Water-soluble germanium nanoparticles cause necrotic cell death and the damage can be attenuated by blocking the transduction of necrotic signaling pathway

(1)

ToxicologyLetters207 (2011) 258–269

ContentslistsavailableatSciVerseScienceDirect

Toxicology

Letters

j o ur na l ho me p ag e :w w w . e l s e v i e r . c o m / l o c a t e / t o x l e t

Water-soluble

germanium

nanoparticles

cause

necrotic

cell

death

and

the

damage

can

be

attenuated

by

blocking

the

transduction

of

necrotic

signaling

pathway

Yu-Hsin

Ma

a

_,

_Chin-Ping

_Huang

b

_,

_Jia-Shiuan

_Tsai

a

_,

_Mo-Yuan

_Shen

b

_,

_Yaw-Kuen

_Li

b,∗∗

_,

_Lih-Yuan

_Lin

a,∗ a_Institute_of_Molecular_and_Cellular_Biology,_and_Department_of_Life_Science,_National_Tsing_Hua_University,_Hsinchu,_Taiwan

b_Department_of_Applied_Chemistry,_National_Chao-Tung_University,_Hsinchu,_Taiwan

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received25May2011 Receivedinrevisedform 20September2011 Accepted21September2011 Available online 29 September 2011 Keywords:

Water-solublegermaniumnanoparticles Cytotoxicity

Calcium

Reactiveoxygenspecies

Mitochondrialmembranepotential Necrosis

a

b

s

t

r

a

c

t

Water-solublegermaniumnanoparticles(wsGeNPs)withallyamine-conjugatedsurfaceswerefabricated andemitblueﬂuorescenceunderultravioletlight.ThewsGeNPwasphysicallyandchemicallystable atvariousexperimentalconditions.CytotoxicityofthefabricatedwsGeNPwasexamined.MTTassay demonstratedthatwsGeNPpossessedhightoxicitytocellsandclonogenicsurvivalassayfurther indi-catedthatthiseffectwasnotresultedfromretardingcellgrowth.Flowcytometricanalysisindicatedthat wsGeNPdidnotalterthecellcycleproﬁlebutthesub-G1fractionwasabsentfromtreatedcells.Results fromDNAfragmentationandpropidiumiodideexclusionassaysalsosuggestedthatapoptoticcelldeath didnotoccurincellstreatedwithwsGeNP.Additionofanecrosisinhibitor,necrostatin-1,attenuatedcell damageandindicatedthatwsGeNPcausednecroticcelldeath.Cellsignalingleadstonecroticdeathwas investigated.Intracellularcalciumandreactiveoxygenspecies(ROS)levelswereincreaseduponwsGeNP treatment.TheseeffectscanbeabrogatedbyBAPTA-AMandN-acetylcysteinerespectively,resultingin areductionincelldamage.Inaddition,wsGeNPcausedadecreaseinmitochondrialmembranepotential (MMP)whichcouldberecoveredbycyclosporineA.ThecellularsignalingeventsrevealedthatwsGeNP increasethecellularcalciumlevelwhichenhancestheproductionofROSandleadstoareductionof MMP,consequentiallyresultsinnecroticcelldeath.

1. Introduction

Germanium(Ge)isa metalloidwithsemiconductorproperty and is dispensablefor human health. Inorganic Ge compound, suchasGeO2, isgenerally non-toxic although renaland neural

damagesare reported after long-term,high-dose consumption. However,organicGecompound,suchasGe-132,isregardedasan elixirinseveralcountriessinceGe-132hasbeenshowntoinhibit cancerdevelopment(Kumanoetal.,1985),induceerythropoietic efﬁciency(Dozonoetal.,1996),exertantimicrobialactivity(Aso etal.,1989)ormodulateimmunopotency(Fukazawaetal.,1994). We have shown that GeO2 blocks cell cycle progression at G2

phaseandcausesradiosensitizingeffectdespitethechemicalitself isverylowincytotoxicity(Chiuetal.,2002).Highconcentrationof GeO2isrequiredtogeneratetheradiosensitizingeffect.However,

GeO2hasalowsolubilitywhichhindersitsbiologicalapplication.

Inaddition,GeO2cannotbedeliveredtoaspeciﬁctargettoexert

∗ Correspondingauthor.Tel.:+88635742693. ∗∗ Correspondingauthor.

E-mailaddress:lylin@life.nthu.edu.tw(L.-Y.Lin).

its radiosensitizing effect. Alternatively, nano-sized germanium particlesmaybeutilizedforthispurpose.

Quantumdots(QDs)aregenerallydefinedasnanometer-sized crystalsfabricatedfrommaterialswithsemiconductorproperties. Duetotheiruniquephysicalproperties,QDsarecurrentlyutilized invarious photoelectronicand biomedicalresearches.Materials withdirectbandgap,suchasCdSe,CdSe/ZnS,InPandPbSe,have photonicproperty underdefined particlesize (Alivisatos, 2004; Michaletetal.,2005).Semiconductor materials(GroupIV)with indirectbandgap,i.e.,Si(0)orGe(0)arerarelycharacterizedsince theydonotemitphotonsorfluoresceeffectivelyatinfraredregion (WarnerandTilley,2006;Zhouetal.,2003).However,germanium nanoparticles(GeNP)withsizes lowerthan therelatively large excitationBohrradius(Rb=11.5nm)exertdirectbandgapand

pro-duceradiativerecombination.HenceGeNPfluorescesatthevisible region.Duetoquantumconfinementandnarrowsizedistribution, GeNPswerealsodefinedasquantumdots(Kauzlarichetal.,2004; WarnerandTilley,2006).

Withasemiconductorproperty,GeorGecompoundsarewidely used in industries. It has been applied to ﬁber-optic systems, infraredoptics,polymerizationcatalyst,variouselectronicdevices and solar cells (Bailey et al., 2002; Rieke, 2007; Thiele, 2001;

(2)

Y.-H.Maetal./ToxicologyLetters207 (2011) 258–269 259

Washio,2003).Withthedevelopmentofnanotechnology, nano-sizedparticlesofGeorGecompoundshavebeenfabricated.Owing tothedistinctphysicalandopticalproperties,GeNPcanpotentially beusedinavarietyofﬁelds(Chiuetal.,2006;Singhetal.,2005; Xieetal.,2009).

FabricationsofGeNPhavebeendescribedusingphysicaland chemical approaches.Physically, GeNP canbeconstructed with reactivelaserablation(Riabininaetal.,2006)orpulselaserwith ionimplantation(Ngiametal.,1994).Chemically,severalmethods employinghightemperaturesandhighreducingenvironmenthave beenreported(ChiuandKauzlarich,2006;Foketal.,2004;Luetal., 2005).Thedrasticreactionconditionscomplicatedtheconcoction procedures.Furthermore,it isdifﬁculttocontrolthesizeofthe fabricatedparticlesormodifytheparticlesurfaces.Thevapor con-densationmethod(WarnerandTilley,2006)forthemanufacturing ofGeNP iseasy tofollowand withoutthehightemperatureor reducingenvironment.Nonetheless,thepreparedparticles aggre-gatedinaquaticphaseanddifﬁculttouseinbiologicalresearch(Lin etal.,2009).Recently,themethodofsynthesizingwater-soluble GeNP(wsGeNP)hasbeenreportedanditpotentiallyavoidedthe defectsofpreviousmethods(Lambertetal.,2007).GeNPproduced byvaporcondensationmethodhaslesstoxicitytocells(Linetal., 2009).ThetoxicityofGeNPfabricatedthroughothermethodshas notbeeninvestigated.

Withthesynthesis ofwsGeNP,thesurfacechemistrycanbe modified.Themodificationsmaybeusefulforpotential biologi-calapplications,suchascelltypespecifictargeting.Sincewehave demonstratedthatGeO2andGeNPareradiosensitizersinChinese

hamsterovary(CHO)K1cells,weexaminedwhetherwsGeNPhad thesame property.We therefore preparedwsGeNP for cellular studies. Thecytotoxicity of the wsGeNP wasexamined. Differ-entfromthatofGeO2 andGeNP, thewsGeNPdamagescells at

lowconcentration.Thetoxicologicalmechanismwasstudiedand chemicalsthatattenuatethetoxicityweresigniﬁed.

2. Materialsandmethods

2.1. Cellcultureandchemicals

CHOK1cellswereculturedasmonolayersat37◦_C_in_McCoy’s_5A_medium

supplemented with 10% heat-inactivated fetal bovine serum, 0.22% sodium bicarbonate,100U/mlampicillin and 100␮g/ml streptomycin,in 5%CO2/95%

airand100% humidity.Reagentsfor cellculturewerepurchasedfromGIBCO (Invitrogen). BAPTA-AM [1,2-bis-(o-aminophenoxy)-ethane-N,N,N_,N_-tetraacetic

acid, tetraacetoxymethyl ester] was obtained from BIOMOL. Necrostatin-1 [5-(indol-3-ylmethyl)-(2-thio-3-methyl)hydantoin] was obtained from Merck. CyclosporinAandFluo-3/AM (1-[2-amino-5-(2,7-dichloro-6-hydroxy-3-oxo-3H-xanthen-9-yl)]-2-(2_-amino-5_{-methylphenoxy)ethane-N,N,N}_,N_-tetraacetic

acid pentaacetoxymethyl ester) were acquired from Kelowna. DiOC6

[3,3_{-dihexloxa-carbocyanine} _iodide] _was _a _product _of _Calbiochem. _MTT

[3-(4,5-dimethylthiazol-2-yl-2, 5-diphenyl tetrazolium bromide)] was purchased fromUSBCrop.ATPLite300assaykitwasaproductofBlossom.Otherchemicals werepurchasedfromSigmaunlessspeciﬁed.

2.2. FabricationofwsGeNP

GeNPwassynthesizedinreversemicellesbyreducingthesolution-phaseGeCl4

(WarnerandTilley,2006).Allreactionswereprocessedunderanitrogen atmo-spheretoslowdowntheoxidationofgermanium.Thereversemicellessolutionwas preparedbystirring100␮lofGeCl4(1.0M,anhydride)and1.5gof

tetraoctylam-moniumbromide(TOAB)in100mlofanhydroustoluenefor30min.GeNPwere formedbyrapidlyadding2mlofthereducingagent(1Mlithium triethylboro-hydride[Li(C2H5)3BH]intetrahydrofuran)tothereversemicellessolution,which

turnedfromcleartoatranslucentyellowcolor.Thesolutionwaslefttoreactfor afurther2hthenquenchedwith20mlofanhydrousmethanol.Inordertoadd aminogroupsontotheGeNPsurface,40␮lofaplatinumcatalyst(0.05MH2PtCl6in

isopropylalcohol)and10mlofallyamineweremixedwiththeGeNPsolutionand stirredfor30min.Aftersurfacecapping,themixturewasremovedfromthe nitro-genenvironmentanddriedinarotaryevaporator,leavingbehindtheTOAB(white powder)andtheGeNP.TheallyaminecappedGeNPwassolubilizedwith50mlof distilledwaterandtheTOABwasremovedbyﬁltrationthrougha0.22␮mﬁlter.The productwasdesignatedaswsGeNP.

2.3. Fouriertransferinfrared(FTIR)spectrophotometricanalysis

Allyamineorallyamine-coatedGeNPwasmixedwithoven-driedFTIRgradeKBr toafinalconcentrationof1%(w/w).Themixturewasgroundedtoafinepowderthen flattenedintothinfilmsunder10tonsinahydraulicpressforFTIRmeasurements. FTIRspectrawereobtainedwithaNicoletAvatar320FTIRspectrometer(Nicolet. InstrumentCo.,Madison,WI,USA).Thirtytwoscanswerecollectedataspectral resolutionof1cm−1_.

2.4. MTTassay

ThecytotoxicityofthesynthesizedwsGeNPwasevaluatedusingMTTassay. Brieﬂy,CHOK1cellswereseededin96-wellplatesatadensityof3.5×103_cells/well

andtreatedwithvariousconcentrationsofwsGeNPat37◦_C_for₂₄_h._Cells_were_then

incubatedwithMTT(ﬁnalconcentration0.4mg/ml)for4hbeforeharvesting.Cells werewashedwithphosphate-bufferedsaline(PBS)and200␮lDMSOwasaddedto eachwell.Theabsorbanceofformazanwasrecordedat565nmwithaThermoMax microplatereader(MolecularDevices,Sunnyvale,CA,USA).

2.5. Clonogenicsurvivalassay

Twohundredandﬁftycellswereseededin60mmdishbeforetreatingwith variousconcentrationsofwsGeNPfor24h.ThewsGeNPwasremovedandcells wereculturedinfreshmediumfor7days,thenﬁxedandstainedwith1%(w/v) crystalvioletdissolvedin30%ethanol.Coloniescontainingatleast50cellswere counted.Thesurvivalfractionofeachtreatmentwasdeterminedbydividingthe numberofcoloniesinthetreatedsamplebythatofthenon-treatedcontrolsample. 2.6. Cellcycleanalysis

CellsweretreatedwithvariousconcentrationsofwsGeNPfor24h.Treated cellswereremovedbytrypsinizationandcollectedbycentrifugationat1500rpm for10min.Thecellpelletswereresuspendedin70%ethanolandstoredat4◦_C

overnight.Cellswerethencentrifugedandresuspendedin1mlPBScontaining RNaseA(100␮g/ml).After30minatroomtemperature,thecellswerespundown andthepelletsstainedwith1mlofpropidiumiodide(PI,20␮g/mlinPBS)for30min. FlowcytometricanalysiswasthencarriedoutonaFACScalibur(BectonDickinson, FranklinLakes,NJ,USA).

2.7. DNAfragmentationassay

ForDNAladderanalysis,cellsweretreatedwith5␮MwsGeNPorirradiated with25J/m2_UV._After_further_incubating_at₃₇◦_C_for₂₄_h,_cells_were_washed_twice

withPBSandthelow-molecular-weightedDNAfragmentswereextractedwithTTE buffer(0.2%TritonX-100,10mMTris,15mMEDTA,pH7.6)for15minatroom temperature.Aftercentrifugingat12,000rpmfor15min,thesupernatantswere transferredtonewtubesbeforeRNAaseA(100␮g/ml)wasaddedandincubated at37◦Cfor1h.DNAwasextractedwith1volumeofphenol/chloroform/isoamyl alcoholandprecipitatedin0.1volumeof3Msodiumacetate(pH5.2)and1volume ofisopropylalcohol.Afterstandingat−70◦_C_for₁₅_min,_the_DNA_was_spun_down_at

12,000rpmfor20min,andwashedwith70%alcohol.TheDNApelletwasdissolved inTEbufferandanalyzedelectrophoreticallyona2%agarosegel.

2.8. Caspase-3activityassay

Cellswerelysedin1%TritonX-100,1%NP-40,2␮g/mlaprotinin,2␮g/ml leupeptin,2mMPMSFandincubatedonicefor10min.Aftercentrifugationat 13,000rpmand4◦Cfor30min,thesupernatantsweretransferredtonewtubes. Proteinconcentrationsweredeterminedusingaproteinassaykit(Bio-Rad).50␮g ofproteinswereincubatedat37◦_C_in₁₀₀_␮l_reaction_buffer₍₁₀_mM_HEPES,₂_mM

EDTA,10mMKCl,1.5mMMgCl2,10mMDTT)containing50␮MAc-DEVD-AFCfor

1h.TheAFCﬂuorescencewasmeasuredatexcitationwavelengthof405nmand emissionwavelengthof505nmwithamicroplatereader(Wallac1420Multilabel Counter,PerkinElmer).

2.9. PIexclusionassay

Theintegrityofplasmamembranewasassessedbydeterminingtheabilityof cellstoexcludePI.Cellsweretrypsinized,collectedbycentrifugation,washedonce withPBSthensuspendedinPBScontaining10␮g/mlPI.Thecellswerestoodatroom temperatureinthedarkfor15min.ThelevelsofPIincorporationweredetermined byﬂowcytometry(FACScalibur,BectonDickinson,FranklinLakes,NJ,USA).The proportionofcellsstainedwithPIwasexpressedaspercentageofPIuptake. 2.10. DeterminationofintracellularCa2+_,_reactive_oxygen_species_(ROS)_and

mitochondrialmembranepotential(MMP)

IntracellularCa2+_,_ROS_and_MMP_were_measured_by_Fluo-3-AM,_H

2DCF-DAand

(3)

260 Y.-H.Maetal./ToxicologyLetters207 (2011) 258–269

Fig.1.CharacterizationofthesyntheticwsGeNP.(A)ThefluorescenceemittedbythefabricatedwsGeNPwasanalyzedwithexcitationandemissionwavelengthsof350 and450nm,respectively.(B)TheFTIRspectrumofthewsGeNPwasanalyzedtoshowthepresenceofallyamineonthesurfaceoftheparticles.(C)Equivalentconcentration (0.1nM)ofwsGeNPwasincubatedatdifferenttemperaturesfor24or48handfluorescenceintensitywascomparedwiththatwithoutincubation.(D)Equivalentconcentration (0.1nM)ofwsGeNPwasincubatedatdifferentpHfor24hatroomtemperature.Fluorescenceintensitywasdeterminedaftertheincubationandcomparedwiththatofthe samplewithoutincubation.(E)Equivalentconcentration(0.5nM)ofwsGeNPwasdissolvedin8NHNO3andtheintensityofthefluorescencewasdetermined.Eachvalue

(4)

Fig.2. AnalysisofcellviabilityandcellcycledistributionafterwsGeNPtreatment.CellsweretreatedwithvariousconcentrationsofwsGeNPfor24handcellviability wasexaminedbyMTT(A)andclonogenicsurvival(B)assay.Cellpopulationdistributedinthecellcyclewasdeterminedbyﬂowcytometry(C).Eachvaluerepresentsa mean±standarddeviationofthreesamples.*Signiﬁcantdifference(p<0.05).

presenceorabsenceofinhibitorsfortheindicatedtimeintervals.Two␮M Fluo-3-AM,5␮MH2DCF-DAor40nMDiCO6wasadded30minbeforecellharvest.Cellular

ﬂuorescencewasmeasuredusingﬂowcytometry(FACScalibur,Becton-Dickinson, FranklinLakes,NJ,USA)withexcitationandemissionwavelengthsof488nmand 530nm,respectively.

3. Results

3.1. Physicalpropertiesofthesynthesizedwater-solubleGeNP GeNP was fabricated by a reduction method. Transmission electronmicroscopicanalysisofthefabricatedGeNPrevealedan averageparticlesizeof4.2±1.2nmandislowerthantherelatively largeexcitationBohrradius(11.2nm,datanotshown).TheGeNP exhibitsbluephotoluminescenceat450nmwithafullwidthat halfmaximumof100nm(Fig.1A).Usinghydridereducingagents intheseexperimentsproducehydrogen-terminatedGeNPsurfaces (WarnerandTilley,2006),whichcanbetreatedwithcompounds containingC CbondandH2PtCl6catalysttoproduceavarietyof

surfacetypes.WemodiﬁedtheGeNPswithallyamine.FourierIR spectrumanalysisdetectedtheallyaminegroupsonthesurfaceof GeNP(Fig.1B).Thetransmittanceat2900and3500cm−1attribute toasymmetricandsymmetricvibrationofC–NH2andC–CH2bonds

of the allyamine while the peak at 1661cm−1 is attributedto theallyamine and clearlyindicative of itsattachment toGeNP.

Themodiﬁednanoparticlesaredesignatedaswater-solubleGeNP (wsGeNP)forthefollowingstudies.

Physical properties associate with subsequent experiments werealsoexaminedpriortoapplyingthewsGeNPtobiological samples.TheﬂuorescenceintensityofwsGeNPisstableas incu-batedatdifferenttemperatures(4,37or50◦C)foratleast48h (Fig.1C).SimilarcharacteristicwasalsoobservedwhenwsGeNP wasincubatedatdifferentpH(3–10)at37◦C for24h(Fig.1D). TheseresultsindicatethatthewsGeNPwasstableenoughunder variousexperimentalconditionsforsubsequentstudies.The emis-sion of blue ﬂuorescence was diminished when wsGeNP was treatedwith8Nnitricacidat65◦Cfor12h,indicatingthecollapse oftheparticlestructure(Fig.1E).Thisconditionwasthenusedto estimatetheGecontentforthefabricatedparticlesbyinductively coupledplasma-massspectrometry.

3.2. CytotoxiceffectofwsGeNP

CytoxicityofthewsGeNPwasexamined.Gecontentinthe solu-tionwasdeterminedbyICP-MSbeforeaddingtocells.Cellviability wasinitiallymeasuredbyMTTassay.Cellsweretreatedwith vari-ousconcentrationsofwsGeNPfor24handsurvivalfractionswere determined.AsshowninFig.2A,cellviabilitydeclinedataslowas 1␮MofwsGeNPandtoxicityobservedathigherthan3␮M.Since

(5)

Fig.3. CharacterizationofwsGeNP-inducedcelldeath.Cellsweretreatedwith5␮MwsGeNPfor24horirradiationwith25J/m2_UV._The_levels_of_DNA_{fragmentation}_(A)

andcaspase3activity(B)wereanalyzed.Cellswereharvested24h(A)or12and24h(B)afterUV-irradiation.ATPlevelwasdeterminedincellsafter24hwsGeNPtreatment (C).Eachvaluerepresentsamean±standarddeviationofthreesamples.*Signiﬁcantdifference(p<0.05).

germaniumoxide(GeO2)solutiondidnotcausesigniﬁcant

cytotox-icityevenatmillimolarlevel(Chiuetal.,2002),wespeculatedthat thetoxiceffectcamefromthechemicalreagentusedinproducing thewsGeNP.ThesameproceduresforwsGeNPsynthesiswerethen performedintheabsenceofgermaniumchloride.Theresulting solutionwasusedtoexaminethecytotoxiceffect.Notoxiceffect wasobservedforthesolvent(datanotshown).Chlorogenicsurvival assaywasusedtoconﬁrmthedecreaseviabilityobservedinMTT assay.AsshowninFig.2B,cellviabilitydroppeddose-dependently withtheincreaseofwsGeNPconcentration.Theresultsindicate thatGeNPsynthesizedinthisworkpossessestoxiceffect.

SincewehaveshownpreviousthatGeO2blockedcellcycle

pro-gression(Chiu etal.,2002),we examinedwhetherthewsGeNP processessimilarcharacteristic.Flowcytometricanalysiswas con-ducted to examine the cell distribution profile. After treating withvariousconcentrationsofwsGeNPfor24h,cellcycleprofile remainedsimilaramongthetreatments(Fig.2C).Nosignificantcell cyclearrestcouldbeobserved.

3.3. wsGeNPcausednecroticcelldeath

Cytometric analysis did not detect cells in sub-G1 fraction (Fig.2C).Thisresultsuggeststhat GeNPtreatmentdidnotlead to apoptotic cell death. To demonstrate this effect, DNA frag-mentation assay was conducted. As indicated in Fig. 3A, DNA fragmentationwasnotnotedinwsGeNP-treatedcells.However, cellsafterUVirradiationweresubjectedtoapoptoticdeathand

have fragmented DNA. Since caspase 3 is the effector enzyme thatcausescellapoptosis,weexamineitsactivityaftertreatment.

Fig.3Bshowsthatcaspase3activitydidnotincreasebutdecreased afterwsGeNPtreatment.Theresultwasagaincontrarytothatof UV-irradiatedcellswhichshowedanincreaseincaspaseactivity. Forapoptoticcells,ATPlevelremainsunchangedwithinaperiodof time(Zamaraevaetal.,2005).However,wsGeNPtreatmentaltered cellularATPleveldespitethatnocorrelationcouldbeestablished betweentheATPlevelanddosesofwsGeNP(Fig.3C).Theseresults suggestthatGeNPcausesnecroticcelldeathinCHOK1cells.

3.4. Necrostatin1rescuescellsfromnecroticdeath

Cellnecrosisleadstothelossofmembraneintegrityandallows PItodiffuseintocells.Relativecellviabilitycanthusbeanalyzedvia thelevelofPIuptake.Aftertreatingcellsfor24h,wsGeNPcauseda dose-dependentlossofcellmembraneintegrity(Fig.4A). Compa-rabletotheMTTassay,asigniﬁcantincreaseofcellularPIuptake wasnotedwhen2.5␮MwsGeNPwasadministeredtocells.A time-coursestudywassubsequentlyconductedandtheresultshows thatincreaseinPIuptakeoccurswithin12–24hafteradding5␮M wsGeNP(Fig.4B).PIexclusionassayisthususedtoestimatecell damageinsubsequentstudies.

Necrotic cell death is currently recognized to proceed in a programmedmanner.Interruptionofthesignaltransductioncan reducecelldamage.Aninhibitor,necrostatin-1(Nec-1),can atten-uatetheoccurrenceofnecroticcelldeath(Degterevetal.,2005).

(6)

Fig.4.wsGeNPcausesnecroticcelldeathandcanberescuedbyNec-1.CellmembraneintegritywasestimatedwiththePIexclusionassaywith:(A)cellstreatedwith variousconcentrationsofwsGeNPfor24h;(B)cellstreatedwith5␮MwsGeNPforvarioustimeintervals;(C)cellstreatedwith5␮MwsGeNPinthepresenceofvarious concentrationsofNec-1for24h.Eachvaluerepresentsamean±standarddeviationofthreesamples.*Signiﬁcantdifference(p<0.05).#_Signiﬁcant_(p_<_0.05)_difference

betweenthepairedgroups.(D)Morphologyofcellsreceived5␮MwsGeNPand/or300␮MNec-1wasrecorded.Magnitudeofampliﬁcation:400×.

Nec-1wasthereforeusedtoverifythedamagingeffect.Asshown inFig.4C,Nec-1isabletoreducethewsGeNP-inducedcytotoxic effect.Therescueofcelldamagecanbecorrelatedwiththe morpho-logicalchangesofcells.CellstreatedwithwsGeNPhavearound-up morphology.WiththeadditionofNec-1,cellmorphologyreturned tothatofuntreatedcells(Fig.4D).Theseresultsdemonstrate fur-therthatwsGeNPcausesnecroticcelldeathandthedamagecanbe blockedbyNec-1.

3.5. Signalingfactorsinvolvedinthecelldamageinducedby wsGeNP

Since wsGeNP-treated cells subjected to necrotic cell death, signalingfactorsinvolvedinthedamagewereinvestigated. Intra-cellular calcium concentration was examined initially because elevationofcalciumcontentisfrequentlyobservedincellsexposed tometal.Cellsweretreatedwith3or5␮MwsGeNPfor24hand calciumcontentwasestimatedbyﬂowcytometry.Fig.5Ashows thatcalciumcontentincreaseswiththetreatment.Atime-course studywassubsequently conductedand showedthat intracellu-larcalciumcontentincreased12hafterthetreatment(Fig.5B). Toanalyzewhethertheelevatedcalciumcontentisrelatedtocell damage,intracellularcalciumchelator(BAPTA-AM)wasaddedto GeNP-treatedcellsandreducedthecellularcalciumlevel(Fig.5C). Thisreductioncorrelatestotheprotectiveroleofthechelatorsince thelevel of PI uptake dropped withthe incrementsof BAPTA-AM(Fig.5D).BAPTA-AMdidnotrecoverthecellmorphologyof

thewsGeNP-treatedcellssincethecalciumchelatoritselfaltered thecellmorphology(Fig.5D).ThisresultsuggeststhatwsGeNP stimulatestherelease ofintracellular calciumthat leadstocell damage.

Reactiveoxygenspecies(ROS)canbeproducedbymetal chal-lenge andstimulatedby elevatingcellularcalciumcontent.We thereforeestimatetheROSlevelafterwsGeNPtreatment.Asshown inFig.6A,adose-dependentincreaseofROSlevelwasobserved in cells after 24h of GeNP exposure. Time-course study shows that theincrease ofROS occurredat15hafterGeNP treatment (Fig.6B).Anantioxidant,N-acetylcysteine(NAC),wasaddedto examine whether the ROS can be removed and consequently reducescelldamage.Fig.6CshowsthattheROSwaseffectively diminishedbytheantioxidantinadose-dependentmanner. Addi-tion of NAC towsGeNP-treated cells alsoabrogated cellular PI uptake (Fig.6D)andrecoveredcellmorphology(Fig.6E).These resultsindicatethatNACpreventscellsfromdamagescausedby wsGeNP.

Alternationofmitochondrialmembranepotential(MMP)can be foundin cells subjected tovarious stresses resultingin cell deaththrougheitherapoptotic ornecroticpathway.Theroleof MMPinwsGeNP-induceddamageistheninvestigated.A reduc-tion in MMP was noted in cells treated with more than 3␮M wsGeNPfor 24h(Fig.7A). Dropin MMP became signiﬁcantat 21hinthepresenceof3␮MwsGeNP(Fig.7B).Tolinkthe alter-nation of MMP with cell damage, cyclosporin A (CsA), which inhibits the openingof the mitochondria membrane transition

(7)

Fig.5. wsGeNPtreatmentelevatedintracellularCa2+_and_caused_cell_damage._(A)_Cells_were_treated_with₃_or₅_␮M_wsGeNP_for₂₄_h_and_{intracellular}_Ca2+_level_was

determined.(B)Cellsweretreatedwith5␮MwsGeNPforvarioustimeintervalsandintracellularCa2+_level_was_determined._Cells_were_treated_with₅_␮M_wsGeNP_for₂₄_h

inthepresenceofvariousconcentrationsofBAPTA-AM.IntracellularCa2+_(C)_or_PI_uptake_(D)_level_was_determined._Each_value_represents_a_mean_±_standard_deviation_of

threesamples.*Signiﬁcantdifference(p<0.05).#_Signiﬁcant_(p_<_0.05)_difference_between_the_paired_groups._(D)_Cell_morphology_in_the_presence_of₅_␮M_wsGeNP_and/or

60␮MBAPTA-AMwasrecorded.Magnitudeofampliﬁcation:400×.

pore,wasusedtoblockthereduction ofMMP. Employmentof theinhibitorcanrecoverthewsGeNP-inducedreductionofMMP (Fig.7C).Moreover,theblockageofMMPreductioncorrelatedwith thelevelofcelldamagesinceCsAreducedalsoPIuptakeincells treatedwithwsGeNPfor24h(Fig.7D).TheeffectivenessofCsA inprotectingcellfromwsGeNPdamagecanalsobenotedbythe alternationofcellmorphology.WiththeadditionofCsA,the mor-phologyofwsGeNP-treatedcellsreturnedtothatofuntreatedcells (Fig.7E).

3.6. SequenceofthesignalingeventsafterGeNPexposure

Itisevidentfromtheaboveresultsthatintracellularcalcium, ROS and MMP are involved in the signaling pathway for the wsGeNP-inducedcelldamage.Thealternationsofthesesignaling factorsoccurrespectivelyat12,15and21h,implyingasequence ofeventoccursfromintracellularcalcium,ROStoMMP.To demon-stratethissequence,calciumchelatorwasﬁrstapplied,thenROS and MMP levels were determined. Fig. 8 shows that ROS was

(8)

Fig.6.wsGeNPtreatmentincreasedROSlevelandcausedcelldamage.(A)CellsweretreatedwithvariousconcentrationsofwsGeNPfor24handROSlevelwasdetermined. (B)Cellsweretreatedwith5␮MwsGeNPforvarioustimeintervalsandROSlevelwasdetermined.Cellsweretreatedwith5␮MwsGeNPfor24hinthepresenceof variousconcentrationsofNAC.ROS(C)orPIuptake(D)levelwasdetermined.Eachvaluerepresentsamean±standarddeviationofthreesamples.*Signiﬁcantdifference (p<0.05).#_Signiﬁcant_(p_<_0.05)_difference_between_the_paired_groups._(D)_Cell_morphology_in_the_presence_of₅_␮M_wsGeNP_and/or₁₀₀_␮M_NAC_was_recorded._Magnitude

ofampliﬁcation:400×.

reduced(Fig.8A)whileMMPwaselevated(Fig.8B)afteradding BAPTA-AMtowsGeNP-treatedcells.ThisresultindicatesthatROS andMMParethedownstreamsignalsfortheincreasedcalcium content.Secondly,NACwasadministratedtowsGeNP-treatedcells

toreduceROS,thencalciumcontentandMMPwereestimated.An elevationinMMPwasnoted,butthecalciumlevelremainedhigh withthetreatment(Fig.8CandD).ThisﬁndingshowsthatROS isdownstreamofcalciumsignalingbutupstreamofMMP.Thirdly,

(9)

Fig.7.GeNPtreatmentreducedMMPlevelandcausedcelldamage.(A)CellsweretreatedwithvariousconcentrationsofwsGeNPfor24handMMPlevelwasdetermined. (B)Cellsweretreatedwith3␮MwsGeNPforvarioustimeintervalsandMMPlevelwasdetermined.Cellsweretreatedwith3␮MwsGeNPfor24hinthepresenceofvarious concentrationsofCsA.MMP(C)orPIuptake(D)levelwasdetermined.Eachvaluerepresentsamean±standarddeviationofthreesamples.*Signiﬁcantdifference(p<0.05).

#_Signiﬁcant_(p_<_0.05)_difference_between_the_paired_groups._(D)_Cell_morphology_in_the_presence_of₃_␮M_wsGeNP_and/or₅_␮M_CsA_was_recorded._Magnitude_of_{ampliﬁcation:}

400×.

CsAwasgiventoattenuatethereductionofMMPbywsGeNP. Intra-cellularcalcium(Fig.8E)andROSlevels(Fig.8F)werenotaffected bytheadditionofCsA,indicatingthatreductionofMMPderives fromanincreaseinintracellularcalciumcontentandthe subse-quentaugmentedROSlevelbythewsGeNPtreatment.Asummary ofthesignalingpathwayofwsGeNP-inducedcelldeathandthe chemicalsprotectingthecellsfromdamagesareshowninFig.9.

4. Discussion

GeNP can beproduced using differentapproaches. Whether GeNPs fabricated by different methods have similar biological propertiesremaintobeinvestigated.Inthisstudy,wefoundthat GeNPpreparedwithdifferentmethodologiesprocesseddissimilar biologicaleffects.WereportedpreviouslythatGeNPfabricatedby

(10)

Fig.8.DeterminationofthesignalsequenceforwsGeNP-inducedcelldamage.(A)ROSand(B)MMPlevelsweredeterminedincellstreatedwith5␮MwsGeNPfor24h inthepresenceorabsenceof60␮MBAPTA-AM;(C)intracellularCa2+_and_(D)_MMP_levels_were_determined_in_cells_treated_with₅_␮M_wsGeNP_for₂₄_h_in_the_presence_or

absenceof50or100␮MNAC;(E)intracellularCa2+_and_(F)_ROS_levels_were_determined_in_cells_treated_with₃_␮M_wsGeNP_for₂₄_h_in_the_presence_or_absence_of_2.5_or₅_␮M

CsA.Eachvaluerepresentsamean±standarddeviationofthreesamples.*Signiﬁcantdifference(p<0.05).#_Signiﬁcant_(p_<_0.05)_difference_between_the_paired_groups.

vaporcondensationmethoddidnotexertcytotoxicitytoCHOK1 cells(Linetal.,2009).Additionally,itretardscellcycleprogression andenhancesradiosensitizingactivityofcells.These characteris-ticsaresimilartothatofGeO2.However,wefoundinthisstudy

that wsGeNP has toxicity to cells. This ﬁnding is unexpected. Weexaminedthephysicalpropertiesofvapor-condensed GeNP (vpGeNP)and observedthat the nanoparticlesform aggregates

afterincubatinginwater.ThevpGeNPcanevenbedecomposed anddissolvedinwaterafterstandinginroomtemperaturewithin 2weeks.ThesolubilizedvpGeNPmayshowthesamepropertyas GeO2solution.Ontheotherhand,wsGeNPisphysicallystable.The

ﬂuorescenceintensityofwsGeNPremainsthesameafterstoring atroomtemperaturefor morethan2 months.Itcansustainat elevatedtemperature(50◦C)oracidity(pH3)forprolongedperiod

(11)

Fig.9.AschematicillustratingthewsGeNP-inducedsignalingpathwaywith chem-icalsprotectingthecellsfromdamageslisted.

without affecting the ﬂuorescence intensity. The wsGeNP may beeasiertoenterandremainintactincells.Thesedifferencesin physicalpropertiesmayattributetoitscytotoxiceffectoncells.

Becauseofthepromptdevelopmentinnanotechnology,a vari-etyofnanoparticles(NPs)weresynthesized.Attemptsweremade toutilizetheseNPsforbiomedicalapplications.Oneofthemajor concernsabouttheapplicationisthetoxicitytoorganisms.Cell damagingeffectsinducedbyNPswerereportedinseveral stud-ies.QuantumdotswithCd–Secoreareknowntobetoxictocells duetothereleaseofhazardousmetalions(Medintzetal.,2004). TreatingRAW264.7cellswithAgNPresultedinareductionof cel-lularGSHcontent,increasedNOproductionandTNF-␣synthesis andsubsequentapoptoticcelldeath(Parketal.,2010).CuNPand MnNPincreaseoxidativestressviaROSproductionafterentering cells,andleadtocelldeath(Vanwinkleetal.,2009).Thesestudies indicatethatcytotoxicityoftheNPscanbederivedfromtherelease oftheconstituentsand/orNP-activatedsignalingpathway(s).Since thedissolvedGeexertslowcelltoxicity(Chiuetal.,2002),the dam-agingeffectofwsGeNPcomesfromtheactivatedsignalcascade. Blockingthetransductionofthesignalsrescuescellsfromnecrotic death.ThisﬁndingindicatesthatentranceofforeignNPsstimulates variouscellularresponses.Damagesmayoccurifthecellsarenot abletomanageandmodulatethepassageofsignals.

The degree of cell damage may also be associated with the entrance rate of the NPs. For example, AuNP enters cells rapidlythroughreceptor-mediatedendocytosisandactivates pro-inflammatorygenessuchasinterleukinI(IL-1),IL-6andTNF-␣(Yen etal.,2009).Owingtohigherlevelofaccumulationincells,AuNP hasmoresignificantcytotoxicand immuneresponsesthanthat inducedbyAgNP,whichenteredcellsthroughnon-specific endocy-tosis.Presently,wedonotknowtheexactmechanismforwsGeNP cellentry.WespeculatethatwsGeNPsentercellsthrough endocy-tosisandtransientlysequesteredinlysosomes.AlthoughthepHof lysosomecanbelowerthan4,GeNPisstableatpHlowerthan3 (Fig.2).ItisunlikelythatthewsGeNPsaredissolvedinlysosome andtheGeionsleakouttothecytoplasmtoproducethetoxic effects.ConsideringthatGeO2athigherthan20mMdoesnotexert

cellularcytotoxicityafter24htreatment(Chiuetal.,2002),cell damagesshouldbeverylimitedevenifwsGeNPdecomposedand releasesfromlysosomes.Remarkably,damagetothesametypeof cellscanbedetectedatlessthan3␮MofwsGeNP.Becausethe ﬂuorescenceofwsGeNPcouldnot bedetectedin thecells, cel-lularlocalization ofthewsGeNPcannot beidentiﬁedpresently. However,intracellularcalciumlevelelevated12hafterwsGeNP exposure(Fig.5B).Thisresultimpliesthattheeffectwasproduced afterwsGeNPenteredcells,butnotaftertheimmediateexposure ofwsGeNPtocells.

Cellssubjectedtochemicaltreatmentsusuallycauseadeﬁned type of cell death. However, different types of cell deathmay occursimultaneouslybyatreatment.Thetypeofcelldeathalso variesunderdifferentconditions;evencellsaretreatedwiththe samechemical.Cellsmaysubjecttoapoptoticdeathunderlower chemicalconcentration,butswitchestonecroticdeathwhenthe doseishigh(MajnoandJoris,1995).Studieshaveindicatedthat apoptoticandnecroticcelldeathscanberegulatedbycommon sig-nalingfactors.AlternationsinintracellularCa2+_,_ROS_and_MMP_are

frequentlyfoundineitherapoptoticornecroticcelldeath. Intracel-lularCa2+_content_was_reportedly_to_determine_the_cell_fate₍_Zong andThompson,2006).Moderateelevationofcalciumcontentleads toapoptotic death(McConkeyand Orrenius,1996).However,a greatincreaseofcalciumionsresultsinmitochondrialCa2+

over-loadandcausesnecroticdeath(RichterandSchlegel,1993).For wsGeNP-treatedcells,a5-foldincreaseinintracellularCa2+

con-tentwasnotedat5␮M treatment.Thisraisein Ca2+_content_is

apparentlysufﬁcienttoinitiatethesignalfornecrosis.Studyalso showedthatATPleveldeterminedthetypeofcelldeath. Reduc-tioninATPcontentmayleadtonecrotic celldeath(Leistetal., 1997).Inourstudy,wedidnotobserveapoptoticcelldeatheven athighwsGeNPconcentration.Althoughwehavedemonstrated thatwsGeNPcausednecroticcelldeath,theATPleveldidnotshow adose-dependentreductionafterthetreatment.Thisresponseis similartothatofperoxynitrite-treatedU931orTHP-1monocytes. Thistreatmentcausednecroticcelldeathbutdidnotaltercellular ATPlevel(Cantonietal.,2005).

ThecytotoxicresponsesofwsGeNPtoCHOK1cellsaresimilarto thatofCd(Yangetal.,2007).Cd-treatedcellsalsorenderednecrotic celldeath.WeshowedthattheCd-inducedcelldamagebeganwith elevationofintracellularCa2+_ions,_followed_by_an_increase_in_ROS

levelthenreductionofMMP;asignalingpathwayexactlythesame asthatofwsGeNP.However,theelevatedCa2+_content_also

stim-ulatescalpainactivityinparallelwiththeROSincrease.Calpain activitywasnotalteredinwsGeNP-treatedcells(datanotshown). Thisvariationisanexampleofdifferentchemicaltreatment stim-ulates apartially overlappingpathway andleadstosimilarcell fate.

Reporthasindicatedthatdifferentsizesofnanoparticlesmay producedifferentcytotoxiceffects(Frohlichetal.,2009).For exam-ple,treatingcellswith1.4nmAuNPresultedinnecroticcelldeath. However,thetoxicity canbereducedbyincreasingtheparticle sizes(Panetal.,2007).SincethevpGeNPaggregatesinwaterand isnotcytotoxic,wespeculatethatthesizeoftheGeNPmayplay rolesindeterminingcytotoxicity.SincethesizeofNPcanbe mod-ulatedbysurfactantinthefabricationprocess(Dungetal.,2009; Natarajanetal.,1996), weincreasedtheconcentrationofTOAQ togeneratewsGeNPwithlargersizes(50and150nmdiameterin average).Areductionincytotoxicity(determinedbyPIexclusion assay)wasnotedwhenthelargersizewsGeNPwasexamined. Sig-niﬁcantincreaseofPIuptakewasobservedafteradding100␮M ofthe150nmwsGeNPfor24h(datanotshown).However,the toxicityisnotreducedinasize-dependentmanner.Nevertheless, theresultsuggeststhatthesizeofwsGeNPattributesinpartthe toxicitytocells.

Besidestoparticlesizes,thecytotoxicityofwsGeNPmayvary indifferentcelltypes.WehaveexaminedthetoxicityofwsGeNP inHEK293cellsandfoundasignificantincreaseinPIuptakewhen theconcentrationreached20␮M.Thisresultindicatesacell spe-cificdamagingeffectofthewsGeNP.Cell-typespecifictoxicityhas beenreportedin severalstudies.ZnONPproduced varied cyto-toxiceffecttohumanimmunecells;lymphocytesaremostresistant whilemonocytesaremostsusceptibletothetoxicity(Kauzlarich etal.,2004).CytotoxicityofsilicaNPalsoshowedastrong depen-denceoncelltypeandparticlesizewhenhumanepithelialcellsand mousemonocyteswerecompared(Diazetal.,2008;Warnerand

(12)

Tilley,2006).WeusedCHOK1cellsinthisstudysuchthattheGe effectcanbecomparedwithourpreviousworks.AlthoughwsGeNP exertslesstoxiceffectinhumankidneycell(HEK293),thetoxicity isstillpresent.

Weestablishedinthisworkthesignalingpathwayof wsGeNP-induced cell damage. A variety of metallic compounds have reportedtotransducecytotoxicityviatheCa2+_/ROS/MMP_pathway.

ThispathwaymayalsobecommontoNP-activatedcelldamages. Signalsthroughthiscommonpathwaycauseeitherapoptoticor necroticdeathdependingonthecelltypeandthepropertyofthe NP.We showherethat wsGeNPcausesnecrotic celldeathand Nec-1canattenuateNP-induced necrotic celldeath. Thisisthe ﬁrststudytoindicatetheeffectivenessofNec-1andthusprovides apotentialantagonisttocounteractthehazardofNP exposure. WehavesuggestedpreviouslythatreductioninMMPafter chem-icaltreatmentmaybeaprerequisiteforNec-1toreducenecrotic celldeath(Hsuetal.,2009).SincewsGeNPtreatmentcausesMMP reduction andleadsto necroticcell death, Nec-1is effectivein rescuingdamagedcells.

In summary, we reported here the toxicological study of wsGeNP.IncontracttoGeO2andvpGeNP,wsGeNPshowedhigh

toxicitytocells.Itcausedcellnecrosisthroughelevating intracellu-larcalciumconcentrationwhichresultsinanincreaseinROSlevel. ROSstimulatesthereductionofMMPandleadscellstonecrotic death.Thisdamagingeffectcanbeattenuatedbyaddingcalcium chelator,ROSscavenger,MPTporeinhibitororNec-1.

Acknowledgements

Thisworkwassupported bygrants NSC95-2627-M-007-006 andNSC96-2627-M-007-006 fromtheNationalScienceCouncil, Taiwan,RepublicofChina.TheauthorsthankDr.M.F.Tamfor crit-icalreadingofthemanuscript.

References

Alivisatos,P.,2004.Theuseofnanocrystalsinbiologicaldetection.Nat.Biotechnol. 22,47–52.

Aso,H.,Suzuki,F.,Ebina,T.,Ishida,N.,1989.Antiviralactivityof carboxyethyl-germaniumsesquioxide(Ge-132)inmiceinfectedwithinﬂuenzavirus.J.Biol. ResponseMod.8,180–189.

Bailey,S.G.,Raffaelle,R.,Emery,K.,2002.Spaceandterrestrialphotovoltaics:synergy anddiversity.Prog.Photovoltaics10,399–406.

Cantoni,O.,Guidarelli,A.,Palomba,L.,Fiorani,M.,2005.U937cellnecrosismediated byperoxynitriteisnotcausedbydepletionofATPandispreventedby arachi-donateviaanATP-dependentmechanism.Mol.Pharmacol.67,1399–1405. Chiu,H.W.,Kauzlarich,S.M.,2006.Investigationofreactionconditionsforoptimal

germaniumnanoparticleproductionbyasimplereductionroute.Chem.Mater. 18,1023–1028.

Chiu,H.W.,Kauzlarich,S.M.,Sutter,E.,2006.Thermalbehaviorandﬁlm forma-tionfromanorganogermaniumpolymer/nanoparticleprecursor.Langmuir22, 5455–5458.

Chiu,S.J.,Lee,M.Y.,Chen,H.W.,Chou,W.G.,Lin,L.Y.,2002.Germaniumoxideinhibits thetransitionfromG2toMphaseofCHOcells.Chem.Biol.Interact.141, 211–228.

Degterev,A.,Huang,Z.,Boyce,M.,Li,Y.,Jagtap,P.,Mizushima,N.,Cuny,G.D., Mitchi-son,T.J.,Moskowitz,M.A.,Yuan,J.,2005.Chemicalinhibitorofnonapoptoticcell deathwiththerapeuticpotentialforischemicbraininjury.Nat.Chem.Biol.1, 112–119.

Diaz,B.,Sanchez-Espinel,C.,Arruebo,M.,Faro,J.,Miguel,E.d.,Magadan,S.,Yague,C., Fernandez-Pacheco,R.,Ibarra,M.,Santamaria,J.,Gonzalez-Fernandez,A.,2008. Assessingmethodsforbloodcellcytotoxicresponsestoinorganicnanoparticles andnanoparticleaggregates.Small4,2025–2034.

Dozono,H.,Ikeda,K.,Onishi,T.,1996.EffectivenessofGe-132torelievepainand smoothhomecareadministrationfortheterminalcancerpatient.GanToKagaku Ryoho23(Suppl.3),291–295.

Dung,T.T.N.,Buu,N.Q.,VietQuang,D.,ThiHa,H.,Bang,L.A.,HoaiChau,N.,ThiLy,N., Trung,N.V.,2009.Synthesisofnanosilverparticlesbyreversemicellemethod andstudyoftheirbactericidalproperties.J.Phys.:Conf.Ser.187,012054. Fok,E.,Shih,M.,Meldrum,A.,Veinot,J.C.,2004.Preparationofalkyl-surface

func-tionalizedgermaniumquantumdotsviathermallyinitiatedhydrogermylation. Chem.Commun.,386–387.

Frohlich,E.,Samberger,C.,Kueznik,T.,Absenger,M.,Roblegg,E.,Zimmer,A.,Pieber, T.R.,2009.Cytotoxicityofnanoparticlesindependentfromoxidativestress.J. Toxicol.Sci.34,363–375.

Fukazawa,H.,Ohashi,Y.,Sekiyama,S.,Hoshi,H.,Abe,M.,Takahashi,M.,Sato,T., 1994.MultidisciplinarytreatmentofheadandneckcancerusingBCG,OK-432, andGE-132asbiologicresponsemodiﬁers.HeadNeck16,30–38.

Hsu,T.S.,Yang,P.M.,Tsai,J.S.,Lin,L.Y.,2009.Attenuationofcadmium-induced necroticcelldeathbynecrostatin-1:potentialnecrostatin-1actingsites.Toxicol. Appl.Pharmacol.235,153–162.

Kauzlarich,S.M.,Zou,J.,Baldwin,R.K.,Pettigrew,K.A.,2004.Solutionsynthesis ofultrastableluminescentsiloxane-coatedsiliconnanoparticles.NanoLett.4, 1181–1186.

Kumano,N.,Ishikawa,T.,Koinumaru,S.,Kikumoto,T.,Suzuki,S.,Nakai,Y.,Konno,K., 1985.AntitumoreffectoftheorganogermaniumcompoundGe-132ontheLewis lungcarcinoma(3LL)inC57BL/6(B6)mice.TohokuJ.Exp.Med.146,97–104. Lambert,T.N.,Andrews,N.L.,Gerung,H.,Boyle,T.J.,Oliver,J.M.,Wilson,B.S.,Han,

S.M.,2007.Water-solublegermanium(0)nanocrystals:cellrecognitionand near-infraredphotothermalconversionproperties.Small3(4),691–699. Leist,M., Single, B., Castoldi,A.F., Kuhnle, S.,Nicotera, P., 1997. Intracellular

adenosinetriphosphate(ATP)concentration:aswitchinthedecisionbetween apoptosisandnecrosis.J.Exp.Med.185,1481–1486.

Lin,M.H.,Hsu,T.S.,Yang,P.M.,Tsai,M.Y.,Perng,T.P.,Lin,L.Y.,2009.Comparisonof organicandinorganicgermaniumcompoundsincellularradiosensitivityand preparationofgermaniumnanoparticlesasaradiosensitizer.Int.J.Radiat.Biol. 85,214–226.

Lu,X.,Korgel,B.A.,Johnston,K.P.,2005.Highyieldofgermaniumnanocrystals syn-thesizedfromgermaniumdiiodideinsolution.Chem.Mater.17,6479–6485. Majno,G.,Joris,I.,1995.Apoptosis,oncosis,andnecrosis.Anoverviewofcelldeath.

Am.J.Pathol.146,3–15.

McConkey,D.J.,Orrenius,S.,1996.Theroleofcalciumintheregulationofapoptosis. J.Leukoc.Biol.59,775–783.

Medintz,I.L.,Trammell,S.A.,Mattoussi,H.,Mauro,J.M.,2004.Reversible modula-tionofquantumdotphotoluminescenceusingaprotein-boundphotochromic ﬂuorescenceresonanceenergytransferacceptor.J.Am.Chem.Soc.126,30–31. Michalet,X.,Pinaud,F.F.,Bentolila,L.A.,Tsay,J.M.,Doose,S.,Li,J.J.,Sundaresan,G., Wu,A.M.,Gambhir,S.S.,Weiss,S.,2005.Quantumdotsforlivecells,invivo imaging,anddiagnostics.Science307,538–544.

Natarajan,U.,Handique,K.,Mehra,A.,Bellare,J.R.,Khilar,K.C.,1996.Ultraﬁnemetal particleformationinreversemicellarsystems:effectsofintermicellarexchange ontheformationofparticles.Langmuir12,2670–2678.

Ngiam,S.-T.,Jensen,K.F.,Kolenbrander,K.D.,1994.SynthesisofGenanocrystals embeddedinaSihostmatrix.J.Appl.Phys.76,8201.

Pan,Y.,Neuss,S.,Leifert,A.,Fischler,M.,Wen,F.,Simon,U.,Schmid,G.,Brandau,W., Jahnen-Dechent,W.,2007.Size-dependentcytotoxicityofgoldnanoparticles. Small3,1941–1949.

Park,E.J.,Yi,J.,Kim,Y.,Choi,K.,Park,K.,2010.Silvernanoparticlesinducecytotoxicity byaTrojan-horsetypemechanism.Toxicol.InVitro24,872–878.

Riabinina,D.,Durand,C.,Chaker,M.,Rowell,N.,Rosei,F.,2006.Anovelapproachto thesynthesisofphotoluminescentgermaniumnanoparticlesbyreactivelaser ablation.Nanotechnology17,2152–2155.

Richter,C.,Schlegel,J.,1993.Mitochondrialcalciumreleaseinducedbyprooxidants. Toxicol.Lett.67,119–127.

Rieke,G.H.,2007.Infrareddetectorarraysforastronomy.Annu.Rev.Astron.Astr. 45,77–115.

Singh,A.K.,Kumar,V.,Kawazoe,Y.,2005.Designofaverythindirect-band-gap semiconductornanotubeofgermaniumwithmetalencapsulation.Phys.Rev. 71,75312.

Thiele,U.K.,2001.Thecurrentstatusofcatalysisandcatalystdevelopmentfor theindustrialprocessofpoly(ethyleneterephthalate)polycondensation.Int.J. Polym.Mater.50,387–394.

Vanwinkle,B.A.,deMesyBentley,K.L.,Malecki,J.M.,Gunter,K.K.,Evans,I.M.,Elder, A.,Finkelstein,J.N.,Oberdorster,G.,Gunter,2009.Nanoparticle(NP)uptakeby typeIalveolarepithelialcellsandtheiroxidantstressresponse.Nanotoxicology 3,307–318.

Warner,J.H.,Tilley,R.D.,2006.Synthesisofphotoluminescencewater-soluble ger-maniumnanocrystals.Nanotechnology17,3745–3749.

Washio,K.,2003.SiGeHBTandBiCMOStechnologiesforopticaltransmissionand wirelesscommunicationsystems.IEEETrans.Electron.Dev.50,656–668. Xie,P.,Hu,Y.,Fang,Y.,Huang,J.,Lieber,C.M.,2009.Diameter-dependentdopant

locationinsiliconandgermaniumnanowires.Proc.Natl.Acad.Sci.U.S.A.106, 15254–15258.

Yang,P.M.,Chen,H.C.,Tsai,J.S.,Lin,L.Y.,2007.CadmiuminducesCa2+-dependent necroticcelldeaththroughcalpain-triggeredmitochondrialdepolarizationand reactiveoxygenspecies-mediatedinhibitionofnuclearfactor-kappaBactivity. Chem.Res.Toxicol.20,406–415.

Yen,H.J.,Hsu,S.H.,Tsai,C.L.,2009.Cytotoxicityandimmunologicalresponseofgold andsilvernanoparticlesofdifferentsizes.Small5,1553–1561.

Zamaraeva,M.V.,Sabirov,R.Z.,Maeno,E.,Ando-Akatsuka,Y.,Bessonova,S.V.,Okada, Y.,2005.CellsdiewithincreasedcytosolicATPduringapoptosis:a biolumines-cencestudywithintracellularluciferase.CellDeathDiffer.12,1390–1397. Zhou,Z.,Brus,L.,Friesner,R.,2003.Electronicstructureandluminescenceof

1.1-and1.4-nmsiliconnanocrystals:oxideshellversushydrogenpassivation.Nano Lett.3,163.