A Functional, New Short Isoform of Death Receptor 4 in Ewing's … · A Functional, New Short...

Cell Cycle, Cell Death, and Senescence

A Functional, New Short Isoform of Death Receptor 4in Ewing's Sarcoma Cell Lines May be Involved in TRAILSensitivity/Resistance Mechanisms

Ga€elle Picarda1,2, Sylvanie Surget3, Romain Guiho1,2, St�ephane T�eletch�ea1,2, Martine Berreur1,2,Franck Tirode4,5, Catherine Pellat-Deceunynck3, Dominique Heymann1,2, Val�erie Trichet1,2, andFrancoise R�edini1,2

AbstractEwing's sarcoma (ES) is a high-grade neoplasm arising in bones of children and adolescents. Survival rate

decreases from greater than 50% to only 20% after 5 years for patients not responding to treatment or presentingmetastases at diagnosis. TRAIL, which has strong antitumoral activity, is a promising therapeutic candidate. Toaddress TRAIL sensitivity, 7 human ES cell lines were used. Cell viability experiments [30[1-(phenylaminocarbo-nyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro-)benzene sulfonic acid hydrate (XTT) assay] showed that 4 of the 7ES cell lines were resistant to TRAIL. Western blotting and flow cytometry analyses revealed that DR5 wasuniformly expressed by all ES cell lines, whereas DR4 levels were higher in sensitive cell lines. In TRAIL-sensitiveTC-71 cells, knockdown of TNFRSF10A/DR4 by short hairpin RNA (shRNA) was associated with a loss ofsensitivity to TRAIL, in spite of DR5 presence. Interestingly, we identified a new transcript variant that results froman alternative splicing and encodes a 310–amino acid protein which corresponds to the 468 aa of DR4 originalisoform but truncated of aa 11 to 168 within the extracellular TRAIL-binding domain. According to modelingstudies, the contact of this newDR4 isoform (bDR4) with TRAIL seemed largely preserved. The overexpression ofbDR4 in a TRAIL-resistant cell line restored TRAIL sensitivity. TRAIL resensitization was also observed afterc-FLIP knockdown by shRNA in two TRAIL-resistant cell lines, as shown by XTT assay and caspase-3 assay. Theresults presented in this study showed thatDR4, both as the complete formor as its new short isoform, is involved inTRAIL sensitivity in ES. Mol Cancer Res; 10(3); 336–46. �2012 AACR.

IntroductionBone sarcomas are rare malignancies diagnosed in fewer

than 700 individuals per year in the United States. Ewing'ssarcoma (ES) is the second most frequent primary bonetumors in children and adolescents. It is a high-gradeneoplasm, representing 2% of childhood cancers with apeak incidence at age 15 (1–3), characterized by a rapidtumor growth and extensive bone destruction that can resultin bone pain and pathologic fracture (4). Ewing's tumorsshow a typical chromosomal translocation inmore than 90%of cases linking the EWS gene on chromosome 22q12 to a

member of the ETS transcription gene family, most com-monly to FLI-1 on 11q24 (5,6), leading to an aberranttranscription factor that promotes tumorigenicity (7–9).Current treatment consists of several rounds of multidrugcytostatic therapy combined with surgery and radiotherapyfor local control.With the use of alkylating agents, long-termsurvival can be achieved in more than 50% of patients withlocalized disease, whereas patients with clinically detectablemetastases at diagnosis or who are not responding to therapyhave a significantly poorer prognosis, with long-term survivaldecreasing to 15% to 25%, especially for patients with bonemetastases (10). As the survival rates for metastatic andrelapsed osteosarcoma or ES patients have not changed inthe past 30 years, despite the use of intensive chemotherapy(11), new therapeutic approaches are therefore needed.Among new strategies, TRAIL or APO-2L is a promising

candidate. TRAIL is a member of the TNF superfamily thatexerts a strong antitumor activity in a wide range of cancercell lines, although being minimally cytotoxic to mostnormal cells and tissues (12). TRAIL binds 5 knownreceptors: 2 death receptors acting as activating receptors[DR4 (TRAIL-R1) and DR5 (TRAIL-R2)] and three decoyreceptors DcR1 (TRAIL-R4), DcR2 (TRAIL-R5) andsecreted osteoprotegerin (OPG, a member of the TNF

Authors' Affiliations: 1INSERM, Equipe labellis�ee LIGUE 2012 UMR 957;2Universit�e de Nantes, Nantes Atlantique Universit�es, Laboratoire dePhysiopathologie de la R�esorption Osseuse et Th�erapie des TumeursOsseuses Primitives, EA3822; 3INSERM UMR 892, IRT-UN, Universit�e deNantes, Nantes; 4Institut Curie, Paris; and 5INSERM, U830, Unit�e deG�en�etique et Biologie des Cancers, Paris, France

Corresponding Author: Francoise R�edini, UMR_S957, Laboratoire dephysiopathologie de la R�esorption Osseuse et Th�erapie des TumeursOsseuses Primitives, EA3822, Facult�e de M�edecine, 1 rue Gaston Veil,44035 Nantes cedex 1, France. Phone: 33-0-2-40-41-29-93; Fax: 33-0-2-40-41-28-60; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-11-0390

�2012 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 10(3) March 2012336

on March 11, 2021. © 2012 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst January 18, 2012; DOI: 10.1158/1541-7786.MCR-11-0390

http://mcr.aacrjournals.org/

receptor superfamily: TNFRSF11B)]; ref. 13). The TRAILdecoy receptors are unable to trigger a death signal becausethey lack a functional death domain (DD; ref. 14). TRAILactivates the extrinsic apoptotic pathway by inducing Fasassociated DD (FADD) protein recruitment to the DD,leading to caspase-8 activation and caspase-3 cleavage (15).In addition, it has been shown that TRAIL can induce themitochondrial intrinsic apoptotic pathway through caspase-8 activation followed by Bid cleavage (16). Recombinanthuman (rh)TRAIL is being clinically evaluated for thetreatment of both solid tumors and hematologic malignan-cies. A phase I trial with rhTRAIL is completed, showing nodose-limiting toxicities or clearly attributable toxicities toTRAIL (17). No anti-rhTRAIL antibodies have beendetected. Clinical trials are also currently in progress testingfully humanized activating monoclonal antibodies (mAb)directed against DR4 or DR5. Several phase I and II studiestargeting DR4 (Mapatumumab, TRM-1) or DR5 (Lexatu-mumab) are carried out in patients with solid malignancies,associated or not to chemotherapy (18–22). Despite limitedtimespan, these studies show promising results on tumorprogression. Kontny and colleagues reported that 80% of EStumors express DR4 whereas all express DR5 (23). Inanother study, it has been shown that 72% of ES tumorsamples express both DR4 and DR5 (of 32 samples), 25%express one receptor, and only 3% express none (24). Theseresults suggest a high proportion of TRAIL sensitivity in EStumors.We have previously reported that TRAIL is indeed a good

candidate for ES therapy, as it inhibited cell progression andincreased animal survival in a xenograft model of ES inducedby TRAIL-sensitive cell lines (25). However, half of the EScell lines were resistant to TRAIL (4 of 7), suggesting thatTRAIL or agonist antibodies cannot be proposed to allpatients. Data from the literature have reported that TRAILresistance occurs in numerous cancer cell lines, which can beattributed to dysfunctions in the TRAIL signaling pathway(14, 26). Among the major causes of TRAIL resistance, oneis linked to the absence of activating DR4 and/or DR5 deathreceptors expressed on the surface of tumor cells, leading toan imbalance of the ratio death/decoy receptors (27). Ini-tiation of death receptors DR4 and/or DR5 triggered apo-ptosis by formation of the death-inducing signaling complex(DISC) after recruitment of FADD, which binds procas-pase-8 leading to caspase-8 activation. However, the pres-ence of internal regulators of apoptotic machinery includingcellular FLICE inhibitory protein (c-FLIP), a structuralanalog of procaspase-8 which inhibits caspase-8 activation,is another important determinant (28).The aim of this study is to analyze the molecular mechan-

isms underlying TRAIL resistance in ES cell lines. For thispurpose, 7 human cell lines were studied, all expressing theEWS-FLI-1 fusion gene but differing in theirmutation statusfor p53, p16, p14, etc. The main finding of this study is theexistence of a functional truncated form of DR4 in ES cells,restoringTRAIL sensitivity when reexpressed in resistant celllines. The other main finding is the specific TRAIL sensi-tivity linked to DR4 expression (as both short and long

forms) in ES, independent of the predominant DR5 expres-sion. We also showed that the downregulation of c-FLIPexpression in resistant ES cells represents another way toresensitize these cells to TRAIL.

Materials and MethodsES cell linesSeven human ES cell lines were used. A673, TC-32, SK-

ES-1, SK-N-MC, andRD-ES cell lines were kindly providedby Dr. S. Burchill (Children's Hospital, Leeds, UnitedKingdom) and the EW24 and TC-71 by Dr. O. Delattre(Institut National de la Sante et de la Recherche MedicaleU830, Paris, France). The cell lines A673, TC-32, SK-ES-1,and RD-ES were cultured in Dulbecco's modified Eagle'smedium (DMEM;BioWhittaker) with 10%FBS (Hyclone)and SK-N-MC, EW24, and TC-71 cells were cultured inRPMI (BioWhittaker) with 10% FBS. In addition, TC-71cells require type I collagen to grow.Cells (2.93 HEK293) derived from human fetal kidney

used in in vitro cell transfection assays (29) were cultured inDMEM supplemented with 10% FBS and 2 mmol/Ll-glutamine.

Mitochondrial activityThe mitochondrial activity was determined by a colori-

metric assay using sodium 30[1-(phenylaminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro-)benzene sulfonic acidhydrate (XTT; Roche Molecular Biochemicals). Two thou-sand cells per well were plated into 96-well plates andcultured for 72 hours in culture medium with 2 treatmentswith 0, 50, or 100 ng/mL of TRAIL (R&D systems). Afterthe culture period, XTT reagent was added to each well andincubated for 5 hours at 37�C; absorbance was then read at490 nm using a 96-multiwell microplate reader.

Caspase activitySubconfluent ES cells cultured in 24-well plates were

treated with 50 ng/mL TRAIL for 1 to 6 hours, washed, andlysed in radioimmunoprecipitation assay (RIPA) buffer [150mmol/LNaCl, 50mmol/L Tris (pH 7.4), 1%NP40, 0.25%sodium deoxycholate, 1 mmol/L Na3VO4, 0.5 mmol/Lphenylmethylsulfonyl fluoride (PMSF), 1 mmol/L NaF,10 mg/mL leupeptin, and 10 mg/mL aprotinin] for 30minutes. The cells were then scraped off and the lysates werecleared of debris by centrifugation at 12,000 � g for 15minutes, after which total protein was quantified using thebicinchoninic acid (BCA) þ copper II sulfate assay (PierceChemical). Caspase-3 activity was assessed in 10 mL of celllysate with the CaspACE assay kit (Promega) following themanufacturer's recommendations.UV-treated cells were usedas positive control for caspase activity.

Flow cytometryCell death was determined by Apo2.7-PE staining as

described previously (30). Briefly, cells were harvested withPBS containing EDTA, washed and stained with anti-APO2.7-PE mAb (Beckman Coulter). Fluorescence was

A Functional Alternative Transcript of Death Receptor 4 in Ewing's Sarcoma

www.aacrjournals.org Mol Cancer Res; 10(3) March 2012 337




analyzed on a FACsCalibur cytometer (BD Biosciences). Todetermine cell surface expression of DR4 or DR5, cells werestained with PE-conjugated anti-DR4 or DR5 (eBioscience)or PE-conjugated mouse as an isotype control (BeckmanCoulter). A minimum of 10,000 cells were analyzed on aFACSCalibur flow cytometer (BD Biosciences). Cell viabil-ity was determined with PE-conjugated Apo2.7 labeling andanalyzed by flow cytometry.

Cell transfectionTransfections with pcMV-SPORT6-DR4 (imaGenes

GmbH) were done with polyethyleneimine (PEI, kindlyprovided by Dr B. Pitard, UMR915, Nantes, France). PEI–DNA complexes were prepared by equivolumetric mixingPEI (charge ratio,�4) in water with plasmid DNA solutionat the desired concentration (4 mg per well) in 150 mmol/LNaCl. Transfections were done at 70% to 80% of conflu-ence in eachwell, by adding 200mL of complex formulationsin 0.5 mL of culture DMEM deprived of FBS. After 6 hoursat 37�C, the transfection medium was replaced by 1 mL ofDMEM containing 10% FBS and 1% penicillin/strepto-mycin (complete medium).

RT-PCR analysisForty-eight hours after transfection, total RNA was

extracted from cells using TRIzol (Invitrogen) followingthe protocol suggested by the manufacturer. Reverse tran-scription PCR (RT-PCR) after total RNA extraction usingTRIzol reagent was then done. First, RNA was reversetranscribed using 400 units MMLV-RT (Invitrogen).Then, 2 mL RT reaction mixture was subjected to PCRusing upstream (50-CACCATGGCGCCACCACCAGC-TAGAGTACATCTA-30) and downstream (50-TTAAC-TCCTAACACCTAAGAGGAAACCTCTGG-30) primers(30 pmol each) of DR4 and Taq polymerase (1.25 units;Eurobio).

Quantitative RT-PCR analysisShort hairpin RNA (shRNA) gene silencingwas assessed in

subconfluent ES cells cultured in 6-well plates. c-FLIP andDR4mRNA expression was determined by quantitative RT-PCR (qRT-PCR) after total RNA extraction using TRIzolreagent. Briefly, 10 ng of cDNA were amplified using the IQSYBR Green Supermix (Bio-Rad) for the DR4, c-FLIP, andDR5 gene. Sense and antisense primers used are as follows(DR4: forward: GGAGGCACAGTGTCTGCTG reverse:CAGCACCATTTGCTGGAAC; c-FLIP: forward: CAG-GAACCCTCACCTTGTTT reverse: CAGATTTATCCA-AATCCTCACCA; DR5: forward: AGAGCCAACAGGT-GTCAACAT, reverse:GCCTCCTCCTCTGAGACCTT).Each sample was analyzed twice and quantified with theanalysis software of the iCycler iQ Real-time PCRDetectionSystem (Bio-Rad).

Western blottingCells were lysed with a lysis buffer (150 mmol/L NaCl,

5% Tris, pH 7.4, 1% Nonidet P-40, 0.25% Na deoxycho-late, 1 mmol/L Na3VO4, 0.5 mmol/L PMSF, 10 mg/mL

leupeptin, and 10 mg/mL aprotinin). After 30 minutes onice, lysates were cleared by centrifugation at 10,000 � g for10 minutes at 4�C. Total protein concentrations weredetermined using the BCA (Sigma) based method. Twentymicrograms of total cell lysate proteins was run on a SDS-PAGE and then electrophoretically transferred to Immobi-lon-P membrane (Millipore Corporation). The membranewas blotted with rabbit antibodies to human c-FLIP (Cellsigalling), rabbit anti-human Caspase-8 (Santa Cruz bio-technology), goat anti-human DR4 (R&D systems), rabbitanti-human DR5 (Cell signaling), and rabbit anti-humanactin was used as loading control (Sigma) in PBS, 0.05%Tween, and 3% bovine serum albumin. The membrane waswashed and probed with the secondary antibody (goat andrabbit, respectively) coupled to horseradish peroxidase.Antibody binding was visualized with the EnhancedChemoLuminescence (ECL) system (SuperSignal WestDura kit; ThermoScientific). To visualize bDR4, we chan-ged the lysis buffer reagents (10 mmol/L Tris-HCL pH ¼7.6, 150 mmol/L NaCl, 5 mmol/L EDTA, 1 mmol/LPMSF, 2 mg/mL aprotinin, 1% Triton X100). Then, after40 minutes on ice, lysates were cleared by centrifugation at10,000 � g for 30 minutes at 4�C. Eighty micrograms ofcleared lysates were then separated by SDS-PAGE (10%acrylamide) and electrotransferred to Immobilon-P mem-brane (Millipore Corporation). Western blot analysis wasdone as before by standard techniques with a different ECLdetection kit (Pierce Perbio Science France) this time.

shRNA silencingStably modified ES cells were obtained by lentiviral cell

transduction essentially as described previously (31). Oligo-nucleotides were designed and cloned into pSUPER toproduce the shRNA shDR4 and shc-FLIP directed, respec-tively, against human DR4 and c-FLIP genes. The targetedsequences were GGAGCAGGGACAAGTTAC (start posi-tion 1055, NM_003879.4) for c-FLIP and GCACGCT-GCTTCTTGGATT (start position 36,NM_003844.3) forDR4). The cloned oligonucleotides were controlled bysequencing (Genome Express) before subcloning them withthe upstreamH1 promoter into the vector pFG12 (32). As acontrol, a vector pFG12 was developed to produce shRNAtargeting LacZ gene as done by Qin and colleagues (2003).The resulting constructs allowed the expression of bothgreen fluorescent protein (GFP) and target-specific shRNA.They were used for lentivirus production following theprotocols provided with theViraPower Lentiviral ExpressionSystem (Invitrogen). A multiplicity of infection of 10 wasused to transduce ES cells. Seven days after transduction,GFPþ ES cells were sorted out using fluorescence-activatedcell sorting (FACS).

Short DR4 overexpressionTotal extract RNA from RD-ES ES cell line was reversed

transcribed using MMLV-RT. Then, 0.5 mg of cDNAwas amplified by PCR, with 1 Unit of PlatimiumTaq High Fidelity according to the manufacturer recom-mendation using the following DR4 primers: forward:

Picarda et al.

Mol Cancer Res; 10(3) March 2012 Molecular Cancer Research338




CACCATGGCGCCACCACCAGCTAGAGTACATC-TA; reverse: TTAACTCCTAACACCTAAGAGGAAAC-CTCTGG. Then 0.5 mL of the PCR mixture containingthe truncated form of DR4 was cloned into the pLENTITOPOplasmid according to themanufacturer protocol of thepLenti6/V5Directional TOPOCloningKit. The SK-N-MCcell line stably expressing the DR4-truncated form wasobtainedby transductionwith20 to150mL lentiviral particlesand selection with 3 mg/mL of blasticidine (Sigma).

Modeling DR4 isoforms and TRAIL interactionThe sequence corresponding to the extracellular domain

of aDR4 (amino acids 120 to 236 of NP_003835.3) wasaligned in Discovery Studio 2.5.5 with its correspondingDR5 domain sequence (69 to 185 of NP_003833.4) andadjusted manually to ensure that the cystein-rich domains(CRD) were correctly positioned (33). DR5 and aDR4extracellular domains share 63% identity and 70% similar-ity. The crystal structure of the DR5–TRAIL complex1D4V (34) was used to carry out the molecular modelingof aDR4 using the DR5 template. Of the 6 disulfide bridgespresent on DR5 extracellular domain, only 4 could bematched and modeled for aDR4. The final models werebuilt using Modeller 9v5 (35). All resulting models wereassessed for violations using the Protein Health module ofDiscovery Studio 2.5.5 (Accelrys Inc.) and corrected whennecessary. The DR4–TRAIL complex was produced bymolecular replacement of DR5 by the best model of aDR4in the 1D4V structure. The root mean square difference onmain-chain atoms was 0.86 A

�between the aDR4 model and

the DR5 structure.

Caspase activitySubconfluent ES cells cultured in 24-well plates were

treated with 50 ng/mL TRAIL for 1 to 6 hours, washed,and lysed in RIPA buffer [150 mmol/L NaCl, 50 mmol/LTris (pH 7.4), 1% NP40, 0.25% sodium deoxycholate, 1mmol/L Na3VO4, 0.5 mmol/L PMSF, 1 mmol/L NaF, 10mg/mL leupeptin, and 10 mg/mL aprotinin] for 30 minutes.The cells were then scraped off and the lysates were cleared ofdebris by centrifugation at 12,000� g for 15 minutes, afterwhich total protein was quantified using the BCAþ copperII sulfate assay (Pierce Chemical). Caspase-3 activity wasassessed in 10 mL of cell lysate with the CaspACE assay kit(Promega) following the manufacturer's recommendations.Cells treated one minute with UV were used as positivecontrol for caspase activity.

ResultsTRAIL resistance in ES cell lines is related to low DR4levels despite high DR5 expressionDR4 and DR5 expression were first analyzed by qRT-

PCR in 7ES cell lines and presented in Fig. 1A. Protein levelsof DR5 and DR4 were further detected byWestern blottingand FACS analysis on the same ES cell lines (Fig. 1B and C,respectively). Levels of DR5 were similar in all cell lines,whereas DR4 levels were higher in RD-ES and TC-71 cells

than in the 5 others. Complementary flow cytometry anal-yses were done to determine the presence of both DR4 andDR5 death receptors at the cell surface of 7 different ES celllines (Fig. 1C). Levels of cell surface DR5 were about 3-foldhigher than those detected for cell surface DR4 (range ratio4.7–6.7 vs. 1.2–2.8, respectively). Confirming the differentlevels of DR4 detection by Western blot, flow cytometryanalysis showed low cell surface DR4 levels for SK-N-MC,TC-32, SKES-1, and EW24 cells, intermediate for A673and high for TC-71 and RDES (Fig. 1C). The sameexperiment was conducted on bone marrow–derived mes-enchymal stem cells, and none of DR4 and DR5 wasdetected (data not shown).Despite the fact that DR5 is the predominant TRAIL-

activating receptor on ES cells, DR4 implication may bequestioned. Indeed, 4 (SK-ES-1, SK-N-MC, EW24, andTC-32 cell lines) of 7 ES cell lines were resistant to TRAIL-induced cell death as shown by XTT cell viability assay (Fig.1D) and low levels of DR4 were observed in these resistantcells (Fig. 1A–C). In contrast, RD-ES and TC-71 cell lineswith high DR4 levels are TRAIL-sensitive as their viability isreduced to 17.6� 2.9% and 34.8� 2.9%, respectively, afterTRAIL treatment (Fig. 1D). In the case of the A673 cell line,cell viability was reduced to 16.4� 0.7% in TRAIL-treatedcells which showed intermediate DR4 levels. In addition, nocorrelation could be shown between decoy receptor (DcR1,DcR2, or OPG) and resistance/sensitivity in all ES cell linesstudied (not shown).Because all TRAIL-resistant cell lines tested expressed low

levels ofDR4,wewonderwhether this death receptormay beinvolved in the regulation of TRAIL resistance in ES cells. Toshow this hypothesis, 2 strategies were envisaged: (i) todecrease DR4 expression in sensitive cells and (ii) to re- oroverexpress it in resistant cells.

DR4/TNFRSF10A knockdown in TRAIL-sensitiveTC-71 cell line induced TRAIL resistance despite DR5expressionSensitive TC-71 cells were modified to stably produce

either control shRNAs (shLacZ with no target in humancells) or targeting DR4 mRNA (shDR4) and were namedshLacZ or shDR4TC-71, respectively. As shown by real-time PCR on reverse transcribed RNAs,DR4 expression wassignificantly reduced (2.6-fold comparedwith shLacZ-trans-duced TC-71 cells) in shDR4TC-71 cells as compared withTC-71 or shLacZ TC-71 (Fig. 2A, left panel). However, aconcomitant 2.1-fold increase of DR5/TNFRSF10B level inshDR4TC-71 cells was noticed, although DR5 level wassimilar in shLacZTC-71 as compared with the parentalTC-71 cell line (Fig. 2A, right panel). DR4 knockdownwas associated with a 50% decrease of DR4 protein level inshDR4TC-71 cells as detected by Western blotting andat the membrane level by flow cytometry analyses(respectively Fig. 2B and C). Moreover, the DR4 reductionwas associated with a loss of TRAIL sensitivity. As shownin Fig. 2D, 104� 1.9%viability was observed in shDR4TC-71 cells versus 39.7� 3.9% and 67.2� 7%, respectively, forthe parental or shLacZTC-71 cell lines after 50 ng/mL






TRAIL treatment, and 102.2� 15.4% versus 19.5� 2.6%and 28 � 3.7%, respectively, after 100 ng/mL TRAILtreatment. These results indicated that DR4 and DR5functions may not be redundant. Indeed, even if the lossof DR4 seemed to be compensated by an augmentation ofDR5 expression, it is not associated with restoration ofTRAIL sensitivity.

Identification of a new DR4 splice transcript encoding aDR4 isoform truncated in the TRAIL-binding domainTo resensitize TRAIL-resistant cell lines to TRAIL by

overexpressing DR4, we carried out DR4 cDNA amplifica-tion for cloning into lentivirus vector. As expected, usingprimers encompassing the start codon, and beyond the stop

codon of the DR4 sequence (accession numberNM_003844), a 1,400 bp fragment was obtained on reversetranscribed RNAs of TC-71 and RD-ES cells (Fig. 3A).Surprisingly, only a smaller fragment (900 bp) was obtainedwhen these primers were used on reverse transcribed RNAsfrom 293FT cells transfected with aDR4-expressing plasmid(pCMV-sport6-DR4). We suspected that this unexpectedisoform could result from an alternative splice site of DR4transcript because (i) the expected 1,400 bp fragment wasamplified with the same primers directly from pCMV-sport6-DR4 DNA plasmid and (ii) an additional fragmentof 900 bp was also noticed within RT-PCR products of TC-71 and RD-ES RNAs (Fig. 3A). The 900 bp DR4 cDNAderived from RD-ES RNAs was cloned and sequenced. The

C

D

55 kDa

42 kDa

TC-3

2

Actin

DR4

45 kDaDR5

SK-E-S

1

SK-N

-MC

EW

24

RD-E

S

A67

3

TC-7

1

0

20

40

60

80

100

120

TC-71TC-32A673RD-ESEW24SK-N-MCSK-ES-1

Via

bilit

y (%

of t

he c

ontr

ol)

TRAIL 0 ng/mL TRAIL 50 ng/mL

B

0

1

2

3

4

5

6

7

8

SK-ES-1 SK-N-MC TC-32EW24 RD-ES A673 * TC-71

Rel

ativ

e ge

ne e

xpre

ssio

n

DR4

DR5

DcR1

DcR2

A

TC-71A673SK-N-MC

DR4

Rel

ativ

e ce

ll nu

mbe

r

DR5

2.8

5.9

1.6

6.7

1.2

4.7

EW24

4.96

1.2

6.7

1.2 1.2

SK-ES-1 TC-32 RD-ES

Resistant cell lines

2.9

2.4

Figure 1. TRAIL death receptors(DR4, DR5) detection and TRAILsensitivity in ES cell lines. A, DR4andDR5 expression was assessedby real-time PCR on reversetranscribed RNAs of 7 ES cell lines.B, DR5 and DR4 detection wasassessed by Western blotting fromtotal lysates of the 7EScell lines. C,DR5 and DR4 detection at the cellsurfaceof 3sensitive (A673, TC-71,and RDES) and 4 resistant (SK-N-MC, TC-32, SK-ES-1, and EW24)cell lines was analyzed by flowcytometry (unshaded peaks),isotype control is shown as shadedpeaks. Results are expressed asthe relative mean fluorescenceintensity (MFI) corresponding tothe ratio: MFI of the sampleanalyzed for DR4 or DR5Ab/MFI with the control isotype. D,TRAIL sensitivity of the 7 ES celllines was determined by XTT assayafter 50 ng/mL TRAIL treatment for72 hours.

Picarda et al.





comparison of both (900 and 1,400 bp) sequences showed a473 bp deletion at the beginning ofDR4 exon 1 to the end ofexon 3 and then sequence identity beyond this deletion untilthe stop codon (Fig. 3B). This deletion including nucleo-tides 138 to 611 ofNM_003844DR4 sequence identifies analternative intron as it shows typical intron feet [gt-ag].The 900 bpDR4 cDNAwas designed variant b (bDR4) as

it could produce a protein shorter than the 1,400 bp onedesigned variant a (aDR4). The bDR4 transcript couldpotentially encode a 310–amino acid protein which corre-sponds to the 468–amino acid aDR4, but truncated ofamino acids 11 to 168 within the extracellular TRAIL-binding domain. However, this truncated region shares45% identity and 59% similarity with the correspondingDR5 region, whereas the remaining extracellular TRAIL-binding domain shares 76% identity and 84% similaritywith the corresponding DR5 one (Fig. 3C).On the basis of DR5-TRAIL crystal structure and DR5-

DR4similarities, the interactionsof aDR4orbDR4toTRAILwere simulated. The DR5-TRAIL–binding interface is 1,398A� 2 wide, whereas the predicted aDR4–TRAIL one is 1,196

A� 2. According to our modeling study, the contact of bDR4with TRAIL seemed largely preserved with a relatively limitedinterface decrease (882 A

� 2 for the bDR4–TRAIL–bindinginterface). This difference is mainly related to the absencewithin bDR4 of Ala163, Pro166, and Thr168 which areinvolved in the predicted aDR4–TRAIL interaction (Fig.3D). However, the molecular dynamics simulation of thebDR4–TRAIL interface indicates that the bDR4–TRAILbinding is conserved even without these amino acids.

TRAIL sensitivity is induced by bDR4 overexpression inTRAIL-resistant SK-N-MC cell lineAfter bDR4 cDNA cloning into lentivirus vector, SK-N-

MC cells were transduced with different quantities of viruscontaining supernatant (20, 50, 100, and 150 mL), thenderived cell lines were designed bDR4-20,-50, -100, or -150SK-N-MC, respectively. As shown in Fig. 4A (left panel),these derived cells produced various levels of bDR4 whichwere correlated to the volumes of virus containing superna-tant. Using higher PAGE resolution, both aDR4 andoverexpressed bDR4 isofoms were identified in bDR4-

Figure 2. DR4 knockdown in TRAIL-sensitive TC-71 cell line. A, DR4 andDR5 expression were assessed byreal-time PCRon reverse transcribedRNAs of parental TC-71 orcounterparts shLacZ or shDR4TC-71 cells which were modified toproduce either shRNAs targeting thebacterial transcript forb-galactosidase (shLacZ) or shRNAstargeting DR4 transcripts (shDR4).DR4 protein detection was assessedby (B) Western blotting of total celllysates of parental TC-71, shLacZ,and shDR4 cells and (C) at the cellmembrane by flow cytometry.D, TRAIL sensitivity of thecorresponding cell lines wasdetermined by XTT assay after 50 ng/mL TRAIL treatment for 72 hours.

TC

-71

shLa

cZT

C-7

1

shD

R4

TC

-71

DR4

Actin

55 kDa

42 kDa

A

B

D

DR4 DR5

0.00

0.50

1.00

1.50

2.00

2.50

3.00

shLacZ TC-71TC-71 shDR4 TC-71Rel

ativ

e g

ene

exp

ress

ion

of

DR

5

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

shLacZ TC-71TC-71 shDR4 TC-71

Rel

ativ

e ge

ne e

xpre

ssio

n of

DR

4

0

20

40

60

80

100

120

140

shLacZ TC-71TC-71 shDR4 TC-71

Via

bili

ty (

% o

f th

e co

ntr

ol)

TRAIL 0 ng/mL

TRAIL 50 ng/mL

TRAIL 100 ng/mL

DR4

Rel

ativ

e ce

ll nu

mbe

r

r = 2.3 r = 1.2

shLacZ TC-71shDR4 TC-71C






150 SK-N-MC cells (Fig. 4A, right panel). In addition,bDR4 was detected as a minor isoform in TC-71 and A673cells and also in the human multiple myeloma cell lines, asshown for an example with the KMM1 cell line. Theproportion of bDR4 relative to aDR4 appeared more abun-dant in A673 cells than in KMM1 cells. SK-N-MC cellstransduced with the truncated isoform bDR4 showed a gainof sensitivity to TRAIL-induced apoptosis. Indeed, around95% of viability was observed after TRAIL treatment forparental, bDR4-20 or bDR4-50 SK-N-MC cells versus 78.9� 2.5% to 50.4� 5.2% for bDR4-100 and -150 SK-N-MCcells, respectively (Fig. 4B). Apoptosis assessed by APO2.7-PE staining in flow cytometry showed a 2-fold increase inTRAIL-induced apoptosis for bDR4-100 SK-N-MC cells ascompared with parental cells (Fig. 4C).

These results indicated that TRAIL was able to induceapoptosis through the bDR4-truncated isoform, whereasDR5 was not able to transduce TRAIL signaling.

c-FLIP knockdown induces TRAIL sensitivity inoriginally resistant SK-N-MC cell lineTRAIL-induced apoptosis pathway can be regulated by

the competitive recruitment of procaspase-8 and c-FLIP, asrespective caspase initiator and negative regulator. As shownin Fig. 5A, procaspase-8 and c-FLIP levels are not correlatedto sensitivity or resistance of ES cell lines. Indeed, procas-pase-8 levels were higher in only 1 of the 3 TRAIL-sensitive .ES cell lines (RD-ES vs. A673 and TC-71) and in 1 TRAIL-resistant cell line as compared with the 3 others (TC-32 vs.EW24, SK-N-MC, SK-ES-1). In addition, similar levels of

A B

C

A163

P166

T168

1,400 bp

900 bp

183 73 1,463

8 9

371 97 114 112 74 96

4 5 6

32

71 2 3

4 5 6 7 8 9

10

DR4 gene

DR4 cDNAs1,400 bp

900 bp aggt

81 94 11397 137119116 129

178170160156153139

D

Figure 3. New DR4 alternative transcript and its predicted bDR4 protein. A, PCR amplification of DR4 cDNA either after reverse transcription of RNAs ofTC-71 and RD-ES ES cell lines, or in 293FT cells transfected with pCMV-sport6-DR4 (pDR4-293FT), or directly from 50 ng of pCMV-sport6-DR4 (pDR4).B, DR4 gene structure is displayed by boxes and triangles for exons and introns, respectively, whose sizes (bp) are indicated. Untranslated regionsare shaded. aDR4 cDNA (1,400 bp; NM_003844) with numbered exons displayed by boxes is compared with bDR4 (900 bp; BankIt1442931 bDR4 JF729355)cDNA. The alternative intron containing part of exon 1, exon 2, and part of exon 3 is displayed by a broken line and the 2 nucleotides at both sides ofthis intron are indicated. C, the aDR4 and bDR4 sequences are aligned against the 69 to 185 amino acids of DR5, which correspond to the resolved DR5crystal structure. Sequences were aligned using Jalview (41). Amino acids are highlighted using the Zappo color scheme (on the basis of theirphysicochemicals properties), the consensus is displayed as a sequence logo. Disulfide bridges are represented by red lines. D, diagrams of predicted TRAIL(cyan) binding to aDR4 (red) or bDR4 (purple) are shown. The amino acids Ala163, Pro166, and Thr168which are present in the aDR4–TRAIL–binding interfacebut absent of the bDR4–TRAIL one are indicated.

Picarda et al.





c-FLIP were detected in the 4 TRAIL-resistant cell lines(EW24, SK-N-MC, SK-ES-1, and TC-32) as comparedwith the 3 TRAIL-sensitive ones (RD-ES, A673, andTC-71; Fig. 5A). However, the implication of c-FLIP in2 TRAIL-resistant cell lines (EW24 and SK-N-MC) wastested by developing their shRNA-mediated c-FLIP knock-down counterparts. As shown by real-time PCR on reversetranscribed RNAs, c-FLIP expression was significantlyreduced (1.8-fold for SK-N-MC and 2.3 for EW24 com-pared with shLac transduced corresponding cell lines) cellsafter transduction with lentivirus-expressing shRNA target-ing c-FLIP (designed shc-FLIP) as comparedwith parental orshLacZ cell lines (Fig. 5B). c-FLIP knockdown was con-firmed at the protein level byWestern blotting analysis in theEW24 ES cell line (Fig. 5C). The same knockdown wasobserved in the SK-N-MC cell line (data not shown). Suchc-FLIP knockdown in EW24 or SK-N-MC cells led to adecrease of TRAIL resistance: 59.1 � 3.1% and 18.7 �2.05% of cell viability were respectively observed for shc-FLIP-EW24 or -SK-N-MC cells, as compared with 105 �6.5% and 89� 6.2% cell viability for parental and shLacZ-EW24 or -SK-N-MC cells (Fig. 5D). Moreover, caspase-3activity was increased following TRAIL treatment in c-FLIPknockdown cell lines, but not in the parental or shLacZcounterparts, either in EW24 or SK-N-MC cell lines (Fig.5E). These results indicated that in spite of similar c-FLIPexpression in resistant/sensitive cell lines, c-FLIP is impli-cated in TRAIL resistance of ES cell lines.

DiscussionIn this study, we showed that all ES cell lines expressed

high levels of DR5, whereas DR4 was expressed only by 2sensitive cell lines of 3, the third one presenting intermediatelevels. None of the 4 resistant cell lines express significantlevels of DR4. Furthermore, no correlation could be estab-lished with decoy receptor expression and TRAIL resistance(not shown). The correlation betweenTRAIL sensitivity and

DR4 expression alone had never been described in ES. Alonely publication reported that 9 of 10 ES cell linesexpressed both DR4 and DR5 (Western blotting), whereasa TRAIL-resistant cell line expressed only DR4 (24). Theauthors reported thatDR4was absent from the cell surface inthe resistant cell line and 2 additional cell lines, and theysuggest that it may be nonfunctional. In this study, we foundDR4 functional in the sensitive cell lines, and its over-expression in TRAIL-resistant cells restored sensitivity. Thestudy fromMitsiades and colleagues concluded that TRAILsensitivity is related to DR5 expression, the resistance of thecell line being overcome by restoration of DR5 levels bytransfection (24). They concluded that TRAIL sensitivity ofES was mainly based on the presence of DR5/DR4, ratherthan the presence of decoy receptors or c-FLIP expression.We share the same conclusion, except that, in this study, weshowed the pivotal role of DR4 rather than DR5 by 2complementary approaches: first, DR4 knockdown inTRAIL-sensitive TC-71 cell line induced TRAIL resistancedespite DR5 expression and second, DR4 overexpression inresistant cells restored their sensitivity. Therefore, overex-pression of DR4 may represent an alternative strategy torestore TRAIL sensitivity in ES cells and models.During DR4 cloning procedure, we evidenced a splice

transcript of DR4 corresponding to a new isoform truncatedin the TRAIL-binding domain. Transcriptional modifica-tion by alternative splicing is known to be involved in theregulation of programmed cell death (36). Thus, alternativesplicing has also been identified in genes of the TRAILsystem, including DR4 and DR5; however, the isoform pre-sented here has never been reported. For example, 2 isoformsof the DR5/TRICK2/TRAIL-R2/KILLER differing by a 29amino acid extension in the extracellular domain are gener-ated by alternative pre-mRNA splicing (37). The insertedsequence may represent a retained intron as it has flankingsequences that match consensus 50 and 30 splice sites. Theauthors hypothesized that the expression of these 2 isoformsof TRICK2/DR5, which have different spacing between the

Figure 4. bDR4 overexpression inTRAIL-resistant SK-N-MC cell line.A, bDR4 isoform was detected byWestern blotting in SK-N-MC cellstransduced with bDR4-expressingvirus contained in 50, 100, or 150 mLof supernant (left). Both aDR4 andbDR4 were distinguished after high-resolution PAGE and Westernblotting in KMM1, TC-71, A673, andbDR4-150 SK-N-MC cell lines (right).B, TRAIL sensitivity was determinedby XTT assay for transduced andparental SK-N-MC and A673 celllines after 72-hour treatment with 50ng/mL TRAIL. C, percentage of celldeath induced by a 48-hourtreatment with 100 ng/mL TRAILwasdetermined by flow cytometryanalysis after Apo2.7-PE staining.

A

0

20

40

60

80

100

1501005020

Via

bilit

y (%

of t

he c

ontr

ol)

TRAIL 0 ng/mL TRAIL 50 ng/mL TRAIL 100 ng/mL

SK-N-MC-bDR4

C SK-N-MC

Ctrl7%

15% TRAIL100 ng/mL

7%

26%

SK-N-MC-bDR4-100

Apo2.7-PE

Rel

ativ

e ce

ll nu

mbe

r

TC-71KMM1 A673 bDR4-150SK-N-MC

aDR4

bDR4

Actin 42 kDa

34 kDa55 kDa

B

bDR4

42 kDaActin

34 kDa

15010050

SK-N-MC

bDR4 0

SK-N-MC A673






transmembrane domain and the TRAIL-binding domain,might regulate the response to TRAIL. In this study, bDR4was detected as a minor form in TC-71 and A673 cells andalso in the human multiple myeloma cell line KMM1. Theproportion of bDR4 relative to aDR4 appeared more abun-dant in the ES cell line A673 than inKMM1 cells, suggestingthat the newDR4 alternative splicing described here could beregulated by cell-specific factors and may not be an aberrantsplicingwhich occurs proportionally toDR4 total expression.

Therefore, it seems thatDR4 splicing is not influenced by theinitial amount of DR4 transcript in ES or in other tumorcells. In ourmodels, the sensitivity of ES cells toTRAIL is notrelated to the aDR4/bDR4 ratio, but rather to the presence ofDR4, expressed as truncated or complete forms, even if DR5is predominant. This result seems to be specific to ES as itnever had been reported in other cancer systems.In our study, we showed that the truncatedDR4 isoform is

functional because its overexpression inTRAIL-resistant SK-

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

EW24SK-N-MC

Rel

ativ

e ge

ne e

xpre

ssio

n of

c-F

LIP CT shLacZ shc-FLIPB

Procaspase-8

c-FLIPL

TC-3

2

SK-E

S-1

SK-N

-MC

EW

24

RD-E

S

A67

3

TC-7

1

Actin

c-FLIPs

A

50 kDa

60 kDa

30 kDa

42 kDa

0

5

10

15

20

25

30

35

40

45

654321UV

Cas

pas

e-3

acti

vity

AU

EW24 shLacZ EW24 shc-FLIP EW24

D

0

20

40

60

80

100

120

Via

bili

ty (

% o

f th

e co

ntr

ol)

TRAIl 0 ng/mL TRAIL 50 ng/mL

SK-N-MC shLacZ shc-FLIP shLacZEW24

EW24 shc-FLIP SK-N-MC SK-N-MC EW24

0

20

40

60

80

100

120

654321UV

Cas

pas

e-3

acti

vity

AU

TRAIL 50 ng/mL (h)TRAIL 50 ng/mL (h)

SKNMC shLacZ SKNMC shc-FLIP SKNMCSK-N-MC shc-FLIP SK-N-MCshLacZ SK-N-MCE

c-FLIPL

c-FLIPs

60 kDa

30 kDa

Actin 42 kDa

EW

24

shL

acZ

EW

24

shL

acZ

c-F

LIP

C

Figure 5. c-FLIP knockdown in TRAIL-resistant EW24 and SK-N-MC cell lines. A, c-FLIP and procaspase-8 were detected by Western blot of total celllysates of 7 ES cell lines. Actin was used as loading control. B, c-FLIP expression was assessed by real-time PCR on reverse transcribed RNAs of EW24 andSK-N-MC cells unmodified or modified with lentivirus-expressing shRNAs which target either b-galactosidase or c-FLIP transcripts (shLacZ and shc-FLIP,respectively). C, c-FLIP protein expression was determined on whole-cell lysates of EW24 cells. D, TRAIL sensitivity was determined by XTT assay fortransduced and parental EW24 or SK-N-MC cell lines after 72-hour treatment with 50 ng/mL TRAIL. E, caspase-3 activity was determined by enzymaticassay after 1 to 6 hours incubation of EW24 (left) and SK-N-MC (right) cells in the presence of 50 ng/mL TRAIL. UV exposition during 1 minute was used aspositive control.

Picarda et al.





N-MC cells induced a gain of sensitivity to TRAIL-inducedcell death. This result implies potential trimerization ofbDR4 as homotrimerization or heterotrimerization withaDR4. At this stage, we have no indication about the trimericnature of the short form in the absence of TRAIL. Becauseboth the truncated and the complete DR4 isoforms are ableto transduceTRAIL signaling, it indicates that the shortDR4isoformdoes not prevent theDR4 trimerization necessary forTRAIL binding. It is known that TRAIL binds with a highspecificity toDR5 via its loopAA0 because the removal of thisloop nearly abolishes DR5-TRAIL recognition (34). ThisTRAIL loop is in direct interaction with the CRD2 domainof DR5/DR4 but not with the CRD1 domain (34). Ourmodeling study indicates that the CRD2 domain involved inTRAIL binding is maintained in the bDR4 isoform. Thissuggests that most of the TRAIL–bDR4–binding specificitycan be attributed to the CRD2 domain. This would explainwhy bDR4 is still able to sensitize cells to TRAIL activity.Another interesting and novel result of this study relies to a

second level of regulation of TRAIL resistance in ES, linked toc-FLIP downregulation. Indeed, even if no correlation couldbe evidenced between c-FLIP expression (or the ratio c-FLIP/caspase-8) and TRAIL sensitivity/resistance in ES cells, weshowed that c-FLIP knockdown induced sensitivity in origi-nally TRAIL-resistant SK-N-MC cell line. It is well knownthat TRAIL-induced apoptosis pathway can be regulated bythe competitive recruitment of procaspase-8 and c-FLIP,respectively, as initiator or repressor of caspase-8 activity.Several publications have reported that inhibition of c-FLIPexpression could represent an interesting strategy to sensitizeTRAIL-resistant cells whatever the initial level of c-FLIPexpression (38–40). Therefore, using a RNA interferencestrategy, we showed that c-FLIP knockdown in TRAIL-resistant EW24 or SK-N-MC cells restored TRAIL sensitivityassociated with an increased caspase-3 activity. We thereforehypothesize that the involvement of c-FLIP in TRAIL resis-tance in ES cells may be linked to its differential recruitmentwithin the DISC. In RD-ES, A673, and TC-71 TRAIL-sensitive cell lines, cells express higher levels of DR4 (includingboth truncated and complete DR4 isoforms) than in resistantcells, but always less thanDR5 levels. The intracellular domainof death receptorsDR5orDR4binds FADDvia theDD, thenFADDbinds caspase-8 through death effector domain (DED)leading to the DISC formation and death signal transduction.The presence of c-FLIP interferes with procaspase-8 throughthe DED domain, acting as a competitor and inhibiting theprocaspase-8 cleavage and thus caspase-8 activation. In sensi-tive ES cells, DR4 is expressed together with DR5, inducing ahigh DISC formation with procaspase-8 recruitment thatcannot be competed by c-FLIP. Because no significant differ-ence have been observed in the procaspase-8/c-FLIP ratio

between resistant and sensitive ES cells as analysed byWesternblotting, other hypotheses could be proposed: (i) the relativeavailability of both protagonists (procaspase-8 or c-FLIP) isdifferent in the cytoplasmnear theDR4orDR5/FADD/DEDcomplex depending on the presence of lipid rafts, (ii) therelative affinity between FADD (linked to the intracellulardomains of death receptors) to procaspase-8 or c-FLIP isdifferent between DR4 and DR5, (iii) the procaspase-8recruitment is more rapid or effective when DR4 is expressedthan with DR5. These hypotheses have been partly confirmedwhen sensitive cells are treatedwith shDR4: the level ofDR4 isdecreased, so procaspase-8 is less recruited inDISC formation,and c-FLIP is therefore able to inhibit caspase-8 cleavage, thento induce TRAIL resistance.TRAIL-resistant cell lines (TC-32, EW24, SK-ES-1, and

SK-N-MC) express low (or no) levels of DR4, but still DR5.The addition of TRAIL induced the DISC formation but ina lesser extend than inDR4-expressing sensitive cells. c-FLIPis therefore able to inhibit caspase-8 cleavage in this complex.When resistant cells are treated with shc-FLIP, apoptosis isrestored as the DISC complex could be activated. The otherstrategy developed to resensitize TRAIL-resistant ES cells isto overexpress DR4 (as truncated or complete isoform),restoring higherDISC formation by procaspase-8 binding toFADD, leading to apoptosis.In conclusion, by using shDR4 in sensitive cells and

overexpressing DR4 in TRAIL-resistant cells, the resultspresented in this study showed that DR4 is involved inTRAIL sensitivity in ES.We also evidenced a new isoform ofDR4 as a consequence of an alternative transcript lackingexons 1 to 3, which is still functional and induced apoptosisthrough TRAIL binding. Finally, we hypothesized that theresistance/sensitivity in ES cells is linked to the DISCformation regulated by c-FLIP, but being also related tothe presence ofDR4 (in sensitive cells) or not (resistant cells).

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

AcknowledgmentsThe authors thank Dr. Susan Burchill (Children's Hospital, Leeds, United

Kingdom) for the kind gift of A673, TC-32, SK-ES-1, SK-N-MC, and RD-ESEwing's sarcoma cell lines and Dr. Olivier Delattre (Institut National de la Sante et dela Recherche Medicale U830, Paris, France) for EW24 and TC-71 cell lines.

Grant SupportThe work was supported by the "F�ed�eration Nationale Enfants et Sant�e," the

"Soci�et�e Francaise de lutte contre les Cancers et les leuc�emies de l'Enfant et del'Adolescent."

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received August 18, 2011; revised December 27, 2011; accepted December 27,2011; published OnlineFirst January 18, 2012.

References1. Ewing J.Diffuse endotheliomaof bone. ProcNYPathol Soc 1921;12:17.2. Hense HW, Ahrens S, Paulussen M, Lehnert M, J€urgens H. Factors

associated with tumor volume and primary metastases in Ewing tu-mors: results from the (EI)CESS studies. Klin Pediatr 1999;211:271–5.

3. Stiller CA, Bielack SS, Jundt G, Steliarova-Foucher E. Bone tumours inEuropean children and adolescents, 1978–1997. Report from theAutomated Childhood Cancer Information System project. EurJ Cancer 2006;42:2124–35.






4. Lau YS, Adamopoulos IE, Sabokbar A, Giele H, Gibbons CLMH,Athanasou NA. Cellular and humoral mechanisms of osteo-clast formation in Ewing's sarcoma. Br J Cancer 2007;96:1716–22.

5. Jaffe R, Santamaria M, Yunis EJ, Tannery NH, Agostini RM Jr, MedinaJ, et al. The neuroectodermal tumor of bone. Am J Surg Pathol1984;8:885–98.

6. Folpe AL, Chand EM, Goldblum JR, Weiss SW. Expression of Fli-1, anuclear transcription factor, distinguishes vascular neoplasms frompotential mimics. Am J Surg Pathol 2001;25:1061–6.

7. May WA, Gishizky ML, Lessnick SL, Lunsford LB, Lewis BC,Delattre O, et al. Ewing sarcoma 11;22 translocation produces achimeric transcription factor that requires the DNA-binding domainencoded by FLI1 for transformation. Proc Natl Acad Sci U S A1993;90:5752–6.

8. Nakatani F, Tanaka K, Sakimura R, Matsumoto Y, Matsunobu T, Li X,et al. Identification of p21WAF1/CIP1 as a direct target of EWS-Fli1oncogenic fusion protein. J Biol Chem 2003;278:15105–15.

9. Dauphinot L, De Oliveira C, Melot T, Sevenet N, Thomas V, WeissmanBE, et al. Analysis of the expressionof cell cycle regulators inEwing celllines: EWS-FLI-1 modulates p57KIP2and c-Myc expression. Onco-gene 2001;20:3258–65.

10. Cohen MD, Bugaieski EM, Haliloglu M, Faught P, Siddiqui AR. Visualpresentation of the staging of pediatric solid tumors. Radiographic1996;16:523–45.

11. Cotterill SJ, Ahrens S, PaulussenM, J€urgens HF, Voute PA, Gadner H,et al. Prognostic factors in Ewing's tumor of bone: analysis of 975patients from the European Intergroup Cooperative Ewing's SarcomaStudy Group. J Clin Oncol 2000;18:3108–14.

12. BouralexisS, FindlayDM, EvdokiouA.Death to the badguys: targetingcancer via Apo2L/TRAIL. Apoptosis 2005;10:35–51.

13. Holen I, ShipmanCM.Role of osteoprotegerin (OPG) in cancer. Clin Sci(Lond) 2006;110:279–91.

14. Zhang L, Fang B. Mechanisms of resistance to TRAIL-induced apo-ptosis in cancer. Cancer Gene Ther 2005;12:228–37.

15. Liabakk NB, Espevik T. Monoclonal antibodies against TRAIL. VitamHorm 2004;12:65–79.

16. Kruyt FA. TRAIL and cancer therapy. Cancer Lett 2008;263:14–25.17. Wu GS. TRAIL as a target in anti-cancer therapy. Cancer Lett 2009;

285:1–5.18. Tolcher AW,MitaM,Meropol NJ, vonMehrenM, Patnaik A, Padavic K,

et al. Phase I pharmacokinetic and biologic correlative study of mapa-tumumab, a fully human monoclonal antibody with agonist activity totumor necrosis factor-related apoptosis-inducing ligand receptor-1.J Clin Oncol 2007;25:1390–5.

19. Leong S, Cohen RB, Gustafson DL, Langer CJ, Camidge DR, PadavicK, et al. Mapatumumab, an antibody targeting TRAIL-R1, in combi-nation with paclitaxel and carboplatin in patients with advanced solidmalignancies: results of a phase I and pharmacokinetic study. J ClinOncol 2009;27:4413–21.

20. Plummer R, Attard G, Pacey S, Li L, Razak A, Perrett R, et al. Phase 1and pharmacokinetic study of lexatumumab in patients with advancedcancers. Clin Cancer Res 2007;13:6187–94.

21. Wakelee HA, Patnaik A, Sikic BI,MitaM, FoxNL,Miceli R, et al. Phase Iandpharmacokinetic studyof lexatumumab (HGS-ETR2) given every 2weeks in patients with advanced solid tumors. Ann Oncol 2010;21:376–81.

22. Adams C, Totpal K, Lawrence D, Marsters S, Pitti R, Yee S, et al.Structural and functional analysis of the interaction between theagonisticmonoclonal antibodyApomaband theproapoptotic receptorDR5. Cell Death Differ 2008;15:751–61.

23. Kontny HU, H€ammerle K, Klein R, Shayan P, Mackall CL, NiemeyerCM. Sensitivity of Ewing's sarcoma to TRAIL-induced apoptosis. CellDeath Differ 2001;8:506–14.

24. Mitsiades N, Poulaki V, Mitsiades C, Tsokos M. Fas-mediated apo-ptosis in neuroblastoma requires mitochondrial activation and isinhibited by FLICE inhibitor protein and Bcl-2. Cancer Res 2001;61:2704–12.

25. PicardaG, Lamoureux F,Geffroy L,DelepineP,Montier T, LaudK, et al.Preclinical evidence that use of TRAIL in Ewing's sarcoma and oste-osarcoma therapy inhibits tumor growth, prevents osteolysis, andincreases animal survival. Clin Cancer Res 2010;16:2363–74.

26. Kurbanov BM, Fecker LF, Geilen CC, Sterry W, Eberle J. Resistance ofmelanoma cells to TRAIL does not result from upregulation of anti-apoptotic proteins by NF-kappaB but is related to downregulation ofinitiator caspases and DR4. Oncogene 2007;26:3364–77.

27. Wang S, El-Deiry WS. TRAIL and apoptosis induction by TNF-familydeath receptors. Oncogene 2003;22:8628–33.

28. Zhang XD, Franco A,Myers K, Gray C, Nguyen T, Hersey P. Relation ofTNF-related apoptosis-inducing ligand (TRAIL) receptor and FLICE-inhibitory protein expression to TRAIL-induced apoptosis of melano-ma. Cancer Res 1999;59:2747–53.

29. Graham FL, Smiley J, Russell WC, Nairn R. Characteristics of a humancell line transformedbyDNA fromhumanadenovirus type5. JGenVirol1977;36:59–74.

30. Menoret E, Gomez-Bougie P, Geffroy-Luseau A, Daniels S, Moreau P,Le Gouill S, et al. Mcl-1L cleavage is involved in TRAIL-R1- and TRAIL-R2-mediated apoptosis induced by HGS-ETR1 and HGS-ETR2human mAbs in myeloma cells. Blood 2006;108:1346–52.

31. Trichet V, Benezech C, Dousset C, Gesnel MC, Bonneville M, Breath-nach R. Complex interplay of activating and inhibitory signals receivedby Vgamma9Vdelta2 T cells revealed by target cell beta2-microglo-bulin knockdown. J Immunol 2006;177:6129–36.

32. Qin XF, An DS, Chen IS, Baltimore D. Inhibiting HIV-1 infection inhuman T cells by lentiviral-mediated delivery of small interfering RNAagainst CCR5. Proc Natl Acad Sci USA 2003;100:183–8.

33. Naismith JH,SprangSR.Modularity in theTNF-receptor family. TrendsBiochem Sci 1998;23:74–9.

34. Mongkolsapaya J, Grimes JM, Chen N, Xu XN, Stuart DI, Jones EY,et al. Structure of the TRAIL-DR5 complex reveals mechanisms con-ferring specificity in apoptotic initiation. Nat Struct Mol Biol 1999;6:1048–53.

35. Sali A, Blundell TL. Comparative protein modelling by satisfaction ofspatial restraints. J Mol Biol 1993;234:779–815.

36. JiangZH,WuJY. Alternative splicing andprogrammedcell death. ProcSoc Exp Biol Med 1999;220:64–72.

37. Screaton GR, Mongkolsapaya J, Xu X-N, Cowper AE, McMichael AJ,Bell JI. TRICK2, a new alternatively spliced receptor that transducesthe cytotoxic signal from TRAIL. Curr Biol 1997;7:693–6.

38. Galligan L, Longley DB, McEwan M, Wilson TR, McLaughlin K, John-ston PG. Chemotherapy and TRAIL-mediated colon cancer cell death:the roles of p53, TRAIL receptors, and c-FLIP. Mol Cancer Ther2005;4:2026–36.

39. Longley DB, Wilson TR, McEwan M, Allen WL, McDermott U, GalliganL, et al. c-FLIP inhibits chemotherapy-induced colorectal cancer celldeath. Oncogene 2006;25:838–48.

40. Wilson TR, McLaughlin KM, McEwan M, Sakai H, Rogers KM, Red-mond KM, et al. c-FLIP: a key regulator of colorectal cancer cell death.Cancer Res 2007;67:5754–62.

41. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. Jal-view Version 2–a multiple sequence alignment editor and analysisworkbench. Bioinformatics 2009;25:1189–91.

Picarda et al.





2012;10:336-346. Published OnlineFirst January 18, 2012.Mol Cancer Res Gaëlle Picarda, Sylvanie Surget, Romain Guiho, et al. Sensitivity/Resistance MechanismsSarcoma Cell Lines May be Involved in TRAIL A Functional, New Short Isoform of Death Receptor 4 in Ewing's

Updated version

10.1158/1541-7786.MCR-11-0390doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/10/3/336.full#ref-list-1

This article cites 41 articles, 14 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/10/3/336To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-11-0390

http://mcr.aacrjournals.org/content/10/3/336.full#ref-list-1

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/10/3/336


A Functional, New Short Isoform of Death Receptor 4 in Ewing's … · A Functional, New Short...

Documents

Transcript of A Functional, New Short Isoform of Death Receptor 4 in Ewing's … · A Functional, New Short...