-induced transactivation of Nur77was investigated in Panc-28 cells transfected with pGAL4 anda chimeric GAL4-Nur77(E/F) construct containing only the E/Fdomain of Nur77. Treatment of Panc-28 cells with differentconcentrations (5–15 �M) of DIM-C-pPhCF3, DIM-C-pPhOCH3,and DIM-C-Ph induced luciferase activity, whereas no re-sponse was observed in cells treated with Nur77-inactive DIM-C-pPhOH (Fig. 2A). These results are the first to identify aseries of compounds that directly activate Nur77(LBD)-dependent transactivation in Panc-28 or any other cancer cellline. The role of Nur77 in mediating transactivation was fur-ther investigated in Panc-28 cells treated with 10 or 20 �M

DIM-C-pPhOCH3 or DIM-C-Ph and transfected with pNurRE,a nonspecific “scrambled” small inhibitory RNA (iScr), or smallinhibitory RNA for Nur77 (iNur77). The results (Fig. 2B)showed decreased Nur77 protein in whole cell lysates and a90–100% decrease in ligand-induced transactivation over thedifferent concentrations of compounds, thus confirming the roleof Nur77 in mediating this response. As noted above, one com-pound that contained a p-hydroxy substituent (DIM-C-pPhOH)did not induce activity (Fig. 1B) and DIM-C-pPhOH was fur-ther investigated as a potential Nur77 antagonist. Panc-28cells were transfected with GAL4-Nur77/pGAL4 and cotreated

with DIM-C-pPhOH and Nur77 agonists DIM-C-pPhCF3, DIM-C-pPhOCH3, and DIM-C-pH (Fig. 2C). The results show thatDIM-C-pPhOH antagonizes activation of Nur77 by all threeC-DIM compounds. The structural specificity of Nur77 antag-onists was further investigated using meta-hydroxy (DIM-C-mPhOH) and ortho-hydroxy (DIM-C-oPhOH) analogs. DIM-C-mPhOH (10 or 20 �M) did not inhibit DIM-C-pPhOCH3- orDIM-C-Ph-induced transactivation (Fig. 2D). DIM-C-oPhOHalso did not exhibit Nur77 antagonist activity (Fig. 2E); how-ever, high doses (20 �M) of both Nur77 agonists and DIM-C-oPhOH were toxic. Thus, activation of Nur77 by C-DIMs wasE/F domain-dependent and Nur77 activation was inhibited byDIM-C-pPhOH; moreover, both activation and inhibition ofNur77-mediated transactivation was dependent on the struc-ture of the C-DIM compounds.

Nur77 DNA Binding and C-DIM-induced Nur77-coactivatorInteractions—Incubation of nuclear extracts from Panc-28 cellstreated with Me2SO or DIM-C-pPhOCH3 with 32P-labeledNBRE and NurRE (lanes 1 and 2, and 5 and 6, respectively)gave retarded bands in EMSA assays (Fig. 3A). Retarded bandintensities were decreased after incubation with 100-fold ex-cess NurRE (lane 3) or NBRE (lane 7) but not by mutantNurRE (lane 4) or mutant NBRE (lane 8) oligonucleotides.These results show that nuclear extracts containing Nur77bind NurRE and NBRE as dimers and monomers, respectively,and this corresponds to their migration in an electrophoreticmobility shift assay. Extracts from cells treated with Nur77-active C-substituted DIMs gave retarded band intensities sim-ilar to those observed for solvent-treated extracts suggestingminimal ligand-dependent loss of nuclear Nur77 in these cells.The retarded band pattern corresponds to that observed inprevious studies using nuclear extracts from cells or in vitrotranslated Nur77 (27, 28).

Ligand-dependent activation of nuclear receptors is depend-ent on interaction of the bound receptor with coactivators (29–31), and Fig. 3 (B–D) summarizes results of a mammaliantwo-hybrid assay in Panc-28 cells transfected with VP-Nur77(ligand binding domain) and GAL4-coactivator chimeras. Li-gand-induced Nur77-coactivator interactions were determinedusing a construct (pGAL4) containing 5 GAL4 response ele-ments. Coactivators used in this study include SRC-1, SRC-2(TIFII), SRC-3 (AIB1), PGC-1, TRAP220, and CARM-1. AGAL4-repressor (SMRT) chimera was also included in the as-say. All three ligands induced transactivation in cells trans-fected with GAL4-SRC-1, GAL4-PGC-1, and GAL4-TRAP220chimeras. DIM-C-pPhOCH3-induced transactivation in cellstransfected with GAL4-SRC-3 and GAL4-CARM-1 was slightlyactivated by DIM-C-pPhOCH3 and DIM-C-pPhCF3. The re-sults demonstrate that there were some ligand-dependent dif-ferences in transactivation observed for GAL4-SRC-3 andGAL4-CARM-1; however, the most significant interactions be-tween VP-Nur77 and GAL4 chimeras expressing SRC-1,PGC-1, and TRAP220 were induced by all three compounds.

Effects of Nur77-active C-DIMs on Cell Survival and Apo-ptosis and Role of Nuclear Nur77—In several cancer cell linestransfected with Nur77-GFP constructs, treatment with apo-ptosis and differentiation-inducing agents results in rapidtranslocation of Nur77 into the cytosol/mitochondria (15–20).Similar results have been observed in BGC-823 human gastriccancer cells where endogenous Nur77 is nuclear and TPA in-duced Nur77 translocation into the cytosol, and this was ac-companied by apoptosis but not by Nur77-dependent transac-tivation (17). Results summarized in Fig. 4A showimmunostaining of Nur77 in the nucleus of Panc-28 cellstreated with Me2SO and Nur77-active DIM-C-pPhCF3, DIM-C-pPhOCH3, and DIM-C-Ph for 6 h, and comparable results

Nuclear Nur77 Induces Transactivation and Apoptosis24906

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

were obtained in Panc-28, MiaPaCa, and LNCaP cells aftertreatment for 6 or 12 h (data not shown). In all cases, Nur77remained in the nucleus, and cells exhibited a compacted nu-clear staining pattern typically observed in cells activated forcell death pathways. In a separate experiment, Panc-28 cellswere treated with 10 or 20 �M DIM-C-pPhCF3, DIM-C-pPhOCH3, and DIM-C-Ph or 10 �M DIM-C-pPhOH for 12 h,and Nur77 protein levels were determined by Western blotanalysis of cytosolic and nuclear extracts (Fig. 4B). These re-sults also confirm that Nur77, in the presence or absence ofC-substituted DIM agonists, is a nuclear protein and ligand-

induced Nur77 translocation from the nucleus is not observed.Sp1 is a nuclear protein and was used as a control to ensureefficient separation of the two extracts, and Sp1 was identifiedonly in the nuclear fraction (Fig. 4B).

Nur77 agonists significantly decreased survival of Panc-28cells (Fig. 5A), and IC50 values for DIM-C-pPhCF3, DIM-C-pPhOCH3, and DIM-C-Ph were between 1 and 5 �M, whereasDIM-C-pPhOH did not affect cell survival. At longer timepoints (4 and 6 days), DIM-C-pPhOH slightly inhibited cellproliferation; however, induction of cell death was not observedfor this compound at concentrations as high as 20 �M. De-

FIG. 2. A, activation of GAL4-Nur77(E/F)/pGAL4. The effects of the various compounds was essentially determined as described in Fig. 1;however, a truncated GAL4 chimera expressing only the E/F domain of Nur77 was used in this experiment. B, effects of iNur77 on transactivation.Cells were treated with DIM-C-pPhOCH3 or DIM-C-Ph, transfected with NuRE and iNur77 or iScr (nonspecific), and luciferase activity determinedas described. Similar results were obtained for DIM-C-pPhCF3. Nur77 antagonist activity of DIM-C-pPhOH (C), DIM-C-mPhOH (D), andDIM-C-oPhOH (E). Cells were transfected with GAL4-Nur77/pGAL4 and different concentrations of the DIM-C-PhOH isomers and the Nur77agonists, and luciferase activity determined as described under “Materials and Methods.” Significant (p � 0.05) inhibition of transactivation isindicated (**). Results are expressed as means � S.E. for at least three separate determinations for each treatment group, and significant (p � 0.05)induction (*) or inhibition by iNur77 or DIM-C-pPhOH (**) is indicated.

Nuclear Nur77 Induces Transactivation and Apoptosis 24907

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

FIG. 3. DNA binding of Nur77 and ligand-induced coactivator-Nur77 interactions. A, gel mobility shift assay. Cells were treated withMe2SO (DMSO) or Nur77 agonists for 0.5 h, nuclear extracts were incubated with 32P-labeled NurRE and NBRE, and formation of retarded bandswas determined in a gel mobility shift assay as described under “Materials and Methods.” Arrows denote the specifically bound bands.GAL4-coactivator interactions with VP-Nur77(E/F) in Panc-28 cells were treated with DIM-C-pPhCF3 (B), DIM-C-pPhOCH3 (C), and DIM-C-Ph(D). Cells were transfected with the pGAL4, VP-Nur77(E/F), and GAL4-coactivator/repressor (chimera) constructs and treated with the Nur77agonists, and luciferase activity was determined as described under “Materials and Methods.” Significant (p � 0.05) induction of luciferase activityis indicated (*), and results are expressed as means � S.E. for three separate determinations for each treatment group.

Nuclear Nur77 Induces Transactivation and Apoptosis24908

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

creased cell survival is also observed for agents that induceapoptosis and/or Nur77 nuclear to cytosolic translocation incancer cells (15–20). Results illustrated in Fig. 5B show thattreatment of Panc-28 cells with Nur77 agonists induced cleav-age of PARP, whereas the Nur77-inactive DIM-C-pPhOH didnot induce this response. PARP cleavage is associated withactivation of cell death pathways; however, this was not accom-panied by changes in levels of bax (Fig. 5B) or bcl-2 proteins(data not shown). Moreover, treatment of Panc-28 cells with 10and 20 �M DIM-C-pPhOCH3 for 8 and 12 h showed a time- anddose-dependent increase of annexin V-stained cells using agreen fluorescent Alexa Fluor 488 probe (Fig. 5C). The effects ofcamptothecin (positive control for apoptosis) and DIM-C-pPhOCH3 were comparable. After treatment with DIM-C-pPhOCH3 for 6 h, annexin V-stained cells were significantlyincreased, plasma membrane blebbing was observed, and therewas minimal PI staining. However, after 12 h, PI staining wasincreased. Induction of PARP cleavage by Nur77 agonists wasalso observed in other pancreatic (MiaPaCa-2), prostate (LN-CaP), and breast (MCF-7) cancer cell lines (Fig. 5D). Inductionof PARP cleavage by the Nur77-active compounds in Panc-28cells was not accompanied by changes in Nur77 expression(Fig. 4B), and this was in contrast to TPA, which activatesnuclear pathways by inducing Nur77 expression (21). Using aprotocol comparable to that outlined in Fig. 5B, the induction ofPARP cleavage by the Nur77 agonists in Panc-28 cells was notaffected by the nuclear export inhibitor leptomycin B (LMB) (1ng/ml) (Fig. 5E). LMB alone slightly induced PARP cleavageand, for some cells cotreated with LMB plus Nur77 agonists,there was enhanced PARP cleavage. In contrast, previous stud-ies showed that LMB inhibits apoptosis in cells treated withapoptosis-inducing agents that activate nuclear-cytosol/mito-chondrial translocation of Nur77 (15, 16). These results dem-onstrate that activation of nuclear Nur77 by C-substituted

DIMs induces apoptosis in Panc-28 and other cancer cell lines;however, evidence for activation of the intrinsic apoptotic path-ways was not observed.

Nur77-active C-DIMs Induce TRAIL—In thymocytes, thereis evidence that Nur77-induced apoptosis is linked to transcrip-tional activation (32), and microarray studies in thymocytesundergoing Nur77-dependent apoptosis identified several apo-ptosis-related genes, including fasL and TRAIL (33). Results inFig. 6A show that Nur77 agonists that induce PARP cleavagealso induce TRAIL (but not fasL) protein expression in Panc-28cells, suggesting that this response may be a direct or indirectdownstream target of Nur77 agonists in cancer cells. TheNur77-inactive DIM-C-pPhOH did not induce TRAIL. In addi-tion, DIM-C-pPhOCH3 or DIM-C-Ph induced TRAIL mRNAlevels in Panc-28 cells (Fig. 6B). Because TRAIL activates theextrinsic apoptosis pathway and activation of caspase 8, wealso investigated the effect of a caspase 8 inhibitor (z-IETD-fmk) and the pan-caspase inhibitor (z-VAD-fmk) on inductionof PARP cleavage by Nur77 agonists (Fig. 6C). The resultsshow that both inhibitors blocked (60–90%) induction of PARPcleavage by Nurr7 agonists.

The role of Nur77 in mediating induction of TRAIL andPARP cleavage by DIM-C-pPhOCH3 was further investigatedin Panc-28 cells transfected with nonspecific RNA (iScr) andiNur77 (Fig. 6D). Levels of Nur77, PARP cleavage, and TRAILproteins were determined by Western blot analysis of wholecell extracts, and the results showed that iNurr significantlydecreased levels of all three proteins. In addition, cotreatmentof Panc-28 cells with DIM-C-pPhOH3 or DIM-C-Ph and theNur77 antagonist DIM-C-pPhOH (Fig. 6E) showed that thelatter compound also inhibited induction of PARP cleavage andTRAIL protein expression induced by Nur77 agonists. Theseresults demonstrate that Nur77 agonists induce apoptosispathways in cancer cells through transcriptional (nuclear)

FIG. 4. Nuclear localization ofNur77. A, immunostaining. Panc-28 cellswere treated with Me2SO or 10 �M Nur77agonists for 6 h, and cells were immuno-stained for Nur77 as described under“Materials and Methods.” Nur77 stainingwas not observed in cells treated withnonspecific IgG. B, nuclear localization insubcellular fractions. Panc-28 cells weretreated with the various compounds for12 h and Nur77 protein expression in cy-tosolic, and nuclear extracts were deter-mined by Western blot analysis. Sp1 pro-tein and a nonspecific (NS) band serve asloading controls, and Sp1, a nuclear pro-tein, also serves as a control for separa-tion of nuclear and cytosolic extracts.

Nuclear Nur77 Induces Transactivation and Apoptosis 24909

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

FIG. 5. Nur77 agonists decrease cell survival and induce apoptosis. A, cell survival. Panc-28 cells were treated with different concen-trations of C-substituted DIMs for 2 days, and cell numbers were determined as described under “Materials and Methods.” Results are expressedas means � S.E. for three separate determinations for each treatment group, and a significant (p � 0.050 decrease in cell survival is indicated byan asterisk. DIM-C-pPhOH inhibited cell growth only after treatment for 96 or 144 h; however, this compound did not induce cell death at any timepoint. B, effects of Nur77 agonists on PARP cleavage in Panc-28. Cells were treated with the different compounds alone or with LMB, and PARPcleavage was determined by Western blot analysis of whole cell lysates as described under “Materials and Methods.” Bax and bcl-2 (not shown)protein levels were not affected by treatment and NS (nonspecific) protein served as a loading control. C, annexin staining. Panc-28 cells weretreated with camptothecin (positive control) or DIM-C-pPhOCH3 for 6 h, and annexin staining was determined as described under “Materials andMethods.” Approximately 30–40% of cells treated with DIM-C-pPhOCH3 exhibited annexin staining. Induction of apoptosis in LNCaP, MiaPaCa-1,and MCF-7 cells (D) or Panc-28 cells (E) treated with Nur77 agonists alone or in combination with LMB, respectively. Cells were treated essentiallyas described (A), and PARP cleavage determined by Western blot analysis as described under “Materials and Methods.”

Nuclear Nur77 Induces Transactivation and Apoptosis24910

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

mechanisms, and at least one of the induced proteins (TRAIL)activates an extrinsic apoptotic pathway. In summary, selectedC-substituted DIMs have now been identified as ligands for the

orphan receptor Nur77, and activation of this receptor is asso-ciated with decreased cancer cell survival, induction of TRAIL,and apoptosis.

FIG. 6. Induction of TRAIL and PARP cleavage in Panc-28 cells is dependent on Nur77. A, induction of TRAIL. Cells were treated withNur77 agonists and levels of TRAIL protein in whole cell lysates were determined by Western blot analysis as described under “Materials andMethods.” Results are expressed as means � S.E. for three separate determinations and significant (p � 0.05) induction is indicated. B, inductionof TRAIL mRNA. Panc-28 cells were treated with 10 or 20 �M DIM-C-pPhOCH3 or DIM-C-Ph for 12 h, and TRAIL mRNA levels were determinedby reverse transcription-PCR as described under “Materials and Methods.” Results are expressed as means � S.E. for three separate determi-nations for each treatment group as significant (p � 0.05) is indicated by an asterisk. C, effects of caspase inhibitors. Cells were treated andanalyzed essentially as described in A; however cells were also cotreated with caspase 8 and pan-caspase inhibitors z-IETD or z-VAD-fmk,respectively. Both inhibitors partially decreased PARP cleavage. D, effects of iNur77 on TRAIL expression and PARP cleavage. Panc-28 cells weretreated with Me2SO or 10 �M DIM-C-pPhOCH3 for 24 h, transfected with iNur77 or iScr (nonspecific), and whole cell lysates were analyzed byWestern blot analysis as described under “Materials and methods. ” Results are expressed as means � S.E. for three separate determinations foreach treatment group and significant (p � 0.05) induction (*) or inhibition by iNur77 (**) is indicated. E, inhibition of induced PARP cleavage andTRAIL by DIM-C-pPhOH. Panc-28 cells were treated with 10 �M DIM-C-pPhOCH3 or DIM-C-Ph alone or in the presence of 20 �M DIM-C-pPhOH,and PARP cleavage and TRAIL protein expression were determined by Western blot analysis as described under “Materials and Methods.”

Nuclear Nur77 Induces Transactivation and Apoptosis 24911

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Inhibition of Tumor Growth in Athymic Nude Mice BearingPanc-28 Cell Xenografts—Approximately 11 days after injec-tion of Panc-28 cells, palpable tumors were detected and themice were administered corn oil (control) or DIM-C-pPhOCH3

(in corn oil) at a dose of 25 mg/kg/day, which was given by oralgavage. The animals were treated every second day, and tumorareas were determined over the duration of the experiment.The results (Fig. 7A) showed that DIM-C-pPhOCH3 signifi-cantly inhibited tumor growth (area), and this was also com-plemented by a parallel decrease in tumor weights (Fig. 7B).Analysis of tumors from control and treated animals (TUNELassay) indicated similar levels of apoptosis. Animal weight gainand organ weights were comparable in both treatment groups,and there were no apparent signs of toxicity in the DIM-C-pPhOCH3-treated mice compared with the corn oil controls.The mouse brain and muscle express relatively high levels ofNur77 (34), and examination of brain regions by H&E stainingdid not indicate any differences between the control (corn oil)and DIM-C-pPhOCH3-treated animals.

DISCUSSION

Nur77 is widely expressed in multiple tissues and has beenidentified as a critical mediator of T-cell receptor-dependentapoptosis in T-lymphocytes and T-hybridoma cells (35, 36).Expression of dominant negative or antisense Nur77 blockedT-cell receptor-mediated apoptosis in T-hybridoma cells andextensive apoptosis of thymocytes was observed in transgenic

mice overexpressing full-length Nur77 (13, 14, 35, 36). Activa-tion of cell death in macrophages is associated with increasedexpression of Nur77, and decreased cell death was observed inNur77-deficient macrophages (37). A recent study (38) showthat cadmium acetate induced apoptosis in WI-38 human lungfibroblasts and A549 human lung carcinoma cells, and this wasalso accompanied by induction of Nur77. Moreover, transfec-tion with dominant-negative Nur77 protected the cells againstcadmium-induced apoptosis.

Ongoing studies in the laboratory with a series of C-substi-tuted DIMs indicate that these compounds inhibit growth orinduce cell death of multiple cancer cell lines, and some of theseanalogs, including DIM-C-pPhCF3, activate PPAR� (22, 26).However, several PPAR�-inactive C-substituted DIMs also de-creased cell survival of several cancer cell lines (e.g. Fig. 5A)and inhibited carcinogen-induced mammary tumor growth infemale Sprague-Dawley rats (data not shown). Nur77 was con-sidered as a possible target for C-DIM compounds based onresults of several studies with retinoids, apoptosis, and differ-entiation-inducing agents that also inhibit cell growth andactivate extranuclear Nur77 (15–20). Initial studies confirmedNur77 protein expression in 12 prostate, colon, bladder, pan-creatic, and breast cancer cell lines (Fig. 1A). Results of screen-ing a panel of structurally diverse C-substituted DIMs showsthat DIM-C-pPhCF3 and two PPAR�-inactive analogs, DIM-C-pPhOCH3 and DIM-C-Ph, activate Nur77-dependent transac-tivation in Panc-28 and other cancer cell lines transfected withGAL4-Nur77 (full-length) or NuRE (Fig. 1, B and C). Moreover,ligand-induced transcriptional activation is observed withGAL4-Nur77(E/F) chimeras (Fig. 2C) in which only the ligandbinding domain of Nur77 is expressed. The role of Nur77 inmediating ligand-dependent transactivation was confirmed instudies showing that these responses were inhibited by eitheriNur77 (small inhibitory RNA) (Fig. 2B) or DIM-C-pPhOH,which exhibited Nur77 antagonist activity (Fig. 2C). Previousstudies on the crystal structure of the mouse Nurr1 LBD (39)and the Drosophila Nurr1 homolog DHR38 (40) show that theligand binding pocket is occupied by bulky hydrophobic aminoacid side chains. Moreover, due to the high sequence homologyamong NGFI-B family proteins, it has been suggested thatNur77, Nurr1, and Nor-1 may represent a class of orphanreceptors that function independently of ligand binding (41). Incontrast, this study shows that selected C-DIM compoundsuniquely induce nuclear Nur77-mediated transactivation; thisinduction response is observed through the E/F domain ofNur77 (Fig. 1C) and can be inhibited by DIM-C-pPhOH, a Nur77antagonist (Figs. 2 and 5). Moreover, both activation of Nur77 andNur77 antagonist activities by C-DIMs were highly structure-de-pendent (Figs. 1 and 2). These results suggest that the activeC-DIM compounds interact with the E/F domain of Nur77 andinduce conformational changes, resulting in binding to the ligandbinding pocket or other sites within the C-terminal region of Nur77.Currently, we are examining critical ligand interaction sites withinthe E/F domain of Nur77 by deletion/mutation analysis and bycrystallization of the Nur77 LBD in the presence or absence of theC-DIM ligands.

Several studies have reported activation of Nur77-dependenttransactivation in different cell lines, and these responses pri-marily involve the AF-1 domain of Nur77 and activation bykinases (28, 42, 43). For example, induction of Nur77-depend-ent transactivation was observed for the coactivator ASC-2 inCV-1 and HeLa cells; however, this effect was dependent oncalcium/calmodulin-dependent protein kinase IV and did notinvolve direct ASC-2-Nur77 interactions (42). Transactivationmediated by Nur77 homodimers is enhanced by protein kinaseA and SRC1–3 in CV-1 and AtT-20 cells and these responses

FIG. 7. Inhibition of tumor growth by DIM-C-pPhOCH3. Tumorareas (A) and weights (B). Male athymic nude mice bearing Panc-28 cellxenografts were treated every second day with corn oil and DIM-C-pPhOCH3 in corn oil as described under “Materials and Methods.” Theoverall dose of DIM-C-pPhOCH3 was 25 mg/kg/day. Significant (p �0.05) inhibition of tumor areas and weights is indicated by an asterisk.H&E stainings of muscle and brain tissue from control and treatedanimals were similar, and tumors from control and treated animalsexhibited apoptosis (data not shown).

Nuclear Nur77 Induces Transactivation and Apoptosis24912

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

were AF-1-dependent (28). Another report also confirmed thatNur77 transactivation in C2C12 and COS-1 cells was enhancedby SRCs and other coactivators, and involved direct interac-tions of coactivators with the A/B (and not E/F) domain ofNur77 (43). These observations are consistent with the crystalstructure of Nurr1, which lacks the “classical binding site forcoactivators” (39). However, ligand-dependent activation ofNur77 E/F domain observed in this study (Fig. 1C) should alsobe accompanied by interactions with some nuclear receptorcoactivators/coregulators. Initial studies showed that, in theabsence of ligand, VP-Nur77(E/F) did not interact with GAL-4-coactivators (PGF-1, CARM-1, SRC1–3, and TRAP220) orGALR-SMRT chimeras (data not shown); however, DIM-C-pPhCF3, DIM-C-pPhOCH3, and DIM-C-Ph induced interac-tions between several common nuclear receptor coactivators(PGC-1, SRC-1, and TRAP220) and the LBD (E/F) of Nur77 inmammalian two-hybrid assays (Fig. 3, B–D). These results areconsistent with other studies on activation of nuclear receptorsby ligands and their interactions with specific coactivatorsthrough binding receptor E/F domains. For example, our recentstudies with PPAR�-active C-substituted DIMs in colon cancercells show that ligand-induced PPAR�(E/F domain)-coactivatorinteractions in mammalian two-hybrid assays primarily in-volved PGC-1 (26), whereas C-DIM-induced Nur77-coactivatorinteractions in Panc-28 cells involve multiple coactivators. Al-though the crystal structure of unliganded Nurr1 shows thatthis receptor does not contain a classic coactivator interactionsite in the E/F domain (helix 12), novel coactivator interactionsurfaces have recently been identified between helices 11 and12 in Nurr1 (44). This region is similar in human Nur77, andcurrent studies are investigating C-DIM-induced interactionsurfaces between the E/F domain of Nur77 and coactivators. Insummary, the transactivation and coactivator-Nur77 inter-actions induced by DIM-C-pPhCF3, DIM-C-pPhOCH3, andDIM-C-Ph are consistent with results obtained for otherligand-activated nuclear receptors suggesting that selected C-substituted DIMs are a novel class of compounds that induceE/F domain-dependent activation of Nur77.

Treatment of Panc-28 cells with Nur77-active C-substitutedDIMs agonists decreased cell survival (Fig. 4A) and inducednuclear condensation within 48 and 24 h, respectively, and thisis typically observed in cells undergoing cell death. We there-fore further examined Nur77-mediated induction of PARPcleavage, which is a well characterized downstream marker ofactivated cell death pathways. PARP cleavage was induced inPanc-28 cells treated with Nur77 agonists (Fig. 5B), and simi-lar results were observed in other pancreatic, prostate, andbreast cancer cell lines (Fig. 5D). Annexin V staining was alsoobserved in Panc-28 cells treated with Nur77-active C-DIMs(Fig. 5C), and these data further confirm induction of apoptosisin these cancer cell lines. Previous studies report that inductionof cell death pathways by apoptosis-inducing agents in somecancer cell lines is accompanied by translocation of Nur77 fromthe nucleus to the cytosol/mitochondria, and this has beenlinked to cytochrome c release and direct interaction of Nur77with bcl-2 (15–20). In contrast, we observed that treatment ofPanc-28 cells with Nur77-active C-DIMs resulted only in for-mation of a nuclear complex (Fig. 4, A and B). Moreover, inhi-bition of nuclear export of Nur77 by LMB did not affect PARPcleavage induced by Nur77-active C-DIMs (Fig. 5E), suggestingthat this response is mediated through nuclear Nur77. Thisnuclear pathway for induction of apoptosis is in contrast to theeffects observed for TPA and CD437, which induce nuclearexport of Nur77 in cancer cell lines, and inhibition of Nur77nuclear export by LMB, which inhibits induction of apoptosis(15, 16). These results clearly distinguish between the induc-

tion of cell death pathways in cancer cells through ligand-de-pendent activation of nuclear Nur77 (this study) and throughinduction of Nur77 nuclear translocation (15–20).

Overexpression of Nur77 in thymocytes induces expressionof several genes associated with apoptosis (33), and at least oneof the genes, TRAIL (protein and mRNA), is also induced byNur77 agonists in Panc-28 cells (Fig. 6, A and B). RNA inter-ference assays with iNur77 (Fig. 6D) and inhibition studieswith the Nur77 antagonist DIM-C-pPhOH (Fig. 6E) demon-strate that induction of TRAIL and PARP cleavage by DIM-C-pPhOCH3 and DIM-C-Ph are Nur77-dependent. Thus, the nu-clear action of Nur77 agonists in cancer cell lines is comparableto the transcriptionally dependent pathway observed in T-cellsoverexpressing Nur77 (33). TRAIL typically activates caspase8, and the extrinsic pathways of apoptosis and the caspase 8inhibitor z-IETD-fmk significantly blocks (�60%) induction ofPARP cleavage by Nur77 agonists (Fig. 5C). The pan-caspaseinhibitor z-VAD-fmk blocked �90% of induced PARP cleavagesuggesting that, although TRAIL may be a major Nur77-in-duced gene in Panc-28 cells, other pro-apoptotic genes may alsobe induced; these are currently being investigated. We alsoobserved in xenograft experiments that DIM-C-pPhOCH3 in-hibited tumor growth in athymic nude mice bearing Panc-28cell xenografts (Fig. 7).

In summary, results of this study have identified a novelgroup of C-substituted DIMs that activate the orphan receptorNur77 through the E/F domain. These results are in contrast toprevious reports showing that kinase/coactivator-dependentactivation of Nur77 was primarily AF-1-dependent (23, 42, 43).It has also been reported that nuclear receptor coactivatorsinteract with the N-terminal A/B but not E/F domains ofNur77, and in the absence of C-DIM compounds, coactivator-Nur(E/F) interactions were not observed in this study. How-ever, DIM-C-pPhCF3, DIM-C-Ph, and DIM-C-pPhOCH3 in-duced coactivator interactions with the E/F domain of Nur77(Fig. 3), and this was consistent with Nur77 (nuclear)-depend-ent transactivation. Activation of Nur77 by selected C-DIMs isassociated with decreased cancer cell survival, induction ofapoptosis, induced expression of the apoptosis gene/proteinTRAIL, and inhibited tumor growth in vivo. These resultssuggest that C-DIM ligands that activate Nur77 are a potentialnew class of anticancer agents. Their activities and mecha-nisms of action in other cancer cell lines are currently beinginvestigated.

REFERENCES

1. Tsai, M. J., and O’Malley, B. W. (1994) Annu. Rev. Biochem. 63, 451–4862. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono,

K., Blumberg, B., Kastner, P., Mark, M., Chambon, P., and Evans, R. M.(1995) Cell 83, 835–839

3. Beato, M., Herrlich, P., and Schutz, G. (1995) Cell 83, 851–8574. Olefsky, J. M. (2001) J. Biol. Chem. 276, 36863–368645. Enmark, E., and Gustafsson, J. A. (1996) Mol. Endocrinol. 10, 1293–13076. Giguere, V. (1999) Endocr. Rev. 20, 689–7257. Mohan, R., and Heyman, R. A. (2003) Curr. Top. Med. Chem. 3, 1637–16478. Milbrandt, J. (1988) Neuron 1, 183–1889. Ryseck, R. P., Macdonald-Bravo, H., Mattei, M. G., Ruppert, S., and Bravo, R.

(1989) EMBO J. 8, 3327–333510. Nakai, A., Kartha, S., Sakurai, A., Toback, F. G., and DeGroot, L. J. (1990)

Mol. Endocrinol. 4, 1438–144311. Winoto, A. (1997) Semin. Immunol. 9, 51–5812. He, Y. W. (2002) J. Leukoc. Biol. 72, 440–44613. Cheng, L. E., Chan, F. K., Cado, D., and Winoto, A. (1997) EMBO J. 16,

1865–187514. Calnan, B. J., Szychowski, S., Chan, F. K., Cado, D., and Winoto, A. (1995)

Immunity 3, 273–28215. Li, H., Kolluri, S. K., Gu, J., Dawson, M. I., Cao, X., Hobbs, P. D., Lin, B., Chen,

G., Lu, J., Lin, F., Xie, Z., Fontana, J. A., Reed, J. C., and Zhang, X. (2000)Science 289, 1159–1164

16. Lin, B., Kolluri, S. K., Lin, F., Liu, W., Han, Y. H., Cao, X., Dawson, M. I., Reed,J. C., and Zhang, X. K. (2004) Cell 116, 527–540

17. Wu, Q., Liu, S., Ye, X. F., Huang, Z. W., and Su, W. J. (2002) Carcinogenesis 23,1583–1592

18. Holmes, W. F., Soprano, D. R., and Soprano, K. J. (2003) Oncogene 22,6377–6386

Nuclear Nur77 Induces Transactivation and Apoptosis 24913

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

19. Holmes, W. F., Soprano, D. R., and Soprano, K. J. (2003) J. Cell. Biochem. 89,262–278

20. Wilson, A. J., Arango, D., Mariadason, J. M., Heerdt, B. G., and Augenlicht,L. H. (2003) Cancer Res. 63, 5401–5407

21. Mu, X., and Chang, C. (2003) J. Biol. Chem. 278, 42840–4284522. Qin, C., Morrow, D., Stewart, J., Spencer, K., Porter, W., Smith, R., III,

Phillips, T., Abdelrahim, M., Samudio, I., and Safe, S. (2004) Mol. CancerTher. 3, 247–259

23. Abdelrahim, M., Samudio, I., Smith, R., Burghardt, R., and Safe, S. (2002)J. Biol. Chem. 277, 28815–28822

24. McDougal, A., Sethi-Gupta, M., Ramamoorthy, K., Sun, G., and Safe, S. (2000)Cancer Letts. 151, 169–179

25. McDougal, A., Wilson, C., and Safe, S. (1997) Cancer Lett. 120, 53–6326. Chintharlapalli, S., Smith, R., III, Samudio, I., Zhang, W., and Safe, S. (2004)

Cancer Res. 64, 5994–600127. Philips, A., Lesage, S., Gingras, R., Maira, M. H., Gauthier, Y., Hugo, P., and

Drouin, J. (1997) Mol. Cell. Biol. 17, 5946–595128. Maira, M., Martens, C., Batsche, E., Gauthier, Y., and Drouin, J. (2003) Mol.

Cell. Biol. 23, 763–77629. Rosenfeld, M. G., and Glass, C. K. (2001) J. Biol. Chem. 276, 36865–3686830. Xu, J., and Li, Q. (2003) Mol. Endocrinol. 17, 1681–169231. Smith, C. L., and O’Malley, B. W. (2004) Endocrine Rev. 25, 45–7132. Kuang, A. A., Cado, D., and Winoto, A. (1999) Eur. J. Immunol. 29, 3722–372833. Rajpal, A., Cho, Y. A., Yelent, B., Koza-Taylor, P. H., Li, D., Chen, E., Whang,

M., Kang, C., Turi, T. G., and Winoto, A. (2003) EMBO J. 22, 6526–6536

34. Law, S. W., Conneely, O. M., DeMayo, F. J., and O’Malley, B. W. (1992) Mol.Endocrinol. 6, 2129–2135

35. Liu, Z. G., Smith, S. W., McLaughlin, K. A., Schwartz, L. M., and Osborne,B. A. (1994) Nature 367, 281–284

36. Woronicz, J. D., Calnan, B., Ngo, V., and Winoto, A. (1994) Nature 367,277–281

37. Kim, S. O., Ono, K., Tobias, P. S., and Han, J. (2003) J. Exp. Med. 197,1441–1452

38. Shin, H. J., Lee, B. H., Yeo, M. G., Oh, S. H., Park, J. D., Park, K. K., Chung,J. H., Moon, C. K., and Lee, M. O. (2004) Carcinogenesis 25, 1467–1475

39. Wang, Z., Benoit, G., Liu, J., Prasad, S., Aarnisalo, P., Liu, X., Xu, H., Walker,N. P., and Perlmann, T. (2003) Nature 423, 555–560

40. Baker, K. D., Shewchuk, L. M., Kozlova, T., Makishima, M., Hassell, A.,Wisely, B., Caravella, J. A., Lambert, M. H., Reinking, J. L., Krause, H.,Thummel, C. S., Willson, T. M., and Mangelsdorf, D. J. (2003) Cell 113,731–742

41. Benoit, G., Malewicz, M., and Perlmann, T. (2004) Trends Cell Biol. 14,369–376

42. Sohn, Y. C., Kwak, E., Na, Y., Lee, J. W., and Lee, S. K. (2001) J. Biol. Chem.276, 43734–43739

43. Wansa, K. D., Harris, J. M., and Muscat, G. E. (2002) J. Biol. Chem. 277,33001–33011

44. Codina, A., Benoit, G., Gooch, J. T., Neuhaus, D., Perlmann, T., and Schwabe,J. W. (2004) J. Biol. Chem. 279, 53338–53345

Nuclear Nur77 Induces Transactivation and Apoptosis24914

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from

Yoon and Stephen SafeSudhakar Chintharlapalli, Robert Burghardt, Sabitha Papineni, Shashi Ramaiah, Kyungsil

Induces Apoptosis through Nuclear Pathways-substituted phenyl)methanesp-indolyl)-1-(′Activation of Nur77 by Selected 1,1-Bis(3

doi: 10.1074/jbc.M500107200 originally published online May 3, 20052005, 280:24903-24914.J. Biol. Chem. 

  10.1074/jbc.M500107200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/280/26/24903.full.html#ref-list-1

This article cites 44 references, 16 of which can be accessed free at

by guest on January 9, 2020http://w

ww

.jbc.org/D

ownloaded from