Apoptogenic peptides from Tityus discrepans scorpion venom acting against the SKBR3 breast cancer

Apoptogenic peptides from Tityus discrepans scorpion venom acting against the SKBR3breast cancer cell lineGina D ’Suze a ,*,Arnaldo Rosales a ,Víctor Salazar b ,Carlos Sevcik aa Laboratory on Cellular Neuropharmacology,Biophysics and Biochemistry Center,Instituto Venezolano de Investigaciones Cienti ficas (IVIC)Caracas,Venezuela bHistology Service,Biophysics and Biochemistry Center,Instituto Venezolano de Investigaciones Cienti ficas (IVIC)Caracas,Venezuelaa r t i c l e i n f oArticle history:Received 27May 2010Received in revised form 14September 2010Accepted 22September 2010Available online 1October 2010Keywords:Scorpion venom SKBR3FasLAntineoplastic natural compounds Apoptosis BcL-2a b s t r a c tTwo novel peptides named neopladine 1and neopladine 2were puri fied from Tityus discrepans scorpion venom and found to be active on human breast carcinoma SKBR3cells.Mass spectrometry molecular masses of neopladine 1and 2were 29918and 30388Da,respectively.Their N -terminal sequences were determined by Edman degradation.The peptides induced apoptosis of SKBR3cells but had a negligible effect on non-malignant MA104monkey kidney cells.Neopladine 1and 2induced 6.3and 4.1%of SKBR3apoptosis,respectively,in 5h of exposure;the effect was larger with more prolonged exposures.Inmunohistochemistry showed that neopladines bind to SKBR3cell surface inducing FasL and BcL-2expression.Ó2010Elsevier Ltd.All rights reserved.1.IntroductionSome scorpion venoms inhibit the growth of various types of cancers.Only a few toxins have been found to be the responsible for these anticancer effects and exert their action by three different mechanism:1)blocking a speci fic ion channel (Jäger et al.,2004);2)inhibiting cancer cells invasion binding to a speci fic site (different from an ion channel)on the plasmatic membrane (Deshane et al.,2003);3)activating intracellular pathways inducing apoptosis (Gupta et al.,2010).Apoptosis is a type of pro-grammed cell death,initiated by several factors involving a cascade of intracellular events,leading to the termination of cell survival (Green and Reed,1998).There are two major pathways initiating apoptosis,the extrinsic way mediatedby a death receptor and the intrinsic pathway mediated by mitochondrial damage (Khan et al.,2007).Apoptosis is the underlying mechanism accounting for the action of clini-cally useful antineoplastic drugs such as vinblastine,cisplatin and paclitaxel (Harmon et al.,1994;Villunger et al.,1997;Reshkin et al.,2003).Recent evidence indi-cates that antineoplastic drugs induce a chain of events where MAP kinases,FasL,matrix metalloproteinases,P38,P53or BcL-2are involved in tumor cell death (Müller et al.,1997;Villunger et al.,1997;Deshane et al.,2003;Reshkin et al.,2003;Vousden,2005).Fas ligand (FasL)expression accompanies tumor cell death;the activation of Fas signaling by induction of FasL expression has been proposed as an extrinsic mechanism of apoptosis (Friesen et al.,1996;Müller et al.,1997;Villunger et al.,1997;Khan et al.,2007;Reshkin et al.,2003).Breast cancer is one of the most common malignancies worldwide.Although several novel therapeutic methods have been utilized,curative effects in advanced breast cancer remains poor.Developing new drugs against breast cancer will be essentials for improving therapeutic*Corresponding author.IVIC CBB,Apartado 20632,Caracas 1020A,Venezuela.Tel.:þ582125041225,þ584129217482(Mobile);fax:þ582125041093.E-mail addresses:gina.dsuze@ ,gdsuze@ivic.gob.ve (G.D ’Suze).Contents lists available at ScienceDirectToxiconjournal hom epage:/locate/toxicon0041-0101/$–see front matter Ó2010Elsevier Ltd.All rights reserved.doi:10.1016/j.toxicon.2010.09.008Toxicon 56(2010)1497–1505intervention and prognosis for this malignancy.Scorpion venoms are complex mixtures which may contain over two hundred different peptidic compounds(Batista et al., 2006).These compounds act directly on a broad spectrum of cell membrane ion channels(Possani et al.,1999),and indirectly by activating cellular metabolic pathways,prob-ably,via secondary messengers(López,1999)or by elec-trostatic interaction with the cell membrane(Corzo et al., 2001;Boman,2003).In recent years there have been many studies(Liu et al.,2002;Fu et al.,2007;Mamelak and Jacoby,2007;Gao et al.,2008;Gupta et al.,2010;Lee et al., 2010)aimed at identifying scorpion peptides with anti-neoplastic activity.Scorpion venoms and toxins isolated from Leiurus quinquiestratus,Buthus martensii and Hetero-metrus bengalensis have been found to kill specifically human glioma and leukemic cells(Soroceanu et al.,1998; Wang and Ji,2005;Gupta et al.,2010).Chlorotoxin, a peptide isolated from L.quinquiestratus,specifically binds to glioma cells and prevents their proliferation(Deshane et al.,2003).Recent advances in nanotechnology have led to the development of biocompatible nanoparticles for in vivo molecular imaging and for targeted therapy using chlorotoxin(Lee et al.,2010).The existence of a chlor-otoxin-like peptide has been reported in the venom of B.martensii(Wang and Ji,2005).Also,bengalin a peptide of 72kDa isolated from the venom of H.bengalensis,has antiproliferative and apoptogenic activities against human leukemic cells(Gupta et al.,2010).In this work,we studied the action of Tityus discrepans scorpion venom and two purified peptidic components on SKBR3human breast cancer cells and on non-malignant MA104cell line.Two purified peptides,called here neo-pladine1and2,were found to be active on SKBR3human breast carcinoma cells and had negligible effects on MA104 cells.This work demonstrated that neopladine1and2, induce apoptosis on SKBR3cells via the activation of Fas signaling by induction of FasL expression.2.Material and methods2.1.Venom source and extractionAdult T.discrepans scorpions were collected around Caracas(Venezuela)and kept in plastic boxes with water and food ad libitum.Adult specimens(z200)were anes-thetized once a month with CO2and milked for venom by electrical stimulation.Venom was dissolved in double distilled water and centrifuged at15,000g for15min at 5 C.Protein content was estimated based on absorbance at 280nm,assuming that1mg/ml protein produces1unit of absorbance per unit path length.The supernatant was freeze-dried and stored atÀ80 C until use.2.2.Peptides isolation proceduresVenom fractionation was accomplished using a SephadexÒG-50column(0.9Â200cm)equilibrated and eluted at25 C with20mM ammonium acetate buffer,pH 4.7(19ml/h/cm2).Fractions were collected,lyophilized and stored atÀ80 C until used.Fraction I from SephadexÒG-50 was further fractionated through a semipreparative C4reversed phase column(250Â10mm,Vydac214TP1010, Hesperia,CA,USA),at aflow-rate of1mL/min using a linear gradient from solution A[water with0.12%trifluoroacetic acid(TFA)]to80%solution B(acetonitrile with0.10%TFA)in 80min.Fractions were collected manually and dried using a Savant Speed-Vac dryer.Active fractions were further purified by analytical reverse phase chromatography on a C4column(250Â4.6mm,Vydac214TP54,Hesperia,CA, USA)using a linear gradient from38to53%solution B in 30min(0.5%solution B/min).2.3.Mass and N-terminal sequence determinationMolecular weights of neopladine1and2were deter-mined at the Harvard University FAS Center for Systems Biology Mass Spectrometry and Proteomics Resource Laboratory(MSPRL),by matrix-assisted laser desorption time-of-flight mass spectrometry(MALDI-TOF)performed on a Waters MALDI Micro MX(Waters Corp.,Milford MA) using a standard steel well plate and sinapic acid matrix in linear positive ion mode.N-terminal amino acid sequences of neopladines were determined by automated Edman degradation on an Applied Biosystems Procise494-HT Protein Sequencer(Life Technologies,Carlsbad CA)using manufacturer recommended chemistry cycles and reagents.2.4.Cell lines and cultureHuman breast carcinoma cell line SKBR3(ATCCÒNumber:HTB-30Ô)and normal monkey kidney cell line MA104(ATCC Number:CRL-2378.1Ô)were obtained from the American Type Culture Collection(Rockville,MD).Cells were grown in Dubelco/Vogt modified Eagle’s minimal essential medium(DMEM)supplemented with10%heat-inactivated fetal bovine serum(FBS)(Sigma Chemical Co., St.Louis,MO),penicillin(100U/mL),gentamicin(48m g/mL) and amphotericin B(3m g/mL)under an humidified atmo-sphere of5%CO2and95%air at37 C.2.5.Viability assaysSKBR3or MA104cell lines were cultured to confluence in96-well plates in DMEM and FBS culture medium.Before exposure to venom,culture medium was changed to FBS free DMEM.Both cell lines were incubated with SephadexÒG-50fractions FI to FV(0.15m g/m L),cell viability was determined after48h using thiazolyl blue tetrazolium bromide(Sigma Chemical Co.,St.Louis,MO)(Mosmann, 1983).Cell death was determined by reading absorbance at570nm.2.6.Morphological assessment and apoptosis–necrosisquantificationMA104and SKBR3cell lines were cultured to confluence on22Â22mm2cover slides kept in culture plates with DMEM-FBS.Cell death was determined as apoptosis or necrosis by epifluorescence microscopy using the DNA binding dyes ethidium bromide(100m g/mL)and acridine orange(100m g/mL)(EB/AO)in PBS(McGahon et al.,1995;G.D’Suze et al./Toxicon56(2010)1497–1505 1498Jolly et al.,1997).This method measures differential dye uptake in viable and nonviable cells.In this assay,the apoptotic index and cell membrane integrity are assessed simultaneously and there is no cellfixation step(Renvoize et al.,1998;Mironova et al.,2007).Venoms’(1,15and30m g/m L)and neopladines’(1m g/m L or z33m M)capacity to induce apoptosis or necrosis were measured on cell lines after6and5h exposition respec-tively.The apoptotic or necrotic effects as function of exposure time to whole venom(5m g/m L)was determined in both cell lines after1,2,3,4,5and6h.Percentages of apoptosis or necrosis were determined on at least811cells counted from randomly selected microscopefields.Controls were exposed to FBS free DMEM.Positive necrosis controls were exposed for1h to1mM H2O2to induce oxidative stress,followed by4h recovery in fresh medium(Dini,2005).Positive apoptosis controls were exposed to6nM paclitaxel(ClitaxelÒ,Nolver C.A., Venezuela)or vinblastine(Vinblastina TN,Elmor S.A.,Cara-cas,Venezuela)during5h;both drugs are known to induce apoptosis in human breast cancer(Reshkin et al.,2003).2.7.ImmunohistochemistrySKBR3monolayers incubated with whole venom,neo-pladines or fresh medium were washed with PBS,scraped and resuspended in FBS free DMEM,centrifuged3times and smeared on subbed chrome-gelatin slides,fixed with Bouins’fluid for5min,washed with PBS,discolored with 0.5%w/v ammonium acetate to eliminated picric acid,and washed with bidistillated water.Slides were dehydrated for 1h in70%ethanol supplemented with0.25%NH4OH,then rehydrated with50%ethanol for10min and transferred to PBS(Baschong et al.,2001).Antigen was retrieved by boiling slides in0.01M citrate buffer(pH6.0)in a500W microwave oven twice for5min, and extensively washed with PBS(Farioli-Vecchioli et al., 2001).Unspecific binding sites were blocked with PBS containing5%skim milk,0.2%triton X-100for1h at25 C. Four slides of experimental and control groups were incu-bated overnight with the primary antibodies at4 C with 0.2mg/mL rabbit polyclonal anti-Tityus venom IgG (1:1000),or with polyclonal sheep anti-recombinant full-length BcL-2oncoprotein IgG(ab16120,Abcam),or with rabbit polyclonal antibody against FasL synthetic peptide (human)IgG(ab15285,Abcam Inc.)all diluted1:1000in PBS containing2.5%skim milk and0.1%triton X-100.All incubation steps were followed by extensive washing in PBS.Then,slides were incubated at25 C for1h with the secondary antibodies(1:5000in PBS):goat polyclonal anti-rabbit IgG coupled with FITC(ab6717Abcam Inc.)for Tityus–venom and FasL,and donkey polyclonal anti-sheep IgG coupled with rhodamine(ab6897Abcam Inc.)for BcL-2.Nuclei were stained with DAPI(1m g/mL)for5min and washed with PBS.Finally slides were mounted with50%v/v glycerol in PBS.Control slides were incubated only with antibodies.Slides were observed using phase contrast illumination and epifluorescence on a Nikon E600light microscope (Nippon Kogahu KK,Tokyo,Japan)with40Âand100Âplan-apochromatic objectives.For epifluorescence a100W mercury lamp light source and cubes UV-2A,B-2A and G-2A(Nikon)were used to visualizes DAPI,FITC and rhoda-mine respectively.Slides were digitally photographed with a Nikon Coolpix8700digital camera.The2560Â1920pixel images were saved as24-bit color TIFFfiles.The ImageJ (version 1.34w)public domain Java image processing package(National Institute of Health,Bethesda,MA,USA) was used to merge images.2.8.Statistical proceduresSome data are presented as medians and ranges(n¼2). Large samples of viable,apoptotic or necrotic cells (n!811),guarantee that the binomial distribution of the proportions is asymptotically Gaussian.Consequently proportions(p)are presented as percentages with bars indicating standard deviation,s(p),calculated as indicated in Fleiss(1973)as follows:sðpÞ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffip$ð1ÀpÞnr:Statistical significance of the differences between proportions was calculated also with the Gaussian asymptote of the binomial distribution as follows:z¼j p1Àp2jffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffis2ðp1Þþs2ðp2Þp¼j DðpÞjffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffis2ðp1Þþs2ðp2Þp;where z is a normalized Gaussian variable;DðpÞwas considered significant if z!1:9601ÀPðzÞ0:025which corresponds to a confidence level of P0.05for a two tailed test(Fleiss,1973).3.Results3.1.Active peptides isolationVenom fractionation on SephadexÒG50columns yiel-ded5fractions called FI to FV(Fig.1A,right inset).Further fractionation of FI by HPLC rendered26fractions,out of which components labeled N1and N2with retention times of57.6and58.3min respectively were preferentially toxic on SKBR3cell line and were called neopladines(Fig.1A). The left inset in Fig.1A,shows sharp HPLC peaks which confirm the homogeneity of N1and N2,this was confirmed by mass spectrometry data where only one molecular mass of29918Da for N1and one of30388Da for N2were obtained;also,in the amino terminal sequencing only one residue per cycle was observed.The N-terminal amino acid sequence of N1was SMG RQD KILV and is different from the N2first15amino terminal residues which were AENLFSMG RQD KCLV(Fig.1B).3.2.Toxicity assaysThe differential EB/AO uptake stains nuclei of living cell in green,nuclei of early apoptotic cells with condensed or fragmented chromatin in bright green,and nuclei of late apoptotic cells with condensed and fragmented chromatin in orange;necrotic cells had orange nuclei.Fig.2A showsG.D’Suze et al./Toxicon56(2010)1497–15051499Fig.1.Venom fractionation and neopladines puri fication A)Right inset :Chromatographic pro file of T.discrepans venom fractionated on a Sephadex ÒG-50column (0.9Â200cm)equilibrated and eluted at 25 C with 20mM ammonium acetate buffer,pH 4.7(19ml/h/cm 2).I ¼FI,II ¼FII,III ¼FIII,IV ¼FIV,V ¼FV,an asterisk identi fies the active fraction.NEV ¼Normalized elution volume.A)Chromatographic pro file of FI eluted through a C4semipreparative column,at 1mL/min using a linear gradient from solution A [water with 0.12%tri fluoroacetic acid (TFA)]to 80%solution B (acetonitrile with 0.10%TFA)in 80min.A)Left insets :N1and N2Active components puri fication by elution through a C4analytical column using a linear gradient from 38to 53%solution B in 30min.B)neopladine 1and 2molecular masses determined by MALDI-TOF.Neopladine 1and 2N -terminal amino acid sequences determined by automated Edman degradation.Other details in Material and Methodssection.Fig.2.Apoptosis and necrosis percentages as function of time and venom concentration .A)Apoptosis -and necrosis A percentages of normal kidney cell (MA104)and human cancer breast cell (SKBR3)lines after 6h incubation with 1,15and 30m g/m L whole venom.B)Apoptosis -or necrosis A percent of MA104and SKBR3cell lines induced by 5m g/m L venom as function of time;insets are controls cell lines exposed to FBS free DMEM.Percentages were determined on >811cells counted from randomly selected microscope fields.Bars indicate the standard deviation of the percentages.Other details in Material and Methods section .G.D ’Suze et al./Toxicon 56(2010)1497–15051500that incubation of SKBR3cells with whole venom(1and15, 30m g/m L)for6h induced a z10fold larger effect than on the non-malignant cell line MA104incubated under similar conditions.The lowest venom concentration studied(1m g/ m L)induced more apoptosis than necrosis on the SKBR3cellline.Fig.2B shows that the SKBR3cell line suffered4.6%of apoptosis and1.4%necrosis after treatment with venom (5m g/m L)after a6h exposure.The same conditions induced 1%of apoptosis and0.1%of necrosis in MA104cells.The SKBR3cell death was time and venom concentration dependent.SKBR3cells incubated with FI,FII,FIII,FIV or FV(0.15m g/ m L)for48h,were killed in proportions of70,7,10,0and 26%,respectively(Fig.3A).After treatment with FI,SKBR3 cell were round,shrunk,had protruding elongated spikes at the cell surface,and died.Cell detachment from the substrate was also observed(Fig.3B right).These changes were not observed in the MA104cell line exposed to FI under same conditions where only16%of the cells were killed.Control SKBR3cells exposed to FBS free DMEM for 48h maintained its morphology and remained organized as a monolayer(Fig.3B left).3.3.Apoptotic and necrotic effectsNeopladine1(1m g/m L,z33m M,5h exposure)induced 2.1Æ0.28%(meanÆsd)apoptosis and0.31Æ0.11%of necrosis in MA104cells(n¼2620).Neopladine2(1m g/m L, z33m M,5h exposure)induced2.02Æ0.26%apoptosis and 0.28Æ0.10%of necrosis in MA104cells(n¼2868).Neo-pladine1(1m g/m L,z33m M,5h exposure)induced 6.31Æ0.48%apoptosis and0.57Æ0.15%of necrosis in SKBR3(n¼2613)cells.Neopladine2(1m g/m L,z33m M,5h exposure)induced4.04Æ0.29%apoptosis and0.24Æ0.07% of necrosis in SKBR3cells(n¼4649).The neopladines were also applied in combination(z33m M of each one,5h exposure);this treatment produced 5.06Æ0.56%of apoptosis and 2.92Æ0.43%of necrosis in SKBR3cells (n¼4649)(Fig.4A).We used two positive control drugs:vinblastine and paclitaxel(both6nM with a5h exposure).Paclitaxel induced0.42Æ0.21%apoptosis and no necrosis in MA104 cells(n¼952);the same drug induced 3.21Æ0.43% apoptosis and0.47Æ0.17%of necrosis in SKBR3cells (n¼1714).Vinblastine induced3.2Æ0.56%apoptosis and 0.2Æ0.14%of necrosis in MA104cells(n¼999);the same drug induced4.69Æ0.50%apoptosis and1.87Æ0.32%of necrosis in SKBR3cells(n¼1814)(Fig.4A).In the micrographies in Fig.4B it is shown that expo-sures to neopladine1or2produced nuclear fragmentation (F),cytoplasm condensation(C),membrane protrusions (P),induction of vesicles(V)and loss of cellular confluence (L)of SKBR3cells.3.4.Immunohistochemical venom localizationFig.5B–D show that whole venom and neopladines bind to SKBR3cell membranes.This is observed as anintenseFig.3.SephadexÒG50fractions effect on malignant cell line Viability.A)Cell death percentages induced on SKBR3confluent layers after48h exposure to SephadexÒG-50fractions(0.15m g/m L).Cell death was determined by absorbance at570nm.These percentages were deduced from the absorbance at570nm of cells in96-well plates after treatment with MTT.The bars are means and ranges(brackets on top of the bars)of two replicates.B)Morphology changes of SKBR3 confluent layers after48h incubation with FBS free DMEM culture medium(control,CNT)and SephadexÒG-50fraction I(0.15m g/m L).Bars¼25m m.Other details in Material and Methods section.G.D’Suze et al./Toxicon56(2010)1497–15051501green fluorescence surrounding the cells,due to FITC-labeled anti-rabbit IgG which recognizes anti-T.discrepans rabbit IgG,coupled to either venom (Fig.5B),neopladine 1(Fig.5C)or neopladine 2(Fig.5D)bound to the cell membrane.The FITC-IgG was absent from control SKBR3cells exposed only to fresh medium (Fig.5A).3.5.Venom enhanced FasL and BcL-2expressionFig.5F through Fig.5H show that SKBR3cells treated with either venom (Fig.5F),neopladine 1(Fig.5G)or neopladine 2(Fig.5H)expressed Fas ligands.This appears as an intense green fluorescence at the cell membrane,produced by FITC-labeled anti-rabbit IgG which recognizes rabbit polyclonal anti-FasL IgG coupled to FasL expressed by the cells.This fluorescence was absent in control SKBR3cells (Fig.5E).Fig.5J through Fig.5L show that BcL-2is expressed by SKBR3cells treated with venom (Fig.5J),neopladine 1(Fig.5K)or neopladine 2(Fig.5L).This is seenas an intense red fluorescence inside the cells,due to rhodamine-labeled anti-sheep IgG recognizing sheep anti-BcL-2IgG which bind to the expressed BcL-2.SKBR3control group did not exhibit fluorescence (Fig.5I).4.DiscussionScorpion venom may contain hundreds of compounds with different pharmacological activities.Venoms from scorpions from the Buthidae family can induce human death.However,not all venom compounds are toxic to humans,specially the ones with high molecular masses.T.discrepans venom induced apoptosis in a concentra-tion similar or lower than other scorpion venoms (Wang and Ji,2005;Gupta et al.,2010);four of the five venom fractions isolated on Sephadex G50were able to kill SKBR3cells.The most potent fraction,FI,contains the highest molecular weight components and is harmless to mammalians (D 0Suze et al.,1995).Neopladines 1and2Fig.4.Human cancer breast cell death induced by puri fied neopladines.A)Apoptosis (crisscrossed bars)and necrosis (open bars)percentages of SKBR3cell lines after 5h treatment with puri fied neopladines 1(N1)and 2(N2)(z 33m M)alone and in combination (N1þN2,z 33m M of each one).Positive apoptosis controls were exposed for 5h to 6nM paclitaxel (PCT)(Clitaxel ÒNolver C.A,Venezuela)or 6nM vinblastine (VBL)(Vinblastina TN ,Elmor S.A.,Venezuela).Control (CNT)group were incubated 5h with FBS free DMEM culture medium.Brackets on top of the bars indicate its standard deviation calculated as indicated in Section 2.8.B)In as little as 5h,neopladine 1and 2induced nuclear fragmentation (arrow labeled F),cytoplasm condensation (arrow labeled C),membrane protrusions (arrow labeled P),induction of vesicles (arrow labeled V)and loss of cellular con fluence (arrow labeled L)on con fluent layers of SKBR3cell lines (1m g/m L,33m M).Control (CNT)group were incubated 5h with FBS free DMEM culture medium.Bars ¼25m m.G.D ’Suze et al./Toxicon 56(2010)1497–15051502z 30KDa eluted with fraction FI.The N -terminal amino acid sequence of neopladine 2was similar to one toxin reported in the proteomic study of T.discrepans venom (Batista et al.,2006),with retention time of 50.88min (z 30kDa).Both neopladines induced nuclear fragmenta-tion,loss of cellular con fluence and apoptosis in SKBR3cell cultures.The apoptotic effects resemble those caused by B.martensii (Karsch)venom on human glioma,and by bengalin,from H.bengalensis venom,on leukemic cells (Wang and Ji,2005;Gupta et al.,2010).Chlorotoxin,iso-lated from L.quinquiestratus venom,inhibits glioma growth by interacting with the plasmalemma matrix metal-loproteinase-2(Deshane et al.,2003).Chlorotoxin effect occurs at concentrations ranging between 10and 20m M,close to the neopladines ’z 33m M.Neopladines act on SKBR3cells,if this interaction occurs by binding to membrane receptors (Ehrenpreis et al.,1969)or if they directly activate second messengers which trigger the expression of FasL,caspases or BcL-2(Villunger et al.,1997;Reshkin et al.,2003),is still unknown.A group of snake venom toxins block malignant cell migration and invasion via an RGD-amino acid sequence dependent interaction (Ritter et al.,2000;Schmitmeier et al.,2000;Sheu et al.,1997).We found that both neo-pladines ’N -terminal have a RQD-amino acid sequence perhaps determining a similar toxin –cell interaction.Neopladine 1and 2bind to SKBR3cell membrane,induce FasL expression and trigger programmed cell death;similar apoptotic mechanisms are known for other drugs (Villunger et al.,1997;Reshkin et al.,2003).FasL expression occurs during methotrexate-or doxorubicin-induced apoptosis in CEM T-ALL cells (Friesen et al.,1996)and bleomycin-and cisplatin-induced apoptosis in HepG2hepatoma cells (Müller et al.,1997).BcL-2,an integral outer mitochondrial membrane protein,suppresses apoptosis in a variety of cell systems;yet,malignant cells under the action of apoptotic drugs,produce BcL-2(Chao and Korsmeyer,1998).Thus,in ef fi-cient apoptotic drugs,the cytotoxic effect prevails over the antiapoptotic BcL-2effect.Neopladines were abletoFig.5.Immunohistochemical localization of venom,neopladine 1,neopladine 2,FasL and BcL-2on malignant treated cell.SKBR3con fluent layers after exposure to fresh medium,whole venom or neopladine 1or 2.Whole venom (B),neopladine 1(C)and neopladine 2(D)was immunohistochemical localized as a fluorescent green color at the malignant cell surface membrane.Similar immunohistochemical localization of FasL is shown as an intense green color at SKBR3cell membrane as response to venom (F),neopladine 1(G)and neopladine 2(H).Immunohistochemical localization of BcL-2is shown as an intense red color inside SKBR3cells as response to venom (J),neopladine 1(K)and neopladine 2(L)treatment.Control SKBR3cells were exposed to either culture medium and to primary and secondary antibodies anti-Tityus (A),anti-FasL (E)and anti-BcL-2(I).Sections were observed by epi fluorescence microscopy.Bars ¼2m m.Other details in Material and Methods section.G.D ’Suze et al./Toxicon 56(2010)1497–15051503promote BcL-2in SKBR3cells,and still were active proa-poptotic compounds.We observed3populations of cells in cultures exposed to antineoplastics:viable cells,apoptotic cells and necrotic cells.With the doses and the observation periods used,only vinblastine(P¼2Â10À7)and neopladine1(P¼0.009), produced significant necrosis on SKBR3cells when compared with the controls.Neither paclitaxel(P¼0.054) nor neopladine2(P¼0.266)necrotic SKBR3cell frequen-cies were statistically different from the controls.Neo-pladine1induced larger percentages of SKBR3apoptosis than paclitaxel(P¼6Â10À7)and vinblastine(P¼0.009). Neopladine2was as apoptogenic for SKBR3as paclitaxel (P¼0.053)and as vinblastine(P¼0.129).The SKBR3 percent apoptosis when the two neopladines were applied together,was weakly significantly lower than the SKBR3 apoptotic effect of neopladine1(P¼0.044)and not different from neopladine2(P¼0.052);still,the SKBR3 necrotic cell percentage of the two neopladines combined was the largest one observed(P¼0.025versus vinblastine, and P<10À7versus any of the other conditions tested).Vinblastine(P¼0.006)increased apoptosis slightly above the control in MA104cells;neopladine1(P¼0.115) and neopladine2(P¼0.152)were not apoptotic in these cells.Neither the neopladines nor vinblastine had necroticeffects above the control values for MA104culturesðP! 0:069Þ,paclitaxel produced no necrosis at all.Taken together with the results discussed in the preceding paragraph the neopladines seem to be apoptotic peptides with selectivity for SKBR3breast cancer cells rather than for non-malignant MA104monkey kidney cells. Promoting apoptosis is the mechanism underlying the effect of efficient chemotherapeutic drugs such as vinblastine(Harmon et al.,1994)cisplatin and paclitaxel (Villunger et al.,1997;Reshkin et al.,2003).We must, however,deal with the bigger necrotic pool observed with an apparently unchanged apoptotic pool,when the neo-pladines were applied together.To explain this apparent paradox,we will consider3possible kinetic relationships between the cell pools under consideration.Two of these models are shown in Fig.6.At the top of thefigure we present what we call a sequential model where the drug induces apoptosis at a rate k A,and the apoptotic cells die, becoming necrotic,at a rate k N.The bottom of Fig.6pres-ents an alternative model(parallel model)where a direct pathway between the viable and necrotic pools exists (dotted line)in parallel with the sequential model.A third alternative could be similar to the parallel model,but with no connection between the apoptotic and necrotic pools; since this implies that the apoptotic cells are immortal,we will disregard this third possibility.From the sequential model it is very easy to explain that a drug,such as the neopladines,at short times and/or at low concentrations,may have a significant apoptotic effect without apparent necrosis.It is also intuitively clear that as the apoptotic pool increases with time,a necrotic pool must become evident.The time and concentrations of drug required depend on f¼k A=k N,the ratio between the rates of apoptosis and necrosis.When the viable pool is very large the model could reach a state where the apoptotic pool seems constant while the necrotic pool increases;implicit in the model is that an apoptotic drug will somehow increase k A and perhaps also k N,thus the sequential model is quite compatible with our results with neopladines1and2,applied alone or in combination.To explain our results with the neopladines1or2not in combination,with the parallel model,the branch repre-sented by the dotted line must correspond to either a very low k N’or to a cooperative process which starts with a delay.Otherwise,some increase in the necrotic pool size should occur in parallel with an increase in the apoptotic pool.We will accept the sequential model as valid,for a reason of simplicity for the time being,until we are able to do a full synergism study of the neopladines.The statistical analysis clearly shows that neopladines are apoptotic peptides on SKBR3cell and have little or no effect on MA104cells.Furthermore,the apoptogenic effects of the neopladines were statistically significantly larger than the apoptogenic effect of paclitaxel and vinblastine. The neopladines also had negligible necrotic effects on SKBR3cells when applied separately.Also the percent apoptosis induced by the neopladines may seem small due to the short observation period,but is completely equiva-lent to the effects of two classical antineoplastic drugs (paclitaxel and vinblastine)studied under equivalent conditions.Regarding the comparison of the pure peptides with the whole venom,please notice that the venom is a cocktail of bioactive peptides with a wide spectrum of toxicities;it is not thus surprising tofind that whole venom would be more toxic to anything than its individual components. Neopladines are a new kind of antineoplastic peptides, which are interesting per se,as tools to understand the physiopathology of cancer,or as templates for novel anti-neoplastic drugs of clinical interest.Given the chlorotoxin success at labeling malignant cells to better remove tumors (Veiseh et al.,2007),peptides such as neopladines,which Fig.6.Two possible models to produce apoptosis and necrosis.We observed3 populations of cells in cultures exposed to antineoplastics:viable cells, apoptotic cells and necrotic cells.At the top of thefigure we present what we call a sequential model where the drug induces apoptosis at a rate k A,and the apoptotic cells die,becoming necrotic,at a rate k N.The bottom offigure presents an alternative model(parallel model)where a direct pathway (dotted line)between the viable and necrotic pools exists in parallel with the sequential model with a rate constant k N0.G.D’Suze et al./Toxicon56(2010)1497–1505 1504。