Sensitization of TRAIL-resistant breast cancer cells by 2-selenium

Carcinogenesis vol.32 no.2 pp.154–167, 2010doi:10.1093/carcin/bgq234Advance Access publication November 16, 2010

2-Tellurium-bridged b-cyclodextrin, a thioredoxin reductase inhibitor, sensitizes humanbreast cancer cells to TRAIL-induced apoptosis through DR5 induction and NF-kBsuppression

Tingting Lin1,2, Zhiying Ding3, Nan Li1, Jiayun Xu1,Guimin Luo4, Junqiu Liu1,� and Jiacong Shen1

1State Key Laboratory of Supramolecular Structure and Materials, JilinUniversity, Changchun 130012, People’s Republic of China, 2College ofInstrumentation and Electrical Engineering, Jilin University, Changchun130061, People’s Republic of China, 3School of Pharmaceutical Sciences,Jilin University, Changchun 130012, People’s Republic of China and 4KeyLaboratory for Molecular Enzymology and Engineering of the Ministry ofEducation, Jilin University, Changchun 130012, People’s Republic of China

�To whom correspondence should be addressed. Department of Chemistry,Jilin University, 2699 Qianjin Road, Changchun 130012, People’s Republic ofChina. Tel: þ86 431 85168452; Fax: þ86 431 85193421;Email: [email protected]

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)exhibits potent antitumor activity via membrane receptors oncancer cells without deleterious side effects for normal tissue. Un-fortunately, breast cancer cells, as many other cancer types, de-velop resistance to TRAIL; therefore, TRAIL sensitizing agentsare currently being explored. 2-Tellurium-bridged b-cyclodextrin(2-TeCD) is a synthetic organotellurium compound, with bothglutathione peroxidase-like catalytic ability and thioredoxin re-ductase inhibitor activity. In the present study, we reported that 2-TeCD sensitized TRAIL-resistant human breast cancer cells andxenograft tumors to undergo apoptosis. In vitro, 2-TeCD effi-ciently sensitized MDA-MB-468 and T47D cells, but not untrans-formed human mammary epithelial cells, to TRAIL-mediatedapoptosis, as evidenced by enhanced caspase activity and poly(adenosine diphosphate-ribose) polymerase cleavage. Froma mechanistic standpoint, we showed that 2-TeCD treatment ofbreast cancer cells significantly upregulated the messenger RNAand protein levels of TRAIL receptor, death receptor (DR) 5, ina transcription factor Sp1-dependent manner. 2-TeCD treatmentalso suppressed TRAIL-induced nuclear factor-kB (NF-kB) pro-survival pathways by preventing cytosolic IkBa degradation, aswell as p65 nuclear translocation. Consequently, the combinedadministration suppressed anti-apoptotic molecules that are tran-scriptionally regulated by NF-kB. In vivo, 2-TeCD and TRAILwere well tolerated in mice and their combination significantlyinhibited growth of MDA-MB-468 xenografts and promoted ap-optosis. Upregulation of DR5 and downregulation of NF-kB bythe dual treatment were also observed in tumor tissues. Overall,2-TeCD sensitizes resistant breast cancer cells to TRAIL-basedapoptosis in vitro and in vivo. These findings provide strong evi-dence for the therapeutic potential of this combination againstbreast cancers.

Introduction

Tumor necrosis factor a-related apoptosis-inducing ligand (TRAIL) isa member of the tumor necrosis factor (TNF) family of cytokines.TRAIL exerts cytotoxic effects on malignant cells without any harmto normal cells (1). To date, five members of the human TNF receptor(TNFR) superfamily have been identified that can bind TRAIL. The

death receptors (DRs), DR4 (TRAIL-R1) and DR5 (TRAIL-R2), con-tain two cysteine-rich extracellular TRAIL-binding domains and a cy-toplasmic death domain that are required for transmitting a cytotoxicsignal (2,3). The decoy receptors, DcR1 (TRAIL-R3) and DcR2(TRAIL-R4), also possess comparable affinity for binding withTRAIL, but they do not transmit apoptogenic signals due to a non-functional death domain (4,5). Finally, TRAIL binds the receptorosteoprotegerin (TNFR11B), which is a soluble protein incapable ofsignaling (6).

Following TRAIL engagement with either DR4 or DR5, the ligatedDRs cluster and microaggregate within the cell membrane, therebyinitiating formation of the death-inducing signaling complex (DISC)(7). The functional DISC is composed minimally of DRs (DR4 andDR5), adapter protein fas-associated death domain and caspase 8 or10 (8). Active caspases 8 and 10 cleave and directly activate down-stream effector caspases (3, 6 and 7), which ultimately cut vital cel-lular substrates and result in apoptosis (9). Meanwhile, TRAILreceptors signaling pathway also leads to the activation of the nuclearfactor-jB (NF-jB), which operates as a negative regulator for DISCformation (10,11). Once stimulated by TRAIL, the subunit of NF-jBcould translocate to nucleus and upregulate anti-apoptotic genes, suchas cellular FLICE-like inhibitory proteins and cellular inhibitors ofapoptosis (c-IAPs), which eventually thwart activation of caspases(12,13).

As a promising therapy agent for treatment of malignancies,TRAIL has been shown to induce apoptosis against breast carcinoma(14). Unfortunately, the majority of cell lines are insensitive toTRAIL-induced apoptosis, and the mechanism of the resistance hasbeen attributed to dysfunction of different steps in the apoptosis path-ways, as well as elevation of survival signals. The former includessuppressed expression of the DRs, fas-associated death domain orcaspase-8 by mutation or imprinting (15). The survival signals consistof overexpression of cellular FLICE-like inhibitory protein, Bcl-2 orBcl-XL and inhibitors of apoptosis (IAPs) or activation of NF-jB(16,17). Therefore, modulation of these points with a chemotherapeu-tic agent would sensitize TRAIL-induced apoptosis in breast cancercells (18,19).

2-Tellurium-bridged b-cyclodextrin (2-TeCD; Figure 1A), a syn-thetic organotellurium compound, has exhibited both glutathione per-oxidase (GPx)-like catalytic ability and thioredoxin reductase (TrxR)inhibitor activity (20,21). As GPx mimics, 2-TeCD share the chemicalcharacteristics with its selenium counterparts 2-selenium-bridgedb-cyclodextrin (22,23). It was considered as a promising substitutefor ebselen, a well-known GPx mimic (24), with improved enzymaticactivity (25,26). In addition, 2-TeCD has been shown to exhibit anti-tumor effects on breast carcinoma and colon carcinoma (21,27–29).The chemopreventive activity has been attributed to its ability to in-hibit TrxR, which is overexpressed in primary tumors and becomingan attractive target for cancer therapy (30). Our recent report demon-strated that 2-TeCD was able to suppress the translocation of NF-jBthat was stimulated by TNFa (31). Considering NF-jB could bea target for TRAIL-sensitization, we hypothesized that a combinationof 2-TeCD and TRAIL could be a promising strategy in the treatmentof refractory breast cancers.

In the present study, we found that 2-TeCD potently sensitized hu-man breast cancer cells to TRAIL-mediated apoptosis via DR5 induc-tion and NF-jB inhibition. Using the MDA-MB-468 xenograft model,we further demonstrated that this combination was well tolerated innude mice and reduced tumor burden. For the first time, we presentclear evidences that 2-TeCD, a TrxR inhibitor, can overcome TRAILresistance in vitro and in vivo; therefore their combination providesa powerful therapeutic option for human breast cancer treatment.

Abbreviations: c-IAP, cellular inhibitor of apoptosis; DISC, death-inducingsignaling complex; DR, death receptor; GPx, glutathione peroxidase; IAP, in-hibitor of apoptosis; mRNA, messenger RNA; NF-jB, nuclear factor-jB;PCR, polymerase chain reaction; siRNA, small interfering RNA; 2-TeCD,2-tellurium-bridged b-cyclodextrin; TNF, tumor necrosis factor; TRAIL, tumornecrosis factor related apoptosis-inducing ligand; TrxR, thioredoxin reductase.

� The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] 154

Downloaded from https://academic.oup.com/carcin/article-abstract/32/2/154/2391749by gueston 09 April 2018

Materials and methods

Cell lines and reagents

BT-474, MCF-7, MDA-MB-231, MDA-MB-453, MDA-MB-468, T47D,SKBr3 and ZR75-1 breast carcinoma cells (American Type Culture Collec-tion) were grown in Dulbecco’s modified Eagle’s medium (Gibco , Rockville,MD) supplemented with 4.5 g/l of glucose, 4 mmol/l of l-glutamine, 100U/ml of penicillin/streptomycin and 10% fetal calf serum (Invitrogen, Carls-bad, CA). Human mammary epithelial cells Cambrex, Walkersville, MD)were grown as described in the manufacturer’s instructions. All cells weregrown at 37�C in a humidified incubator with 5% CO2 atmosphere. 2-TeCDwas prepared according to our previous work (20). Soluble RecombinantHuman TRAIL was purchased from R&D Systems (Minneapolis, MN).Sp1 reporter constructs were purchased from Clontech (Mountain View,

CA). Mithramycin A was purchased from Sigma Chemical Co (St. Louis,MO). The pan-caspase inhibitor z-VAD-fmk and the caspase-8 inhibitorZ-IETD-FMK were purchased from Alexis Corporation (San Diego, CA).

Cell viability assay

Cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2-5-diphenyltetrazolium bromide (MTT; Chemicon, Temecula, CA) colori-metric method. Following treatment of cells (5 � 103 per well) with variousconcentrations of 2-TeCD and/or TRAIL in 96-well plates for 24 h, 20 ll/well of MTT (5 mg/ml) solution was added. After a 4 h incubation at 37�C,the formazan crystals were solubilized with 150 ll/well dimethylsulfoxidefor 10 min. The absorbance values of the solution in each well were mea-sured at 570 nm using an ELx808 microplate reader (BioTek Instruments,Winooski, VT).

Fig. 1. 2-TeCD sensitizes human breast cancer cells to TRAIL-induced cytotoxicity in vitro. (A) The structural formula of 2-TeCD. (B) Cytotoxic effects of2-TeCD and/or TRAIL on normal and malignant human breast cell lines. Eight different human breast cell lines and human mammary epithelial cells(HMECs) (C) were treated with 2-TeCD for 30 min and then further treated with TRAIL for 24 h at the indicated concentrations. Cellular viability wasdetermined by MTT assay. Columns, average of three independent experiments; bars, standard deviation. �P , 0.05, ��P , 0.01 or ��P , 0.001 versusuntreated cells.

2-TeCD sensitizes TRAIL-induced apoptosis

155


Apoptosis assay

Cells (6 � 105) were seeded in six-well plates and the next day treated with 2-TeCD, TRAIL or combination of the two reagents. After 24 h, both adherentand floating cells were harvested using trypsinization, washed once with warmphosphate-buffered saline and then subjected to caspase activity assay andAnnexin V/PI staining. For caspase activity assay, cells were exposed toCaspase-Glo 3/7 and Caspase-Glo 8 assay kits (Promega Corporation, Madi-son, WI) for 30 min at room temperature, and luminescence was quantified ina plate-reading luminometer (Thermo Labsystems, Helsinki, Finland). Annex-in V assays were done using the Annexin V-FITC apoptosis detection kit(Clontech) per instructions of the manufacturer. As a standard, 10 000 cellsper treatment condition were analyzed by FACSCalibur (Becton Dickinson,Franklin lakes, NJ) using a Cell Quest Software.

Flow cytometry of DRs

Following 24 h incubation with 2-TeCD or 2-TeCD þ TRAIL, adherent cells(1 � 106) were stained with 200 ll phosphate-buffered saline containing satu-rating amounts of anti-DR4 or anti-DR5 antibody (R&D Systems) on ice for30 min. Then, cells were reacted with FITC-conjugated rabbit anti-goat IgG(Sigma, Beijing, China) on ice for 30 min. After washing with phosphate-buffered saline, the expressions of these DRs were analyzed by FACSCalibur(Becton Dickinson). A total of 10 000 events were analyzed for each treatment.

Reverse transcription–polymerase chain reaction analysis

Total RNA was isolated from cells treated with vehicle or various concentra-tions of 2-TeCD using the RNeasy Mini Kit (Qiagen, Valencia, CA). A semi-quantitative reverse transcription –polymerase chain reaction (PCR) methodwas used to assess DR4 and DR5 messenger RNA (mRNA) expression. Thefollowing primers were used: DR4 (forward: 5#-CCGCGGCCACACCCA-GAAAGT-3# and reverse: 5#-GTACATGGGAGGCAAGCAAACAAA-3#);DR5 (forward: 5#-GAGCTAAGTCCCTGCACCAC-3# and reverse: 5#-AAT-CACCGACCTTGACCATC-3#); and glyceraldehyde-3-phosphate dehydroge-nase primers, all purchased from Clontech. The PCR products were separatedby electrophoresis on a 2% agarose gel and visualized by ethidium bromidestaining.

Quantitative real-time PCR was performed to determine the mRNA expres-sion levels of DRs and IAPs. Primers for IAPs were designed as: c-IAP1(forward: 5#-CAGCCTGAGCAGCTTGCAA-3# and reverse: 5#-CAAGC-CACCATCACAACAAAA-3#); XIAP (forward: 5#-AGTGGTAGTCCTGT-TTCAGCATCA-3# and reverse: 5#-CCGCACGGTATCTCCTTCA-3#) andsurvivin (forward: 5#-TGCCTGGCAGCCCTTTC-3# and reverse: 5#-CC-TCCAAGAAGGGCCAGTTC-3#). For quantitative PCR, 1 ll of gene primerswith SYBR Green (Applied Biosystems, Foster City, CA) in 20 ll of reactionvolume was applied. Temperature cycling and real-time fluorescence measure-ment were done using an ABI prism 7300 Sequence Detection System (Ap-plied Biosystems). The relative expression level for each target gene mRNAwas calculated using the following formula:

�2�ðCTtarget�CTGAPDHÞ�� 100%,

where CT is the threshold cycle.

Small interfering RNA

The 25-nucleotide small interfering RNA (siRNA) duplexes were purchasedfrom Invitrogen and had the following sequences: DR5, UACAAUACCGAC-CUUGACCAUCCC and green fluorescent protein, AAGACCCGCGCCGSG-GUGAAG. For transfection, cells were seeded in 10 cm dishes at 30%confluency, and siRNA duplexes (200 nM) were introduced into the cells usingLipofectamineTM 2000 (Invitrogen) according to the manufacturer’s recom-mendations.

Total, cytosolic and nuclear proteins extraction

Adherent cells were collected after 24 h treatment and pretreated with 25 ll ofprotease inhibitor cocktail (Pierce, Rockford, IL). NE-PER Nuclear and Cy-toplasmic Extraction kit (Pierce) was used to extract nuclear and cytoplasmiccontents. Samples were concentrated with PEG 8000, and protein concentra-tions were estimated by using a Micro BCA kit (Pierce).

Western blot analysis

Samples (30 lg protein) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and then transferred to nitrocellulosepaper (Immobilon-NC; Millipore, Billerica, MA) soaked in a Tris (20 mM),glycine (150 mM) and methanol (20%) buffer at 55 V for 4 h. After washing,the blots were incubated overnight at 4�C with the following primary antibodies:mouse monoclonal anti-RelA (p65), anti-IjBa, anti-Sp1 and anti-poly(adeno-sine diphosphate-ribose) polymerase (PARP), anti-histone H4 (diluted 1:1000vol/vol; Santa Cruz Biotechnology, Santa Cruz, CA); anti-Akt, anti-phospho-Akt(p-Akt; Ser473), anti-caspase-3, anti-caspase-8, anti-TRAIL-R2, anti-TRAIL-R1, anti-c-IAP1, anti-XIAP, anti-survivin (1:2000 vol/vol; Stressgen, British

Columbia, Canada) and mouse monoclonal anti-b-actin, anti-a-tubulin(1:2000 vol/vol; Sigma). Following incubation, the secondary antibodies(diluted\ 1:2000 vol/vol; Sigma) were added for 1 h at room temperature. Pro-teins were visualized using an enhance chemiluminesence system (Santa Cruz)and captured on X-ray films.

Reporter plasmid construction

The genomic DNA of T47D was extracted using the DNeasy kit (Qiagen)according to the manufacturer’s protocol. The 5#-flanking fragment of DR5gene corresponding to the nucleotides �1152/�2 (numbering from the ATGsite) was amplified by PCR using the T47D genomic DNA as template and thefollowing primers: sense 5#-GAG CTCAGGAACAAACTCCAGACACG-3#and antisense 5#-CTCGAGGCGGTAGGGAACGCTCTTAT-3#. The PCRproduct was purified by 1% (wt/vol) agarose gel electrophoresis and subclonedinto the firefly luciferase-based pGL3-basic vector (Promega) between SacIand XhoI sites. A series of deletion mutants containing DR5 promoter weregenerated by PCR amplification using pGL3-basic/�1152 plasmid as templateand the following primers and subcloned into the pGL3-basic vector to gen-erate the reporter plasmid series of pGL3-basic/�949, pGL3-basic/�744,pGL3-basic/�552, pGL3-basic/�377, pGL3-basic/�270, pGL3-basic/�207and pGL3-basic/�144, respectively. Point mutations were introduced to thepGL3-5#�1152 by PCR-based site-directed mutagenesis technology using theprimers: 5#-GGATCTGATTCGCCAAGT TCC GAATGACGCC-3# and 5#-GGCGTCATTCGGAACTTGGCGAATCAGATC C-3# for Sp1-binding site1 and 5#-GAAAGTACAGCCGCGAAGTTCCAAGTCAGCCTG-3# and 5#-CAGGCTGACTTGGAACTTCGCGGCTGTACTTTC-3# for Sp1-binding site2. The sequences of all constructs were confirmed by nucleotide-sequencinganalysis.

Transfections and luciferase assays

The T47D cells cultured in 96-well plates were cotransfected with 5 ng of phRL-TK, which encodes Renilla luciferase as internal control (Promega) and 200 ngof the various reporter plasmids by using Lipofectamine 2000 (Invitrogen). After24 h of transfection, the cells were incubated with increasing doses of 2-TeCD ormedia (control) for another 24 h. The cells were lysed with protein lysis buffer(Promega). Both firefly and Renilla luciferase activities in the cell lysates werequantitated using Dual-Luciferase reporter assay system (Promega) according tothe manufacturer’s instructions. In case of NF-jB activity assay, MDA-MB-468and T47D cells were transiently cotransfected with an NF-jB reporter construct(pNF-jB) or a control reporter plasmid (pControl) (Panomics, Fremont, CA),together with a b-galactosidase reporter vector (Promega), which was used tonormalize NF-jB reporter gene activity.

In vivo tumor growth model

MDA-MB-468 cells (5 � 106) resuspended in 0.1 ml serum-free Dulbecco’smodified Eagle’s medium were subcutaneously injected into the right axilla ofthe 6-week-old female Balb/c nu/nu mice (National Academy of MedicalSciences). When the average size of tumors reached �100 mm3 (about 14days), animals were randomly separated into four groups (n 5 10 per group)to receive treatment of intraperitoneal injection of vehicle control (100 ll of0.9% NaCl), 2-TeCD (5 mg/kg/day–0.1 mg/day for 20 g mouse in a maximalvolume of 100 ll 0.9% NaCl), TRAIL (5 mg/kg/day) and the combination of 2-TeCD plus TRAIL (6 h later after 2-TeCD treatment). Mice were treated for 3weeks, and tumor size was measured with a caliper and calculated by thefollowing formula: (long axis � short axis2)/2. Physical parameters, includingbody weight, food and water intake, were studied throughout the treatment.

Tissue and blood analysis

For toxicology studies, tail blood samples (20 ll each time) were taken dailyfor 7 days (day 15th–21st) after the initiation of the treatment, and thereafter atleast once a week. These samples were analyzed using Advia 1200 automatedhemocytometry (Bayer Diagnostics, Puteaux, France) to give a hemoglobinlevel and white cell and platelet counts. At week 6 after last treatment, half ofthe randomly selected mice (n 5 5) from each group were killed under anes-thesia using avertin; blood samples were subsequently collected for the serumchemistry and toxicology studies. Tumor tissues were then immediately re-moved, fixed in paraformaldehyde at room temperature for 48 h and thenembedded in paraffin. For immunohistochemistry, dewaxed tissue sections(5.0 lm) were fixed and incubated with primary antibodies: proliferating cellnuclear antigen, Ki-67, p65-NF-jB and TRAIL-R2/DR5 (Santa Cruz). Thedetection of nuclei with fragmented DNA by terminal deoxyribonucleotidyltransferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL)was accomplished using an in situ cell death detection kit (POD; Roche Diag-nostics, Mannheim, Germany) according to the instructions of the manufac-turer. Tumor volume was measured in the remaining mice (n 5 5) for anadditional 4 weeks before the experiment was terminated.

T.Lin et al.

156


Statistical analysis

The mean and standard deviation were calculated for each experimental group.Differences between groups were analyzed by one- or two-way analysis ofvariance followed by Bonferroni’s multiple comparison tests using PRISMstatistical analysis software (GraphPad Software version 5.0). Significant dif-ferences among groups were calculated at P , 0.05.

Results

2-TeCD sensitizes resistant human breast cancer cells toTRAIL-induced cytotoxicity

We initially examined a panel of breast cancer cell lines (BT-474,MCF-7, MDA-MB-231, MDA-MB-453, MDA-MB-468, SKBr3,T47D and ZR75-1) for TRAIL sensitivity using different concentra-tions of recombinant human soluble TRAIL (Figure 1B). Consistentwith the previous studies (19), only MDA-MB-231 and ZR75-1 cellswere sensitive to TRAIL-induced cytotoxicity, whereas the rest of thecell lines were resistant. Next, we examined the cytotoxic effects of 2-TeCD alone or in combination with TRAIL in these cells. 2-TeCDalone induced a limited cell death (,10%) at concentrations up to10 lM. In contrast, cell viability was significantly reduced by com-bined treatment under conditions of a fixed TRAIL concentration andvaried 2-TeCD concentrations as well as the reverse in all cell lines.The tested cell lines are differing in several variables, such as expres-sion of oncogenes, presence of estrogen receptors and p53 status, thesensitization to TRAIL-induced cytotoxicity by 2-TeCD is thereforea general phenomenon of breast tumor cells. Finally, combined treat-ment with 2-TeCD and TRAIL of various concentrations had no sig-nificant effect on the viability of normal human mammary epithelialcells, indicating 2-TeCD did not abrogate the potential tumor selec-tivity of TRAIL (Figure 1C). Taken together, these results suggest that2-TeCD effectively sensitizes resistant breast cancer cells, but notuntransformed human mammary epithelial cells, to TRAIL-inducedcell death in vitro.

The sensitization of TRAIL-induced cytotoxicity by 2-TeCD isassociated with apoptosis

TRAIL-induced cancer cell death is mainly apoptotic. We next exam-ined whether cell death caused by TRAIL plus 2-TeCD was associ-ated with apoptosis. MDA-MB-468 or T47D cells were cultured with2-TeCD (10 lM) and/or TRAIL (200 ng/ml) for 24 h and then sub-jected to Annexin V/propidium iodide staining. Using the acceptedcriterion that apoptotic cells are Annexin V-positive/propidiumiodide-negative, we found that treated with TRAIL or 2-TeCD alonewas not able to induce apoptosis in either MDA-MB-468 or T47Dcells. However, significant increase of apoptotic cells were observedin both of the cancer cell lines, upon exposure to the combining of thetwo agents (Figure 2A). These results support our previous MTTanalysis, confirming that 2-TeCD is essential for the sensitization ofcancer cells to TRAIL-induced apoptosis. Finally, the apoptosis in-duced by the combined treatment of 2-TeCD and TRAIL werecompletely blocked when cells were preincubated with 100 lMz-VAD-fmk, a broad-spectrum caspase inhibitor, suggesting a caspase-dependent mechanism (Figure 2B).

Next, we analyzed the caspase-8 cleavage, a proximal event in theTRAIL-induced caspase cascade. As shown in Figure 2C, left, TRAILalone or 2-TeCD alone failed to induce detectable caspase-8 process-ing when compared with control groups. In contrast, cleaved-caspase-8 was clearly detected in both of the cancer cells when treated with2-TeCD plus TRAIL. The immunoblotting results were further sub-stantiated by caspase activity assay (Figure 2C, right). Compared withagent-alone group, the combination group significantly increased cas-pase-8 activity in both of the MDA-MB-468 and T47D cells. Collec-tively, these data indicate that treatment of resistant cancer cellswith the combination of 2-TeCD and TRAIL significantly elevatescaspase-8 activity.

The essential apoptosis executor caspase-3 and the caspase-3 sub-strate PARP were also analyzed to substantiate the above results.

Caspase-3 and PARP cleavages were barely observed in both of theTRAIL-resistant cells when cultured with TRAIL or 2-TeCD alone.By comparison, the cleavages were clearly detected in cells whentreated with 2-TeCD plus TRAIL (Figure 2C). However, the caspase-3 and PARP cleavages were abrogated when cells were incubatedwith Z-IETD-FMK (the caspase-8 inhibitor, data not shown), confirm-ing the involvement of caspase-8 activity in the 2-TeCD-enhancedsensitization.

DR5 upregulation is important for 2-TeCD-mediated sensitization ofbreast cancer cells to TRAIL-induced apoptosis

The activation of the caspase-8 in combined treatment suggests that 2-TeCD targets an early step in the TRAIL-induced apoptosis pathway.Therefore, we evaluated the effects of 2-TeCD on expression of var-ious proteins involved in TRAIL-initiated DISC components, includ-ing fas-associated death domain, death-associated protein 3, DcR1,DcR2 and critical DRs (DR4/DR5). Although 2-TeCD had minimaleffects on the expression levels of the vast majority of these proteins(Figure 3 and data not shown), it notably increased DR5 protein levelsin resistant breast cancer cells.

As shown in Figure 3A, 2-TeCD increased DR5 expression ina concentration-dependent manner. Note that the DR5 antibody usedhere recognized two DR5 splice variants, with approximate molecularweights of 43 kDa [DR5S(Short)] and 48 kDa [DR5L(Long)]. The in-crease in DR5S(Short) was more striking than DR5L(Long), as densitom-etry analysis indicated that 2-TeCD increased DR5S protein levels by�4.0- to �5.5-fold relative to untreated cells. In contrast, 2-TeCD didnot detectably alter expression levels of DR4, even at the highestconcentration (10 lM). Moreover, flow cytometric analysis of DRs,considered as a more sensitive method, was performed in both of thecell lines. Compared with increased expression of DR5, DR4 levelswere not affected by 2-TeCD (Figure 3B). Besides, the surface ex-pression of DR5 was also increased in response to cotreatment with2-TeCD plus TRAIL.

To determine whether 2-TeCD could regulate gene expression ofDR4 and DR5, we treated MDA-MB-468 and T47D cells with vary-ing concentrations of 2-TeCD for 24 h and thereafter analyzedthe expression of DRs transcripts by means of semiquantitative re-verse transcription–PCR and real-time PCR. 2-TeCD treatment dosedependently increased DR5 mRNA levels in both of the cell lines(Figure 3C). However, 2-TeCD has no significant effects on the expres-sion of DR4 mRNA, which are in agreement with the above results.

To clarify the functional significance of DR5 in the 2-TeCD-mediated sensitization of TRAIL-induced apoptosis, we assessedthe effects of DR5 knockdown by transfecting a siRNAs into cells.Through detection of caspase activation as well as quantification ofapoptotic cells, we found that siRNA-mediated suppression ofDR5 effectively blocked 2-TeCD-stimulated TRAIL-induced apoptosis(Figure 3D). Taken together, these data confirm that 2-TeCD-mediatedDR5 upregulation is essential for TRAIL sensitivity in resistant breastcancer cells.

Sp1 mediates 2-TeCD-induced DR5 upregulation

To further understand the transcriptional regulation of DR5 expres-sion, the luciferase reporter gene system was used to determine thespecific element responsible for the 2-TeCD-augmented DR5 expres-sion. The 2-TeCD-responsive element in the DR5 promoter activitywas evaluated by using a series of 5#-deletion mutants, respectively.As shown in Figure 4A, 2-TeCD significantly facilitated the luciferaseactivity in the pGL3-basic/�1152-transfected cells, indicating thatthere is at least one transcriptional element in the 5#-flanking regionof DR5 gene responsible for 2-TeCD-enhanced DR5 expression. The5#-deletion to �377 did not change the luciferase activity in the cellstreated with 2-TeCD, but further deletion to �207 reduced the lucif-erase activity significantly, suggesting that the responsible element for2-TeCD-enhanced DR5 expression might exist in the fragment span-ning from �377 to �2 in the DR5 promoter region. By more detaileddeletion analysis, we observed that the deletion of the �270/�207


157


fragment reduced some of the luciferase activity, however, similarratios (3-fold) of the luciferase activities were kept in the T47Dcells treated with or without 2-TeCD, suggesting that 2-TeCDcould still regulate DR5 gene expression in the cells with the deletionof �270/�207 in DR5 promoter. Further deletion demonstrated thatlack of the fragment of �207/�144 blocked DR5 promoter activitysignificantly, confirming that 2-TeCD-responsive element seemed lo-cated in a 63 bp region between �207 and �144, although we cannotrule out the possibility that other responsive regions might also exist.

The region of the DR5 promoter spanning nucleotides �207 to �144 contains two putative Sp1-binding sites (32), we thereforetested the possible involvement of Sp1 in 2-TeCD-mediated DR5upregulation. As shown in Figure 4B, treatment of T47D cells with2-TeCD increased Sp1-mediated transcriptional and translationalactivity in a dose-dependent manner. Furthermore, pretreatmentwith a specific inhibitor of Sp1, mithramycin A, dose dependentlyattenuated the 2-TeCD-mediated upregulation of both DR5 pro-moter activity and DR5 protein levels (Figure 4C). These results

Fig. 2. The sensitization of TRAIL-induced cytotoxicity by 2-TeCD is associated with apoptosis. (A) Cells were treated with 2-TeCD (10 lM) and/or TRAIL(200 ng/ml) for 24 h and then assayed for apoptosis by Annexin V (x-axis)/propidium iodide (PI) (y-axis) staining. (B) Apoptosis are mostly blocked by the caspaseinhibitor. Cells were incubated with 2-TeCD (10 lM) and TRAIL (200 ng/ml) for 24 h after 1 h pretreatment with (þ)/without (�) z-VAD-fmk (100 lM).Columns, mean percentage of apoptotic (Annexin Vþ/PI�) cells, bars, standard deviation. �P , 0.05 versus control group. ��P , 0.05 versus 2-TeCD þ TRAILgroup. (C) Caspase-8, caspase-3 and PARP were analyzed by immunoblotting (left) and activity assay (right). Columns, mean fold change of activity relative tocontrol group (n 5 3); bars, standard deviation. �P , 0.05 or ��P , 0.01 versus control group. C, T and Te denote control, TRAIL and 2-TeCD, respectively.

T.Lin et al.

158


Fig. 3. DR5 upregulation is important for 2-TeCD-mediated sensitization of breast cancer cells to TRAIL-induced apoptosis (A) 2-TeCD treatment increases DR5protein expression, but not DR4. Immunoblotting was used to quantify total DR4 and DR5 protein levels in whole cell lysates following 24 h of 2-TeCD treatment(top). Graph indicates relative DR5S and DR5L expression (normalized to both loading and vehicle controls), as measured by scanning densitometry (bottom).(B) 2-TeCD alone or 2-TeCD plus TRAIL increase the surface expression of DR5. MDA-MB-468 and T47D cells were incubated with 2-TeCD (10 lM) in thepresence or absence of 200 ng/ml TRAIL for 24 h, and the surface expression of DR5 and DR4 proteins was analyzed by flow cytometry. x-Axis, fluorescenceintensity; y-axis, relative number of cells. Black histograms, cells treated with 2-TeCD alone or 2-TeCD plus TRAIL; white histograms, control cells. (C) 2-TeCDtreatment increases the DR5 mRNA levels. Semiquantitative (top) and real-time reverse transcription–PCR (bottom) was used to quantify the DR4 and DR5mRNA levels following 24 h of 2-TeCD treatment. Fold increase of gene expression was calculated by dividing the normalized gene expression activity by that ofthe untreated control. (D) Suppression of DR5 expression by siRNA reduces 2-TeCD-stimulated TRAIL-induced apoptosis in breast cancer cells. MDA-MB-468


159


Fig. 4. Sp1 mediates 2-TeCD-induced DR5 upregulation. (A) Identification of the responsible element in DR5 promoter for the 2-TeCD-enhanced DR5expression. The T47D cells were transfected with the indicated luciferase reporter plasmids and subsequently incubated in the presence or absence of 2-TeCD(10 lM). The cells were lysed and the luciferase activities were analyzed. The panels show that the 5# fragments of DR5 promoter in the luciferase reporterplasmids were deleted gradually. Columns, means of triplicate experiments; bars, standard deviation. �P , 0.05 or ��P , 0.01. (B) Effect of 2-TeCD on thetranscriptional and translational activities of Sp1. Left, a reporter vector (pSp1-Luc) was transfected into T47D cells. Transfected cells were treated with 2-TeCDat the indicated concentrations for 24 h, and lysed and luciferase activity was measured. Columns, means of triplicate experiments; bars, standard deviation.�P , 0.05 compared with untreated cells. Right, Sp1 protein levels were measured by western blot with increasing concentrations of 2-TeCD. (C) Inhibition of Sp1by treatment with mithramycin A abolishes 2-TeCD-mediated increase in DR5 promoter activity and protein levels. Left, T47D cells were transfected withpDR5-207 promoter construct and further treated with 2-TeCD, together with or without mithramycin A at the indicated concentrations for 24 h. Cells were lysed,and luciferase activity was measured. Columns, means of triplicate experiments; bars, standard deviation. �P , 0.05 compared with untreated cells. Right,T47D cells were treated with 10 lM 2-TeCD, together with or without mithramycin A at the indicated concentrations for 24 h. Cell extracts were prepared forwestern blotting of DR5. (D) Luciferase reporter plasmid constructs containing Sp1 mutations were transfected into T47D cells. hTwenty-four hours aftertransfection, cells were cultured in the presence of 2-TeCD for another 24 h. Luciferase activity was measured using a dual luciferase reporter assay system.Firefly luciferase activity was normalized to Renilla luciferase activity.

and T47D cells were transfected with the control GFP siRNA or DR5 siRNA. hTwenty-four hours later, the cells were treated with TRAIL (200 ng/ml) and/or 2-TeCD (10 lM) for another 24 h treatment. Top, cytosolic fractions were then subjected to western blot analysis with anti-caspase-8, anti-capsase 3 and anti-PARP,respectively. Western bolts of DR5 were done to confirm the knockdown of DR5 by siRNA transfection. Bottom, represented the percentage of apoptotic (AnnexinVþ/PI�) cells that were evaluated by FACS. Columns, mean of three independent experiments; bars, standard deviation.

T.Lin et al.

160


confirm that Sp1 is critical factor promoting 2-TeCD-induced DR5upregulation.

We next examined which Sp1 site(s) in �207/�145 of DR5 pro-moter is responsible for 2-TeCD-induced DR5 upregulation. EachSp1 site was mutated alone or in combination with the other (dou-ble). Analysis of the luciferase activity revealed that the upstreamSp1 site located at �198 is required for 2-TeCD-induced DR5 pro-moter activity, whereas the downstream Sp1 site located at �149does not play a role (Figure 4D). This is supported by the fact thatwhen both Sp1 sites were mutated, the promoter activity was un-changed as compared with only the upstream (�198) Sp1 site beingmutated. Thus, it can be implied from these data that 2-TeCD in-duces an increase in Sp1 levels, which activates the DR5 promoter(primarily at the �198 site) and increases DR5 mRNA and proteinexpression on the surface of the cells, allowing them to respond toTRAIL stimulation.

2-TeCD inhibits TRAIL-induced NF-jB activation

NF-jB usually plays protective roles in apoptosis of various cells inresponse to TNF superfamily molecules (11,33). Previously, it hasbeen demonstrated that 2-TeCD can inhibit the NF-jB signaling path-way in human umbilical vein endothelial cells (31). We, therefore,sought to examine whether 2-TeCD might regulate NF-jB to thesensitization of TRAIL-mediated apoptosis. Figure 5A revealed thatthe nuclear levels of p65 were elevated in TRAIL-treated cancer cellswhen compared with control. However, 2-TeCD treatments signifi-cantly attenuated p65 expression in a dose-dependent manner andalmost abolished its nuclear translocation at 10 lM. Moreover, thefindings of the cytoplasmic extracts indicated that, 2-TeCD supple-mentation completely blocked TRAIL-induced degradation of IjBa,the NF-jB inhibitory subunit (34), in both of the cell lines. Hence, thelack of IjBa degradation with escalating doses of 2-TeCD can becorrelated with the minimal nuclear accumulation of p65. Consistentwith the above observations, the decrease in translocated NF-jBprotein levels correlated with decreased phosphorylation of Akt(Figure 5B).

To confirm that the increased sensitivity to TRAIL-mediated apo-ptosis was due to inhibition of p65, cells were transiently transfectedwith the pCMV4-p65 vector or the control vector (pCMV4-neo).Immunoblot analysis of whole cell extracts confirmed that MDA-MB-468-RelA cells and T47D-RelA cells expressed higher levels ofRelA as compared with corresponding neo-cells (Figure 5C). Then,we examined the effects of 2-TeCD in TRAIL-induced apoptosis.Consistent with the above results, 2-TeCD enhanced TRAIL-inducedapoptosis in both of the cell lines, however, the sensitization could begreatly reversed by ectopic expression of p65/RelA. This result in-dicated that 2-TeCD-mediated potentiation of TRAIL-induced apo-ptosis was partially through the inhibition of RelA-mediatedactivation of NF-jB.

To further validate that 2-TeCD could inhibit NF-jB pathway, weanalyzed the effects of 2-TeCD on NF-jB activation induced byTRAIL in cells using luciferase-based NF-jB reporter assay. Asshown in Figure 5D, 24 h TRAIL treatment resulted in �5-foldNF-jB activation compared with control group. However, at this timepoint, 2-TeCD inhibited TRAIL-induced NF-jB activation in a dose-dependent manner and reduced the basal activity of NF-jB in both ofthe cell lines at 10 lM.

As discussed above, NF-jB translocation contributes to theTRAIL-resistance in cancer cells via its ability to upregulate IAPsfamily (13). Hence, we evaluated the expression of mRNA and proteinexpression of these IAPs in cells treated with TRAIL in the absenceand presence of 2-TeCD. As shown in Figure 5E, TRAIL alone in-creased mRNA and protein expression levels of c-IAP1, XIAP as wellas survivin after 24 h incubation, whereas 2-TeCD cotreatment sig-nificantly reduced the three IAPs gene and protein expression even ata lower concentration (2.5 lM). This result indicates that 2-TeCDcould suppress expression of anti-apoptotic molecules that are tran-scriptionally regulated by NF-jB.

2-TeCD and TRAIL results in tumor growth inhibition and apoptosisin vivo

Based on the above findings, we next sought to evaluate whether2-TeCD in combination with TRAIL could inhibit tumor growthin vivo. MDA-MB-468 tumor-bearing mice were treated with vehicle,2-TeCD alone (5 mg/kg/day), TRAIL alone (5 mg/kg/day) or 2-TeCDplus TRAIL for 3 weeks. As shown in Figure 6A, neither TRAIL nor2-TeCD alone suppressed tumor growth compared with vehicle-treated mice. In contrast, the combination of 2-TeCD and TRAILexhibited significant inhibitory effects on xenografts compared withagent alone or vehicle control groups. At week 6 after last treatment,half of the randomly selected mice (n 5 5) from each group werekilled. Tumor volume was measured in the remaining mice (n 5 5) foran additional 4 weeks before the experiment was terminated. Tumorsin the 2-TeCD and TRAIL treatment group resumed growth, indicat-ing persistent need for 2-TeCD and TRAIL to maintain suppression oftumor growth in this xenograft model (data not shown).

We next evaluated the effects of agent treatment on tumor cellproliferation and apoptosis (Figure 6B). Very few apoptotic cells werefound in the vascularized tumor tissues derived from control mice,indicating an environment to support the establishment of xenografts.Scoring immunohistochemical examination of tumor tissues revealedthat the combination of 2-TeCD and TRAIL effectively inhibited cellproliferation (proliferating cell nuclear antigen and Ki-67 staining)and induced apoptosis (TUNEL staining). On the contrary, neither ofthe single agents had significant effects on tumor cell proliferation orapoptosis, compared with control group.

2-TeCD in combination with TRAIL regulates the expression ofp65-NF-jB and DR5 in vivo

We next sought to confirm the effects of 2-TeCD in combination withTRAIL on p65-NF-jB and DR5 expression levels in vivo. Xenograftsections from each of the treatment groups were immunostained withanti-p65 and anti-DR5 antibodies, respectively. As shown in Figure6C, TRAIL alone had no significant effect on the expressions ofp65-NF-jB and DR5. By comparison, treatment of mice with a com-bination of 2-TeCD and TRAIL significantly showed more expressionof DR5 and less expression of p65-NF-jB proteins than that of micetreated with vehicles or TRAIL, which were all consistent within vitro results.

TRAIL in combination with low dose of 2-TeCD is well tolerated innude mice

In vivo toxicity in nude mice associated with combination treatment of2-TeCD and TRAIL was recorded during the entire experiments,using several different examinations, including animal weight, ap-pearance and behavior, as well as histopathological analysis (hema-toxylin and eosin staining). Animal weight changes were recordedover 3 weeks of combination 2-TeCD and TRAIL treatment andduring the 4 weeks follow-up. On the 15th day after first injectionof 2-TeCD þ TRAIL, the animal activities and food intake decreased(supplementary Figure 1 is available at Carcinogenesis Online), butthere were no over reactions to the injection. Hematological toxicityevaluation showed that there were no statistically significant differ-ences (P . 0.05) in the cotreatment group with the comparison of thatin the control group (data not shown). Hence, the reduction in bodyweight on the day 17th might be due to the decrease of animal activ-ities and food intake. Finally, by day 28th, the body weight hadreturned and surpassed baseline (Figure 6D). After termination oftreatment, no significant weight changes were noted during the 4weeks follow up (data not shown). Besides, animal appearance andbehavior (fur/lethargy/ruffled) were also observed over the entire ex-periment and no abnormal phenomenon was noticed under synergistictreatment. Serum chemistries, including electrolytes (sodium, potas-sium), glucose, liver transaminases (alanine aminotransferase, aspar-tate aminotransferase), alkaline phosphatase and markers of renalfunction (blood urea nitrogen, creatinine), were further analyzed in


161


supplementary Figure 1

tumor-bearing mice after the experiment termination. No significantdifferences were noted when comparing 2-TeCD treatment groups(2-TeCD, 2-TeCD þ TRAIL) with vehicle control group (Figure 6D).Finally, hematoxylin and eosin staining was also used to ascertain

toxicity to mouse tissues [brain, liver, spleen, lung and kidney (n 55 per group)] at the end of treatment (data not shown). The resultssuggested that tissues were all maintained normal morphology with-out significant differences among four treatment groups.

Fig. 5. 2-TeCD inhibits TRAIL-induced NF-jB activation. (A) Effects of TRAIL combined with 2-TeCD on p65 and IjBa expression. Cells were treated with200 ng/ml of TRAIL for the indicated time (top). Then, cells were treated with 200 ng/ml TRAIL in combination with various concentration of 2-TeCD (2.5–10 lM) for 24 h (bottom). Nuclear proteins and cytosolic fractions were extracted and assayed for p65 and IjBa by western blotting. H4 and b-actin were used asinternal controls in nuclear and cytosol specimen, respectively. (B) Effects of 2-TeCD on TRAIL-induced Akt phosphorylation. The cells were then treated with2-TeCD (10 lM) and/or TRAIL (200 ng/ml) for the indicated time intervals. Cytosolic fractions were extracted and assayed for Akt and p-Akt by western blotting.(C) RelA (p65) overexpression decreases 2-TeCD-mediated sensitization of breast cancer cells to TRAIL. MDA-MB-468 and T47D cells were transientlytransfected with neo or p65/RelA vectors, respectively. The cells were then treated with 2-TeCD (10 lM) and/or TRAIL (200 ng/ml) for 24 h. After incubation,western blots were performed to determine the overexpression of p65/RelA (top). The apoptotic cells (Annexin Vþ/PI�) in each treatment condition wereevaluated by FACS (bottom). (D) 2-TeCD inhibits TRAIL-induced NF-jB activation in NF-jB luciferase reporter assay. MDA-MB-468 and T47D weretransiently cotransfected with pNF-jB or pControl together with b-galactosidase plasmid and then treated with TRAIL for the indicated time (left).Transfected cells were treated with 200 ng/ml TRAIL in the presence of 2-TeCD (2.5–10 lM) for 24 h (right). Data are expressed as fold increase relative tountreated cells. Columns, mean of three independent experiments; bars, standard deviation. ��P , 0.01 versus untreated cells. (E) Treatment with 2-TeCDdecreases gene and protein expression levels of NF-jB targets IAPs in breast cancer cells. Cells were treated with 200 ng/ml TRAIL in combination with variousconcentration of 2-TeCD (2.5–10 lM) for 24 h. After incubation, mRNA and protein levels of IAPs (IAP-1, XIAP and survivin) were analyzed using real-timePCR (top) and western blot (bottom), respectively.

T.Lin et al.

162


The safety of TRAIL is well documented by preclinical animalstudies (35). However, much less is known about the in vivo safetyof 2-TeCD. To determine whether higher 2-TeCD doses (.5 mg/kg/day) are safe in vivo, tumor-bearing mice were treated for 21 consec-utive days with intraperitoneal injections of 10.0 mg/kg 2-TeCD,15.0 mg/kg 2-TeCD and 20.0 mg/kg 2-TeCD or vehicle (n 5 10);then, on day 22, blood and tissue samples were collected for toxicalanalysis. 2-TeCD delivered at 20.0 mg/kg/day for 21 days was welltolerated in mice; only slight anemia (red blood cell counts belownormal) was observed (supplementary Table 1 is available at Carci-nogenesis Online), without changes in white blood cell or plateletcounts, serum chemistries (supplementary Table 2 is available at Car-cinogenesis Online) or tissue histology (supplementary Figure 2 isavailable at Carcinogenesis Online). No changes in gross appearance

or behavior (ruffled fur or lethargy) were noted over the 21 days ofthese dosing. When taken together with the aforementioned experi-ments using 2-TeCD at 5 mg/kg/day, these data suggest that 2-TeCD,at doses of 5–20 mg/kg daily, is well tolerated in mice. No inhibitionof MDA-MB-468 xenograft growth was seen in animals treatedwith 2-TeCD (10.0–20.0 mg/kg/day) compared with vehicle con-trol-treated mice (data not shown), thus emphasizing the importanceof including TRAIL in combination with 2-TeCD to achieve tumorgrowth suppression in this breast cancer model.

Discussion

The advantage of the clinical application of TRAIL is that normalcells are resistant to its cytotoxic activity and thus it specifically

Fig. 5. Continue.


163


Fig. 6. Effects of 2-TeCD and TRAIL on MDA-MB-468 xenografts in nude mice. (A) MDA-MB-468 cells (5 � 106/100 ll) were injected subcutaneously into theright axilla of each mouse. The animals were randomly divided into four groups (n 5 10) when tumor volume reached 100 mm3 (day 14) and received dailyintraperitoneal injection of vehicle control (0.9% NaCl), 2-TeCD (5 mg/kg), TRAIL (5 mg/kg) or combination of 2-TeCD and TRAIL (6 h after 2-TeCD treatment)for 3 weeks. At week 6 after last treatment, half of the randomly selected mice (n 5 5) from each group were killed. Tumor volumes were measured in theremaining mice (n 5 5) for an additional 4 weeks before the experiment was terminated. Points, mean of 10 mice in each group; bars, standard deviation.�P , 0.05 versus control group. Right, representative images of tumors in TRAIL alone group and 2-TeCD þ TRAIL groups. (B) Effects of 2-TeCD and/orTRAIL on cell apoptosis and proliferation. Immunohistochemistry was performed to measure apoptosis (TUNEL assay) and expression of PCNA and Ki67 intumor tissues derived from control and treated mice on week 6 (top). Quantification of TUNEL, PCNA and Ki-67-positive tumor cells (bottom). Values denoted by� are significantly different from respective control groups. (C) Effects of 2-TeCD and/or TRAIL on p65-NFjB and DR5 and quantification of positive tumor cells.Values denoted by � are significantly different from their respective control groups. Tumor slides of different treatment groups were visualized under microscope,and the positive cells were quantified. Original magnification �40. (D) Toxicology analysis in tumor-bearing mice. Top, animal weights were recorded during thethree weeks of treatment. Points, mean (n 5 10/group); bars, standard deviation. In the combination of 2-TeCD and TRAIL treatment group, statisticallysignificant differences (�P , 0.05) were found on days 17 when compared with baseline weight. Bottom, blood chemistry variables were analyzed aftertermination of treatment, including: sodium (Naþ), potassium (Kþ), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase(ALP), blood urea nitrogen (BUN), creatinine (Crea) and glucose (GLU). Columns, mean (n 5 10/group); bars, standard deviation. (E) Working model of 2-TeCDin sensitizing TRAIL-induced apoptosis. Induction of DR5 is responsible for sensitization of breast cancer cells to TRAIL-induced apoptosis by 2-TeCD. Inaddition, 2-TeCD suppresses TRAIL-induced NF-jB activation by preventing IjBa degradation and RelA (p65) nuclear translocation. As a consequence,blockade of NF-jB–IAPs signaling by 2-TeCD abolishes counteraction of pro-survival factors on TRAIL-mediated apoptosis.

T.Lin et al.

164


targets cancer cells. However, TRAIL will probably not be viable asa single agent since the majority of tumor cells are resistance toTRAIL. The combination therapy (chemotherapy or radiation) istherefore essential for the use of TRAIL against refractory tumors.In this report, we found that 2-TeCD, an organotellurium compound,was capable of sensitizing highly resistant breast cancer cells toTRAIL-mediated apoptosis both in vitro and in vivo.

The sensitization of tumor cells to TRAIL by 2-TeCD was associ-ated with the downregulation of antiapoptotic protein NF-jB andupregulation of DR5 (summarized in Figure 6E). We speculate thatboth of these mechanisms affect TRAIL sensitivity in breast cancercells. NF-jB normally resides in the cytoplasm as an inactivated formin a complex with IjBa. Once triggered by different pathogenic stim-uli (chemical agents or cytokine), IjBa started to degrade, and NF-jBcould translocate into the nucleus to upregulate a number of genesinvolved in survival responses (34). To date, numerous evidencessuggest that NF-jB is a negative regulator of TRAIL-mediated cancerapoptosis and a functionally important target for TRAIL-sensitizingactions (11,36). In the present study, we clearly demonstrated that 2-TeCD dose dependently inhibited the nuclear translocation of p65-NF-jB and stabilized its physiological inhibitor IjBa. This is consistentwith our earlier observation that 2-TeCD was able to suppress the trans-

location of NF-jB that was stimulated by TNFa (31). According to ourknowledge, both the GPx-like ability and TrxR inhibitory activity of 2-TeCD would be possible explanations for these results. GPx and itsmimics have been demonstrated to inhibit NF-jB translocation by in-creasing the IjBa half-life, and thus exhibited a significant role inpreventing cancer cell growth (37,38). Furthermore, other evidencesalso pointed out that GPx overexpression could specifically impairedNF-jB activation, p65 phosphorylation and its nuclear translocationthrough a redox-regulated mechanism (39,40). In addition, 2-TeCDcould also downregulate NF-jB by inhibiting the activity of TrxR,which have been demonstrated to activate NF-jB through directly orindirectly pathways (41,42). Taken together, our findings indicate thatthe TRAIL-sensitizing effects of 2-TeCD are mediated at least in part byNF-jB downregulation, which are also strongly supported by critical invivo proofs.

DRs TRAIL-R1/DR4 and TRAIL-R2/DR5 are selectively expressedin cancer cells and thus offer an advantage for TRAIL-targetedtherapy and prevention. In present study, it is clearly evident that2-TeCD caused upregulation of DR5 in a dose-dependent manner,thus creating a more TRAIL-sensitive environment. Since p53, ROS,CHOP, Yin Yang 1 and Sp1 (43–46) have been implicated in regulat-ing DR5 expression, we examined their possible involvement in

Fig. 6. Continue.


165


2-TeCD-induced DR5 upregulation. DR5 was significantly upregu-lated in not only breast carcinoma cells with wild-type p53 (e.g.MCF-7 and ZR75-1) and but also those with mutant p53 (e.g.MDA-MB-468, SkBr3 and T47D), suggesting that p53 is not criticalfor 2-TeCD-induced DR5 upregulation in breast carcinoma cells (datanot shown). Furthermore, pretreatment with the antioxidant, N-acetylcysteine, did not reduce the 2-TeCD induced DR5 upregulation,suggesting that 2-TeCD-induced DR5 upregulation is independent ofROS (data not shown). Therefore, we tested the possible involvementof transcriptional factors in 2-TeCD-mediated DR5 upregulation. Byreporter gene analysis, we initially identified �207/�145 region ofthe DR5 promoter, which appeared to be important in the regulation ofDR5/TRAIL promoter activity by 2-TeCD. This 5#-flanking sequencecontains two putative Sp1-binding sites, suggesting that Sp1 mightlead to the augment of DR5 transcriptional activity. We then, however,developed a number of lines of evidence clearly demonstrating theinvolvement of the Sp1 transcription factor in 2-TeCD-induced DR5upregulation, including the findings that: (i) 2-TeCD treatment dosedependently increased the Sp1 transcriptional and translational activ-ity in breast cancer cells, (ii) inhibition of Sp1 activity by mithramycinA abolished 2-TeCD-induced DR5 upregulation, and (iii) 2-TeCD-induced activation of DR5 promoter activity was abrogated by themutations in Sp1-binding sites (primarily at the �198 site) of the DR5promoter. Our data is consistent with previous studies indicating thatbinding of Sp1 in the promoter regions tightly regulates DR5 transcrip-tion in a variety of cancer cells, and the required specific Sp1 site wasmainly found at �207/�144 to the transcription start site (45,47–50).However, how does 2-TeCD affect the signaling pathway eventuallyleading to the augment of Sp1 transcriptional and translationalactivity remains an area of intense research.

The encouraging synergistic effects in vitro promoted us to evaluateantitumor activity of 2-TeCD plus TRAIL in vivo. Our result sug-gested that 2-TeCD significantly sensitized MDA-MB-468 xenograftsto undergo TRAIL-mediated apoptosis by inducing apoptosis(TUNEL staining) and inhibiting proliferation (proliferating cell nu-clear antigen and Ki67 staining). Moreover, upregulation of DR5 anddownregulation of NF-jB by the dual treatment in xenografted tumorswere consistent with the in vitro findings. Of note, this report providesthe first preclinical data concerning the in vivo efficacy and safety oftellurium compound in combination of TRAIL. At doses resulting insignificant suppression of tumor xenograft growth, the combination of2-TeCD and TRAIL was well tolerated in mice, without apparenttoxicity to normal tissues.

In summary, here we have highlighted a novel function of 2-TeCD:sensitizes human cancer cells to TRAIL-induced apoptosis via upre-gulation of DR5 and downregulation of NF-jB. Our studies providestrong clinical evidence that 2-TeCD in combination of TRAIL can beused for the treatment of breast cancer.

Supplementary material

Supplementary Figures 1 and 2 and Tables 1 and 2 can be found athttp://carcin.oxfordjournals.org/

Funding

Natural Science Foundation of China (20534030, 20874036); NSFCfor Outstanding Younger Scientist (20725415); National BasicResearch Program (2007CB808006).

Acknowledgements

Conflict of Interest Statement: None declared.

References

1.Walczak,H. et al. (1999) Tumoricidal activity of tumor necrosis factor-

related apoptosis-inducing ligand in vivo. Nat. Med., 5, 157–163.

2.Pan,G. et al. (1997) The receptor for the cytotoxic ligand TRAIL. Science,276, 111–113.

3.Walczak,H. et al. (1997) TRAIL-R2: a novel apoptosis-mediating receptorfor TRAIL. EMBO J., 16, 5386–5397.

4.Degli-Esposti,M.A. et al. (1997) Cloning and characterization of TRAIL-R3, a novel member of the emerging TRAIL receptor family. J. Exp. Med.,186, 1165–1170.

5.Degli-Esposti,M.A. et al. (1997) The novel receptor TRAIL-R4 inducesNF-jB and protects against TRAIL-mediated apoptosis, yet retains an in-complete death domain. Immunity, 7, 813–820.

6.Emery,J.G. et al. (1998) Osteoprotegerin is a receptor for the cytotoxicligand TRAIL. J. Biol. Chem., 273, 14363–14367.

7.Kischkel,F.C. et al. (1995) Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) withthe receptor. EMBO J, 14, 5579–5588.

8.Kischkel,F.C. et al. (2000) Apo2L/TRAIL-dependent recruitment of endoge-nous FADD and caspase-8 to death receptors 4 and 5. Immunity, 12, 611–620.

9.Boatright,K.M. et al. (2003) A unified model for apical caspase activation.Mol. Cell., 11, 529–541.

10.Schneider,P. et al. (1997) TRAIL receptors 1 (DR4) and 2 (DR5) signalFADD-dependent apoptosis and activate NF-kappaB. Immunity, 7, 831–836.

11.Keane,M.M. et al. (2000) Inhibition of NF-kappaB activity enhancesTRAIL mediated apoptosis in breast cancer cell lines. Breast CancerRes. Treat., 64, 211–219.

12.Kreuz,S. et al. (2001) NF-kappaB inducers upregulate cFLIP,a cycloheximide-sensitive inhibitor of death receptor signaling. Mol.Cell. Biol., 21, 3964–3973.

13.Wang,C.Y. et al. (1998) NF-kappaB antiapoptosis: induction of TRAF1 andTRAF2 and c-IAP1 and c-IAP2 to suppress caspase-8 activation. Science,281, 1680–1683.

14.Rahman,M. et al. (2009) The TRAIL to targeted therapy of breast cancer.Adv. Cancer Res., 103, 43–73.

15.Zhang,Y. et al. (2008) TRAIL resistance of breast cancer cells is associatedwith constitutive endocytosis of death receptors 4 and 5. Mol. Cancer Res.,6, 1861–1871.

16.Palacios,C. et al. (2006) Flavopiridol induces cellular FLICE-inhibitoryprotein degradation by the proteasome and promotes TRAIL-induced earlysignaling and apoptosis in breast tumor cells. Cancer Res., 66, 8858–8869.

17.Fulda,S. et al. (2002) Inhibition of TRAIL-induced apoptosis by Bcl-2overexpression. Oncogene, 21, 2283–2294.

18.Keane,M.M. et al. (1999) Chemotherapy augments TRAIL-induced apo-ptosis in breast cell lines. Cancer Res., 59, 734–741.

19.Singh,T.R. et al. (2003) Synergistic interactions of chemotherapeutic drugsand tumor necrosis factor-related apoptosis-inducing ligand/Apo-2 ligandon apoptosis and on regression of breast carcinoma in vivo. Cancer Res., 63,5390–5400.

20.Ren,X. et al. (2002) A novel cyclodextrin-derived tellurium compound withglutathione peroxidase activity. Chembiochem, 3, 356–363.

21.McNaughton,M. et al. (2004) Cyclodextrin-derived diorganyl tellurides asglutathione peroxidase mimics and inhibitors of thioredoxin reductase andcancer cell growth. J. Med. Chem., 47, 233–239.

22.Sun,Y. et al. (2005) The molecular mechanism of protecting cells againstoxidative stress by 2-selenium-bridged beta-cyclodextrin with glutathioneperoxidase activity. Biochim. Biophys. Acta., 1743, 199–204.

23.Luo,G.M. et al. (2003) Towards more efficient glutathione peroxidasemimics: substrate recognition and catalytic group assembly. Curr. Med.Chem., 10, 1151–1183.

24.Sies,H. (1993) Ebselen, a selenoorganic compound as glutathione peroxi-dase mimic. Free Radic. Biol. Med., 14, 313–323.

25.Dong,Z.Y. et al. (2006) Cyclodextrin-derived mimic of glutathione perox-idase exhibiting enzymatic specificity and high catalytic efficiency. Chem-istry, 12, 3575–3579.

26.Dong,Z. et al. (2004) Aryl thiol substrate 3-carboxy-4-nitrobenzenethiolstrongly stimulating thiol peroxidase activity of glutathione peroxidasemimic 2,2’-ditellurobis(2-deoxy-beta-cyclodextrin). J. Am. Chem. Soc.,126, 16395–16404.

27.Engman,L. et al. (2000) Water-soluble organotellurium compounds inhibitthioredoxin reductase and the growth of human cancer cells. AnticancerDrug Des., 15, 323–330.

28.Urig,S. et al. (2006) On the potential of thioredoxin reductase inhibitors forcancer therapy. Semin. Cancer Biol., 16, 452–465.

29.Nogueira,C.W. et al. (2004) Organoselenium and organotellurium com-pounds: toxicology and pharmacology. Chem. Rev., 104, 6255–6285.

30.Pennington,J.D. et al. (2007) Thioredoxin and thioredoxin reductase asredox-sensitive molecular targets for cancer therapy. Curr. Pharm. Des.,13, 3368–3377.

T.Lin et al.

166


Supplementary Figures 1

Tables 1

http://carcin.oxfordjournals.org/

31.Wang,K. et al. (2009) Effect of 2-TeCD on the expression of adhesionmolecules in human umbilical vein endothelial cells under the stimulationof tumor necrosis factor-alpha. Int. Immunopharmacol., 9, 1087–1091.

32.Yoshida,T. et al. (2001) Promoter structure and transcription initiation sitesof the human death receptor 5/TRAIL-R2 gene. FEBS Lett., 507, 381–385.

33.Aggarwal,B.B. (2000) Signalling pathways of the TNF superfamily: adouble-edged sword. Nat. Rev. Immunol., 3, 745–756.

34.Ghosh,S. et al. (1990) Activation in vitro of NF-kappa B by phosphoryla-tion of its inhibitor I kappa B. Nature, 344, 678–682.

35.Kelley,S.K. et al. (2001) Preclinical studies to predict the disposition ofApo2L/tumor necrosis factor-related apoptosis-inducing ligand in humans:characterization of in vivo efficacy, pharmacokinetics, and safety. J. Phar-macol. Exp. Ther., 299, 31–38.

36.Oya,M. et al. (2001) Constitutive activation of nuclear factor-kappaB pre-vents TRAIL-induced apoptosis in renal cancer cells. Oncogene, 20, 3888–3896.

37.Li,S. et al. (2000) The role of cellular glutathione peroxidase redox regu-lation in the suppression of tumor cell growth by manganese superoxidedismutase. Cancer Res., 60, 3927–3939.

38.Sharma,V. et al. (2008) Ebselen sensitizes glioblastoma cells to TumorNecrosis Factor (TNFalpha)-induced apoptosis through two distinct path-ways involving NF-kappaB downregulation and Fas-mediated formation ofdeath inducing signaling complex. Int. J. Cancer., 123, 2204–2212.

39.Kretz-Remy,C. et al. (1996) Inhibition of IkB-a phosphorylation and deg-radation and subsequent NF-kB activation by glutathione peroxidase over-expression. J. Cell Biol., 133, 1083–1093.

40.Li,Q. et al. (2001) GPx-1 gene delivery modulates NFkappaB activationfollowing diverse environmental injuries through a specific subunit of theIKK complex. Antioxid. Redox Signal., 3, 415–432.

41.Das,K.C. (2001) c-Jun NH2-terminal kinase-mediated redox-dependentdegradation of IkappaB: role of thioredoxin in NF-kappaB activation.J. Biol. Chem., 276, 4662–4670.

42.Hayashi,T. et al. (1993) Oxidoreductive regulation of nuclear factor kappaB. Involvement of a cellular reducing catalyst thioredoxin. J. Biol. Chem.,268, 11380–11388.

43.Wu,G.S. et al. (1997) KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene. Nat. Genet., 17, 141–143.

44.Su,R.Y. et al. (2008) 15-deoxy-Delta12,14-prostaglandin J2 up-regulatesdeath receptor 5 gene expression in HCT116 cells: involvement of reactiveoxygen species and C/EBP homologous transcription factor gene transcrip-tion. Mol. Cancer Ther., 7, 3429–3440.

45.Kim,Y.H. et al. (2004) Sodium butyrate sensitizes TRAIL-mediated apo-ptosis by induction of transcription from the DR5 gene promoter throughSp1 sites in colon cancer cells. Carcinogenesis, 25, 1813–1820.

46.Baritaki,S. et al. (2008) Inhibition of Yin Yang 1- dependent repressoractivity of DR5 transcription and expression by the novel proteasome in-hibitor NPI-0052 contributes to its TRAIL enhanced apoptosis in cancercells. J. Immunol., 180, 6199–6210.

47.Higuchi,H. et al. (2004) Bile acids up-regulate death receptor 5/TRAIL-receptor 2 expression via a c-Jun N-terminal kinase-dependent pathwayinvolving Sp1. J. Biol. Chem., 279, 51–60.

48.VanOosten,R.L. et al. (2005) Histone deacetylase inhibitors modulate renalcell carcinoma sensitivity to TRAIL/Apo-2L-induced apoptosis by enhanc-ing TRAIL-R2 expression. Cancer Biol. Ther., 4, 1104–1112.

49.Sun,M. et al. (2008) Sp1 is involved in 8-chloro-adenosine-upregulateddeath receptor 5 expression in human hepatoma cells. Oncol. Rep., 19,177–185.

50.Kim,J.Y. et al. (2008) Quercetin sensitizes human hepatoma cells toTRAIL-induced apoptosis via Sp1-mediated DR5 up-regulation andproteasome-mediated c-FLIPS down-regulation. J. Cell. Biochem., 105,1386–1398.

Received May 9, 2010; revised November 1, 2010;accepted November 7, 2010


167


Documents

Sensitization of TRAIL-resistant breast cancer cells by 2-selenium