Accumulation of Cytosolic Calcium Induces Necroptotic Cell ... · Following treatment with HVJ-E or A23187, SK-N-SH cells or the removed xenografts were ﬁxedwith4%paraformal- dehyde,

Molecular and Cellular Pathobiology

Accumulation of Cytosolic Calcium Induces NecroptoticCell Death in Human Neuroblastoma

Motonari Nomura1,2, Ayumi Ueno1, Kotaro Saga1, Masahiro Fukuzawa2, and Yasufumi Kaneda1

AbstractNecrosis has been studied extensively since the early days of medicine, with some patterns of necrosis found

to be programmed like apoptotic cell death. However, mechanisms of programmed necrosis (necroptosis) areyet to be fully elucidated. In this study, we investigated how the hemagglutinating virus of Japan-envelope(HVJ-E) induces necrosis in mouse xenografts of human neuroblastoma cells. HVJ-E–induced necrosis inthis system was found to depend on phosphorylation of the death receptor kinase receptor interacting proteinkinase 1 (RIP1) and on the production of reactive oxygen species. This process was interpreted as necroptosis,based on its suppression by the small molecule necrostatin-1, and it did not involve the TNF-a receptorpathway. We also demonstrated that increased concentrations of cytoplasmic calcium triggered necroptosisby activating calcium-calmodulin kinase (CaMK) II. Finally, we determined that RIP1 phosphorylation wasmediated by CaMK II activation. Together, our results define an upstream pathway for the activationof necroptosis in neuroblastoma cells, with potential therapeutic implications. Cancer Res; 74(4); 1056–66.�2013 AACR.

IntroductionThe way in which a cell dies is decided according to the

death stimuli and the endogenous expression level of deathsignaling effectors. Among the different mechanisms of celldeath, there is much more information on apoptosis thannecrosis, pyroptosis, or autophagy. Apoptosis is a well-known form of programmed cell death induced by theactivation of caspase-8 or -9. Pyroptosis is also a form ofcaspase-dependent cell death like apoptosis, but the deathstimuli are different from apoptosis. In contrast to apopto-sis, caspase-1 is activated during the process of pyroptosis bythe formation of inflammasome complex, which is inducedby the recognition of Salmonella and Shigella species (1). Inaddition, a mechanism of cell death was recently identifiedthat is morphologically necrotic but is induced by the samestimuli as apoptosis. This programmed necrotic cell death,which is referred to as necroptosis, is thought to be inducedby apoptotic death stimuli, such as TNF-a and Fas ligand.The signal transduction for necroptosis is known to becaspase independent, and its mechanism is as follows: whena ligand binds to the death receptor in the cell, receptor

interacting protein kinase 1 (RIP1) is deubiquitinated byCYLD and forms a complex with receptor interacting proteinkinase 3 (RIP3), TNF receptor–associated death domain(TRADD), Fas-associated death domain (FADD), and cas-pase-8. Most human cells express caspase-8, which sup-presses RIP1 or RIP3, but if the endogenous expression ofcaspase-8 is absent, RIP1 remains activated via phosphor-ylation of serine 161 (2–8). A necrosome consisting of RIP1,RIP3, FADD, and TRADD induces mitochondrial complexI–mediated reactive oxygen species (ROS) production, whichis an executor of necroptosis (2, 9). However, the upstreamactions of the necroptotic signaling pathway following deathreceptor stimulation, i.e., the mechanism by which RIP1 isphosphorylated, have not yet been clarified.

Neuroblastoma is one of the most common malignant solidtumors in children and is responsible for 12% of deathsassociated with childhood cancer (10). Many genetic featuresand prognostic factors of neuroblastoma have been revealed.Moreover, most aggressive neuroblastoma cells reportedly donot express caspase-8 (11, 12), a key molecule in the extrinsicpathway of apoptosis.We recently reported a novel strategy fortreating aggressive neuroblastoma with UV-treated, nonrepli-cating Sendai virus [also known as hemagglutinating virus ofJapan-envelope (HVJ-E)] particles (13). We also reported thatHVJ-E induces necrosis in the xenografts derived from humanneuroblastoma cells in severe combined immunodeficient(SCID) mice (13), but themechanism of this necrotic cell deathremains to be solved. Therefore, we reasoned that humanneuroblastoma cell lines would be the ideal experimentalmaterials for investigating the mechanism of necroptosis.

In this study, our first objective was to confirm whether theHVJ-E–induced cell death of human neuroblastoma cells isnecroptosis, and our second objective was to investigate the

Authors' Affiliations: 1Division ofGeneTherapyScience and 2Departmentof Pediatric Surgery, Graduate School of Medicine, Osaka University,Osaka, Japan

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Yasufumi Kaneda, Division of Gene TherapyScience, Graduate School of Medicine, Osaka University, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan. Phone: 81-6-6879-3901; Fax: 81-6-6879-3909; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-13-1283

�2013 American Association for Cancer Research.

CancerResearch

Cancer Res; 74(4) February 15, 20141056

on August 24, 2019. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst December 26, 2013; DOI: 10.1158/0008-5472.CAN-13-1283

http://cancerres.aacrjournals.org/

mechanism of RIP1 activation during necroptosis in cancercells.We concluded that caspase-8–deficient cancer cells, like

most neuroblastoma cells, are induced to undergo necroptosisby HVJ-E. The mechanism by which this occurs involves thephosphorylation of RIP1 by calcium-calmodulin kinase (CaMK)II, which is activated by membrane fusion with HVJ-E.

Materials and MethodsCell linesThe human neuroblastoma cell lines SK-N-SH and SK-N-AS

were obtained from the European Collection of Animal CellCultures, and themonkey kidney cell line LLCMK2 and humanprostate cancer cell line PC3 were from the American TypeCulture Collection (ATCC). The SK-N-SH, SK-N-AS, and PC3cells were maintained in Dulbecco's Modified Eagle Medium(DMEM; Nacalai Tesque Inc.), and the LLCMK2 cells weremaintained in Minimum Essential Medium (Gibco-BRL). Allmedia was supplemented with 10% FBS, 100 IU/mL penicillin,and 100 mg/mL streptomycin. The cells were incubated at 37�Cin a humidified atmosphere of 95% air and 5% CO2.

Preparation of HVJ-EHVJ (VR-105 parainfluenza1 Sendai/52, Z strain from the

ATCC) was amplified in the chorioallantoic fluid of 10- to14-day-old chicken eggs and then purified by centrifugationand inactivated by UV irradiation (99 mJ/cm2), as previouslydescribed (14). The inactivated virus lost its ability to engage ingenome replication and protein synthesis but retained itsmembrane fusion activity (14).

Generation of WT-HVJ-E, DHN–HVJ-E, andF1/F2-WT-HVJ-EThe majority of the experimental procedures have been

previously reported (15). For wild-type (WT)–HVJ-E, the cul-ture medium of HVJ-infected LLCMK2 cells was passedthrough a filter and then centrifuged at 100,000� g for 2 hoursat 4�C to precipitate the WT-HVJ particles. WT-HVJ wasinactivated by UV irradiation. For D hemagglutinin neuramin-idase (HN)–HVJ-E, the WT-HVJ particles were heated at 65�Cfor 30 minutes, thus denaturing the HN protein. F1/F2-WT-HVJ-E was prepared by treating WT-HVJ-E with 5 mg/mLtrypsin for 30 minutes at 37�C.

Transmission electron microscopyCells were fixed with 2.5% glutaraldehyde in 0.1 mol/L

phosphate buffer (pH 7.4) at 4�C and postfixed in 1% OsO4

solution at 4�C for 1 hour. The samples were then dehydratedin a graded ethanol series and embedded in Quetol 812 epoxyresin (Nissin EM). Ultrathin sections (80 nm) were cut on aReichert ultramicrotome (Ultracut E; Leica Microsystems),stained with uranyl acetate and lead citrate, and examinedwith a Hitachi H-7650 electron microscope (Hitachi).

Cell proliferation assayA MTS assay was performed with the CellTiter 96 AQueous

One Solution Cell Proliferation Assay Kit (Promega) to evaluate

cell viability. Cells were seeded in 24-well plates (5 � 104 cellsper well in 500 mL of medium). Twenty-four hours later, thecells were treated with HVJ-E [multiplicity of infection (MOI)of 100–10,000], 0.1 to 10 ng/mL of human recombinant TNF-a(BD Biosciences), or 0.1 to 10 mmol/L of the calcium ionophoreA23187 (Sigma-Aldrich Japan, Inc.). Twenty-four hours afterthis treatment, 100 mL of CellTiter 96 AQueous One SolutionReagent was added to each well, and the plates were incubatedat 37�C. After transferring 100 mL of the incubation mediumfrom each well to a new 96-well plate, the absorbance wasmeasured at 490 nm.

Western blottingAnti–caspase-3, anti–caspase-8, anti–caspase-9, anti-PARP,

anti-RIP1, anti-TNFR1, anti–pan-CaMK II, and anti–phospho-CaMK II antibodies were purchased from Cell Signaling JapanTechnology K.K., and the anti–b-actin antibody was purchasedfrom Sigma-Aldrich Japan, Inc. The anti-RIP3 antibody waspurchased from Santa Cruz Biotechnology, Inc. The anti-HNand -F antibodies were from Scrum Inc. Horseradish peroxi-dase–conjugated donkey anti-rabbit IgG (GE Healthcare) wasused as the secondary antibody for the detection of caspase-3,caspase-9, PARP, RIP1, TNFR1, pan-CaMK II, phospho-CaMKII, HN, and F. Horseradish peroxidase–conjugated sheep anti-mouse IgG (GE Healthcare) was used as the secondary anti-body for the detection of caspase-8 and b-actin. Horseradishperoxidase–conjugated chicken anti-goat IgG (R&D Systems)was used as the secondary antibody for the detection of RIP3.Cell lysates were separated using polyacrylamide gels, and theproteins were transferred onto polyvinylidene difluoridemem-branes. The membranes were then blocked with blockingbuffer (TBS containing 0.1% Tween 20 and 3% skim milk) for1 hour and incubated with primary antibody in blocking bufferat 4�C overnight. The membranes were then washed withwashing buffer (TBS containing 0.1% Tween 20) and incubatedwith the horseradish peroxidase–conjugated anti-mouse, anti-goat, or anti-rabbit secondary antibody (GE Healthcare) inblocking buffer for 1 hour. Signals were detected with Chemi-Lumi One (Nakalai Tesque Inc.) according to the manufac-turer's instructions.

Measurement of ROS productionROS generation was detected using a fluorescence probe

called MitoSOX, which detects mitochondrial superoxideproduction. The cells were pretreated with 5 mmol/L MitoSOXat 37�C for 10 minutes according to the manufacturer'sinstructions.

Estimation of the intracellular Ca2þ concentrationThe intracellular Ca2þ concentration was estimated with

the Fura 2 Kit (DOJINDO) according to the manufacturer'sprotocol. Briefly, SK-N-SH cells (1� 104 cells) were seeded in a96-well plate and incubated overnight. Then, the medium wasremoved, and the manufacturer's loading buffer was addedto the cells. The cells were incubated at 37�C for 1 hour, andthe buffer was replaced with recording buffer containingFura 2-acetoxymethyl ester. Next, 1,000 MOI of HVJ-E withWT HN–inactivated F (WT.HN-F0) or WT HN-activated

Activated CaMK II Induces Necroptosis in Human Neuroblastoma

www.aacrjournals.org Cancer Res; 74(4) February 15, 2014 1057




F (WT.HN-F1/F2) were added to each well, and the resultingfluorescence intensity was estimated by recording the absor-bance at the peak excitation wavelength of 340/380 nm andthe peak emission wavelength of 510 nm.

Detection of apoptosis and necrosisFollowing treatment with HVJ-E or A23187, SK-N-SH cells

or the removed xenografts were fixed with 4% paraformal-dehyde, and apoptosis or necrosis was detected with theGFP-Certified Apoptosis/Necrosis Detection System Kit(Enzo Life Science Inc.) according to the manufacturer'sprotocol.

Pretreatment of cells with inhibitorsTo inhibit the activity of pan-caspase, 20 mmol/L Z-VAD-

FMK (Medical & Biological Laboratories) was added tothe cells 1 hour before treatment with HVJ-E. To inhibitthe activity of RIP1, 20 mmol/L necrostatin-1 (Enzo LifeScience Inc.) was added to the cells 24 hours before expo-sure to HVJ-E. To inhibit the increase of cytoplasmic Ca2þ,10 mmol/L BAPTA-AM (Sigma-Aldrich Japan, Inc.) wasadded to the cells 30 minutes before exposure to HVJ-E orA23187. To inhibit the activity of CaMK II, 20 mmol/L CK59(EMD Millipore) was added to the cells 1 hour before HVJ-Eexposure.

Cell transfection with siRNASiGENOME SMARTpool Human RIP1 (M-004445-02-0005;

Thermo Fisher Scientific), siGENOME SMARTpool HumanRIP3 (M-003534-01-0005; Thermo Fisher Scientific), orON-TARGET plus Human CAMK IIa siRNA SMARTpool(L-004942-00-0005; Thermo Fisher Scientific) was used totransfect SK-N-SH cells. The siGENOME Non-TargetingsiRNA Pool (D-001206-13-05; Thermo Fisher Scientific) wasused as a control. The siRNA transfection was performed asfollows: siRNA and Lipofectamine RNAiMAX (Invitrogen)were mixed in Opti-MEM and incubated for 20 minutes atroom temperature, and then the mixture was added to cul-tured SK-N-SH cells in 6- or 24-well plates. The cells wereincubated for 4 hours at 37�C, and the medium was exchang-ed with DMEM. The siRNA dosage was 100 or 20 pmol/well.The knockdown efficiency was confirmed by Western blot-ting at 72 hours after transfection.

Labeling of cancer cells with 32P-orthophosphate andimmunoprecipitation

SK-N-SH cells [1 � 106 cells per immunoprecipitation (IP)sample] were seeded 1 day before the labeling experiment,resuspended in phosphate-free medium, and incubated for40 minutes at 37�C. The cells were centrifuged and resus-pended in phosphate-free medium containing 0.1 mCi of32P-orthophosphate (Perkin Elmer, Inc.) and incubated for2 hours before stimulation with 1,000 MOI of HVJ-E for 2hours. In some experiments, 20 mmol/L necrostatin-1 or 20mmol/L CK-59 was added to the cells before HVJ-E stimulation.The cells were treated with HVJ-E for 2 hours before cell lysis.For immunoprecipitation, the cell lysates were preclearedusing 5mg of the Normal Rabbit IgG–Agarose Conjugate (Santa

Cruz Biotechnology, Inc.) plus 30 mL of Protein A Agarose FastFlow (Millipore) for 1 hour at 4�C. The agarose beads andnonspecifically associating proteins were removed by centri-fugation at 12,000 � g for 10 minutes at 4�C. The supernatantwas incubated with 5 mg of anti-RIP1 antibody or RabbitControl IgG (ChiP grade; Abcam) plus 30 mL of ProteinA Agarose Fast Flow for 12 hours at 4�C. The agarose beadswere centrifuged at 3,000� g for 2minutes at 4�C,washed threetimes, and centrifuged as described above. The washedbeads were boiled three times for 5 minutes at 95�C andresolved by SDS–PAGE. The gel was dried and subjected toautoradiography.

Statistical analysesThe data are expressed as the means � SD. A two-tailed

unpaired Student t test was used to determine the statisticalsignificance of the difference between two groups. Proba-bility values of P < 0.05 were considered to be statisticallysignificant.

ResultsHVJ-E induces necrotic cell death in neuroblastoma cells

We first analyzed the morphologic changes induced byHVJ-E in SK-N-SH cells. Twenty-four hours after the treat-ment of SK-N-SH cells with 1,000 MOI of HVJ-E, we observedswollen nuclei by transmission electron microscopy (Fig. 1A).Whether the whole cell body was swollen by HVJ-E could notbe confirmed because the cells were fused together (Supple-mentary Fig. S1A). A MTS assay showed that the cell viabilitysignificantly decreased in a dose-dependent manner follow-ing treatment with different MOIs (100, 1,000, or 10,000) ofHVJ-E in SK-N-SH and SK-N-AS cells (Fig. 1B). The HVJ-E–induced cell death was not significantly suppressed by thepretreatment with cytochalasin D, which binds to actinfilaments and thus blocks actin polymerization, suggestingthat cell fusion does not contribute to this cell death (Sup-plementary Fig. S1B).

HVJ-E–induced cell death in neuroblastoma cells is notapoptosis but necroptosis

Although we observed morphologic changes in necroticSK-N-SH cells, we next needed to analyze the degree atwhich apoptosis contributes to this cell death. The cleavageof caspase-9, caspase-3, and PARP was detected by Westernblot analysis at 12 hours after treatment with 1,000 MOIof HVJ-E, but caspase-8 was not endogenously expressed(Fig. 2A). However, the inhibition of PARP cleavage by Z-VAD-FMK could not significantly suppress HVJ-E–inducedcell death (Fig. 2B and C). These results imply that this celldeath is not caspase-dependent apoptosis. We could notnarrow down the key factors responsible for HVJ-E–inducedcell death. Therefore, we exhaustively investigated the chan-ges in gene expression during this process by microarrayanalysis. However, no remarkable changes in the expressionof individual genes were detected. Hence, we subjected theexpressed genes to pathway analysis using MetaCore v6.5.The pathway analysis showed significant changes in some

Nomura et al.

Cancer Res; 74(4) February 15, 2014 Cancer Research1058




pathways, including oxidative phosphorylation, ubiquinonemetabolism, cytoskeleton remodeling, the immune re-sponse, and others (data not shown). Of these processes,

we focused on the oxidative phosphorylation pathwaybecause it produces ROS, which is one of the key executorsof necrosis. Taking into consideration our findings on the

Figure 1. Neuroblastoma cell deathinducedbyHVJ-E. A, 24hours afterthe treatment of SK-N-SH cellswith 1,000 MOI of HVJ-E, theswollen nuclei (arrowhead) of thecells were observed by electronmicroscopy. Scale bar, 10 mm. B,the treatment of SK-N-SH andSK-N-AS cells (3 � 105 cells) withHVJ-E for 24 hours resulted in thecell death in a dose-dependentmanner. Each survival value (mean� SD; n ¼ 4) was the ratio of thevalue with treatment to the valuewithout treatment. �, P < 0.05.

Figure 2. HVJ-E–induced cell deathin SK-N-SH cells is not caspase-dependent apoptosis. A, SK-N-SHcells (3 � 105 cells) were treatedwith 1,000 MOI of HVJ-E for 6, 12,or 24 hours. The cleavage ofcaspase-9, caspase-3, and PARPwas observed 12 hours after HVJ-Eexposure, but caspase-8 was notendogenously expressed. B, theHVJ-E–induced cleavage of PARPwas inhibited by pretreatment withZ-VAD-FMK. C, the HVJ-E–induced cell death was notsignificantly suppressed bypretreatment with Z-VAD-FMK.The experiments were performedin triplicate, and representativeresults are shown. Each survivalvalue (mean � SD; n ¼ 4) was theratio of the value with treatment tothe value without treatment.






necrotic morphologic changes, nonapoptotic cell death andcaspase-8 defects, we hypothesized that the HVJ-E–induceddeath of SK-N-SH cells is actually necroptosis. To confirmthis hypothesis, we first used the RIP1 inhibitor necrostatin-1. The pretreatment of SK-N-SH cells with necrostatin-1before HVJ-E exposure significantly suppressed cell deathby inhibiting ROS production without inhibiting PARPcleavage (Fig. 3A–C). We confirmed that the HVJ-E–inducedcell death was suppressed by pretreatment with necrosta-tin-1 also in SK-N-AS cells (Fig. 3C). To show that the lack ofcaspase-8 sensitizes the cells to necroptosis, caspase-8cDNA was transferred to SK-N-SH cells. The restored cas-pase-8 in SK-N-SH cells resulted in the significant suppres-sion of HVJ-E–induced cell death (Supplementary Fig. S2Aand S2B). From these data, we suspected that HVJ-E–induced cell death might associate with the RIP1–RIP3necrosome in caspase-8–deficient neuroblastoma cells. Weused necrosulfonamide as the inhibitor of necroptosis,which blocks further downstream of necroptosis. The pre-treatment of SK-N-SH cells with necrosulfonamide beforeHVJ-E exposure also significantly suppressed the cell death

as well as necrostatin-1 (Supplementary Fig. S3). Knock-down of RIP3 using siRNA also suppressed the HVJ-E–induced cell death (Supplementary Fig. S4A and S4B). Tofurther evaluate the involvement of RIP1 and RIP3 in HVJ-E–induced cell death, we demonstrated the HVJ-E–inducedformation of RIP1–RIP3 necrosome by in situ proximityligation assay, which can detect the protein–protein inter-actions using two antibodies derived from different species(Supplementary Fig. S5). These results imply that HVJ-Einduces necroptosis in SK-N-SH cells.

HVJ-E induces cell death in neuroblastoma cells byenhancing the cytoplasmic Ca2þ concentration

We next attempted to unveil the upstream actions of HVJ-E–induced necroptosis in SK-N-SH cells. Necroptosisrequires the phosphorylation of RIP1, which forms a com-plex with the TNF-a receptor in its ubiquitinated formbefore it is deubiquitinated by CYLD (9). By using 32Plabeling, we demonstrated that RIP1 phosphorylation wasinduced by HVJ-E treatment for 2 hours, and this phosphor-ylation was suppressed by RIP1 knockdown (Fig. 3D). If this

Figure 3. HVJ-E inducesnecroptosis in SK-N-SH cells viaRIP1 phosphorylation. A, SK-N-SHcells (3 � 105 cells) with or withoutpretreatment with 20 mmol/L ofnecrostatin-1 (Nec-1) wereexposed to 1,000 MOI of HVJ-E for24 hours. Western blot analysisshowed that the expression ofRIP1was slightly suppressed bypretreatment with necrostatin-1,but the cleavage of PARP was notinhibited. B, ROS production wasdetected by MitoSOX Red staining(red). The treatment of SK-N-SHcells with 1,000 MOI of HVJ-Einduced ROS production, whichwas inhibited by pretreatment withnecrostatin-1. The nuclei werestainedwith DAPI (blue). C, HVJ-E–induced cell deathwassignificantlysuppressed by pretreatment withnecrostatin-1 both in SK-N-SHandSK-N-AS cells. D, SK-N-SH cells(1 � 106 cells) were transfectedwith RIP1 siRNA for 24 hours. Twohours after labeling of thetransfected cells with 3.7 MBq of32P, the cells were treated with1,000MOI of HVJ-E for 2 hours andthen immunoprecipitated with anti-RIP1 antibody. HVJ-E inducedRIP1 phosphorylation, which wassuppressed by the knockdown ofRIP1. Each survival value (mean �SD; n¼ 4) was the ratio of the valuewith HVJ-E treatment to thevalue without HVJ-E treatment.�, P < 0.05.

Nomura et al.





SK-N-SH cell death is mediated by the TNF receptor, necrop-tosis should be induced by exposure to TNF-a. However,TNF-a was not able to induce cell death in SK-N-SH cells,and moreover, the TNF receptor was not endogenouslyexpressed in SK-N-SH cells (Supplementary Fig. S6A andS6B), implying that this HVJ-E–induced cell death is notmediated by TNF receptor signaling. We confirmed that SK-N-SH cells could be killed only by HVJ-E with the WT HNprotein (WT.HN) and HVJ-E with an activated fusion (F)protein (F1/F2) but not by HVJ-E with either denatured HN(DHN) or inactivated F (F0; Fig. 4A). Moreover, HVJ-E withF1/F2 significantly increased the cytoplasmic Ca2þ concen-tration compared with HVJ-E with F0 at 20 minutes aftertreatment of the SK-N-SH cells with 1,000 MOI of HVJ-E (Fig.4B). This HVJ-E–induced cell death was significantly sup-pressed by pretreatment with the Ca2þ-chelating agentBAPTA-AM (Fig. 4C), suggesting that a cytoplasmic Ca2þ

increase induces cell death and that this increase in Ca2þ

requires the fusion of HVJ-E to its target cells by theactivated F protein.

RIP1 phosphorylation in necroptosis is mediated byCaMK II

These results imply that a cytoplasmic Ca2þ increase upre-gulates HVJ-E–induced necroptosis. RIP1 activation requiresthe phosphorylation of serine 161. Among the serine/threoninekinases, protein kinase C (PKC) and CaMK II are downstreamof Ca2þ signaling and, therefore, they may be candidates forRIP1 phosphorylation. Pretreatment with the PKC inhibitorbisindolylmaleimide before HVJ-E exposure did not signifi-cantly suppress HVJ-E–induced cell death, suggesting thatPKC is not related to this type of cell death (SupplementaryFig. S7). Next, we detected the phosphorylation of CaMK IIafter treating SK-N-SH and SK-N-AS cells with HVJ-E(Fig. 5A). The HVJ-E–induced phosphorylation of CaMK IIwas inhibited by pretreatment with BAPTA-AM, implyingthat the increase in intracellular Ca2þ activates CaMK II(Fig. 5B). Knockdown of CaMK IIa resulted in the significantsuppression of HVJ-E–induced cell death (SupplementaryFig. S8A and S8B). The endogenous expression of CaMK IIbwas not detected (data not shown). Moreover, the inhibition

Figure 4. HVJ-E induces cell deathin SK-N-SH cells by enhancing theintracellular Ca2þ concentration.HVJ particles produced by LLC-MK2 cells were either treated withtrypsin for cleaving the F0 proteinto form F1/F2 protein or heated todenature the HN protein. A, theexpression of the HN, F0, or F1proteins was detected by Westernblot analysis. WT.HN, wild-typeHN; DHN, denatured HN. Twenty-four hours after the treatment ofSK-N-SH cells (3 � 105 cells) with1,000 or 10,000 MOI of the varioustypes of HVJ-E, only the HVJ-Ewith WT.HN-F1/F2 protein couldkill the cancer cells. B, SK-N-SHcells (1 � 104 cells) were treatedwith 1,000 MOI of HVJ-E withWT.HN-F0 or WT.HN-F1/F2protein. Only the HVJ-E with WT.HN-F1/F2 could enhance theintracellular Ca2þ concentration. C,HVJ-E–induced cell death wasinhibited by pretreatment withBAPTA-AM. Each survivalvalue (mean � SD;n ¼ 4) or intracellular Ca2þ

concentration reflects the ratio ofthe value with HVJ-E treatment tothe value without HVJ-E treatment.The experiments were performedin triplicate, and representativeresults are shown. �, P < 0.05.






of CaMK II phosphorylation by pretreatment with the CaMK IIinhibitor CK59 significantly suppressed HVJ-E–induced celldeath in SK-N-SH and SK-N-AS cells (Fig. 5C and D), implyingthat CaMK II phosphorylation is related to this type of celldeath.

The final objective of this study was to investigate therelationship between necroptosis and CaMK II phosphor-ylation. The HVJ-E–induced phosphorylation of RIP1, whichwas labeled using 32P, was significantly suppressed bypretreatment with necrostatin-1 and CK59 (Fig. 6A), sug-gesting that CaMK II phosphorylation is necessary for RIP1phosphorylation. Moreover, using 7-Amino-Actinomycin D(AAD) staining, we confirmed that HVJ-E induced necrop-tosis in SK-N-SH cells and showed that this necroptosis wasinhibited by CK59 (Fig. 6B). The possibility that phosphor-ylated CaMK II is an upstream component of RIP1-depen-dent necroptosis was also supported by our results showingthat CK59 pretreatment before HVJ-E exposure decreasedROS production and increased intracellular ATP (Supple-mentary Fig. S9A and S9B). This HVJ-E–induced necroptosis

was also observed in xenograft tumors derived from SK-N-SH cells in SCID mice using AnnexinV-EnzoGold and 7-AAD(Supplementary Fig. S10).

In summary, we demonstrated that HVJ-E increases thecytoplasmic Ca2þ concentration followed by CaMK II phos-phorylation, and that this activated CaMK II induces RIP1phosphorylation and necroptosis (Fig. 7). To show that theincrease of cytoplasmic Ca2þ induces necroptosis, we usedCa2þ ionophore A23187 instead of HVJ-E. As shown in Sup-plementary Fig. S11A, A23187 induced SK-N-SH cell death in adose-dependent manner, and this cell death was significantlysuppressed by pretreatment with the Ca2þ chelating agentBAPTA-AM. Using 7-AAD staining, we demonstrated that thisA23187-induced cell death was necroptosis (SupplementaryFig. S11B).

DiscussionNecroptosis is inducible in many types of cells if apo-

ptotic death signaling is inhibited by pretreatment with Z-

Figure 5. HVJ-E induces cell death via CaMK IIphosphorylation. A, SK-N-SH and SK-N-AScells (3� 105 cells) were treatedwith 1,000MOIof HVJ-E, and CaMK II phosphorylation wasdetected by Western blot analysis. B, the HVJ-E–induced phosphorylation of CaMK II wasinhibited by pretreatment with BAPTA-AM inSK-N-SK cells. C, the HVJ-E–inducedphosphorylation of CaMK II was inhibited bypretreatment with CK59, which was confirmedby Western blot analysis. D, the HVJ-E–induced cell death was significantlysuppressed by pretreatment with CK59 both inSK-N-SH and SK-N-AS cells. Each survivalvalue (mean � SD; n ¼ 4) was the ratio of thevaluewithHVJ-E treatment to the valuewithoutHVJ-E treatment. Experimentswere performedin triplicate, and representative results areshown. �, P < 0.05.

Nomura et al.





VAD-FMK and cycloheximide before being exposed todeath ligands, such as TNF-a and the Fas ligand, and theresulting morphologic features are similar to those ofnecrosis (2, 3). In the ischemic brain, this type of celldeath tends to be induced, and necrostatin-1 was firstreported to be the agent that suppresses ischemic braininjury in mice through a mechanism that is distinct fromapoptosis (16). Necrostatin-1 was later discovered to be aspecific inhibitor of RIP1 (17). Therefore, cell death that isrescued by necrostatin-1 can be considered to be necrop-tosis, and RIP1 is believed to be the key necroptosis factor.It was recently reported that in addition to RIP1, RIP3 isessential for necroptosis because RIP3 is regulated by thecaspase-8–FLIP complex (18) and mediates the embryoniclethality of caspase-8–deficient animals (19). Moreover,RIP3-knockout mice are very vulnerable to some viruses(6). Thus, necroptosis plays an important role in theinflammatory response or innate immune response to virusinfection. The representative stimulator of necroptosis isTNF-a, and the mechanism of TNF-a–induced necroptosishas been well studied, but we could not study this factorhere because its receptor is absent in SK-N-SH cells.Similar to HVJ-E in this study, other agents have been

reported to stimulate necroptosis, such as kuguaglycosideC, a constituent of Momordica charantia (20). We choseRIP1 as the necroptosis marker for this study because aspecific inhibitor of RIP3 was not available. The key execu-tors of necroptosis are RIP1 and RIP3, and the downstreamfactors of these kinases are ROS and apoptosis-inducingfactor (AIF). AIF is a Janus protein that exerts redoxactivity in the mitochondria and proapoptotic activity inthe nucleus, but it can also regulate necroptosis (21). ROSalso regulates apoptosis through other mechanisms involv-ing AIF. Thus, the absolute markers of necroptosis thatexist downstream of RIP1 and RIP3 have not yet beendiscovered. However, we chose ROS and ATP rather thanAIF as markers of HVJ-E–induced necroptosis because achange in the oxidative phosphorylation pathway wasdetected in our pathway analysis. We have herein demon-strated that HVJ-E induces necroptosis in the neuroblas-toma cell lines SK-N-SH and SK-N-AS.

We previously reported that HVJ-E induces apoptosis incastration-resistant human prostate cancer cells andhuman glioblastoma cells but does not cause toxicity innormal cells (22–24). Furthermore, we reported that HVJ-Estimulates an antitumor immune response by activating

Figure 6. HVJ-E–inducednecroptosis is upregulated byCaMK II. A, SK-N-SH cells (1� 106

cells) were pretreated withnecrostatin-1 or CK59. Two hoursafter treatment of the cells with 3.7MBq of 32P, the cells were treatedwith1,000MOIofHVJ-E for 2 hoursand immunoprecipitated with anti-RIP1 antibody. The HVJ-E–induced phosphorylation of RIP1was inhibited by pretreatment withnecrostatin-1 and CK59. B, theHVJ-E–induced necroptosis, asdetected by 7-AAD staining (red),was inhibited by pretreatment withnecrostatin-1 andCK59. Thenucleiwere stained with DAPI (blue).






cytotoxic T lymphocytes and natural killer cells and sup-pressing regulatory T cells (25, 26). Among these direct andindirect HVJ-E antitumor activities, we recently discoveredthat HVJ-E induces apoptosis via activation of the retinoicacid–inducible gene-I (RIG-I)/mitochondrial antiviral-sig-naling protein pathway followed by activation of interferonregulatory transcription factor (IRF)3 or IRF7 in prostatecancer cells (27). HVJ is a mouse parainfluenza virus thatbelongs to the paramixoviridae genus. Two glycoproteins,HN and F, are present on the viral envelope (28). Aninfection occurs when HN binds to its receptor followedby the fusion of the hydrophobic region of the F protein tothe lipid bilayer through its association with lipid mole-cules, such as cholesterol (29). After this first infection step,the HVJ virus genome is recognized by RIG-I, and deathsignaling effectors are activated (30, 31). However, no RIG-Iexpression was detected in the SK-N-SH cells in this study(data not shown). Therefore, we hypothesized that somemechanism that does not involve the innate immunesystem is associated with HVJ-E–induced necroptosis. Ithas been reported that HVJ-E enhances the cytoplasmicCa2þ concentration of target cells upon fusion with the cell

membrane, (28, 32–35) and we could confirm this increasein intracellular Ca2þ.

Many intracellular Ca2þ effects are mediated by proteinphosphorylation events that are catalyzed by a family ofserine/threonine protein kinases called CaM kinases. SomeCaM kinases phosphorylate gene regulatory proteins, suchas CREB, and they regulate the transcription of specificgenes. One of the best-studied CaM kinases is CaMK II,which is expressed in most animal cells and is especiallyabundant in the nervous system. It contributes up to 2% ofthe total protein of some regions in the brain and is highlyconcentrated in the synapses. CaMK II has a remarkableproperty in that it can function as a molecular memorydevice, switching to an activated state when exposed to Ca2þ

/calmodulin and then remaining activated (36, 37). This isbecause the kinase phosphorylates itself (autophosphoryla-tion) and it also phosphorylates other proteins followingactivation by Ca2þ/calmodulin. In this autophosphorylatedstate, the enzyme remains activated even in the absence ofCa2þ, thereby prolonging the duration of its kinase activity.The enzyme maintains this activity until serine/threonineprotein phosphatases suppress phosphorylated CaMK II andshut it off. CaMK II activation can thereby serve as a memorytrace of a prior Ca2þ pulse, and it seems to have a role insome types of memory and learning in the vertebrate ner-vous system. We herein demonstrated that HVJ-E enhancesthe cytoplasmic Ca2þ concentration for a short time fol-lowed by the phosphorylation of CaMK II for a longer period.This finding suggests that the temporal increase in cyto-plasmic Ca2þ can trigger CaMK II autophosphorylation,thereby causing necroptosis via RIP1 phosphorylation incaspase-8–deficient neuroblastoma cells. The first step inRIP1 activation is the deubiquitination of RIP1 by CYLD.RIP1 can then freely migrate to the cytoplasm and form acomplex with RIP3. Although little is known about how thiscomplex formation is regulated, it was recently reported thatRIP1–RIP3 complex formation requires RIP1 deacetylationby NAD-dependent deacetylase SIRT2 (38). The final step ofRIP1 activation requires the phosphorylation of serine 161 toproduce ROS through the mitochondrial oxidative phos-phorylation pathway, but the factor that phosphorylatesRIP1 has not yet been discovered. Although it is not clearwhether the regulation of RIP1 by CaMK II is direct orindirect, we are the first to discover one of the downstreamCa2þ signaling pathways associated with necroptotic celldeath.

In conclusion, an increase in the cytoplasmic Ca2þ concen-tration induces necroptosis via CaMK II phosphorylation incaspase-8–deficient neuroblastoma cells.

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

Authors' ContributionsConception and design: M. Nomura, M. Fukuzawa, Y. KanedaDevelopment of methodology: M. Nomura, Y. KanedaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. Nomura, A. UenoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): M. Nomura, Y. Kaneda

Figure 7. Schematic representation of death signaling stimulated byHVJ-E. HVJ-E enhances the cytoplasmic Ca2þ concentration bybinding and fusing the target cells. Then, CaMK II is phosphorylatedand thus activated. Once CaMK II is phosphorylated, it remainsactivated for a long time and exerts its kinase activity on its targetproteins. Among the downstream effects of activated CaMK II, RIP1-dependent necroptosis is induced in the absence of endogenouslyexpressed caspase-8.

Nomura et al.





Writing, review, and/or revision of the manuscript:M. Nomura, Y. KanedaAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M. Nomura, K. Saga, Y. KanedaStudy supervision: Y. Kaneda

AcknowledgmentsThe authors thank Kikuya Kato (Research Institute, OsakaMedical Center for

Cancer and Cardiovascular Diseases, Osaka, Japan) for performing the micro-array analysis and Dr. Ryo Matoba (DNA Chip Research Inc., Yokohama, Japan)for performing the pathway analysis.

Grant SupportThis study was supported by the Program for the Promotion of Funda-

mental Studies in Health Sciences of the National Institute of BiomedicalInnovation (Project ID: 10-03).

The costs of publication of this article were defrayed in part by the pay-ment of page charges. This article must therefore be hereby marked adver-tisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 3, 2013; revised November 8, 2013; accepted November 29, 2013;published OnlineFirst December 26, 2013.

References1. Mariathasan S, Monack DM. Inflammasome adaptors and sensors:

intracellular regulators of infection and inflammation. Nat Rev Immunol2007;7:31–40.

2. Vandenabeele P, Galluzzi L, Vanden Berghe T, Kroemer G. Molecularmechanisms of necroptosis: an ordered cellular explosion. Nat RevMol Cell Biol 2010;11:700–14.

3. Holler N, Zaru R,MicheauO, ThomeM, Attinger A, Valitutti S, et al. Fastriggers an alternative, caspase-8–independent cell death pathwayusing the kinase RIP as effector molecule. Nat Immunol 2000;1:489–95.

4. Peter ME. Programmed cell death: apoptosis meets necrosis. Nature2011;471:310–2.

5. Galluzzi L, Kroemer G. Necroptosis: a specialized pathway of pro-grammed necrosis. Cell 2008;135:1161–3.

6. Cho YS, Challa S, Moquin D, Genga R, Ray TD, Guildford M, et al.Phosphorylation-driven assembly of theRIP1-RIP3 complex regulatesprogrammed necrosis and virus-induced inflammation. Cell 2009;137:1112–23.

7. Declercq W, Vanden Berghe T, Vandenabeele P. RIP kinases at thecrossroads of cell death and survival. Cell 2009;138:229–32.

8. Green DR, Oberst A, Dillon CP, Weinlich R, Salvesen GS. RIPK-dependent necrosis and its regulation by caspases: a mystery in fiveacts. Mol Cell 2011;44:9–16.

9. Vandenabeele P, Declercq W, Van Herreweghe F, Vanden Berghe T.The role of the kinases RIP1 and RIP3 in TNF-induced necrosis. SciSignal 2010;3:re4.

10. Surveillance, Epidemiology, and End Results (SEER) Program. SEER*-Stat mortality database: total U.S. (1969–2006), National CancerInstitute, DCCPS, Surveillance Research Program [database on theInternet]. Atlanta, GA: Centers for Disease Control and Prevention.Available from: www.cdc.gov/nchs. Accessed August 3, 2012.

11. Haase GM, Perez C, Atkinson JB. Current aspects of biology, riskassessment, and treatment of neuroblastoma Semin Surg Oncol1999;16:91–104.

12. Zauli G, Milani D, Rimondi E, Baldini G, Nicolin V, Grill V, et al. TRAILactivates a caspase 9/7-dependent pathway in caspase 8/10-defective SK-N-SH neuroblastoma cells with two functional endpoints: induction of apoptosis and PGE2 release. Neoplasia 2003;5:457–66.

13. Nomura M, Shimbo T, Miyamoto Y, Fukuzawa M, Kaneda Y. 13-Cisretinoic acid can enhance the antitumor activity of non-replicatingSendai virus particle against neuroblastoma. Cancer Sci 2013;104:238–44.

14. Kaneda Y, Nakajima T, Nishikawa T, Yamamoto S, Ikegami H, SuzukiN, et al. Hemagglutinating virus of Japan (HVJ) envelope vector as aversatile gene delivery system. Mol Ther 2002;6:219–26.

15. Saga K, Tamai K, Yamazaki T, Kaneda Y. Systemic administration of anovel immune-stimulatory pseudovirion suppresses lung metastaticmelanoma by regionally enhancing IFN-g production. Clin Cancer Res2013;19:668–79.

16. Degterev A, Huang Z, Boyce M, Li Y, Jagtap P, Mizushima N, et al.Chemical inhibitor of nonapoptotic cell death with thera-peutic potential for ischemic brain injury. Nat Chem Biol 2005;1:112–9.

17. Degterev A,Hitomi J, GermscheidM,Ch'en IL, KorkinaO, TengX, et al.Identification of RIP1 kinase as a specific cellular target of necrostatinsNat Chem Biol 2008;4:313–21.

18. Oberst A, Dillon CP, Weinlich R, McCormick LL, Fitzgerald P, Pop C,et al. Catalytic activity of the caspase-8-FLIP(L) complex inhibitsRIPK3-dependent necrosis. Nature 2011;471:363–7.

19. Kaiser WJ, Upton JW, Long AB, Livingston-Rosanoff D, Daley-BauerLP, Hakem R, et al. RIP3 mediates the embryonic lethality of caspase-8-deficient mice. Nature 2011;471:368–72.

20. Tabata K, Hamano A, Akihisa T, Suzuki T. Kuguaglycoside C, aconstituent of Momordica charantia, induces caspase-indepen-dent cell death of neuroblastoma cells. Cancer Sci 2012;103:2153–8.

21. Delavall�ee L, Cabon L, Gal�an-Malo P, Lorenzo HK, Susin SA. AIF-mediated caspase-independent necroptosis: a new chance for tar-geted therapeutics. IUBMB Life 2011;63:221–32.

22. Kawaguchi Y, Miyamoto Y, Inoue T, Kaneda Y. Efficient eradication ofhormone-resistant human prostate cancers by inactivated Sendaivirus particle. Int J Cancer 2009;124:2478–87.

23. Tanaka M, Shimbo T, Kikuchi Y, Matsuda M, Kaneda Y. Sterilealpha motif containing domain 9 (SAMD9) is involved in deathsignaling of malignant glioma treated with inactivated Sendaivirus particle (HVJ-E) or type I interferon. Int J Cancer 2010;126:1982–91.

24. Kaneda Y. Update on non-viral delivery methods for cancertherapy; possibilities of DDS with anti-cancer activities beyonddelivery as a new therapeutic tool. Expert Opin Drug Deliv 2010;9:1079–93.

25. Kurooka M, Kaneda Y. Inactivated Sendai virus particles eradicatetumors by inducing immune responses through blocking regulatoryT cells. Cancer Res 2007;67:227–36.

26. Fujihara A, Kurooka M, Miki T, Kaneda Y. Intratumoral injection ofinactivated Sendai virus particles elicits strong antitumor activity byenhancing local CXCL10 expression and systemic NK cell activation.Cancer Immunol Immunother 2008;57:73–84.

27. Matsushima-Miyagi T, Hatano K, Nomura M, Li-Wen L, Nishikawa T,Saga K, et al. TRAIL and Noxa are selectively upregulated inprostate cancer cells downstream of the RIG-I/MAVS signalingpathway by nonreplicating Sendai virus particles. Clin Cancer Res2012;18:6271–83.

28. Okada Y. Sendai virus-induced cell fusion. Methods Enzymol 1993;221:18–41.

29. Kaneda Y. Applications of hemagglutinating virus of Japan intherapeutic delivery systems. Expert Opin Drug Deliv 2008;5:221–33.

30. Takeuchi O, Akira S. Pattern recognition receptors and inflammation.Cell 2010;140:805–20.

31. Rehwinkel J, Tan CP, Goubau D, Schulz O, Pichlmair A, Bier K, et al.RIG-I detects viral genomic RNA during negative-strand RNA virusinfection. Cell 2010;140:397–408.

32. Hallett MB, Fuchs P, Campbell AK. Sendai virus causes a rise inintracellular free Ca2þ before cell fusion. Biochem J 1982;206:671–4.

33. KunzelmannK,K€onig J, Sun J,MarkovichD,KingNJ, KarupiahG, et al.Acute effects of parainfluenza virus on epithelial electrolyte transport.J Biol Chem 2004;279:48760–6.

34. Falvo JV, Uglialoro AM, Brinkman BM, Merika M, Parekh BS, Tsai EY,et al. Stimulus-specific assembly of enhancer complexes on thetumor necrosis factor alpha gene promoter. Mol Cell Biol 2000;20:2239–47.






35. Zhu J, Coyne CB, Sarkar SN. PKC alpha regulates Sendai virus-mediated interferon induction through HDAC6 and b-catenin. EMBOJ 2011;30:4838–49.

36. BayerKU,DeKoninckP, LeonardAS,Hell JW,SchulmanH. Interactionwith the NMDA receptor locks CaMK II in an active conformation.Nature 2001;411:801–5.

37. Meyer T, Hanson PI, Stryer L, Schulman H. Calmodulin trapping bycalcium-calmodulin-dependent protein kinase. Science 1992;256:1199–202.

38. Narayan N, Lee IH, Borenstein R, Sun J, Wong R, Tong G, et al. TheNAD-dependent deacetylase SIRT2 is required for programmednecrosis. Nature 2012;492:199–204.

Nomura et al.





2014;74:1056-1066. Published OnlineFirst December 26, 2013.Cancer Res Motonari Nomura, Ayumi Ueno, Kotaro Saga, et al. in Human NeuroblastomaAccumulation of Cytosolic Calcium Induces Necroptotic Cell Death

Updated version

10.1158/0008-5472.CAN-13-1283doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/12/30/0008-5472.CAN-13-1283.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/4/1056.full#ref-list-1

This article cites 37 articles, 9 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/4/1056.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/74/4/1056To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-1283

http://cancerres.aacrjournals.org/content/suppl/2013/12/30/0008-5472.CAN-13-1283.DC1

http://cancerres.aacrjournals.org/content/74/4/1056.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/4/1056.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/4/1056


Documents

Accumulation of Cytosolic Calcium Induces Necroptotic Cell ... · Following treatment with HVJ-E or A23187, SK-N-SH cells or the removed xenografts were ﬁxedwith4%paraformal- dehyde,