Original Articles

Lobatin B inhibits NPMALK and NF-κB attenuating anaplastic-large-cell-lymphomagenesis and lymphendothelial tumour intravasationIzabella Kiss ab1 Christine Unger a1 Chi Nguyen Huu b Atanas Georgiev Atanasov cNina Kramer a Waranya Chatruphonprasert de Stefan Brenner d Ruxandra McKinnon cAndrea Peschel b Andrea Vasas f Ildiko Lajter f Renate Kain b Philipp Saiko gThomas Szekeres g Lukas Kenner bhi Melanie R Hassler b Rene Diaz j Richard Frisch jVerena M Dirsch c Walter Jaumlger d Rainer de Martin k Valery N Bochkov lClaus M Passreiter m Barbara Peter-Voumlroumlsmarty n Robert M Mader o Michael Grusch nHelmut Dolznig a Brigitte Kopp c Istvan Zupko p Judit Hohmann f Georg Krupitza ba Institute of Medical Genetics Medical University of Vienna Waehringer Strasse 10 A-1090 Vienna Austriab Clinical Institute of Pathology Medical University of Vienna Waehringer Guertel 18-20 A-1090 Vienna Austriac Department of Pharmacognosy University of Vienna Althanstrasse 14 A-1090 Vienna Austriad Department of Clinical Pharmacy and Diagnostics University of Vienna Althanstrasse 14 A-1090 Vienna Austriae Department of Preclinic Faculty of Medicine Mahasarakham University Mahasarakham 44000 Thailandf Department of Pharmacognosy University of Szeged Eotvos Str 6 H-6720 Szeged Hungaryg Department of Medical and Chemical Laboratory Diagnostics Medical University of Vienna Waehringer Guertel 18-20 Austriah Ludwig Boltzmann Institute for Cancer Research LBI-CR Waehringerstrasse 13a 1090 Vienna Austriai Unit of Pathology of Laboratory Animals University of Veterinary Medicine Vienna 1210 Vienna Austriaj Institute for Ethnobiology Playa Diana San Joseacute Peteacuten Guatemalak Department of Vascular Biology and Thrombosis Research Center of Biomolecular Medicine and Pharmacology Medical University of ViennaSchwarzspanierstraszlige 17 A-1090 Vienna Austrial Institute of Pharmaceutical Sciences University of Graz Schubertstraszlige 1 A-8010 Graz Austriam Institute of Pharmaceutical Biology and Biotechnology Heinrich-Heine-University Duumlsseldorf Universitaumltsstrasse 1 D-40225 Duumlsseldorf Germanyn Department of Medicine I Division Institute of Cancer Research Comprehensive Cancer Center Medical University Vienna Borschkegasse 8a A-1090Vienna Austriao Department of Medicine I Comprehensive Cancer Center Medical University Vienna Waehringer Guertel 18-20 A-1090 Vienna Austriap Department of Pharmacodynamics and Biopharmacy University of Szeged H-6720 Szeged Hungary

A R T I C L E I N F O

Article historyReceived 26 August 2014Received in revised form 8 November 2014Accepted 11 November 2014

KeywordsLobatinNPMALKALCLLymphendothelial intravasation3D-compound testing

A B S T R A C T

An apolar extract of the traditional medicinal plant Neurolaena lobata inhibited the expression of the NPMALK chimera which is causal for the majority of anaplastic large cell lymphomas (ALCLs) Therefore anactive principle of the extract the furanoheliangolide sesquiterpene lactone lobatin B was isolated andtested regarding the inhibition of ALCL expansion and tumour cell intravasation through thelymphendothelium

ALCL cell lines HL-60 cells and PBMCs were treated with plant compounds and the ALK inhibitor TAE-684 to measure mitochondrial activity proliferation and cell cycle progression and to correlate the resultswith protein- and mRNA-expression of selected gene products Several endpoints indicative for cell deathwere analysed after lobatin B treatment Tumour cell intravasation through lymphendothelial monolay-ers was measured and potential causal mechanisms were investigated analysing NF-κB- and cytochromeP450 activity and 12(S)-HETE production

Abbreviations ALCL anaplastic large cell lymphoma ALOX lipoxygenase A CCID circular chemorepellent induced defect CYP cytochrome P450 DCM dichloromethaneextract EROD ethoxyresorufin-O-deethylase HOPI Hoechst 33258propidium iodide LEC lymph endothelial cell MYPT1 myosin phosphatase 1 target subunit 1 NF-κBnuclear factor kappa B NPMALK nucleophosminanaplastic lymphoma kinasethe t(25)(p23q35) chromosomal translocation PARP poly ADP-ribose polymerase PBMCperipheral blood mononuclear cell PDGF-Rβ platelet derived growth factor receptor p21 tumour suppressor protein 21 3D 3-dimensional 12(S)-HETE 12(S)hydroxyeicosatetraenoic acid

Corresponding author Tel 0043 1 40400 34870 fax 0043 1 40400 37070E-mail address georgkrupitzameduniwienacat (G Krupitza)

1 Equal contribution

httpdxdoiorg101016jcanlet2014110190304-3835copy 2014 Elsevier Ireland Ltd All rights reserved

Cancer Letters 356 (2015) 994ndash1006

Contents lists available at ScienceDirect

Cancer Letters

journal homepage wwwelseviercom locate canlet

Lobatin B inhibited the expression of NPMALK JunB and PDGF-Rβ and attenuated proliferation ofALCL cells by arresting them in late M phase Mitochondrial activity remained largely unaffected uponlobatin B treatment Nevertheless caspase 3 became activated in ALCL cells Also HL-60 cell prolifera-tion was attenuated whereas PBMCs of healthy donors were not affected by lobatin B Additionally tumourcell intravasation which partly depends on NF-κB was significantly suppressed by lobatin B most likelydue to its NF-κB-inhibitory property

Lobatin B which was isolated from a plant used in ethnomedicine targets malignant cells by at leasttwo properties

I) inhibition of NPMALK thereby providing high specificity in combating this most prevalent fusionprotein occurring in ALCL

II) inhibition of NF-κB thereby not affecting normal cells with low constitutive NF-κB activity Thisproperty also inhibits tumour cell intravasation into the lymphatic system and may provide an optionto manage this early step of metastatic progression

copy 2014 Elsevier Ireland Ltd All rights reserved

Introduction

About 60 of currently used pharmaceutical drugs are mostlyderived from natural products Plant metabolites comprise a con-tinuing source of new structural leads for drug discovery anddevelopment because of the vast chemical diversity and ability tointeract with multiple cellular target proteins but only a small pro-portion of them have been investigated regarding their therapeuticvalue [1] Also plants used in ethnomedicine are not extensivelystudied Therefore traditional medicinal plants may lead to new ther-apeutic compounds against a variety of hard-to-cure diseases dueto their evident benefit and safe use throughout centuries of em-pirical testing Due to these reasons we recently investigated thedichloromethane (DCM) extract of Neurolaena lobata (L) R Br exCass (Asteraceae) and reported on its particular property to down-regulate the lymphoma-causing t(25)(p23q35) translocation NPMALK [2] that gives rise to ALK-positive anaplastic large cell lymphoma(ALK+ALCL) [3] Of particular relevance to the continuation of thisstudy was the demonstration that the EtOH leaf extract and thedichloromethane (DCM) fraction of the methanolic leaf extractshowed activity in the carrageenan-induced mouse- and rat pawoedema models (respectively) [45] manifesting that the extractsstill possessed active principles that were effective in intact organ-isms [6] Among three furanoheliangolide sesquiterpene lactones[7] lobatin B was isolated from the DCM fraction and its activity wascharacterised in NPMALK positive ALCL lines Lobatin B has beenisolated and tested before in human cancer cell lines exhibitingstrong anti-neoplastic activity [89] and here we report that lobatinB inhibits NPMALK expression in ALCL cells The inhibition of NPMALK signalling via the recently demonstrated pathway is a successfulclinical approach in the treatment of NPMALK positive ALCL [10]Given the youth of the vast majority of ALCL patients a careful se-lection of drugs is warranted to avoid the development of secondarymalignancies decades after the initial treatment with genotoxic drugsbut currently the choice of ALK-specific therapies is extremely limited[11] Therefore we tried to elucidate the NPMALK-targeting prop-erties of lobatin B In addition lobatin B was studied in a validatedmodel resembling the intravasation of tumour emboli through thelymphatic vasculature which is an early step of the metastaticprocess [12] As there are currently no therapies available thatprevent lymph node metastasis the inhibition of this process bylobatin B may serve as lead to develop anti-intravasative treat-ment concepts

Materials and methods

Plant material fine chemicals and antibodies

Extraction isolation and quantification of N lobata furanoheliangolide sesquit-erpene lactones were described by McKinnon et al [5] N lobata compounds weredissolved and prepared in DMSO (Sigma-Aldrich St Louis MO USA) as

concentrated stock solutions ALK-inhibitor NVP-TAE-684 (TAE-684) was fromSelleckchem (Houston TX USA)

CD246 anti-ALK protein mouse monoclonal antibody (mAB) and anti-nucleophosmin mouse mAB were purchased from Dako Cytomation (GlostrupDenmark) PDGF-Rβ rabbit mAB caspase 3 polyclonal antibody (pAB) histone H3rabbit mAb and phospho-histone H3 rabbit pAB were purchased from Cell Signal-ing (Cambridge UK) PARP-1 mouse mAB JunB rabbit pAB JunD rabbit pAB c-Junrabbit pAB p21 rabbit pAB cyclin B1 rabbit pAB and GAPDH mouse mAB were pur-chased from Santa Cruz Biotechnology Inc (Santa Cruz CA USA) Anti-szlig-actin (ascitesfluid) mouse mAB was ordered from Sigma (St Louis MO USA)

Cell culture

SR-786 NPMALK positive human ALCL (anaplastic large cell lymphoma) cellswere from DSMZ (Braunschweig Germany) CD-417 NPMALK positive mouse ALCLcells were isolated from CD4-NPMALK mice HL60 (human promyelocytic leuke-mia cells) were obtained from ATCC (Manassas VA USA) All cells were grown inRPMI 1640 medium (Life Technologies Carlsbad California USA) supplemented with10 heat inactivated fetal calf serum (FCS Life Technologies Carlsbad CaliforniaUSA) 1 L-glutamine (Lonza Verviers Belgium) and 1 antibiotics (penicillinstreptomycin (PS) Sigma-Aldrich St Louis MO USA) and maintained in a humidifiedatmosphere containing 5 CO2 at 37 degC

Isolation of peripheral blood mononuclear cells (PBMCs)

With the informed consent of the donors PBMCs were isolated from human pe-ripheral blood as described earlier [13]

Proliferation assay

The proliferation of SR-786 CD-417 PBMC and HL60 was determined by count-ing cells with a Casy cell counter (Roche Innovatis AG Bielefeld Germany) as describedbefore [2]

Western blotting

SR-786 cells were seeded at a concentration of 2 times 105 cellsml and CD-417 ata concentration of 106 cellsml in 6 cm dishes After treating cells with 3 μM ofN lobata compounds for the indicated times they were harvested and lysed in RIPAbuffer (150 mM NaCl 50 mM Tris pH 76 1 Triton 01 SDS 05 Sodium deoxy-cholate) containing 1 mM phenylmethylsulphonyl (PSMF Sigma-Aldrich St LouisMO USA) and 1 mM protease inhibitor mixture (PIM consists of 2 μgml leupeptin2 μgml aprotinin 03 μgml benzamidine chloride and 10 μgml trypsin inhibitorSigma-Aldrich St Louis MO USA) followed by a short incubation of 5 min on iceLysates were treated stored electrophoretically separated and analysed by Westernblotting as described by Unger et al [2] Chemiluminescence was developed by ECLdetection kit (Thermo Scientific Waltham MA USA) and membranes were exposedto Amersham Hyperfilms (GE Healthcare Buckinghamshire UK) or CL-XPosure films(Thermo Scientific Rockford IL USA) Membranes were stripped in 75 ml buffercontaining 45 ml 1 M TrisndashHCL pH 64 75 ml 20 SDS 05 ml β-mercaptoethanolfor 6ndash15 min shaking in a 55 degC water bath and afterwards the membranes werewashed

Quantitative RT-PCR

SR-786 cells were seeded in a 24-well plate at a concentration of 2 times 105 cellsml and incubated overnight before treatment with 3 μM of N lobata compoundsFor RNA preparation ReliaPrep RNA Cell Miniprep System Kit (Promega MadisonWI) was used and RNA content was measured using a NanoDrop Fluorospectrometer(Thermo Fisher Scientific Waltham MA USA) First-strand cDNA (150 ng RNA as

995I Kiss et alCancer Letters 356 (2015) 994ndash1006

template) was synthesised using GoScriptTM Reverse Transcription System Kit(Promega Madison WI)

Transcript expression was examined by real-time PCR (polymerase chain reac-tion) using a SYBR Green detection system (Promega Madison WI) For each sample10 μl GoTaq qPCR Master Mix (premixed solution containing GoTaq DNA poly-merase GoTaq Reaction Buffer dNTPs and Mg2+ Promega Madison WI) 2 μl forwardprimer and 2 μl reverse primer (see sequences below) 5 μl nuclease free water and1 μl cDNA were added to the wells of a 96-well optical reaction plate The cycle pro-gramme was 50 degC for 2 min 95 degC for 10 min to activate polymerase 40 cycles of95 degC for 15 s and 60 degC for 1 min (Thermocycler Primus25 advanced Peqlab Er-langen Germany) The following primers were used for RT-PCR

NPMALK (fwd 5prime-GTG GTC TTA AGG TTG AAG TGT GGT T-3prime rev 5prime-GCT TCCGGC GGT ACA CTA CTA A-3prime)

nucleophosmin (fwd 5prime-TCC CTT GGG GGC TTT GAA ATA ACA CC-3prime rev 5prime-TGG AAC CTT GCT ACC ACC TC-3prime)

JunB (fwd 5prime-GCT CGG TTT CAG GAG TTT GT-3prime rev 5prime-ATA CAC AGC TAC GGGATA CGG-3prime)

GAPDH (fwd 5prime- AAC AGC GAC ACC CAC TCC TC -3prime rev 5prime- CAT ACC AGG AAATGA GCT TGA CAA -3prime)

To analyse qPCR data the Ct (ΔΔCt) method [14] for relative quantification ofgene expression was used To quantify relative expression of the target genes NPMALK nucleophosmin and JunB the following formula was used ΔCt = Ct target gene

(NPMALK nucleophosmin JunB) ndash X Ct control gene (GAPDH) ΔΔCt = ΔCt drug

treatment ndash X ΔCt control sample Ratio = 2minusΔΔCt

Cell cycle progression (FACS-analysis)

SR-786 cells were seeded in a 6-well plate at a concentration of 2 times 105 cellsml After 8 h of treatment cells were harvested and centrifuged at 300 times g for 5 minat 4 degC and processed as described earlier [2] and analysed on a FACS Calibur flowcytometer (BD Bioscience Franklin Lakes New Jersey USA)

Cytotoxicity mitochondrial activity assay

To measure mitochondrial activity CellTiter-Blue assay (Promega Madison WI)was used according to the manufacturerrsquos instructions For this SR-786 PBMC andHL60 cells were seeded into 96-well plates at concentrations of 2 times 105 5 times 105 and1 times 105 cellsml respectively The compounds were added at the indicated concen-trations and compared to solvent-treated controls Fluorescence was measured at570 nm using a multi-detection reader (Synergy HT Bio-Tek Instrument Win-ooski VT USA)

Cell death analysis ndash (HOPI staining)

Hoechst 33258 (HO) and propidium iodide (PI) double staining (Sigma-Aldrich St Louis MO USA) allows to measure cell death [15] and was performedas described earlier [2] using a fluorescence microscope equipped with a TRITC andDAPI filter (Olympus IX51 Shinjuku Tokyo Japan)

Caspase 37 activity assay

SR-786 cells were seeded in 35 cm dishes at a concentration of 2 times 105 cellmland after incubation of 1 h at 37 degC cells were treated with 3 μgml of N lobata com-pounds for 8 16 and 24 h when they were analysed by the Apo-ONE HomogeneousCaspase-37 assay (Promega Madison WI) according to the manufacturerrsquos instruc-tions Fluorescence was measured by using a multi-detection reader (excitation at499 nm and emission at 521 nm)

NF-κB transactivation assay

The transactivation of a NF-κB-driven luciferase reporter was quantified inHEK293NF-κB-luc cells (Panomics RC0014) as previously described [1617] usinga GeniosPro plate reader (Tecan Groumldig Austria) Parthenolide (SigmandashAldrich ViennaAustria) was used as a positive control

Circular chemorepellent induced defect (CCID) assay

The analysis of tumour intravasation through the lymphendothelial barrier wasdone as described before [1218ndash25] and CCID areas were measured using ZEN 2012software (Zeiss Jena Germany) During the experiments which were short termwe did not observe toxic effects of the tested compounds (monitored by HOPI stain-ing) [15]

12(S)-HETE assay

MCF-7 cells were seeded in 35 cm dishes and grown in 25 ml complete MEMmedium (Gibco 10370-047) The next day the medium was changed to FCS-freemedium and cells were kept at 37 degC for 24 h Then cells were treated with 10 μM

arachidonic acid (A3555 Sigma-Aldrich Munich Germany) and the indicated com-pounds for 24 h The concentration of 12(S)-HETE in the cellular supernatant wasmeasured with minor modifications as described previously [2324] using the 12(S)-HETE enzyme immunoassay kit (EIA ADI-900-050 Enzo Life Sciences LausenSwitzerland) Absorbance was measured with a Wallac 1420 Victor 2 multilabel platereader (Perkin Elmer Life and Analytical Sciences)

Ethoxyresorufin-O-deethylase (EROD) assay selective for CYP1A1 activity

MCF-7 breast cancer cells were grown in phenol red-free DMEMF12 medium(Gibco Karlsruhe Germany) containing 10 FCS and 1 PS (Invitrogen KarlsruheGermany) Before treatment the cells were transferred to DMEMF12 medium supple-mented with 10 charcoal-stripped FCS (PAN Biotech Aldenbach Germany) and 1PS After 24 h of treatment CYP1A1 activity was measured with minor modifica-tions as previously described [22] Briefly ethoxyresorufin (final concentration 50 μMSigma-Aldrich Munich Germany) was added and 04 ml aliquots of the medium weresampled after 180 min and the formation of resorufin was analysed by spectrofluo-rometry (PerkinElmer LS50B Waltham MA USA) with an excitation wavelength of530 nm and an emission wavelength of 585 nm

Statistical analysis

For statistical analyses Excel 2003 software and Prism 5 software package(GraphPad San Diego CA USA) were used The values were expressed as mean plusmn SDand Studentrsquos t-test or ANOVA and Dunnett-post-test were used to evaluate statis-tical significance (P lt 005)

Results

Anti-proliferative effects of N lobata furanoheliangolidesesquiterpene lactones in ALCL cells

In order to explore the effects of isolated N lobata compounds(Fig 1a) on cell growth of ALK-positive ALCL murine CD-417 (Fig 1b)and human SR-786 cells (Fig 1c) cells were treated with 1 and 3 μMlobatin B 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) and8β-isovaleryloxy-9α-acetoxy-calyculatolide (OAc-CAL) A concen-tration of 3 μM lobatin B inhibited proliferation of murine CD-417cells and led to their eradication after 24 h Human SR-786 cellgrowth was inhibited by 1 μM lobatin B 3 μM OAc-CAL slightly in-hibited SR-786 cell growth after 24 h whereas OH-CAL did not inhibitgrowth of both cell lines After 72 h lobatin B OAc-CAL and OH-CAL inhibited SR-768 cell proliferation with an IC50 (the concentrationwhich inhibits cell proliferation by 50 compared to control) of21 μM 80 μM and 243 μM respectively (Fig 1d) Hence furtherexperiments were performed with lobatin B to characterise the cy-totoxic mechanisms and were compared to OH-CAL which did notshow anti-neoplastic effects Interestingly OH-CAL and OAc-CALslightly but consistently induced the growth of CD-417 cells after16 h of treatment

Mitochondrial activity and cell cycle distribution upon lobatin B andOH-CAL treatment

The mitochondrial metabolism of SR-786 cells which was mea-sured by CellTiter-Blue assay was only weakly affected by lobatinB and OH-CAL (Fig 2a) Next the effect of lobatin B on cell cycle dis-tribution of SR-786 was evaluated by flow cytometric analysis (FACS)Treatment with 3 μM lobatin B for 8 h caused the accumulation ofSR-786 cells in G2M phase at the expense of cells in G1 (Fig 2b)Therefore SR-786 cells were still able to pass through S-phase uponlobatin B treatment because no accumulation of S-phase cells wasobserved This suggested that lobatin B inhibited proliferation byarresting SR-786 cells in G2M Neither cyclin B levels were el-evated (Fig 2c) which is indicative for G2 [26] nor was thephosphorylation of serine 10 of histone 3 detectable (not shown)which is tightly associated with the condensation of chromatin untilanaphase of mitosis [27] Therefore we conclude that cells accu-mulated after anaphase in the telophase of mitosis when chromatinde-condenses before cytokinesis This is in concordance with the

996 I Kiss et alCancer Letters 356 (2015) 994ndash1006

a

b

c

dLobatin B IC50=21 μM (72 h) OAc-CAL IC50=80 μM (72 h) OH-CAL IC50=243 μM (72 h)

0 1 2 3 4 5 6 7 8 9 1 0 1 10

5 0

1 0 0

1 5 0

microM

S

R-7

86 c

ell n

umbe

r

S

R-7

86 c

ell n

umbe

r

S

R-7

86 c

ell n

umbe

r

0 5 1 0 1 5 2 0 2 5 3 00

5 0

1 0 0

1 5 0

microM

0 5 1 0 1 5 2 0 2 5 3 00

5 0

1 0 0

1 5 0

microM

Fig 1 Anti-proliferative effects of (a) N lobata compounds in (b) murine CD-417 and (c d) human SR-786 cells (b c) After 8 16 and 24 h of treatment with 1 μM and3 μM or (d) after 72 h of treatment with indicated concentrations of lobatin B 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) and 8β-isovaleryloxy-9α-acetoxy-calyculatolide (OAc-CAL) cells were counted using a Casy cell counter The relative cell number is presented as percent of control Experiments were performed in triplicateerror bars indicate means plusmn SD and asterisks significance (P lt 005 ANOVA followed by Dunnett-post-test)

997I Kiss et alCancer Letters 356 (2015) 994ndash1006

fact that histone 3 protein level was slightly increased after 8 h whichis necessary to structure the duplicated chromatin OH-CAL treat-ment had no effect on cell cycle distribution

Lobatin B inhibits NPMALK expression in SR-786 cells

In ALCL cells the NPMALK chimera is driving proliferation There-fore it was tested whether lobatin B affected the expression of NPMALK Lobatin B treatment strongly suppressed the level of NPMALK after 8 h and 24 h whereas OH-CAL did not reduce NPMALK(Fig 3) Thus lobatin B specifically abrogated the expression of NPMALK and this was most likely causal for growth inhibition of SR-786 cells Interestingly lobatin B caused an oscillation innucleophosmin expression and also OH-CAL suppressednucleophosmin expression after 24 h

To investigate at which stage the expression of NPMALK becamedown-regulated by lobatin B the transcript levels were analysedNPMALK- and also nucleophosmin mRNAs were reduced uponlobatin B treatment (Fig 4ab) hence giving a clue as to how lobatinB mediated the regulation of NPMALK ie by interfering with afactor or a site regulating nucleophosmin transcription The factthat nucleophosmin protein level remained high upon lobatin Btreatment might have been due to high stability of the

polypeptide Yet there was still a discrepancy because ldquoinactiverdquo OH-CAL treatment decreased the protein expression of nucleophosminand slightly that of NPMALK after 24 h Therefore the way of tran-scriptional regulation of NPMALK by lobatin B has to besubstantiated by future investigations

The transcription of the JunB proto-oncogene was shown to beregulated by NPMALK [1028] and accordingly lobatin B treat-ment suppressed JunB mRNA levels (Fig 4c) The Jun family oftranscription factors are components of the AP-1 transcription factorcomplex and AP-1 (activator protein 1) is involved in cell prolifer-ation and apoptosis [29] which provides a mechanistic link betweenlobatin B treatment the down-regulation of NPMALK and subse-quently of JunB and the inhibition of cell proliferationinductionof apoptosis

Lobatin B affects expression of Jun family members and induces thetumour suppressor p21

The expression of the Jun family members was further analysedat the protein level JunB is the main transcription factor in the AP-1complex induced by NPMALK [2830] and lobatin B treatment in-hibited the expression of JunB protein (Fig 5a) which is consistentwith suppression of its mRNA Lobatin B induced c-Jun which was

Fig 2 Potential mechanisms of proliferation inhibition by lobatin B in SR-786 cells (a) Cells were treated with 3 μM lobatin B and OH-CAL for 8 h and 24 h respectivelywhen CellTiter-Blue reagent was added and absorbance measured at 570 nm using a multi-well plate reader The relative cell number is presented as percent of control (b)Cell cycle distribution upon treatment with lobatin B and OH-CAL SR-786 cells were incubated with 3 μM of either compound for 8 h and then subjected to FACS analysisExperiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test) (c) Effect of Lobatin B on cyclin B and histone 3 SR-786 cells were treated with 3 μM lobatin B for 1 2 and 8 h harvested and subjected to Western blot analysis using the indicated antibodies Densitometer readings facilitatedthe comparison of relative protein expression levels with untreated control (which was set as ldquo1rdquo) The values were standardised to GAPDH expression which was used tomonitor equal sample loading

998 I Kiss et alCancer Letters 356 (2015) 994ndash1006

fact that histone 3 protein level was slightly increased after 8 h whichis necessary to structure the duplicated chromatin OH-CAL treat-ment had no effect on cell cycle distribution

Lobatin B inhibits NPMALK expression in SR-786 cells

In ALCL cells the NPMALK chimera is driving proliferation There-fore it was tested whether lobatin B affected the expression of NPMALK Lobatin B treatment strongly suppressed the level of NPMALK after 8 h and 24 h whereas OH-CAL did not reduce NPMALK(Fig 3) Thus lobatin B specifically abrogated the expression of NPMALK and this was most likely causal for growth inhibition of SR-786 cells Interestingly lobatin B caused an oscillation innucleophosmin expression and also OH-CAL suppressednucleophosmin expression after 24 h

To investigate at which stage the expression of NPMALK becamedown-regulated by lobatin B the transcript levels were analysedNPMALK- and also nucleophosmin mRNAs were reduced uponlobatin B treatment (Fig 4ab) hence giving a clue as to how lobatinB mediated the regulation of NPMALK ie by interfering with afactor or a site regulating nucleophosmin transcription The factthat nucleophosmin protein level remained high upon lobatin Btreatment might have been due to high stability of the

polypeptide Yet there was still a discrepancy because ldquoinactiverdquo OH-CAL treatment decreased the protein expression of nucleophosminand slightly that of NPMALK after 24 h Therefore the way of tran-scriptional regulation of NPMALK by lobatin B has to besubstantiated by future investigations

The transcription of the JunB proto-oncogene was shown to beregulated by NPMALK [1028] and accordingly lobatin B treat-ment suppressed JunB mRNA levels (Fig 4c) The Jun family oftranscription factors are components of the AP-1 transcription factorcomplex and AP-1 (activator protein 1) is involved in cell prolifer-ation and apoptosis [29] which provides a mechanistic link betweenlobatin B treatment the down-regulation of NPMALK and subse-quently of JunB and the inhibition of cell proliferationinductionof apoptosis

Lobatin B affects expression of Jun family members and induces thetumour suppressor p21

The expression of the Jun family members was further analysedat the protein level JunB is the main transcription factor in the AP-1complex induced by NPMALK [2830] and lobatin B treatment in-hibited the expression of JunB protein (Fig 5a) which is consistentwith suppression of its mRNA Lobatin B induced c-Jun which was

Fig 2 Potential mechanisms of proliferation inhibition by lobatin B in SR-786 cells (a) Cells were treated with 3 μM lobatin B and OH-CAL for 8 h and 24 h respectivelywhen CellTiter-Blue reagent was added and absorbance measured at 570 nm using a multi-well plate reader The relative cell number is presented as percent of control (b)Cell cycle distribution upon treatment with lobatin B and OH-CAL SR-786 cells were incubated with 3 μM of either compound for 8 h and then subjected to FACS analysisExperiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test) (c) Effect of Lobatin B on cyclin B and histone 3 SR-786 cells were treated with 3 μM lobatin B for 1 2 and 8 h harvested and subjected to Western blot analysis using the indicated antibodies Densitometer readings facilitatedthe comparison of relative protein expression levels with untreated control (which was set as ldquo1rdquo) The values were standardised to GAPDH expression which was used tomonitor equal sample loading

998 I Kiss et alCancer Letters 356 (2015) 994ndash1006

accompanied by an induction of p21 It was shown that c-Jun to-gether with the ubiquitous transcription factor SP1 transactivatesp21 expression [31] On the other hand also silencing of c-Jun bysiRNA caused the upregulation of p21 and accumulation of NPMALK positive ALCL cells in G2M at the expense of S-phase cells [32]Therefore both scenarios ndash c-Jun induction and c-Jun inhibition ndashmay cause p21-mediated G2M arrest In contrast to the observa-tions of Leventaki et al [32] which showed that c-Jun down-regulation is accompanied by a loss of S-phase cells we here reportthat upregulation of c-Jun is accompanied by a loss of G1-phase cells

Lobatin B enhanced c-Jun protein expression in a similar wayas did treatment with the DCM fraction of N lobata [2] Howeverthe regulation of c-Jun by lobatin B remained unclear As c-Jun cansubstitute for JunB the upregulation of c-Jun might be part of a com-pensatory feedback loop in response to JunB inhibition InterestinglyJunD levels oscillated upon lobatin B treatment in a similar way asobserved for nucleophosmin levels

OH-CAL neither induced c-Jun nor p21 (Fig 5a) and tran-siently suppressed JunB independently of NPMALK becauseNPMALK remained expressed at the time point when JunB de-creased However it is possible that just the activity of NPMALKbut not its expression level was compromised This shortdownregulation was not substantial and had no effect on the cellcycle Jun family members especially JunB promote ALCL devel-opment through transcriptional activation of PDGFR-szlig as shownin an ALCL mouse model [10] In the human SR-786 ALCL cellline PDGFR-szlig is not expressed Therefore the murine CD-417ALCL cell line was used to test the effect of lobatin B on PDGFR-szligexpression Lobatin B treatment first inhibited NPMALK (2 h)and subsequently JunB and PDGFR-β was downregulated (Fig 5b)Hence lobatin B inhibited the recently discovered NPMALKsignal transduction cascade down to the level of JunB and PDGFR-szlig[10] As in SR-786 cells lobatin B induced c-Jun also in CD-417cells

Fig 3 Lobatin B downregulates NPM-ALK expression in SR-786 Cells were treated with 3 μM lobatin B (a) or 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL b) for1 2 8 and 24 h harvested and subjected to Western blot analysis using the indicated antibodies Densitometer readings facilitated the comparison of relative protein ex-pression levels with untreated control (which was set as ldquo1rdquo) The values were standardised to β-actin expression which was used to monitor equal sample loading

Fig 4 Quantitative PCR analysis SR-786 cells were treated with 3 μM lobatin B for the indicated times and the mRNA expression of (a) NPMALK (b) nucleophosmin and(c) JunB was measured and normalised to GAPDH mRNA Experiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005t-test)

999I Kiss et alCancer Letters 356 (2015) 994ndash1006

Fig 5 Effect of lobatin B or OH-CAL on Jun-family members PDGFR-szlig and p21 in ALK-positive ALCL cells SR-786 cells were treated with 3 μM lobatin B or 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) (a) for 2 4 6 and 8 h and CD-417 cells (b) were treated with lobatin B Then cells were harvested and subjected to Western blot analysisusing the indicated antibodies Densitometer readings facilitated the comparison of relative protein expression levels with untreated control (which was set as ldquo1rdquo) In thecase of JunD the expression levels of both forms were added together The values were standardised to β-actin expression which was used to monitor equal sample loading

1000 I Kiss et alCancer Letters 356 (2015) 994ndash1006

Fig 6 Apoptoticnecrotic cell death of SR-786 cells treated with N lobata compounds Cells were treated with 3 μM lobatin B and after 24 h cell death was measured by (a)HOPI staining which enables the identification of apoptotic and necrotic cells (b) Cells were treated with 3 μM of lobatin B and after 8 16 and 24 h ApoOne reagent wasadded and caspase 37 activity was measured Experiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test) (c) Cellswere treated with 3 μM lobatin B (left panel) and 8β-isovaleryloxy-9α-hydroxy-calyculatolide OH-CAL (right panel) for 1 2 8 and 24 h harvested and subjected to Westernblot analysis using the indicated antibodies Densitometer readings facilitated the comparison of relative PARP full length protein expression levels which were set to 100(upper band) with the respective cleaved forms of PARP (lower band) No densitometer readings were performed for caspase 3 expression because no signature-type cleav-age band (indicating fully activated caspase 3) appeared in the untreated control thereby making relative comparisons impossible β-actin expression served as control forequal sample loading

1001I Kiss et alCancer Letters 356 (2015) 994ndash1006

Lobatin B triggers SR-786 cell death

Lobatin B induced cell death and more apoptotic than necroticcells were counted by HOPI staining (Fig 6a) This was confirmedby detecting an induction of Caspase 37 activity within 16 h oflobatin B treatment which decreased thereafter (Fig 6b) Further-more caspase 3 activation was confirmed by its proteolytic cleavageafter 24 h and concomitant signature-type degradation of its targetPARP (Fig 6c) OH-CAL treatment had a minor effect on caspase 3pre-activation Apparently this was due to the rather similar struc-tures of lobatin B and OH-CAL However OH-CAL treatment did notseriously affect the survival of SR-786 cells

Impact of lobatin B on NPMALK negative cell types

To assess the specificity of lobatin B towards lymphoma cellsPBMCs from a healthy volunteer were treated with 3 μM lobatin andOH-CAL Interestingly the number of PBMCs increased after 8 h oflobatin B- and OH-CAL treatment (Fig 7a) and this was accompa-nied by an increased mitochondrial activity (Fig 7b) Then PBMCnumbers returned to control levels after 24 h and 48 h indicatingthat the initially propagating cell mass was finally subjected to areduction process which was paralleled by a significantly reducedmitochondrial metabolism upon lobatin B treatment for 24 h SinceOH-CAL did not increase the metabolic activity after 8 h the ob-served correlation between PBMC number and their mitochondrialactivity (also by lobatin B treatment) was coincidental

Furthermore HL60 leukaemia cells which do not harbour theNPMALK translocation were tested to study the specificity of lobatinB towards NPMALK HL60 cell number was reduced by ~60 uponlobatin B treatment (Fig 7a) which severely inhibited HL60 mito-chondrial metabolism (Fig 7b) This showed that lobatin B exhibitedadditional effects beyond NPMALK inhibition targeting leukae-mia cells but not PBMCs Hence the anti-proliferative effects oflobatin B were specific for neoplastic cells (ie lymphoma and leu-kaemia cells) but with a higher specificity to those cells harbouringthe NPMALK translocation because SR-786 cells and CD417 cellswere more sensitive towards lobatin B than HL60 cells

Specificity of the ALK inhibitor TAE-684

To estimate the impact of NPMALK on cell proliferation andthe specificity of lobatin B regarding this mechanism SR-786 ALCLcells ALK-negative HL60 cells and normal PBMCs were treated withthe specific NPMALK inhibitor TAE-684 [11] and the effect on cellproliferation was compared TAE-786 dose dependently inhibitedthe proliferation of SR-786 cells but not that of HL60 and PBMCs(Fig 8a) Hence NPMALK is driving proliferation and TAE-684 ismore specific than lobatin B regarding its property to target solelyALK TAE-684 did not reduce the mitochondrial activity of HL60cells but inhibited PBMC- and SR-786 mitochondrial metabolism(Fig 8b) Obviously mitochondrial activities did not correlatewith cell proliferation rates and were thus independent of eachother

Fig 7 Treatment of PBMC and HL60 cells with N lobata compounds (a) Effects on cell number after treatment of PBMCs and HL60 cells with 3 μM lobatin B and 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) for 8 h 24 h and 48 h Cell number was measured by Casy cell counter (b) Cells were treated with 3 μM N lobata compoundsand after 8 h and 24 h of incubation with lobatin B and OH-CAL CellTiter-Blue reagent was added and absorbance was measured at 570 nm Experiments were performedin triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test)

1002 I Kiss et alCancer Letters 356 (2015) 994ndash1006

Fig 6 Apoptoticnecrotic cell death of SR-786 cells treated with N lobata compounds Cells were treated with 3 μM lobatin B and after 24 h cell death was measured by (a)HOPI staining which enables the identification of apoptotic and necrotic cells (b) Cells were treated with 3 μM of lobatin B and after 8 16 and 24 h ApoOne reagent wasadded and caspase 37 activity was measured Experiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test) (c) Cellswere treated with 3 μM lobatin B (left panel) and 8β-isovaleryloxy-9α-hydroxy-calyculatolide OH-CAL (right panel) for 1 2 8 and 24 h harvested and subjected to Westernblot analysis using the indicated antibodies Densitometer readings facilitated the comparison of relative PARP full length protein expression levels which were set to 100(upper band) with the respective cleaved forms of PARP (lower band) No densitometer readings were performed for caspase 3 expression because no signature-type cleav-age band (indicating fully activated caspase 3) appeared in the untreated control thereby making relative comparisons impossible β-actin expression served as control forequal sample loading

1001I Kiss et alCancer Letters 356 (2015) 994ndash1006

Lobatin B triggers SR-786 cell death

Lobatin B induced cell death and more apoptotic than necroticcells were counted by HOPI staining (Fig 6a) This was confirmedby detecting an induction of Caspase 37 activity within 16 h oflobatin B treatment which decreased thereafter (Fig 6b) Further-more caspase 3 activation was confirmed by its proteolytic cleavageafter 24 h and concomitant signature-type degradation of its targetPARP (Fig 6c) OH-CAL treatment had a minor effect on caspase 3pre-activation Apparently this was due to the rather similar struc-tures of lobatin B and OH-CAL However OH-CAL treatment did notseriously affect the survival of SR-786 cells

Impact of lobatin B on NPMALK negative cell types

To assess the specificity of lobatin B towards lymphoma cellsPBMCs from a healthy volunteer were treated with 3 μM lobatin andOH-CAL Interestingly the number of PBMCs increased after 8 h oflobatin B- and OH-CAL treatment (Fig 7a) and this was accompa-nied by an increased mitochondrial activity (Fig 7b) Then PBMCnumbers returned to control levels after 24 h and 48 h indicatingthat the initially propagating cell mass was finally subjected to areduction process which was paralleled by a significantly reducedmitochondrial metabolism upon lobatin B treatment for 24 h SinceOH-CAL did not increase the metabolic activity after 8 h the ob-served correlation between PBMC number and their mitochondrialactivity (also by lobatin B treatment) was coincidental

Furthermore HL60 leukaemia cells which do not harbour theNPMALK translocation were tested to study the specificity of lobatinB towards NPMALK HL60 cell number was reduced by ~60 uponlobatin B treatment (Fig 7a) which severely inhibited HL60 mito-chondrial metabolism (Fig 7b) This showed that lobatin B exhibitedadditional effects beyond NPMALK inhibition targeting leukae-mia cells but not PBMCs Hence the anti-proliferative effects oflobatin B were specific for neoplastic cells (ie lymphoma and leu-kaemia cells) but with a higher specificity to those cells harbouringthe NPMALK translocation because SR-786 cells and CD417 cellswere more sensitive towards lobatin B than HL60 cells

Specificity of the ALK inhibitor TAE-684

To estimate the impact of NPMALK on cell proliferation andthe specificity of lobatin B regarding this mechanism SR-786 ALCLcells ALK-negative HL60 cells and normal PBMCs were treated withthe specific NPMALK inhibitor TAE-684 [11] and the effect on cellproliferation was compared TAE-786 dose dependently inhibitedthe proliferation of SR-786 cells but not that of HL60 and PBMCs(Fig 8a) Hence NPMALK is driving proliferation and TAE-684 ismore specific than lobatin B regarding its property to target solelyALK TAE-684 did not reduce the mitochondrial activity of HL60cells but inhibited PBMC- and SR-786 mitochondrial metabolism(Fig 8b) Obviously mitochondrial activities did not correlatewith cell proliferation rates and were thus independent of eachother

Fig 7 Treatment of PBMC and HL60 cells with N lobata compounds (a) Effects on cell number after treatment of PBMCs and HL60 cells with 3 μM lobatin B and 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) for 8 h 24 h and 48 h Cell number was measured by Casy cell counter (b) Cells were treated with 3 μM N lobata compoundsand after 8 h and 24 h of incubation with lobatin B and OH-CAL CellTiter-Blue reagent was added and absorbance was measured at 570 nm Experiments were performedin triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test)

1002 I Kiss et alCancer Letters 356 (2015) 994ndash1006

Lobatin B triggers SR-786 cell death

Lobatin B induced cell death and more apoptotic than necroticcells were counted by HOPI staining (Fig 6a) This was confirmedby detecting an induction of Caspase 37 activity within 16 h oflobatin B treatment which decreased thereafter (Fig 6b) Further-more caspase 3 activation was confirmed by its proteolytic cleavageafter 24 h and concomitant signature-type degradation of its targetPARP (Fig 6c) OH-CAL treatment had a minor effect on caspase 3pre-activation Apparently this was due to the rather similar struc-tures of lobatin B and OH-CAL However OH-CAL treatment did notseriously affect the survival of SR-786 cells

Impact of lobatin B on NPMALK negative cell types

To assess the specificity of lobatin B towards lymphoma cellsPBMCs from a healthy volunteer were treated with 3 μM lobatin andOH-CAL Interestingly the number of PBMCs increased after 8 h oflobatin B- and OH-CAL treatment (Fig 7a) and this was accompa-nied by an increased mitochondrial activity (Fig 7b) Then PBMCnumbers returned to control levels after 24 h and 48 h indicatingthat the initially propagating cell mass was finally subjected to areduction process which was paralleled by a significantly reducedmitochondrial metabolism upon lobatin B treatment for 24 h SinceOH-CAL did not increase the metabolic activity after 8 h the ob-served correlation between PBMC number and their mitochondrialactivity (also by lobatin B treatment) was coincidental

Furthermore HL60 leukaemia cells which do not harbour theNPMALK translocation were tested to study the specificity of lobatinB towards NPMALK HL60 cell number was reduced by ~60 uponlobatin B treatment (Fig 7a) which severely inhibited HL60 mito-chondrial metabolism (Fig 7b) This showed that lobatin B exhibitedadditional effects beyond NPMALK inhibition targeting leukae-mia cells but not PBMCs Hence the anti-proliferative effects oflobatin B were specific for neoplastic cells (ie lymphoma and leu-kaemia cells) but with a higher specificity to those cells harbouringthe NPMALK translocation because SR-786 cells and CD417 cellswere more sensitive towards lobatin B than HL60 cells

Specificity of the ALK inhibitor TAE-684

To estimate the impact of NPMALK on cell proliferation andthe specificity of lobatin B regarding this mechanism SR-786 ALCLcells ALK-negative HL60 cells and normal PBMCs were treated withthe specific NPMALK inhibitor TAE-684 [11] and the effect on cellproliferation was compared TAE-786 dose dependently inhibitedthe proliferation of SR-786 cells but not that of HL60 and PBMCs(Fig 8a) Hence NPMALK is driving proliferation and TAE-684 ismore specific than lobatin B regarding its property to target solelyALK TAE-684 did not reduce the mitochondrial activity of HL60cells but inhibited PBMC- and SR-786 mitochondrial metabolism(Fig 8b) Obviously mitochondrial activities did not correlatewith cell proliferation rates and were thus independent of eachother

Fig 7 Treatment of PBMC and HL60 cells with N lobata compounds (a) Effects on cell number after treatment of PBMCs and HL60 cells with 3 μM lobatin B and 8β-isovaleryloxy-9α-hydroxy-calyculatolide (OH-CAL) for 8 h 24 h and 48 h Cell number was measured by Casy cell counter (b) Cells were treated with 3 μM N lobata compoundsand after 8 h and 24 h of incubation with lobatin B and OH-CAL CellTiter-Blue reagent was added and absorbance was measured at 570 nm Experiments were performedin triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test)

1002 I Kiss et alCancer Letters 356 (2015) 994ndash1006

Lobatin B inhibits NF-κB and the intravasation of tumour spheroidsthrough the lymphendothelial barrier

Lobatin B reduced the number of ALK-positive SR-786 lym-phoma cells by downregulating NPMALK and of ALK-negative HL60leukaemia cells by an unknown mechanism In HL60 cells NF-κBsignalling is constitutively activated at high levels and counteractsmonocytic differentiation [33] NF-κB ensures cell survival by keepingup the transcription of IAPs which are proteins intercepting caspaseactivity Therefore we tested whether NF-κB activation was inhib-ited by lobatin B For this modified HEK293 cells which stablyexpress NF-κB recognition sequences linked to luciferase weretreated with lobatin B to report whether NF-κB activity was modu-lated Lobatin B treatment attenuated TNFα-induced luciferaseexpression after 4 h in a dose dependent manner and hence NF-κB activation was suppressed (Fig 9a) This may explain thesusceptibility of HL60 cells to lobatin B treatment PBMCs re-mained unaffected by lobatin B treatment because in normal cellsNF-κB expression is low

In addition to anti-apoptotic signalling NF-κB plays a signifi-cant role when tumour cells intravasate lymphendothelial barriers[1819] In a validated three-dimensional co-culture model in whichMCF-7 breast cancer spheroids are placed on top of lymphendothelialcell (LEC) monolayers the tumour spheroids stimulate the retrac-tion of adjacent LECs [203435] This leads to cell-free areas so calledldquocircular chemorepellent induced defectsrdquo (CCIDs) through whichtumours intravasate lymphatics [12] Lobatin B dose-dependentlyinhibited this complex pro-metastatic process resulting in signifi-cantly reduced CCID formation (Fig 9b) Besides NF-κB also ALOX12and ALOX15 which are the major enzymes generating 12(S)-HETE (ldquoendothelial retraction factorrdquo) [35] and cytochromes P450(CYPs) contribute to CCID formation Therefore 12(S)-HETE pro-duction and CYP activity were studied by respective assays (Fig 9cd)

Neither 12(S)-HETE synthesis nor CYP1A1 activity were signifi-cantly inhibited by 3 and 5 μM lobatin B

Discussion

The DCM extract of N lobata [7] was formerly shown todownregulate NPMALK induce apoptosis in ALCL cell lines [2] andinhibit inflammation in a carrageenan-induced rat paw oedemamodel [5] The work presented here demonstrates that lobatin B isan active principle isolated from the DCM fraction of the methanolicextract of N lobata and suppresses the NPMALK transcript andprotein and also nucleophosmin which in its truncated form isthe 5-prime fusion partner of the NPMALK t(25)(p23q35)translocation [3] This suggested that a transcriptional mecha-nism responsible for nucleophosmin expression was hampered bylobatin B

NPMALK was shown to induce JunB which is a transcriptionfactor of the tyrosine receptor kinase PDGF-Rβ [1028] and lobatinB inhibited the expression of JunB and PDGF-Rβ subsequently toNPMALK down-regulation Interestingly the inactive sesquiter-pene lactone OH-CAL downregulated JunB independently of NPMALK and hence also the effect of lobatin B on JunB might be morecomplex than just inhibiting gene expression downstream of NPMALK signaling JunB transcript suppression by OH-CAL was fairlytransient and also the marginal pre-activation of caspase 3 was notsufficient to affect SR-786 cell viability The specific effect of lobatinB on ALCL cells caused only negligible perturbations of mitochon-drial activity which is otherwise a measure for the general toxicityof a vast variety of stressors and considered as a major trigger ofapoptotic cell death The weak inhibition of mitochondrial activi-ty was in obvious contrast to the strong inhibition of cell proliferationThis showed that growth inhibition was not due to a general tox-icity that was imposed on mitochondrial function but supposedly

Fig 8 (a) Anti-proliferative effects of TAE-684 on SR-786- HL60 cells and on PBMCs Cells were treated with the indicated TAE-684 concentrations for 8 h 24 h and 48 hand then counted by Casy (b) Effect of TAE-684 treatment on SR-786- HL60 cells and PBMC mitochondrial activity Cells were treated with 10 nM TAE-684 for 24 h and48 h and then CellTiter-Blue reagent was added and measured at 570 nm Experiments were performed in triplicate error bars indicate means plusmn SD and asterisks signifi-cance (P lt 005 ANOVA followed by Dunnett-post-test t-test)

1003I Kiss et alCancer Letters 356 (2015) 994ndash1006

to a more specificanti-proliferative activity Alternatively since lobatinB treatment caused a substantial increase of presumably telo-phase cells these may contain more mitochondria compensating theaffected mitochondrial activity of the decreased cell number

When activated by LPS NF-κB-dependent expression of E-selectin[5] and TNFα [36] was inhibited by all N lobata sesquiterpene lac-tones tested in these studies among which were also lobatin B OH-CAL and OAc-CAL The α-methylene-γ-lactone ring common to allsesquiterpene lactones of N lobata and also of the bona fide NF-κBinhibitor parthenolide was reported to cause alkylation of a cyste-ine residue in the activation loop of IκB kinaseβ [3637] therebypreventing the degradation of IκB and hence the translocation ofNF-κB into the nucleus and expression of inflammatory cytokines

Thus the inhibition of NF-κB was responsible for the anti-inflammatory property of lobatin B in a THP-1 monocyte model andin HUVEC [524] and traditional medicine makes use of it whenutilising N lobata [3839] Here we also demonstrated that lobatinB inhibited TNFα-induced NF-κB activation which was most likelyresponsible for the toxicity towards HL60 leukaemia cells This isin agreement with the fact that PBMCs remained unaffected bylobatin B because in contrast to HL60 [33] under normal cell cultureconditions NF-κB is not activated in PBMCs

Structurally the anti-inflammatory activity of the sesquiter-pene lactones was tied to the acetyl group at C-9 and the doublebonds at C-45 and C-23 [5] but this did not correlate with theanti-neoplastic property described here which also involved

Fig 9 (a) Effect on NF-κB activity HEK293-NFκB-Luc cells were stained by incubation for 1 h in serum-free medium supplemented with 2 μM Cell Tracker Green CMFDAThe cells were then reseeded in 96-well plates at a density of 4 times 104 cellswell in phenol red-free and serum-free DMEM On the next day cells were treated with 5 μMparthenolide (Parth) as a specific inhibitor of NF-κB 3 μM and 5 μM lobatin B or solvent (02 DMSO Co) 1 h after treatment cells were stimulated with 2 ngml humanrecombinant TNFα for an additional 4 h The luciferase-derived signal from the NF-κB reporter was normalised by the Cell Tracker Green CMFDA-derived fluorescence toaccount for differences in the cell number (b) Effect of lobatin B on the size of circular chemorepellet-induced defects (CCIDs) in LEC monolayers triggered by MCF-7 cellspheroids Cell cultures were pre-treated for 20 min with the indicated compound concentrations and then MCF-7 spheroids were placed on top of LEC monolayers andco-cultivated for 4 h As control (Co) CCIDs of solvent treated (02 DMSO) co-cultures were measured The CCIDs underneath 15ndash25 spheroids were analysed for each con-dition using an Axiovert microscope and Axiovision Rel 45 software from Zeiss (c) Effect on 12(S)-HETE synthesis MCF-7 cells were seeded in 35 cm dishes and grown to70 confluence and treated with 10 μM arachidonic acid together with the indicated concentrations of lobatin B for 24 h 02 DMSO was used as control (Co) The 12(S)-HETE concentration in the cell culture supernatant was determined by EIA (d) Effect on CYP1A1 activity in MCF-7 cells MCF-7 cells were kept under steroid-free conditionsand treated with the indicated concentrations of lobatin B or solvent (02 DMSO Co) 5 μM ethoxyresorufin was added and after 180 min the formation of resorufin wasanalysed which is specific for CYP1A1 activity Experiments were performed in triplicate error bars indicate means plusmn SD and asterisks significance (P lt 005 t-test ANOVAfollowed by Dunnett-post-test)

1004 I Kiss et alCancer Letters 356 (2015) 994ndash1006

NF-κB because OAc-CAL (acetyl group at C-9) did not inhibitproliferation

The inhibition of NF-κB activity was shown to prevent adhe-sion of tumour emboli to lymphendothelial cells (LECs) [19] and thisstep is necessary for the subsequent retraction of the LEC barrierallowing the tumour to transmigrate Although the structurendashactivity-relationship was not addressed in this investigation blockingNF-κB activity by lobatin B or related sesquiterpenes opens a newstrategy for the management of early steps of metastasis that doesnot exist so far

Lobatin B specifically targets cancer cells by two independentmechanisms inhibition of NF-κB and of NPMALK and does not affectnormal cells

Acknowledgments

We wish to thank Gerhard F Ecker for critically revising themanuscript and Toni Jaumlger and Sigurd Krieger for figures and datapresentations Further we thank the Austrian Exchange Service(OeAD) for a fellow-ship to CNH and CW The work was partiallysupported by a grants S10713-B13 and S10704-B13 from the Aus-trian Science Fund (FWF) to VNB and VMD a grant of theHerzfelder Family Foundation to GK and PS and a grantldquoBioProMotionrdquo Bioactivity and Metabolism from the University ofVienna Austria to SB

Conflict of interest

The authors do not have any conflict of interest

References

[1] GM Cragg DJ Newman Natural products a continuing source of novel drugleads Biochim Biophys Acta 1830 (2013) 3670ndash3695

[2] C Unger R Popescu B Giessrigl D Laimer S Heider M Seelinger et al Thedichloromethane extract of the ethnomedicinal plant Neurolaena lobata inhibitsNPMALK expression which is causal for anaplastic large cell lymphomagenesisInt J Oncol 42 (2013) 338ndash348

[3] SW Morris MN Kirstein MB Valentine KG Dittmer DN Shapiro DLSaltman et al Fusion of a kinase gene ALK to a nucleolar protein gene NPMin non-Hodgkinrsquos lymphoma Science 263 (1994) 1281ndash1284

[4] B de las Heras K Slowing J Benediacute E Carretero T Ortega C Toledo et alAntiinflammatory and antioxidant activity of plants used in traditional medicinein Ecuador J Ethnopharmacol 61 (1998) 161ndash166

[5] R McKinnon M Binder I Zupkoacute T Afonyushkin I Lajter A Vasas et alPharmacological insight into the anti-inflammatory activity of sesquiterpenelactones from Neurolaena lobata (L) RBr ex Cass Phytomedicine 21 (12) (2014)1695ndash1701

[6] V Butterweck A Nahrstedt What is the best strategy for preclinical testingof botanicals A critical perspective Planta Med 78 (2012) 747ndash754

[7] CM Passreiter D Wendisch D Gondol Sequiterpene lactones from Neurolaenalobata Phytochemistry 39 (1995) 133ndash137

[8] I Lajter A Vasas Z Beacuteni P Forgo M Binder V Bochkov et al Sesquiterpenesfrom Neurolaena lobata and their antiproliferative and anti-inflammatoryactivities J Nat Prod 77 (2014) 576ndash582

[9] G Franccedilois C Passreiter H Woerdenbag van Looveren M Antiplasmodialactivities and cytotoxic effects of aqueous extracts and sesquiterpene lactonesfrom Neurolaena lobata Planta Med 62 (1996) 126ndash129

[10] D Laimer H Dolznig K Kollmann PW Vesely M Schlederer O Merkel et alPDGFR blockade is a rational and effective therapy for NPM-ALK-drivenlymphomas Nat Med 18 (2012) 1699ndash1704

[11] AV Galkin JS Melnick S Kim TL Hood N Li L Li et al Identification ofNVP-TAE684 a potent selective and efficacious inhibitor of NPM-ALK ProcNatl Acad Sci USA 104 (2007) 270ndash275

[12] D Kerjaschki Z Bago-Horvath M Rudas V Sexl C SchneckenleithnerS Wolbank et al Lipoxygenase mediates invasion of intrametastaticlymphatic vessels and propagates lymph node metastasis of humanmammary carcinoma xenografts in mouse J Clin Invest 121 (2011) 2000ndash2012

[13] R Popescu EH Heiss F Ferk A Peschel S Knasmueller VM Dirsch et alIkarugamycin induces DNA damage intracellular calcium increase p38 MAPkinase activation and apoptosis in HL-60 human promyelocytic leukemia cellsMutat Res 10 (2011) 709ndash710

[14] KJ Livak TD Schmittgen Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(-delta delta C(T)) method Methods 25(2001) 402ndash408

[15] M Grusch D Polgar S Gfatter K Leuhuber S Huettenbrenner C Leisser et alMaintenance of ATP favours apoptosis over necrosis triggered by benzamideriboside Cell Death Differ 9 (2002) 169ndash178

[16] E Rozema AG Atanasov N Fakhrudin J Singhuber U Namduang EH Heisset al Selected extracts of Chinese herbal medicines their effect on NF-κBPPARα and PPARγ and the respective bioactive compounds Evid BasedComplement Alternat Med (2012) 983023

[17] LW Xie AG Atanasov DA Guo C Malainer JX Zhang M Zehl et alActivity-guided isolation of NF-κB inhibitors and PPARγ agonistsfrom the root bark of Lycium chinense Miller J Ethnopharmacol 152 (2014)470ndash477

[18] C Vonach K Viola B Giessrigl N Huttary I Raab R Kalt et al NF-κB mediatesthe 12(S)-HETE-induced endothelial to mesenchymal transition oflymphendothelial cells during the intravasation of breast carcinoma cells BrJ Cancer 105 (2011) 263ndash271

[19] K Viola S Kopf N Huttary C Vonach N Kretschy M Teichmann et alBay11-7082 inhibits the disintegration of the lymphendothelial barrier triggeredby MCF-7 breast cancer spheroids the role of ICAM-1 and adhesion Br J Cancer108 (2013) 564ndash569

[20] S Madlener P Saiko C Vonach K Viola N Huttary N Stark et al Multifactorialanticancer effects of digalloyl-resveratrol encompass apoptosis cell-cycle arrestand inhibition of lymphendothelial gap formation in vitro Br J Cancer 102(2010) 1361ndash1370

[21] B Giessrigl G Yazici M Teichmann S Kopf S Ghassemi AG Atanasov et alEffects of Scrophularia extracts on tumor cell proliferation death andintravasation through lymphoendothelial cell barriers Int J Oncol 40 (2012)2063ndash2074

[22] K Viola S Kopf L Rarova K Jarukamjorn N Kretschy M Teichmann et alXanthohumol attenuates tumour cell-mediated breaching of thelymphendothelial barrier and prevents intravasation and metastasis ArchToxicol 87 (2013) 1301ndash1312

[23] N Kretschy M Teichmann S Kopf AG Atanasov P Saiko C Vonach et alIn vitro inhibition of breast cancer spheroid-induced lymphendothelialdefects resembling intravasation into the lymphatic vasculature byacetohexamide isoxsuprine nifedipin and proadifen Br J Cancer 108 (2013)570ndash578

[24] S Kopf K Viola AG Atanasov K Jarukamjorn L Rarova N Kretschy et alIn vitro characterisation of the anti-intravasative properties of the marineproduct heteronemin Arch Toxicol 87 (2013) 1851ndash1861

[25] M Teichmann N Kretschy S Kopf K Jarukamjorn AG Atanasov K Viola et alInhibition of tumour spheroid-induced prometastatic intravasation gates in thelymph endothelial cell barrier by carbamazepine drug testing in a 3D modelArch Toxicol 88 (2014) 691ndash699

[26] T Hunt Cyclins and their partners from a simple idea to complicated realitySemin Cell Biol 2 (1991) 213ndash222

[27] MJ Hendzel Y Wei MA Mancini A Van Hooser T Ranalli BR Brinkley et alMitosis-specific phosphorylation of histone H3 initiates primarily withinpericentromeric heterochromatin during G2 and spreads in an ordered fashioncoincident with mitotic chromosome condensation Chromosoma 106 (1997)348ndash360

[28] PB Staber P Vesely N Haq RG Ott K Funato I Bambach et al Theoncoprotein NPM-ALK of anaplastic large-cell lymphoma induces JUNBtranscription via ERK12 and JunB translation via mTOR signaling Blood 110(2007) 3374ndash3383

[29] JD Pearson JK Lee JT Bacani R Lai RJ Ingham NPM-ALK and the JunBtranscription factor regulate the expression of cytotoxic molecules in ALK-positive anaplastic large cell lymphoma Int J Clin Exp Pathol 4 (2011)124ndash133

[30] S Mathas M Hinz I Anagnostopoulos D Krappmann A Lietz F Jundt et alAberrantly expressed c-Jun and JunB are a hallmark of Hodgkin lymphoma cellsstimulate proliferation and synergize with NF-kappa B EMBO J 21 (2002)4104ndash4113

[31] D Kardassis P Papakosta K Pardali A Moustakas C-Jun transactivates thepromoter of the human p21WAF1Cip1 gene by acting as a superactivatorof the ubiquitous transcription factor Sp1 J Biol Chem 274 (1999) 29572ndash29581

[32] V Leventaki E Drakos LJ Medeiros MS Lim KS Elaenitoba-Johnson FXClaret et al NPM-ALK oncogenic kinase promotes cell-cycle progressionthrough activation of JNKcJun signaling in anaplastic large-cell lymphomaBlood 110 (2007) 1621ndash1630

[33] SN Kang SH Kim SW Chung MH Lee HJ Kim TS Kim Enhancement of125-dihydroxyvitamin D3-induced differentiation of human leukaemia HL-60cells into monocytes by parthenolide via inhibition of NF-kB activity Br JPharmacol 135 (2002) 1235ndash1244

[34] K Uchide M Sakon H Ariyoshi S Nakamori M Tokunaga M Monden Cancercells cause vascular endothelial cell retraction via 12(S)-HETE secretion thepossible role of cancer cell derived microparticle Ann Surg Oncol 14 (2007)862ndash868

[35] KV Honn DG Tang I Grossi ZM Duniec J Timar C Renaud et al Tumourcell-derived 12(S)-hydroxyeicosatetraenoic acid induces microvascularendothelial cell retraction Cancer Res 54 (1994) 565ndash574

[36] B Walshe-Roussel C Choueiri A Saleem M Asim F Caal V Cal et al Potentanti-inflammatory activity of sesquiterpene lactones from Neurolaena lobata

1005I Kiss et alCancer Letters 356 (2015) 994ndash1006

(L) R Br ex Cass a Qrsquoeqchirsquo Maya traditional medicine Phytochemistry 92(2013) 122ndash127

[37] BHB Kwok B Koh MI Ndubuisi M Elofsson CM Crews Theanti-inflammatory natural product parthenolide from the medicinalherb feverfew directly binds to and inhibits IkB kinase Chem Biol 8 (2001)759ndash766

[38] VT Amiguet JT Arnason P Maquin V Cal P Sanchez Vindas L Poveda Aconsensus ethnobotany of the Qrsquoeqchirsquo Maya of southern Belize Econ Bot 59(2005) 29ndash42

[39] R Arvigo M Balick Rainforest Remedies second ed Lotus Press Twin LakesWI 1998

1006 I Kiss et alCancer Letters 356 (2015) 994ndash1006