Sulforaphane targets pancreatic tumour-initiating cells by NF- B-induced antiapoptotic signalling

doi: 10.1136/gut.2008.149039 2009 58: 949-963 originally published online October 1, 2008Gut

 G Kallifatidis, V Rausch, B Baumann, et al. antiapoptotic signalling

B-inducedκtumour-initiating cells by NF-Sulforaphane targets pancreatic

http://gut.bmj.com/content/58/7/949.full.htmlUpdated information and services can be found at:

These include:

Data Supplement http://gut.bmj.com/content/suppl/2009/06/01/58.7.949.DC1.html

"Web only appendices"

References

http://gut.bmj.com/content/58/7/949.full.html#related-urlsArticle cited in:  

http://gut.bmj.com/content/58/7/949.full.html#ref-list-1This article cites 49 articles, 19 of which can be accessed free at:

serviceEmail alerting

box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in the

Topic collections

(773 articles)Pancreatic cancer   � (6483 articles)Pancreas and biliary tract   �

 Articles on similar topics can be found in the following collections

Notes

http://gut.bmj.com/cgi/reprintformTo order reprints of this article go to:

http://gut.bmj.com/subscriptions go to: GutTo subscribe to

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

Sulforaphane targets pancreatic tumour-initiatingcells by NF-kB-induced antiapoptotic signalling

G Kallifatidis,1 V Rausch,1 B Baumann,2 A Apel,1 B M Beckermann,1 A Groth,1

J Mattern,1 Z Li,3 A Kolb,3 G Moldenhauer,4 P Altevogt,5 T Wirth,2 J Werner,3

P Schemmer,3 M W Buchler,3 A V Salnikov,4 I Herr1

See Commentary, p 900

c Additional figures arepublished online only at http://gut.bmj.com/content/vol58/issue7

1 Molecular OncoSurgery Group,University of Heidelberg andGerman Cancer ResearchCenter, Heidelberg, Germany;2 Institute of PhysiologicalChemistry, University of Ulm,Ulm, Germany; 3 Department ofGeneral Surgery, University ofHeidelberg, Heidelberg,Germany; 4 Department ofMolecular Immunology, GermanCancer Research Center,Heidelberg, Germany; 5 TumourImmunology Program, GermanCancer Research Center,Heidelberg, Germany

Correspondence to:Professor I Herr, Department ofGeneral Surgery, University ofHeidelberg and DKFZ-G403,Molecular OncoSurgery, ImNeuenheimer Feld 365, D-69120Heidelberg, Germany; [email protected]

Revised 11 July 2008Accepted 2 September 2008Published Online First1 October 2008

ABSTRACTBackground and aims: Emerging evidence suggeststhat highly treatment-resistant tumour-initiating cells(TICs) play a central role in the pathogenesis of pancreaticcancer. Tumour necrosis factor-related apoptosis-inducingligand (TRAIL) is considered to be a novel anticanceragent; however, recent studies have shown that manypancreatic cancer cells are resistant to apoptosisinduction by TRAIL due to TRAIL-activated nuclear factor-kB (NF-kB) signalling. Several chemopreventive agentsare able to inhibit NF-kB, and favourable results havebeen obtained—for example, for the broccoli compoundsulforaphane—in preventing metastasis in clinical studies.The aim of the study was to identify TICs in pancreaticcarcinoma for analysis of resistance mechanisms and fordefinition of sensitising agents.Methods: TICs were defined by expression patterns of aCD44+/CD242, CD44+/CD24+ or CD44+/CD133+ phenotypeand correlation to growth in immunodeficient mice,differentiation grade, clonogenic growth, sphere forma-tion, aldehyde dehydrogenase (ALDH) activity and therapyresistance.Results: Mechanistically, specific binding of transcrip-tionally active cRel-containing NF-kB complexes in TICswas observed. Sulforaphane prevented NF-kB binding,downregulated apoptosis inhibitors and induced apopto-sis, together with prevention of clonogenicity.Gemcitabine, the chemopreventive agents resveratrol andwogonin, and the death ligand TRAIL were less effective.In a xenograft model, sulforaphane strongly blockedtumour growth and angiogenesis, while combination withTRAIL had an additive effect without obvious cytotoxicityin normal cells. Freshly isolated patient tumour cellsexpressing markers for TICs could be sensitised bysulforaphane for TRAIL-induced cytotoxity.Conclusion: The data provide new insights intoresistance mechanisms of TICs and suggest thecombination of sulforaphane with TRAIL as a promisingstrategy for targeting of pancreatic TICs.

Pancreatic adenocarcinoma is an aggressive malig-nancy usually diagnosed when in an advancedstate, for which there are few or no effectivetreatments. It has the worst prognosis of anymajor malignancy, and the vast majority ofpatients die within the first year after diagnosis,and ,1% of patients are alive after 5 years.1 One ofthe major hallmarks of pancreatic cancer is itsextensive local tumour invasion and early systemicdissemination. The molecular explanations forthese characteristics of pancreatic cancer areincompletely understood. Emerging evidence hasshown that the capacity of a tumour to grow and

propagate is dependent on a small subset ofdistinct tumour-initiating cells (TICs).2 3 TICs haverecently been identified in several tumour entitiesincluding pancreatic cancer.4–6 The current con-sensus definition describes a TIC as a cell within atumour that is able to self-renew and to producethe heterogeneous lineages of cancer cells thatcomprise the tumour bulk.7 The implementation ofthis concept explains the use of alternative terms,such as ‘‘cancer stem cell’’ and ‘‘tumourigenic cell’’to describe tumour-initiating cells.2 3 8–10 Commonanticancer treatments such as radiation andchemotherapy do not eradicate the majority ofhighly resistant TICs,11 suggesting the need fornew therapeutic options.

The human tumour necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is con-sidered to be a novel anticancer agent; however,recent studies have shown that many pancreaticcancer cells are resistant to the apoptosis-inducingeffects of TRAIL due to constitutive and TRAIL-activated nuclear factor-kB (NF-kB) signalling.12

Homodimeric or heterodimeric NF-kB complexesconsist of proteins present in the cytoplasm,namely p50, p52, RelA/p65, RelB and cRel.Dimers are kept in an inactive state by a familyof IkB (inhibitor of NF-kB) proteins.13 A pivotalstep in the activation of NF-kB is the phosphoryla-tion of IkB molecules, which in turn leads to theirproteasome-dependent degradation and allows thetranslocation of NF-kB dimers to the nucleuswhere they can bind to specific DNA responseelements.14 Several chemopreventive agents areable to inhibit NF-kB and therefore are proapopto-tic. Cruciferous vegetables, such as broccoli, have ahigh content of glucosinolate-derived compounds,the precursors of anticarcinogenic isothiocyanatesulforaphane (SF). SF protects from DNA damage,induces apoptosis and inhibits NF-kB, as well asblocking cell proliferation.15 Other anticarcinogenicplant substances with similar properties are thepolyphenol resveratrol (RE), present in high levelsin grapes and derived products such as red wine,16

and the flavonoid wogonin (WO), a popular herbalremedy in China and several other Asian coun-tries.17

Why some pancreatic cancer cell lines haveelevated NF-kB signalling, while others have not,is unknown. Therefore, it is tempting to spec-ulate that this might be due to the presence ofTIC populations within pancreatic tumourswhose pronounced resistance might be due toactivated NF-kB signalling. In the present study,

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 949

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

we identified TIC-like populations in established pancreaticcancer cell lines, and in paraffin sections of patient tumours.The presence of TICs correlated with resistance towardsgemcitabine or TRAIL due to binding of transactivationpotent NF-kB dimers in TIChigh but not in TIClow pancreaticcancer cells. The chemopreventive agent SF was able toabrogate resistance by interfering with NF-kB binding. Wepropose that SF is a promising agent for TIC-targetedtherapies in pancreatic carcinoma.

MATERIALS AND METHODSEstablished cell linesAsPC-1, BxPc-3, Capan-1 and MIA-PaCa2 pancreatic cell lineswere obtained from the American Type Culture Collection(Manassas, Virginia, USA) and were cultured as described.18 Themurine plasmocytoma cell line S107 has been described and wasgrown in Dulbecco’s modified Eagle’s medium (DMEM;Invitrogen, Karlsruhe, Germany) containing 10% heat-inacti-vated fetal calf serum (PAN Systems, Aidenbach, Germany).

Primary cell linesPrimary fibroblasts from skin were kindly provided by Dr H-JStark (DKFZ, Heidelberg, Germany). Mesenchymal stem cellsand fresh tumour cells from resected pancreatic tumours wereisolated as described.18 19 Patient material was obtained with theapproval of the ethical committee of the University ofHeidelberg. Diagnoses were established by conventional clinicaland histological criteria according to the World HealthOrganization (WHO). All surgical resections were indicated bythe principles and practice of oncological therapy.

Nude mice and xenograftsMIA-PaCa2 cells (86106 in 150 ml) were injected subcutaneouslyinto the right anterior flank of 4- to 6-week-old NMRI (nu/nu)male mice (day 0). After the tumours had reached a meandiameter of about 8–10 mm, mice carrying MIA-PaCa2 xeno-grafts were randomly divided into groups of six animals eachand treatment was initiated. The mice were treated intraper-itoneally with SF at a dose of 4.4 mg/kg and/or recombinantSuper Killer TRAIL (25 ng/tumour) intratumourally on day 4, 5and 6 after tumour transplantation. In detail, we diluted 1 ml ofTRAIL stock solution (500 ng/ml) in 1 ml of phosphate-bufferedsaline (PBS) (500 ng TRAIL/ml) and injected 50 ml of thissolution per tumour. Tumour growth was monitored bymeasuring two diameters daily with calipers, and tumourvolumes (V) were calculated using the formula V = K

(length6width2). Mice were euthanised at tumour sizes.1500 mm3. For evaluation of tumourigenicity, cells weretrypsinised, counted, diluted to appropriate injection dosesand mixed with Matrigel Matrix (BD Biosciences, Bedford,Massachusetts, USA) at a 1:1 ratio. TIChigh MIA-PaCa2 andTIClow BxPc-3 cells were injected subcutaneously (16103 cells/flank, total volume 200 ml) at opposite flanks of the sameanimals. Animal experiments were carried out in the animalfacilities of the DKFZ after approval by the authorities(Regierungsprasidium Karlsruhe).

Analysis of plasma liver enzymesPeripheral blood was taken before liver perfusion under deepanaesthesia. One drop of heparin was added and, after centrifuga-tion, supernatants were collected. The levels of plasma lactatedehydrogenase (LDH), glutamate-oxalacetate-transaminase(GOT/AST) and glutamate-pyruvate-transaminase (GPT/ALT)

were measured using a DRY-CHEM FDC3500 (Fuji Medical,Tokyo, Japan) according to the manufacturer’s guidelines.

Cytotoxic agentsGemcitabine (a kind gift from Eli Lilly, Indianapolis, Indiana,USA) was diluted in PBS to a 100 mM stock. Recombinanthuman TRAIL/Apo2 ligand produced in Escherichia coli wasfrom Axxora (Lorrach, Germany). Stock solutions of SF and RE(both from Sigma, Deisenhofen, Germany) were prepared inethanol and a stock solution of WO (Phytolab,Vestenbergsgreuth, Germany), phorbol 12-myristate 13-acetate(PMA; Calbiochem, San Diego, California, USA) and Go6983(Sigma-Aldrich Chemie, Steinheim, Germany) in dimethylsul-foxde (DMSO). Final concentrations of the solvents in themedia were (0.1%.

Measurement of apoptosisCells were stained with fluorescein isothiocyanate (FITC)-conjugated annexin V (BD Biosciences, Heidelberg, Germany)and externalisation of phosphatidylserine as well as the forwardside scatter profile were identified by flow cytometry (FACScan,BD Biosciences). DNA fragmentation was detected by theNicoletti method.18

MTT assayTumour cells were resuspended at a density of 36104 to 105/mlin 96-well microplates, 100 ml per well. After treatment,the MTT (3-(4,5-dimethylthiazo-2-yl)-2,5-diphenyl tetrasodiumbromide) assay was performed as described.18

Spheroid assayCells were cultured in NeuroCult NS-A basal serum-freemedium (human) (StemCell Technologies, Vancouver, BC,Canada) supplemented with 2 mg/ml heparin (StemCellTechnologies), 20 ng/ml human epidemal growth factor(hEGF; R&D Systems, Wiesbaden-Nordenstadt, Germany),10 ng/ml human basic fibroblast growth factor (hFGF-b;PeproTech GmbH, Hamburg, Germany) and NeuroCult NS-Aproliferation supplements (human) (StemCell Technologies.).Cells were seeded at low densities (16104 cells/ml) in 12-welllow adhesion plates, 1 ml per well. Upon formation ofspheroids, cells were reseeded at 16104/ml in order to evaluatethe potential for formation of secondary spheroids.

Colony-forming assaysCells were seeded at a density of 36104 in 12-well tissue cultureplates (BD Falcon, San Jose, California, USA). After 24 h thecultures were treated. Forty-eight hours after treatment, thecultures were trypsinised, plated at a density of either 400, 500,600 or 800 cells per well in 6-well tissue culture plates in paralleland incubated for 10 days without changing the medium. Fordetermination of colony formation, cultures were fixed (3.7%paraformaldehyde and 70% ethanol) and stained with 0.05%Coomassie blue. The number of colonies with .50 cells wascounted under a dissecting microscope. The percentage cellsurvival was calculated (plating efficiency of non-treatedcultures = 1, or relative survival rate).

Detection of aldehyde dehydrogenase (ALDH) activityA total of 16106 cells were treated with 5 ml/ml ALDEFLUORsubstrate (Aldagen, Durham, North Carolina, USA), incubatedfor 30 min at 37uC and analysed by flow cytometry accordingto the manufacturer’s instructions.

Pancreas

950 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

Detection of caspase activityA kit providing a fluorochrome inhibitor of caspases(FAM-Asp(OMe)-Glu(OMe)-Val-Asp(OMe)-fluoromethylketone(FLICA)) was used according to the manufacturer’s protocol(Immunochemistry Technologies, Bloomington, Minnesota,USA). FLICA binds covalently to specific active caspases (thusthe fluorochrome accumulates in cells which have activecaspases). For red fluorescence, a sulforhodamine-labelled fluor-omethyl ketone peptide inhibitor of caspases, and for greenfluorescence carboxyfluorescein-labelled inhibitors were used. Inshort, FLICA specific for caspases 3, 7, 8 or 9 was added to cellmedia and incubated for 1 h at 37uC in 5% CO2. Cells were thenwashed in washing buffer and analysed by immunofluorescencemicroscopy.

Whole-cell extract preparation for electrophoretic mobility shiftassayCells were washed twice in PBS and resuspended in threepacked cell volumes of buffer C as described.20 The samples weresubjected to three cycles of freezing in liquid nitrogen andsubsequent thawing on ice. After centrifugation, the super-natant was used as whole-cell extract.

Electrophoretic mobility shift assay (EMSA)Protein extracts (5 mg) were incubated for 30 min at roomtemperature with 3 mg of poly(dI/dC), 10 mg of bovine serumalbumin in buffer containing 50 mM NaCl, 1 mM dithiothrei-tol, 10 mM Tris–HCl, 1 mM EDTA, 5% glycerol and radi-olabelled double-stranded oligonucleotides containing animmunoglobulin (Ig)-k enhancer consensus NF-kB site.21 TheDNA–protein complexes formed were then separated from freeoligonucleotides on a native 4% polyacrylamide gel. For super-shift experiments, 2.5 mg of protein extract were preincubatedfor 30 min with specific antibody before being treated asdescribed before. Antibodies used for supershift experimentswere from Santa Cruz Biotechnology (Santa Cruz, California,USA).

Detection of stem cell markers by flow cytometryThe expression of surface markers was analysed with two-colour flow cytometry using a FACScan flow cytometer(Becton-Dickinson, Heidelberg, Germany). A total of 16106

cells were incubated with Venimmun (Aventis-Behring,Marburg, Germany) at 4uC for 20 min to inhibit unspecificbinding of antibodies. After washing with PBS/5% fetal calfserum (FCS), cells were incubated with unconjugated or withFITC- or phycoerythrin (PE)-conjugated primary antibody.After washing, cells were incubated with FITC- or PE-labelledsecondary antibodies at 4uC for 30 min to detect unconjugatedprimary antibody. Specific antibodies were anti-CD44 mono-clonal antibody (mAb) at a dilution of 1:50 (Clone G44-26,Pharmingen, Heidelberg, Germany), antiepithelial-specific anti-gen (ESA) mAb at a dilution of 1:50 (Biozol, Eching, Germany),anti-CD24 undiluted (ML5 or SWA11 from non-purifiedhybridoma supernatant)22 and anti-CD133 mAb diluted 1:10(Miltenyi Biotec, Bergisch Gladbach, Germany). PE-conjugatedgoat antimouse IgG (BD Pharmingen, Heidelberg, Germany)or FITC-conjugated goat antimouse IgG (JacksonImmunoResearch, Suffolk, UK) were used as secondaryantibodies. The data were analysed using CELLQuest software(Becton-Dickinson, Heidelberg, Germany). PE- or FITC-labelledmouse IgG (BD Pharmingen) served as isotype controls. Gatingwas implemented based on negative control staining profiles.

Transfection of small interfering (siRNA)siRNA oligonucleotides were obtained from Santa CruzBiotechnology (Heidelberg, Germany). One day before transfec-tion, 16105 cells per well were seeded in 12-well tissue cultureplates. One hour before transfection, medium was replaced byantibiotic-free DMEM and transfection of siRNA was per-formed with Lipofectamine 2000 according to the protocolprovided by the manufacturer (Invitrogen, Karlsruhe,Germany).

Protein isolation and western blot analysisWhole-cell extracts were prepared by a standard protocol andproteins were detected by western blot analysis using polyclonalrabbit antibodies anti-IkBa, anti-cRel (SC-371 and SC-70, SantaCruz Biotechnology, Heidelberg, Germany), anti-cIAP1, anti-FLIP or mAb anti-XIAP (R&D Systems, Wiesbaden-Nordenstadt, Germany). Goat antirabbit or antimouse (SantaCruz) secondary antibody conjugated to horseradish peroxidase(HRP) and an enhanced chemiluminescence detection system(Super Signal West Femto, Pierce, Rockford, Illinois, USA) wereused for detection.

Immunohistochemistry and immunocytochemistryImmunohistochemistry on 5 mm frozen or paraffin-embeddedtissue sections was performed using the ABC Elite kit fromLinaris (Wertheim-Bettingen, Germany), the DAB (3,39 diami-nobenzidine) staining kit from Invitrogen (Karlsruhe, Germany)or ZytoChem-Plus HRP polymer kit with AEC (3-amino-9-ethylcarbazole) as a chromogen (Zytomed Systems, Berlin,Germany) according to the manufacturers’ instructions. IkBa,RelA and cRel were detected using rabbit polyclonal antibodyfrom Santa Cruz Biotechnology (Heidelberg, Germany) andgoat antirabbit biotinylated IgG (Vector, Burlingame,California, USA) as a secondary antibody. The signal wasamplified with the ABC kit, and DAB was used as a chromogen.Samples were counterstained with haematoxylin, dehydrated ingraded alcohol, rinsed in xylene and mounted in Entellan(Merck, Darmstadt, Germany). CD133 in tumour tissues wasdetected by rabbit polyclonal anti-CD133 antibody from Biozol(Eching, Germany) and the colour was developed using theZytoChem-Plus HRP polymer (mouse/rabbit) kit with AEC as achromogen. Samples were directly mounted in water-solublemounting medium. To detect CD133+/CD44+ cells, CD44 wasdetected by a mouse mAb (clone G44-26, Pharmingen,Heidelberg, Germany). Tissue sections were blocked with 10%normal goat serum and incubated with primary mouse anti-CD44 mAb followed by washing in PBS and incubation withsecondary goat antimouse Alexa 488 IgG (Molecular Probes,Karlsruhe, Germany). After repeated washing, sections wereincubated with rabbit anti-CD133 antibody followed by goatantirabbit Alexa 594 IgG (Molecular Probes), washed andmounted in Fluoromount G. CD24 was detected by immuno-fluorescence with mouse mAb from non-purified and undilutedhybridoma supernatant22 and goat anti-mouse Alexa 594 IgG(Molecular Probes). Omission of primary antibody served as anegative control. Randomly chosen fields were examined at6400 magnification using a Leica DMRB microscope. Imageswere captured using a Kappa CF 20/4 DX digital colourcamera (Kappa, Gleichen, Germany) and analysed with KappaImageBase 2.2 software. Similar staining protocols were used todetect RelA and cRel in pancreatic cancer cell lines and tissuesamples of xenografts. Microvessel density in acetone-fixedfrozen xenograft sections (5 mm) was deteced by staining with

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 951

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

rat anti-mouse CD31 mAb (PharMingen, San Diego, Califonia,USA) as described previously.30 Randomly chosen areas oftumours were examined under 6400 magnification andcounted. Any distinct area of positive staining for CD31 wascounted as a single vessel. Results were expressed as the meannumber of vessels (SE) per high-power field.

Statistical evaluationsFor MTT and fluorescence-activated cell sorting (FACS)measurements, statistical evaluations are presented as the mean(SD). Data were analysed using the Student t test for statisticalsignificance. p Values were considered significant if they were,0.05. For xenografts on nude mice, a distribution-free test fortumour growth curve analyses for treatment experiments withxenografted cancer cells was used as described by Koziol et al.23

RESULTS

TIC marker expression correlates with resistance,differentiation, sphere and colony formation, ALDH activity andgrowth on nude miceTIC-like characteristics in four human pancreatic carcinoma celllines were defined by expression of characteristic surfacemarkers, the potential to grow in immunodeficient mice,apoptosis resistance, enrichment of a CD44+/CD242 phenotypeby chemotherapy, degree of differentiation, colony- andspheroid-forming capacity, invasion pattern and differentiationpotential (fig 1, table 1 and Supplementary fig 1).

In this way we defined the CD44+/CD242 population as theTIC-like phenotype in pancreatic cancer cell lines. In detail,double staining with specific CD44 and CD24 antibodiesrevealed that AsPC-1, MIA-PaCa2 and Capan-1 cells contained.90% CD44+/CD242 cells, in contrast to BxPc-3 cells, whichcontained only 17.5% (fig 1A). All examined cells also expressedESA, some of which were double positive for CD44 and somenot, while expression of CD133 was undetectable. While an ESAsingle- or an ESA+/CD44+ double-positive phenotype did notcorrelate with any TIC property, the CD44+/CD242 phenotypecorrelated with colony- and spheroid-forming capacity (fig 1B),resistance towards gemcitabine and TRAIL (fig 1C), ALDHactivity (fig 1D) and a poor to moderate differentiation grade(table 1) as well as with the documented ability to grow asxenografts in nude mice.24–26 It has been reported that BxPc-3cells grow very slowly in nude mice,27 consistent with theirlowest content of putative TICs. We confirmed this finding bytransplanting 103 MIA-PaCa2 cells into the right flank and 103

BxPc-3 cells into the left flank of nude mice. While MIA-PaCa2xenografts became visible 17 days after injection, BxPc-3xenografts needed 28 days to form tumours (Supplementaryfig 2). An additional hint of TIC features such as multipotencyis the published finding that xenografted Capan-1 or AsPC-1cells differentiate into phenotypically diverse populations whichshow morphological, biological and biochemical characteristicssimilar to those of the tumour of origin.28 29 The differentiationpotential of TICs present in the cultured cell line is confirmedby our spheroid assays in which we observed that some cells,the putative TICs, acquired the ability to grow in primary andsecondary spheres, while other cells still grow adherent. Thesefeatures correlate with differentiation of TICs into functionallydifferent cells. Since Capan-1, MIA-PaCa2 and AsPC-1 cellsharbour .90% of putative TICs, we did not enrich them furtherbut examined treatment resistance in these TIChigh cellscompared with BxPc-3 TIClow cells. According to the recentsuggestion that chemotherapy treatment of tumours may lead

to enrichment of TICs,6 repeated treatment of BxPc-3 cells with25 nM gemcitabine followed by treatment with 50 nMgemcitabine for 4 days enhanced the percentage of CD44+/CD242 cells from an initial 17% to 42% (fig 1E). Finally, wewould like to point out that TRAIL resistance in TIChigh

pancreatic cancer cell lines is not due to the absence of TRAILreceptor expression, since all cell lines have already been shownto express the TRAIL death receptor protein.30 31 In controlexperiments we confirmed that TRAIL exhibited no pro-nounced cytotoxicity to normal cells such as primary skinfibroblasts and to normal stem cells such as mesenchymal stemcells isolated from human bone marrow (fig 1C).

SF abrogates resistance of TIChigh cells without cytotoxicity tonormal cellsTo test whether chemopreventive agents might overcome theresistance of TICs, MIA-PaCa2, AsPC-1, Capan-1 and BxPc-3cells were treated with RE, WO or SF. After 48 h, viability wasexamined by MTT assay. While RE or WO alone had nopronounced effect on viability of TIChigh MIA-PaCa2 cells, SFstrongly reduced viability already at low concentrations in MIA-PaCa2 and Capan-1 cells and in medium to high concentrationsalso in AsPC-1 cells (fig 2A,B, Supplementary Fig. 3). Mostimportantly, all chemopreventive agents used exhibited nopronounced toxicity in low and medium concentrations tonormal cells as tested in primary skin fibroblasts and mesench-ymal stem cells (Supplementary Fig. 4). When used incombination with recombinant TRAIL, an additive effect wasobserved in MIA-PaCa2, BxPc-3 and mesenchymal stem cells,but not in AsPC-1, Capan-1 cells and skin fibroblasts. Ten daysafter treatment, SF and, to a lesser extent, also RE and WO,strongly inhibited clonogenic survival, which was furtherreduced in combination with TRAIL (fig 2C). The results ofthe MTT and clonogenic assays are corroborated by apoptosismeasurement using annexin staining combined with FACSanalysis (fig 3A), detection of caspase activity (fig 3B) anddownregulation of the apoptosis inhibitors XIAP, cIAP1 andFLIP (fig 3C,D). Since SF was most potent in inducing apoptosiseven in resistant TIChigh cell lines, we examined whether thischemopreventive agent might influence the activity of proteinkinase C (PKC), known to have antiapoptotic activity.Therefore, MIA-PaCa2 cells treated with SF, TRAIL or bothagents together were incubated with a PKC inhibitor (Go6983)or a PKC activator (PMA). While inhibition of PKC stronglyincreased apoptosis induction by SF, TRAIL and both agentstogether, activation of PKC strongly diminished it (fig 3E). Thisresult suggests that regulation of the PKC pathway is involvedin TRAIL- and SF-induced apoptosis.

SF reduces binding of transactivation-competent NF-kB dimersin TIChigh cellsSince PKC may lead to NF-kB activity, we analysed the status ofthis transcription factor. EMSA experiments were performedwith extracts of MIA-PaCa2, AsPC-1, Capan-1 and BxPc-3 cells,pretreated with SF and subsequently co-stimulated with TRAILor incubated with TRAIL alone. EMSA with an NF-kB-specificprobe revealed distinct basal DNA-binding activity in TIChigh

and TIClow cell lines (fig 4A). Compared with the negative andpositive control (unstimulated or TNF-induced S107 cells),binding was elevated, indicating an activated NF-kB system inall pancreatic tumour cell lines used. In TIChigh cells, binding oftwo complexes (3 and 2) was observed, of which in particularthe faster migrating band 2 was diminished in MIA-PaCa2 cells

Pancreas

952 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

Figure 1 A CD44+/CD242 phenotype represents a pancreatic tumour-initiating cell (TIC)-like population. (A) The expression of proposed surfacemarkers for TICs was examined by fluorescence-activated cell sorting (FACS) analysis. AsPC-1, Capan-1, MIA-PaCa2 and BxPc-3 cells were doublestained with fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)-conjugated anti-CD44 monoclonal antibody (mAb), PE-conjugated anti-CD133,FITC-conjugated antiepithelial-specific antigen (ESA) or with unconjugated anti-CD24 antibodies (Abs) as indicated. PE- or FITC-conjugated goat anti-mouse immunoglobulin G (IgG) was used as secondary Ab. The results shown in the upper panel are representative of three independent experiments.The lower panel shows means of three experiments including the SD. (B) Larger pictures: cells were seeded at a density of 1600 cells per well in a 6-well plate. Ten days later, colony formation was analysed by fixing the cells in paraformaldehyde and staining with Coomassie blue. Smaller pictures:spheroid formation was induced as described in the Materials and methods section and visualised by microscopy under6200 magnification. (C) In theupper panel, cells were left untreated (CO) or were treated with gemcitabine at the indicated concentrations. After 72 h DNA fragmentation wasmeasured by Nicoletti staining and FACS analysis. In the middle panel, cells were left untreated (CO) or were treated with recombinant tumour necrosisfactor-related apoptosis-inducing ligand (TR: 5, 25, 50 ng/ml) and 24 h later mitochondrial activity/viability was measured by MTT assay. In the lowerpanel, human primary skin fibroblasts (Fibroblasts) or mesenchymal stem cells (MSC) were treated and analysed as described above for the middlepanel. The experiments were performed three times with identical outcome and the means (SD) are shown. (D) MIA-PaCa2 and BxPc-3 cells weretreated with ALDEFLUOR substrate and analysed by flow cytometry for detection of aldehyde dehydrogenase (ALDH) activity. (E) Parental BxPc-3 cellswere repeatedly treated with 25 nM gemcitabine followed by treatment with 50 nM gemcitabine (BxPc-3 GEM). Four days later, expression of CD24and CD44 was examined by FACS analysis of surviving cells. SSC, sideward scatter.

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 953

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

upon SF treatment. In TIClow BxPc-3 cells, we found aprominent complex 2 while complex 3 was missing. TRAILstimulation did not enhance NF-kB DNA-binding activity inMIA-PaCa2 and AsPC-1 cells but was able to induce NF-kB-binding in BxPc-3 (TIClow) and Capan-1 (TIChigh) cells. In allcells, SF pretreatment resulted in inhibition of NF-kB when cellswere co-treated with TRAIL. Interestingly, we could detect apronounced difference between TIChigh and TIClow cell lines.Only BxPc-3 cells showed a prominent faster migrating band 1which was not affected by SF.

To characterice further the composition of complexes 1–3 weperformed supershift assays with specific antibodies recognisingthe various NF-kB subunits (fig 4B, table 2).

Of the three complexes, the fastest migrating band wascompletely shifted by p50-specific antibodies in BxPc-3 (TIClow)cells, indicating that these cells have a high amount of p50/p50homodimers. In all cell lines, complex 2 was composed of p50/RelA and p52/RelA heterodimers. Interestingly, complex 3 wasonly observed in TIChigh cells and was almost completelyreduced by a cRel antibody, suggesting that this binding activityis composed of cRel/cRel homodimers. RelB-containing dimerscould be detected only in BxPc-3 cells. Immunostaining withcRel- or RelA-specific antibody revealed nuclear translocation ofthese NF-kB subunits already in untreated MIA-PaCa2 cells, andneither TRAIL nor SF had an obvious effect on nucleartranslocation as analysed by immunostaining and immuno-fluorescence microscopy (fig 5A) or enhanced expression of IkBaas demonstrated by western blot analysis (fig 5B). To elucidatefurther the mechanisms of SF-induced NF-kB inhibition, weemployed siRNA specific for cRel in MIA-PaCa2 cells to preventexpression and thus binding of cRel. In the presence of cRelsiRNA we found strong reduction of cRel expression, asexamined by western blot analysis (fig 5C) while a nonsensesiRNA oligonucleotide did not affect cRel expression.Transfected cells were analysed for apoptosis in the presenceof SF, TRAIL and both agents together. In line with the resultsabove, we observed in nonsense siRNA-transfected cells specificSF-mediated apoptosis and to a minor extent TRAIL-mediateddeath. Administration of both agents together resulted in thehighest level of apoptosis. In cRel siRNA-transfected cells, SF,TRAIL or both agents together were able to induce apoptosis toan even greater extent, supporting our hypothesis that cRelbinding is a critical event in elevating NF-kB activity andapoptosis resistance in TIChigh cells. As expected, the opposite

result was obtained by inactivation of IkBa with siRNA (fig 5D).In summary, our data provide strong evidence that TIChigh, incontrast to TIClow, cells largely lack transactivation-deficientp50/p50 complexes and are enriched in transactivation-compe-tent p50/RelA and p52/RelA dimers. Importantly, transactiva-tion-competent cRel homodimers are specifically found inTIChigh cells. Thus, we conclude that the apoptosis resistanceof TIChigh cells is most probably due to the enhanced expressionof NF-kB target genes mediating apoptosis resistance.Furthermore, SF reduces the DNA binding of such transactiva-tion-competent NF-kB dimers but not their nuclear localisationand thereby may impair the expression of NF-kB target genesXIAP, cIAP1 and FLIP with antiapoptotic effects (comparefig 3C,D).

SF overcomes resistance of TIChigh tumours xenografted to nudemiceTo address whether SF might influence sensitivity of TIChigh

tumours in vivo, we xenografted MIA-PaCa2 cells subcuta-neously into nude mice and measured tumour growth during aperiod of 7 days in untreated mice or upon treatment with SF,TRAIL or both agents together (fig 6A). While administration ofSF or TRAIL alone resulted in strong growth retardation, co-treatment with SF and TRAIL had an additive effect on theinhibition of tumour growth which was, however, notstatistically significant compared with SF or TRAIL alone.However, SF in the presence of TRAIL totally repressed tumourgrowth during the days of treatment. Most importantly, wefound that SF or TRAIL treatment were only minimallycytotoxic to mice since no change in body weight occurred(fig 6B) but only a transient elevation in plasma levels of LDHand GOT/AST 24 h after treatment with TRAIL but not aftertreatment with SF. LDH, GOT/AST and GPT/ALT were notelevated when measured in mice 5 days after three consecutivetreatments (fig 6C). Since we found no necrotic liver tissue inparaffin sections of perfused mouse livers (data not shown),these data suggest that therapy with SF, TRAIL or both agentstogether is well tolerated in vivo.

To examine, whether SF might influence the IkB status ornuclear translocation of RelA and cRel in xenografts in vivo, weperformed immunohistochemistry on tumour tissue samples.No difference in expression or nuclear localisation of RelA, cRelor IkBa proteins was detectable between xenografts ofuntreated and treated mice (fig 6D). These results underlineour in vitro finding and suggest that SF sensitises cells forTRAIL-induced apoptosis mainly by preventing promoterbinding of transactivation-competent NF-kB complexes.

Our in vivo finding significantly differs from the in vitroresults by the finding that TRAIL prevented growth of TIChigh

MIA-PaCa2 tumour cells much more effectively in the mousexenograft model. This may be due to an antiangiogenic effect ofTRAIL, as recently described.32 We examined vessel density ofxenograft tissue samples by staining with antimouse CD31 andcounting positive cells. TRAIL and SF strongly reduced bloodvessel density, suggesting that both agents have antiangiogenicactivity. Furthermore, application of both agents together hadan additive effect on prevention of blood vessel formation(Supplementary fig 5).

Together, these data demonstrate that SF alone or combinedwith TRAIL potentially reduces growth of tumours enriched inTICs by repression of NF-kB activity, tumour angiogenesis,proliferation and induction of apoptosis without exhibitingtoxicity to normal tissue.

Table 1 TIC characteristics of human pancreatic adenocarcinoma celllines

AsPC-1 MIA-PaCa2 Capan-1 BxPc-3

ATCC no CRL-1682 CRL-1420 HTB 79 CRL-1687

Source Ascites Primarytumour

Livermetastasis

Biopsy primarytumour

Degree ofdifferentiation

Moderate–poor

Poor Moderate Well

Degree of resistance High High–moderate

High Low

Colony-formingcapacity

High High High Low

Spheroid-formingcapacity

Low High Moderate None

ALDH activity 4% 37% 1% 0%

Growth on nude mice Yes Yes Yes Yes, but slow

CD44+/CD242 surfaceexpression

93.5 (5.3) 95.5 (3.7) 95.3 (4.5) 17.5 (5.7)

ALDH, aldehyde dehydrogenase; ATCC, American Type Culture Collection;TIC, tumour-infiltrating cell.

Pancreas

954 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

SF reduces viability of tumour cells from patients harbouringCD44+/CD133+ TICsTo see whether chemopreventive agents may be able toovercome resistance of tumours freshly resected from patientswith pancreatic cancer, we isolated primary tumour cells,treated them with the chemopreventive agents RE, WO or SF,and TRAIL alone, or the chemopreventive agents and TRAIL

together. As measured 48 h later by MTT assay (fig 7A), cellsfrom tumour 2 were sensitive towards chemopreventive agentsand TRAIL, while cells from tumour 1 were only sensitivetowards SF. The high grade of resistance of tumour 1corresponded to a moderate to poorly differentiated ductaladenocarcinoma and the presence of rare populations of CD44+/CD133+ (7.5% of total cellular mass) and CD44+/CD24+ (25% of

Figure 2 Sulforaphane (SF) sensitisestumour-initiating cells (TICs) with nopronounced toxicity to normal cells. MIA-PaCa2 and BxPc-3 cells were pretreatedwith resveratrol (RE), wogonin (WO) orSF in the concentrations indicated. After24 h, tumour necrosis factor-relatedapoptosis-inducing ligand (TR; 5 ng/ml)was added (black bars) or not (whitebars) and (A) cells were stained withCoomassie and photographed, or(B) mitochondrial activity/viability wasanalysed 24 h later by MTT assay.(C) AsPC-1 and MIA-PaCa2 cells wereseeded at a density of 36 104 cells perwell in a 12-well plate and were treatedwith SF (10 mM), RE (50 mM) or WO(50 mM) in the presence or absence of TR(5 ng/ml) as described above. Ten daysafter treatment, clonogenic survival wasexamined by fixing the cells inparaformaldehyde and staining withCoomassie blue. The number of colonieswith .50 cells was counted under adissecting microscope. The numbers ofthe control (CO) groups were set to 1 andthe survival fraction is shown. Theexperiments were performed three timeswith identical outcome and the means(SD) are shown. An asterisk markssignificant difference from thecorresponding control, p,0.05 (t test).

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 955

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

total cellular mass) cells (fig 7B). These TIC marker-expressingpopulations were also present in carcinoma-free tissue from twoother patients with chronic pancreatitis. Patient and tumourcharacteristics are summarised in table 3.

Since our in vitro data demonstrate enhanced NF-kB activityas a reason for resistance, we examined expression andlocalisation of IkBa and cRel proteins in untreated tissuesections of tumours 1 and 2. IkBa and cRel expression wasdetectable and no obvious difference was observed betweentumour 1 expressing TIC markers and tumour 2 withoutmarkers (fig 7C). These data confirm our hypothesis thatenhanced binding of cRel and not necessarily expression or

translocation of NF-kB subunits is the underlying reason forenhanced NF-kB activity in TIC-like tumour cells.

DISCUSSIONOur report provides evidence that treatment resistance of TICsis due to increased binding of transactivation-competent NF-kBcomplexes, which prevent induction of apoptosis. We demon-strate that chemopreventive agents such as WO, RE and SF areable to re-sensitise TICs by interfering with NF-kB activity toconfer reduced vitality and clonogenicity, induction of apoptosisand caspase activity. This effect is strongest for SF and ourstudy is the first to demonstrate that SF can selectively

Figure 3 Sulforaphane (SF) sensitises tumour-initiating cells (TICs) by inducing apoptosis and caspase activity. Pancreatic cancer cell lines were leftuntreated (CO) or were treated with (RE), wogonin (WO) or SF alone or in combination with tumour necrosis factor-related apoptosis-inducing ligand(TRAIL) as described in fig 2C. (A) Apoptosis was measured by annexin staining and flow cytometry. The percentage of specific apoptosis wascalculated as follows: 1006(experimental apoptosis (%)–spontaneous apoptosis in the control (%))/(100%–spontaneous apoptosis in the control (%)).The experiments were performed three times with identical outcome and the SD is shown. (B) MIA-PaCa2 cells were left untreated (CO), or weretreated with SF (10 mM). After 24 h, TRAIL (TR, 5 ng/ml) was added to untreated (TR) or pre-treated cells (SF+TR) and the activity of caspase 8, 9 and3+7 was analysed by fluorochrome-linked inhibitors of caspases (FLICA). Fluorescence was detected by fluorescence microscopy using an OlympusIX70 microscope at a magnification of 6400. The shape of the cells under phase contrast microscopy at a magnification of 6200 is shown. (C) XIAP,cIAP1 and FLIP expression was detected by western blot analysis in untreated MIA-PaCa2 and BxPc-3 cells or in MIA-PaCa2 cells treated with SF forthe times indicated. (D) XIAP protein expression in MIA-PaCa2 and BxPc-3 cells upon treatment with SF (10 mM) for 48 h, TR (5 ng/ml) for 24 h or acombination of both agents together (SF+TR). (E) MIA-PaCa2 cell were pretreated with the protein kinase C (PKC) inhibitor Go6983 (20 mM) for 45 minor with the PKC activator phorbol 12-myristate 13-acetate (PMA; 200 nM) for 15 min followed by treatment with SF, TR or SF plus TR as described infig 2. Apoptosis was measured by annexin staining and flow cytometry. CASP, caspase.

Pancreas

956 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

potentiate apoptotic effects in TICs without exhibitingpronounced cytotoxic side effects.

Definition of TIC populations in pancreatic carcinomaWe found .90% of a CD44+/CD242 population in highlyresistant established pancreatic cancer cell lines MIA-PaCa2,AsPC-1 and Capan-1 but only 17.5% in sensitive BxPc-3 cells.This CD44+/CD242 population corresponds to a similar expres-sion pattern suggested for TICs detected in aggressive breastcancer.2 A percentage of .90% TICs within established tumourcell lines is extremely high but mimics the aggressive phenotypeof pancreatic cancer. Another research group identified the samepercentage of 90% CD133+ TICs in the established human livercancer cell line Hep3B.33 A strong indication for the TIC-likenature of the CD44+/CD242 population in our study is thatTIChigh pancreatic cancer cell lines AsPC-1 and Capan-1 wereoriginally isolated from low differentiated patient tumour cellsderived from ascites or metastases. These cells, together withMIA-PaCa2, but not BxPc-3 have a high ability to form coloniesand spheroids, are ALDH positive and highly treatmentresistant. In contrast, the TIClow BxPc-3 pancreatic cancer cell

line harbouring only 17.5% of CD44+/CD242 cells was isolatedfrom a biopsy of a patient tumour of a well-differentiatedpancreatic carcinoma and this pathological classification may bea hint as to why they contain a low amount of TICs.Accordingly, we found in BxPx-3 cells a low degree of resistance,low colony formation, no ALDH activity and no sphereformation. However, gemcitabine treatment of BxPc-3 cellsled to enrichment of the CD44+/CD242-expressing cellscombined with enhanced treatment resistance. This findingsupports the recent idea that cytotoxic cancer treatment mayeliminate the majority of normal tumour cells, but not thehighly resistant TICs, which survive cytotoxic treatment andare enriched in this way.34 The TIC characteristics are furtherunderlined by the documented capacity of all our pancreatic celllines used to grow in immunodeficient mice.24–26 Notably, BxPc-3 cells were found to grow very slowly,27 and this was confirmedin our experiments, where 103 BxPc-3 cells needed 28 days toform xenograft tumours in nude mice while the same amountof MIA-PaCa2 cells formed tumours in 17 days. A hint at themultipotency of TIChigh pancreatic cancer cell lines is thereported finding that xenografted Capan-1 and AsPC-1 cellsdifferentiate into phenotypically diverse populations whichshow morphological, biological and biochemical characteristicssimilar to the tumour of origin.28 29 These data are supported byour in vitro spheroid-forming assays: only part of the whole cellpopulation of Capan-1, AsPC-1 and MIA-PaCa2 acquired theability to grow detached in spheroids; other cells from the sameculture did not and continued to grow adherent. These resultsstrongly hint at the differentiation of TICs to daughter cellswith different properties. However, additional markers seem tobe expressed by pancreatic TICs. ABCG2 is such a candidate

Figure 4 Sulforaphane (SF) inhibits binding of transactivation-competent nuclear factor-kB (NF-kB) dimers in tumour-initiating cell (TIC)high cells.(A) Pancreatic cancer cell lines were left untreated (CO) or were treated with SF (10 mM) for 24 h or for the times indicated. Twenty-four hours latertumour necrosis factor-related apoptosis-inducing ligand (TRAIL) (TR, 5 ng/ml) was added to non SF-pretreated cells (TR) or to cells pretreated with SF(SF+TR) and incubated for an additional 24 h or for the times indicated. Untreated murine plasmocytoma S107 cells served as negative control(NegCo). S107 cells treated with recombinant tumour necrosis factor served as positive control (PosCo). After incubation, nuclear proteins wereprepared and DNA binding was analysed by electrophorstic mobility shift assay (EMSA) using a specific 32P-labelled oligonucleotide probe for NF-kB.Three different bands became visible corresponding to three different NF-kB subunits binding to the oligonucleotide (1, 2 and 3). (B) Nuclear proteinsderived from untreated control cells were preincubated for 30 min with specific antibodies followed by EMSA analysis. In some cases, two differentantibodies for the same proteins were used: RelA/RelA(2), cRel/cRel(2), p50/p50(2), p52/p52(2). Broken circles mark the disappearance of shiftedcomplexes due to specific interaction with co-incubated antibodies. The arrows mark a shift of RelB exclusively observed in BxPc-3 cells.PI, preimmune serum.

Table 2 Binding of different transactivation-competent NF-kBcomplexes in TIChigh and TIClow cells

Cell line Complex 1 Complex 2 Complex 3

TIChigh AsPC-1 np p50, p52, RelA cRel

MIA-PaCa2 np p50, p52, RelA cRel

Capan-1 np p50, p52, RelA cRel

TIClow BxPc-3 p50 p50, p52, RelA, RelB np

NF-kB, nuclear factor-kB; np, not present; TIC, tumour-initiating cell.

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 957

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

marker since this high capacity drug transporter mediates effluxof fluorescent dyes such as rhodamine 123 or Hoechst 33342which often correlates with a multidrug-resistant phenotype.4 35

In the same way, the pancreatic cell lines Panc1, Panc89,Colo357, PancTu1 and A818-6 expressed the ABCG2 protein onthe cell surface.4 Additionally, PancTu1 and A818-6 cells isolatedfrom patient tumours with a poor differentiation grade werepositive for CD133. These data are confirmed by Hermann andcolleagues,6 who found a small population of CD133+ cells onthe established human pancreatic cancer cell lines L3.6pl andMIA-PaCa2 and were able to link CD133 expression toinvasiveness. We did not examine ABCG2 expression and couldnot detect expression of CD133 in our pancreatic cancer celllines. This might be due to differences in antibodies available.However, the CD133 antibody used in our studies is functionaland gave positive staining in paraffin sections of three of fourexamined tumour/pancreatitis tissue samples from patients.This higher expression of CD133 in tissue could be bestexplained by clonal selection during conditions of prolongedcell culture and by the missing stem cell niche in vitro. Weassume that CD133 expression might have been lost undernormoxic in vitro conditions, since it has been reported thatCD133 expression as well as invasiveness can be induced byhypoxia.36 In view of the fact that a microenvironment of lowoxygen is a pronounced feature of pancreatic cancer,37 hypoxiamight be the reason for expression of CD133 in patient tumoursbut not in established cells lines. In addition to expression ofCD133, we found CD44+/CD24+ and CD133+/CD44+ double-positive cells in three of four examined tissues. The CD44+/CD24+ phenotype differs from the CD44+/CD242 phenotypewhich we identified to characterise a TIC population in vitro.We assume that the in vivo CD44+/CD24+ population comprisesTICs as well as non-TICs, and cell culture conditions may haveled to loss of CD24 expression in the TIC population.Interestingly, primary fresh cells isolated from marker-positivetissues were totally resistant towards ex vivo treatment withTRAIL but could be sensitised by SF. Two very recent studies inpancreatic cancer have utilised a marker combination of CD44+/CD24+/ESA+5 and of CD133+/CXCR4+6 in patient-derivedtumours after serial transplantation on mice. FACS analysisshowed that these two populations overlap but are notidentical.6 We did not examine expression of ESA in patienttumours, due to a lack of correlation with a TIC phenotype inour in vitro experiments and the correlation between ESA andCD24 expression. Interestingly, CD44+ cells have been shown torepresent a highly tumourigenic TIC-like population in breast,prostate, and head and neck cancer.2 38 39 In this context, avariant of glycoprotein CD44 has been known since 1991 to beexpressed only in metastasising cell lines and whose overexpres-sion was sufficient to establish full metastatic behaviour of a non-metastasising pancreatic cancer cell line.40 The same CD44 variantis expressed on B and T lymphocytes and is required for thelymphatic spread of tumour cells.41 This finding supports therecent idea that metastasising tumour cells, which most probablyrepresent TICs, may mimic lymphocyte behaviour.

Binding of distinct NF-kB subunits in TIChigh and TIClow cellsFrom basic studies with highly resistant tumour cells, we knowthat two types of events are necessary to induce cell death:(1) inhibition of NF-kB-dependent survival signals and(2) activation of the cellular stress response.42 In several studies

Figure 5 Sulforaphane (SF) does not influence translocation of nuclearfactor-kB (NF-kB) and degradation of IkBa (inhibitor of NF-kB). (A) MIA-PaCa2 cells were left untreated (CO) or were treated with tumournecrosis factor-related apoptosis-inducing ligand (TRAIL) (TR, 5 ng/ml)or SF (10 mM), and 6 h later expression of cRel and RelA was analysedby immunocytochemistry at a magnification of6400 or61000 (inserts).(B) IkBa expression was detected in MIA-PaCa2 and BxPc-3 cellstreated as described above by western blot analysis. (C and D) MIA-PaCa2 cells were transfected with non-specific (NS) or specific smallinterfering RNAs (siRNAs) directed towards IkBa or cRel. Five dayslater, cells were either treated with SF alone for 48 h or pretreated withSF for 24 h, followed by co-treatment with TRAIL for an additional 24 h.Specific apoptosis was determined as described in fig 3A. For detectionof siRNA-mediated inhibition of protein expression, cellular proteins wereharvested 5 days after transfection. Expression of IkBa and cRel wasexamined by western blot analysis, and b-actin served as internalstandard.

Pancreas

958 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

Figure 6 Sulforaphane (SF) and tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) are effective and safe in vivo. (A) MIA-PaCa2 cellswere injected subcutaneously into nude mice. After the tumours had reached a mean diameter of ,8–10 mm, treatment was started. Mice receivedSF, TRAIL (TR) or both agents together at days 4–6 after implantation as described in the Materials and methods section. The tumour volumes weremeasured daily for the duration of the experiment. Relative increase in tumour size was calculated as follows: 1006(experimental tumour size–initialtumour size)/initial tumour size. Mice were euthanised on day 8. Data are presented as the mean (SE) of six animals. An asterisk marks a significantdifference from control (CO), p,0.05 (t test). In addition, tumour growth curves were analysed according to Koziol et al23, and the p value for CO vsSF+TR was 0.007. (B) Mice were weighed during the experiment. The body weight of each individual mouse was set to 100% before treatment andbody weight after treatment was correlated to initial weight. (C) Heparinised blood of mice was examined 24 h after a single treatment with TR, SF orboth agents together. Likewise, blood from mice treated three times with TR, SF or both agents together was examined 5 days after the last treatment(120 h). The levels of plasma lactate dehydrogenase (LDH), glutamate-oxalacetate-transaminase (GOT/AST) and glutamate-pyruvate-transaminase(GPT/ALT) were measured using a DRY-CHEM FCD3500. (D) Twenty-four hours after single treatment of mice, xenograft tumour tissue samples wereanalysed by immunohistochemistry for expression of IkBa, cRel and RelA (6200 or 6400 (inserts)).

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 959

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

Figure 7 (A) Tumour cells from patients with pancreatic cancer were isolated immediately after resection from tissue and cultivated in 96-well platesat 56105/ml. As indicated, cells were left untreated (CO) or were treated with resveratrol (RE; 50 mM), wogonin (WO; 50 mM) or sulforaphane (SF;10 mM). After 24 h, recombinant tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) was added to untreated (TR, 5 ng/ml), RE-treated(TR+RE), WO-treated (TR+WO) or SF-treated (TR+SF) cells and 48 h later mitochondrial activity/viability was measured by MTT assay. Due tolimitations in the amount of patient material, experiments were performed only once but in octuplicate, and the SD is shown. An asterisk marks asignificant difference from the corresponding control, p,0.05 (t test). (B) Paraffin-embedded pancreatic tumour tissue sections were subjected todouble immunofluorescence staining. CD24 expression was detected by mouse monoclonal antibodies (mAbs) and goat antimouse Alexa 488-conjugated immunoglobulin G (IgG; green). CD44 expression was analysed using mouse mAb and visualised by goat antimouse Alexa 594–IgG (red).CD44+/CD24+ cells were detected by rabbit anti-CD44 mAb with goat anti-rabbit Alexa 594–IgG (red) and CD24 mouse mAb with goat antimouse Alexa488–IgG (green). CD133 expression was detected by rabbit polyclonal antibody (Ab) and visualised by Alexa 594-conjugated goat antirabbit IgG (red).For double staining, mouse mAb CD44 conjugated to goat antimouse Alexa 488–IgG (green) was used. (C) IkBa and cRel were detected in tumoursamples by regular immunohistochemistry with rabbit polyclonal Abs, and 3,39 diaminobenzidine (DAB) was used as a chromogen. Tissue sampleswere analysed under 6400 magnification and insets are shown under 61000 magnification. Sec Ab CO, secondary antibody control.

Pancreas

960 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

we have demonstrated that chemotherapy or radiation inducesthe cellular stress response, which, however, alone may not beeffective to induce apoptosis in a cellular environment ofincreased NF-kB activity. Elevated NF-kB DNA-binding activityof transactivation-competent dimers may be involved in thepronounced resistance in TICs. By performing EMSA, weidentified binding of distinct NF-kB complexes in TIChigh andTIClow pancreatic carcinoma cell lines. In supershift experimentswe documented that complex 3, most probably composed oftransactivation-potent cRel homodimers, binds only in TIChigh

cells. In contrast, complex 1, which consists of transactivation-inactive p50 proteins, was selectively found in TIClow cells.Complex 2, which represents RelA-containing heterodimers, isdetected in both TIChigh and TIClow cells, with a higherabundance in TIClow cells. Taken together, these data suggestthat in TIClow cells competition between transactivation-competent and -incompetent NF-kB dimers exists, which mostprobably leads to a lower level of NF-kB-dependent genetranscription. In contrast, in TIChigh cells, this competition israther low, leading to enhanced NF-kB target gene expression,such as that found in the present study for XIAP, cIAP and FLIP(see fig 3; and data not shown).

SF inhibits DNA binding of transactivating NF-kB in TIChigh cellsMany plant products have been proven to inhibit NF-kBtogether with inducing apoptosis of tumour cells. NF-kB is atarget of SF, a natural isothiocianate found in high concentra-tion in broccoli. We found impaired DNA binding of transacti-vation-potent NF-kB subunits in SF-treated cell lines and thesedata were confirmed by SF-mediated downregulation ofexpression of antiapoptotic NF-kB target genes. Since we couldenhance SF-induced apoptosis by siRNA-mediated downregula-tion of cRel or, vice versa, were able to block SF-inducedapoptosis completely by additional activation of NF-kB viasiRNA-mediated inhibition of IkBa, NF-kB appears to be the keystep by which SF abrogates resistance of TICs. Quite similarresults were demonstrated in murine macrophages,43 where SFselectively reduced DNA binding of NF-kB without interferingwith nuclear translocation of NF-kB. Moreover, a recent reportindirectly confirms our finding that SF is able to sensitise TICs.In that study, SF is described to induce death in chemotherapy-and TRAIL-resistant Hep3B hepatoma cells.44 This is the samecell line which has just been described to harbour 90% ofCD133+ TICs.33 The suitability of SF for eradication of TICs isunderscored by our experiments in which we could alsosensitise TIChigh pancreatic carcinoma cells to TRAIL-inducedapoptosis by pretreament with RE or WO, although the effectswere less pronounced than those we observed with SF. Mostimportantly, another recent publication suggests that the plant-derived compound parthenolide from traditional Mexican

medicine is useful to kill leukaemia and AML (acute myeloidleukaemia) stem cells while sparing normal haematopoieticcells. Similarly to our results, parthenolide eradicated TICs viainhibition of NF-kB without obvious side effects on normalcells.45

Involvement of PKC in SF-mediated apoptosisTrauzold et al46 described a TRAIL receptor-mediated activationof PKC in the pancreatic adenocarcinoma cell line PancTuI,which contributes to resistance through inhibition of themitochondrial apoptosis pathway and activation of NF-kB.From this point of view, it was interesting to evaluate whetherSF overcomes resistance of TIChigh cells upon activation of PKC.As expected, SF-mediated apoptosis was reduced in the presenceof the PKC activator PMA, but not, however, completelyblocked. These data indicate that SF-mediated apoptosisinvolves a block of PKC, which may result in the observeddownregulation of NF-kB signalling and antiapoptotic targetgenes.

In vivo applicability of TRAIL and SFWe used TRAIL in our study for targeting of TICs since clinicalevaluation in a phase I study confirmed that TRAIL is welltolerated by patients with advanced solid malignancies such ascolorectal, lung, prostate, ovarian or renal cancer.47 These resultsare reflected by our data, in which TRAIL was non-toxic tofibroblasts and mesenchymal stem cells while it inducedapoptosis and reduced viability in TIClow, but not in TIChigh

pancreatic cancer cell lines after single treatment. Only afterprolonged culture in clonogenic assays was TRAIL-inducedcytotoxicity seen in TIChigh cells. Accordingly, in our studieswith TIChigh cells xenografted to nude mice, repeated intratu-moural application of recombinant TRAIL protein was able toinhibit tumour growth without apparent significant effects onbody weight or liver function. In contrast, a recent reportsuggests that TRAIL promotes invasion and metastasis in anapoptosis-resistant orthotopically xenografted Colo357 pancrea-tic ductal adenocarcinoma whose growth has been observedduring 40 days.48 In our subcutaneous MIA-PaCa2 xenograftmodel no micrometastasis could be detected during theobservation period of 10 days. This was obvious from analysisof human EpCAM (epithelial cell adhesion molecule) andhuman major histocompatibility complex (MHC) class I. Nopositive cells were seen in the liver, spleen and lungs ofuntreated or treated xenografted mice as examined byimmunohistochemistry and flow cytometry (data not shown).The difference between our findings and the results of Trauzoldet al48 may be due to different models used and to the fact thatmetastasis in subcutaneous mouse xenograft models used by usis rather untypical. However, TRAIL alone did not completely

Table 3 Characteristics of patient ductal adenocarcinomas

No Gender Age TNM Metastasis Differentiation

Ex vivoTRAILresistance

Marker expression

CD44+/CD24+ CD44+/CD133+

1 F 69 T3, N1 1/8 Moderate–poor Total + +2 F 68 T3, N1 4/24 Well–moderate Moderate – –

3 M 32 Carcinoma-free, chronic pancreatitis Total + +4 M 58 Hyperplasia, chronic pancreatitis Total + +

Resistance was measured ex vivo with freshly isolated tumour cells treated with TRAIL.Surface marker expression was determined in paraffin sections of tumour tissue by immunostaining.F, female; No, number of patient tumour; TNM, cancer staging system (T3 = extent of the primary tumour is .5 cm; N1 = tumourcells spread to closest or a small number of regional lymph nodes); TRAIL, tumour necrosis factor-related apoptosis-inducingligand.

Pancreas

Gut 2009;58:949–963. doi:10.1136/gut.2008.149039 961

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

eliminate pancreatic TICs after three consecutive injections invivo. A more pronounced reduction of tumour growth wasobtained with SF in combination with TRAIL. This is mostprobably due to the SF-mediated strong inhibition of NF-kBbinding to DNA. Thus, SF itself was quite effective ineradication of TICs and increased the antitumour effects ofTRAIL. Besides, our results in mice reflect data of a clinicalphase I study showing that daily consumption of 100 mmol ofglucosinolate, the precursor of SF, is well tolerated and has noadverse side effects—for example, on body weight or liverfunction.49 Most importantly, another prospective study of fruitand vegetable intake in 1338 patients with prostate canceramong 29 361 men (average follow-up 4.2 years) demonstratedthat among several phytonutrients only a high intake ofcruciferous vegetables, including broccoli and cauliflower, wasassociated with reduced risk of aggressive prostate cancer,particularly extraprostatic disease.50

In conclusion, these findings indicate the effectiveness of SFin eradication of pancreatic TICs and its ability to potentiatethe antitumour effects of TRAIL by repression of NF-kB activityand downregulation of expression of antiapoptotic genes,leading to the inhibition of viability, clonogenicity and tumourgrowth, with inhibition of angiogenesis and induction ofcaspase activity followed by apoptosis. The dose of SF(4.4 mg/kg/day) used in the current animal studies is quiterelevant to that used in human subjects. Therefore, the combina-tion of SF with, for example, gemcitabine and/or TRAIL, hassignificant potential as an effective therapy for pancreatic cancerthat can overcome chemoresistance of TICs and enhance thetherapeutic effect. Further clinical studies are necessary to confirmour findings in patients with pancreatic cancer.

Acknowledgements: We thank Dr W Rittgen for statistical analysis, Dr R Saffrich forhelp with immunofluorescence microscopy, and Dr H-J Stark for providing humanprimary skin fibroblasts. This study was supported by grants from theBundesministerium fur Bildung und Forschung (IH: 01GU0611), TumorzentrumHeidelberg/Mannheim (IH: D10027(6)350), Stiftung Chirurgie Heidelberg (IH) andDeutsche Krebshilfe (IH: 107254), and by the Deutsche Forschungsgemeinschaft (BB:SFB 518/A17).

Competing interests: None.

Ethics approval: Patient material was obtained with the approval of the ethicalcommittee of the University of Heidelberg.

REFERENCES1. Neoptolemos JP, Dunn JA, Stocken DD, et al. Adjuvant chemoradiotherapy and

chemotherapy in resectable pancreatic cancer: a randomised controlled trial. Lancet2001;358:1576–85.

2. Al-Hajj M, Wicha MS, Benito-Hernandez A, et al. Prospective identification oftumorigenic breast cancer cells. Proc Natl Acad Sci USA 2003;100:3983–8.

3. Singh SK, Hawkins C, Clarke ID, et al. Identification of human brain tumour initiatingcells. Nature 2004;432:396–401.

4. Olempska M, Eisenach PA, Ammerpohl O, et al. Detection of tumor stem cell markersin pancreatic carcinoma cell lines. Hepatobiliary Pancreat Dis Int 2007;6:92–7.

5. Li C, Heidt DG, Dalerba P, et al. Identification of pancreatic cancer stem cells. CancerRes 2007;67:1030–7.

6. Hermann PC, Huber SL, Herrler T, et al. Distinct populations of cancer stem cellsdetermine tumor growth and metastatic activity in human pancreatic cancer. CellStem Cell 2007;1:313–23.

7. Clarke MF, Dick JE, Dirks PB, et al. Cancer stem cells—perspectives on current status andfuture directions: AACR workshop on cancer stem cells. Cancer Res 2006;66:9339–44.

8. Kim CF, Jackson EL, Woolfenden AE, et al. Identification of bronchioalveolar stemcells in normal lung and lung cancer. Cell 2005;121:823–35.

9. Ricci-Vitiani L, Lombardi DG, Pilozzi E, et al. Identification and expansion of humancolon-cancer-initiating cells. Nature 2007;445:111–5.

10. O’Brien CA, Pollett A, Gallinger S, et al. A human colon cancer cell capable ofinitiating tumour growth in immunodeficient mice. Nature 2007;445:106–10.

11. Jones RJ, Matsui WH, Smith BD. Cancer stem cells: are we missing the target?J Natl Cancer Inst 2004;96:583–5.

12. Ibrahim SM, Ringel J, Schmidt C, et al. Pancreatic adenocarcinoma cell lines showvariable susceptibility to TRAIL-mediated cell death. Pancreas 2001;23:72–9.

13. Aggarwal BB. Nuclear factor-kappaB: the enemy within. Cancer Cell 2004;6:203–8.

14. Nakanishi C, Toi M. Nuclear factor-kappaB inhibitors as sensitizers to anticancerdrugs. Nat Rev Cancer 2005;5:297–309.

15. Juge N, Mithen RF, Traka M. Molecular basis for chemoprevention by sulforaphane:a comprehensive review. Cell Mol Life Sci 2007;64:1105–27.

16. Fulda S, Debatin KM. Resveratrol modulation of signal transduction in apoptosis andcell survival: a mini-review. Cancer Detect Prev 2006;30:217–23.

17. Fas SC, Baumann S, Zhu JY, et al. Wogonin sensitizes resistant malignant cells toTNFalpha- and TRAIL-induced apoptosis. Blood 2006;108:3700–6.

18. Zhang C, Kolb A, Buchler P, et al. Corticosteroid co-treatment induces resistance tochemotherapy in surgical resections, xenografts and established cell lines ofpancreatic cancer. BMC Cancer 2006;6:61.

19. Kallifatidis G, Beckermann B, Groth A, et al. Improved lentiviral transduction ofmesenchymal stem cells for therapeutic intervention in pancreatic cancer. CancerGene Ther 2007;4;231–234.

20. Maier HJ, Marienfeld R, Wirth T, et al. Critical role of RelB serine 368 fordimerization and p100 stabilization. J Biol Chem 2003;278:39242–50.

21. Lernbecher T, Muller U, Wirth T. Distinct NF-kappa B/Rel transcription factors areresponsible for tissue-specific and inducible gene activation. Nature 1993;365:767–70.

22. Weber E, Lehmann HP, Beck-Sickinger AG, et al. Antibodies to the protein core ofthe small cell lung cancer workshop antigen cluster-w4 and to the leucocyteworkshop antigen CD24 recognize the same short protein sequence leucine-alanine-proline. Clin Exp Immunol 1993;93:279–85.

23. Koziol JA, Maxwell DA, Fukushima M, et al. A distribution-free test for tumor growthcurve analyses with application to an animal tumor immunotherapy experiment.Biometrics 1981;37:383–90.

24. Buchler P, Reber HA, Ullrich A, et al. Pancreatic cancer growth is inhibited byblockade of VEGF-RII. Surgery 2003;134:772–82.

25. Nozawa F, Itami A, Saruc M, et al. The combination of tumor necrosis factor-relatedapoptosis-inducing ligand (TRAIL/Apo2L) and Genistein is effective in inhibitingpancreatic cancer growth. Pancreas 2004;29:45–52.

26. Karikari CA, Roy I, Tryggestad E, et al. Targeting the apoptotic machinery inpancreatic cancers using small-molecule antagonists of the X-linked inhibitor ofapoptosis protein. Mol Cancer Ther 2007;6:957–66.

27. Raut CP, Takamori RK, Davis DW, et al. Direct effects of recombinant humanendostatin on tumor cell IL-8 production are associated with increased endothelial cellapoptosis in an orthotopic model of human pancreatic cancer. Cancer Biol Ther2004;3:679–87.

28. Kyriazis AP, Kyriazis AA, Scarpelli DG, et al. Human pancreatic adenocarcinoma lineCapan-1 in tissue culture and the nude mouse: morphologic, biologic, and biochemicalcharacteristics. Am J Pathol 1982;106:250–60.

29. Chen WH, Horoszewicz JS, Leong SS, et al. Human pancreatic adenocarcinoma: invitro and in vivo morphology of a new tumor line established from ascites. In Vitro1982;18:24–34.

30. Ozawa F, Friess H, Kleeff J, et al. Effects and expression of TRAIL and its apoptosis-promoting receptors in human pancreatic cancer. Cancer Lett 2001;163:71–81.

31. Matsuzaki H, Schmied BM, Ulrich A, et al. Combination of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and actinomycin D induces apoptosis evenin TRAIL-resistant human pancreatic cancer cells. Clin Cancer Res 2001;7:407–14.

32. Cantarella G, Risuglia N, Dell’eva R, et al. TRAIL inhibits angiogenesis stimulated byVEGF expression in human glioblastoma cells. Br J Cancer 2006;94:1428–35.

33. Ma S, Chan KW, Hu L, et al. Identification and characterization of tumorigenic livercancer stem/progenitor cells. Gastroenterology 2007;132:2542–56.

34. Abbott A. Cancer: the root of the problem. Nature 2006;442:742–3.35. Patrawala L, Calhoun T, Schneider-Broussard R, et al. Side population is enriched in

tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2– cancer cells aresimilarly tumorigenic. Cancer Res 2005;65:6207–19.

36. Blazek ER, Foutch JL, Maki G. Daoy medulloblastoma cells that express CD133 areradioresistant relative to CD133– cells, and the CD133+ sector is enlarged byhypoxia. Int J Radiat Oncol Biol Phys 2007;67:1–5.

37. Brown JM, Giaccia AJ. The unique physiology of solid tumors: opportunities (andproblems) for cancer therapy. Cancer Res 1998;58:1408–16.

38. Patrawala L, Calhoun T, Schneider-Broussard R, et al. Highly purified CD44+prostate cancer cells from xenograft human tumors are enriched in tumorigenic andmetastatic progenitor cells. Oncogene 2006;25:1696–708.

39. Prince ME, Sivanandan R, Kaczorowski A, et al. Identification of a subpopulation ofcells with cancer stem cell properties in head and neck squamous cell carcinoma.Proc Natl Acad Sci USA 2007;104:973–8.

40. Gunthert U, Hofmann M, Rudy W, et al. A new variant of glycoprotein CD44 confersmetastatic potential to rat carcinoma cells. Cell 1991;65:13–24.

41. Arch R, Wirth K, Hofmann M, et al. Participation in normal immune responses of ametastasis-inducing splice variant of CD44. Science 1992;257:682–5.

42. Herr I, Debatin K-M. Cellular stress response and apoptosis in cancer therapy. Blood2001;98:2603–14.

43. Heiss E, Herhaus C, Klimo K, et al. Nuclear factor kappa B is a molecular target forsulforaphane-mediated anti-inflammatory mechanisms. J Biol Chem2001;276:32008–15.

44. Kim H, Kim EH, Eom YW, et al. Sulforaphane sensitizes tumor necrosis factor-relatedapoptosis-inducing ligand (TRAIL)-resistant hepatoma cells to TRAIL-inducedapoptosis through reactive oxygen species-mediated up-regulation of DR5. CancerRes 2006;66:1740–50.

45. Guzman ML, Rossi RM, Karnischky L, et al. The sesquiterpene lactone parthenolideinduces apoptosis of human acute myelogenous leukemia stem and progenitor cells.Blood 2005;105:4163–9.

Pancreas

962 Gut 2009;58:949–963. doi:10.1136/gut.2008.149039

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from

46. Trauzold A, Wermann H, Arlt A, et al. CD95 and TRAIL receptor-mediated activationof protein kinase C and NF-kappaB contributes to apoptosis resistance in ductalpancreatic adenocarcinoma cells. Oncogene 2001;20:4258–69.

47. Tolcher AW, Mita M, Meropol NJ, et al. Phase I pharmacokinetic and biologiccorrelative study of mapatumumab, a fully human monoclonal antibody with agonistactivity to tumor necrosis factor-related apoptosis-inducing ligand receptor-1. J ClinOncol 2007;25:1390–5.

48. Trauzold A, Siegmund D, Schniewind B, et al. TRAIL promotes metastasis of humanpancreatic ductal adenocarcinoma. Oncogene 2006;25:7434–9.

49. Shapiro M, Wasik MA, Junkins-Hopkins JM, et al. Complete remission in advancedblastic NK-cell lymphoma/leukemia in elderly patients using the hyper-CVAD regimen.Am J Hematol 2003;74:46–51.

50. Kirsh VA, Peters U, Mayne ST, et al. Prospective study of fruit and vegetable intakeand risk of prostate cancer. J Natl Cancer Inst 2007;99:1200–9.

Robin Spiller, editor

Gastric polypoid lesion at the antrum

CLINICAL PRESENTATIONA 79-year-old man underwent upper endoscopy 5 months agobecause of dyspepsia and anaemia, with haemoglobin of 7.2 g/dl.Upper endoscopy revealed an ulcer 12 mm in diameter at theantrum (fig 1A) and biopsies showed benign ulceration and werenegative for Helicobacter pylori infection. The patient was treatedwith a 4-month course of omeprazole with relief of symptoms.After the treatment, endoscopy was performed and revealed a15 mm diameter, broad-based polypoid lesion with a villous-appearing surface at the previous ulcer site at the antrum (fig 1B).Endoscopic ultrasonography showed a hypoechoic homogenouslesion within the first and secondary echo layer of the stomach(fig 1C). Endoscopic mucosectomy was carried out for this lesion.Colonoscopy showed no abnormality.

QUESTIONSWhat pathology does the endoscopic image suggest? What isthe diagnosis?

See page 1024 for the answer.This case is submitted by:

C-J Chen,1 C-W Chang,1 H-L Chen,1 Y-J Chan,2 M-J Chen1

1 Division of Gastroenterology, Department of Internal Medicine, Mackay MemorialHospital and Mackay Medicine, Nursing and Management College, Taipei, Taiwan;2 Department of Pathology, Mackay Memorial Hospital, Taipei, Taiwan

Correspondence to: Dr M-J Chen, Division of Gastroenterology, Department ofInternal Medicine, Mackay Memorial Hospital, No 92, Sec 2, Chung-Shan North Road,Taipei, Taiwan; [email protected]

Competing interests: None.

Patient consent: Obtained.

Gut 2009;58:963. doi:10.1136/gut.2008.172551

Figure 1 Upper endoscopy revealed (A) an ulcer 12 mm in diameter at the antrum and (B) a 15 mm diameter broad-based polypoid lesion with avillous-appearing surface at the site of the previous ulcer after treatment. (C) Endoscopic ultrasonography showed a hypoechoic homogenous lesionwithin the first and secondary layer of the stomach.

Editor’s quiz: GI snapshot

Pancreas

Gut July 2009 Vol 58 No 7 963

group.bmj.com on September 28, 2010 - Published by gut.bmj.comDownloaded from