biro1, a cell-permeable bh3 peptide, promotes ... · bh3 peptide–based approach uses the prodeath...

Cell Death and Survival

BIRO1, a Cell-Permeable BH3 Peptide, PromotesMitochondrial Fragmentation and Death ofRetinoblastoma CellsNathalie Allaman-Pillet1, Anne Oberson1, and Daniel F. Schorderet1,2,3

Abstract

Retinoblastoma is the most common pediatric intraocular neo-plasm. While retinoblastoma development requires the inactiva-tion of both alleles of the retinoblastoma tumor suppressor gene(RB1) in the developing retina, additional genomic changes areinvolved in tumor progression, which progressively lead to resis-tance of tumor cells to death. Therapeutics acting at very down-stream levels of death signaling pathways should therefore beinteresting in killing retinoblastoma cells. The BH3-only proteinspromote apoptosis bymodulating the interactionbetween thepro-and antiapoptotic members of the BCL2 protein family, and thiseffect can be recapitulated by the BH3 domains. This reportanalyzes the effect of various BH3 peptides, corresponding todifferent BH3-only proteins, on two retinoblastoma cell lines,

Y79 and WERI-Rb, as well as on the photoreceptor cell line661W. The BH3 peptide BIRO1, derived from the BCL2L11 deathdomain, was very effective in promoting Y79 and WERI-Rb celldeath without affecting the 661W photoreceptor cells. This celldeath was efficient even in absence of BAX and was shown to becaspase independent.WhileROSproductionorAIF releasewas notdetected from mitochondria of treated cells, BIRO1 initiatedmitochondria fragmentation in a short period of time followingtreatment.

Implications: The BIRO1 peptide is highly effective at killingretinoblastoma cells and has potential as a peptidomimetic. MolCancer Res; 13(1); 86–97. �2014 AACR.

IntroductionRetinoblastoma (RB) is a malignant tumor of the developing

retina that manifests in early childhood, affecting 1 in 15,000 livebirths. Retinoblastoma is initiated by inactivation of both allelesof the RB1 tumor suppressor gene in the developing retina (1).Loss of both alleles appears however to be insufficient to initiatetumor formation, and further genomic changes may drive tohighly proliferative retinoblastoma, exhibiting altered gene copynumbers andmodulated expression of oncogenes (MYCN, E2F3,DEK, KIF14, and MDM4) and tumor suppressor genes (CDH11and NGFR; refs. 2–5).

Much progress has been done in the treatment of retino-blastoma and the survival rate in developed countries is morethan 95% with appropriate treatment, that is, through chemo-therapy with adjuvant local treatments. The incidence of met-astatic disease is low. Unfortunately, when retinoblastoma isdetected at later stages of the disease, especially in childrenrepresenting the first case in a family or in developing countries,the prognosis is poor for visual function and even for survival,as only 50% of the children survive (6). In these situations,aggressive treatments, including enucleation of the most affect-

ed eye, radio/brachy/cryotherapy and sometimes systemic orlocal chemotherapy are involved. These treatments, in additionto having a poor efficiency, have major drawbacks in pediatricpatients, such as bone marrow suppression, second canceroccurrence, cataracts, and retinopathy. Therefore, finding saferand more efficient treatments remains a major challenge.

During retinoblastoma spreading, RB1 inactivation is fol-lowed by additional genomic modifications which progressivelylead to resistance of tumor cells to death. To counteract anytumoral modification and death resistance, drugs that act atdownstream levels of death signaling pathways should be effec-tive in killing retinoblastoma cells. The emerging small antican-cer drugs which target the antiapoptotic members of the mito-chondrial machinery offer a possibility to overcome the deathresistance of cancer cells. Mitochondria are a key downstreamelement in the different programmed cell death pathways, thatis, apoptosis, autophagy, and necrosis. Mitochondrial integrity isunder the control of the proteins of the BCL2 family (7). Theseproteins function as a life/death switch that integrates diverseinter- and intracellular events to determine whether the apo-ptotic pathway should be activated (8). The switch operatesthrough the interactions between pro- and antiapoptotic mem-bers of the family. The prosurvival family members (BCL2,BCL2L1, BCL2L2, MCL1, BCL2A1) are critical for cell survival,as loss of any of them causes cell death in certain cell types (9).Consistently, BCL2 overexpression induces drug resistance invarious tumors cell types. The proapoptotic BCL2 family mem-bers are divided into 2 classes: the BAX-like proteins (BAX, BAK1,and BOK) and the BH3-only proteins (BCL2L11, BID, BBC2,PMAIP1, BMF, BAD, HRK, and BIK; refs. 10, 11). Structuralstudies have revealed that the 20 amino acids BH3 domain of theBH3-only proteins is able to interact with specific residues of the

1Institute for Research in Ophthalmology, Sion, Switzerland. 2Facultyof Biology and Medicine, University of Lausanne, Lausanne, Switzer-land. 3Ecole Polytechnique F�ed�erale de Lausanne, Faculty of LifeSciences, Lausanne, Switzerland.

Corresponding Author: Nathalie Allaman-Pillet, Institute for Research inOphthalmology, Av. de Grand-Champsec 64, Sion 1950, Switzerland. Phone:41-27-2057903; Fax: 41-27-2057901; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-14-0253

�2014 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 13(1) January 201586

on September 27, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst August 20, 2014; DOI: 10.1158/1541-7786.MCR-14-0253

http://mcr.aacrjournals.org/

BH1, BH2, and BH4 domains of the prosurvival members (9),leading to their neutralization. Relating to BAX and BAK1, theyhave been shown to be essential to the proapoptotic function ofBH3-only proteins (12). The lack of both BAX and BAK1 inducesperinatal lethality in mice, whereas cells knockdown for boththese proteins are resistant to overexpression of BH3-only pro-teins as well as to various stimuli known to activate the intrinsicapoptotic pathway.

Manipulating the balance between the pro- and antiapoptoticBCL2 family members presents a great opportunity to restoreapoptosis in cancer cells (13–21).

BH3 peptide–based approach uses the prodeath BH3 min-imal death domains to re-establish mitochondrial sensitivityin tumoral cells (13, 14). Small peptides derived fromBH3 domains can retain an a-helical structure (14) and areable to recapitulate the proapoptotic effects. In cell-free assaysconsisting of isolated mitochondria, BH3 peptides disruptcomplexes formed between pro- and antiapoptotic BCL2 fam-ily proteins (22) and induce oligomerization of BAX andBAK1 followed by mitochondrial membrane permeabilizationand release of cytochrome c. These effects could be abolishedby using BAK1�/� mitochondria or by BCL2 overexpression(22). BH3 peptides delivered into tumor cells have beenshown to engage apoptosis through BAX activation and cyto-chrome c release (18, 23). Therefore, BH3 peptides can providemolecular targeting of antiapoptotic members of BCL2 proteinfamily and potentially improve traditional therapy of cancer.

In this report, we studied the death potential of different BH3-derived peptides on 2 human retinoblastoma cell lines, Y79 andWERI-Rb, as well as on primarymouse retinoblastoma. Amassivecell death was promoted by the BH3 peptide from BCL2L11through a mechanism that is caspase-independent and that hasnecrotic characteristic.

Materials and MethodsChemicals

The different BH3 peptides were synthetized by NeoMPS andAuspep. zVAD was purchased from Promega; 3-methyladenine(3-MA) and Necrostatin-1 were from Calbiochem; and butylatedhydroxyanisole (BHA) was from Sigma-Aldrich. The primaryantibodies used for Western blotting, immunoprecipitation, andimmunofluorescent experiments were as follows: BAX (Santa-Cruz sc-493), BAX conformation (BD Pharmingen, 556467),BAK1 (Cell Signaling, 3814), BCL2 (Cell Signaling, 2876),BCL2L1 (Cell Signaling, 2764), BCL2L11 (Cell Signaling,2819), BID (Cell Signaling, 2002), MCL1 (Cell Signaling,4572), BCL2A1 (sc-8351), Noxa (sc-30209), BAD (sc-942), BBC2(Cell signaling 4976), VDAC1 (Abcam, Ab16816-100), actin(Sigma, A5441), PARP (Santa-Cruz sc-7150), AIF (Santa-Cruzsc-5589), DNM1L (BD Bioscience, 611113), LC3 (Cell Signaling,2775), PGAM5 (Santa-Cruz sc-161156), Ezrin (sc-20773), andOPA1 (BD Bioscience 612606).

WERI-Rb, Y79, and 661W cell culturesAll the cell lines were cultured in RPMI-1640 medium

supplemented with 100 mg/mL streptomycin, 100 units/mLpenicillin, 1 mmol/L sodium pyruvate, 2 mmol/L glutamine,and 10% FCS (20% for Y79). 661W cells were generouslyprovided by Dr. M. Al-Ubaidi (University of Oklahoma, Okla-homa City, OK).

Isolation of primary tumoral cellsThe SV40-LT transgenic mice developing retinoblastoma are a

gift of J.M. O'Brien (Salk Institute for Biological Studies, La Jolla,CA; ref. 24). The eye of sickmice was removed and tumormaterialwas isolated. Primary tumor cells were immediately transferredinto RPMI-1640 without FBS, minced, and washed with PBS toremove any residual blood. Thewashed cellswere then cultured inRPMI-1640 containing 10% FBS as above.

Cell viability assaysFor viability assays, cells were seeded at a density of 10,000 cells

per well in 96-well plates, incubated overnight in 10% FBS/RPMI(20% for Y79), and then treated for varying lengths of time withincreasing doses of ABT-737. Following treatment, ATP content wasmeasured using the ATPlite assay, a luminescence ATP detectionassay system developed by PerkinElmer using a microplate reader(Bio-Tek Instruments).

Hoechst/Propidium iodide stainingDying cells were discriminated from normal cells by the

Hoechst/propidium iodide (PI) technique. Briefly, tumors wereincubatedwithHoechst 33342 (12.5mg/mL) andPI (5mg/mL) for5 minutes before visualization under a fluorescence microscope.

FACS analysisCell viability was evaluated by the Annexin V/PI (BD Bios-

ciences) double staining assay following the manufacturer'sinstructions. Y79 cells were centrifuged at the end of treatment,rinsed twice with PBS, and stained with Annexin V-FITC apoptosisdetection kit I (BD Biosciences). Analysis was performed on theFACS Calibur using CellQuest software.

Caspases-7 and -3 activationCells were seeded at a density of 10,000 cells per well in 96-well

plates, incubated overnight in 10%FBS/RPMI (20% for Y79), andthen treated for varying lengths of timewith increasing doses ABT-737 (1, 3, and 10 mmol/L), in presence or absence of cisplatin andetoposide. Following treatment, caspase-3 and -7 activity wasmeasured using the Caspase-Glo 3/7 Assay from Promega.

The cleavage of PARP, a downstream substrate of caspase-3,wasstudied by Western blotting experiment using an antibody thatrecognizes the full-length protein (105 kDa), aswell as the cleavedform (85 kDa).

Measurement of reactive oxygen species levelFollowing exposure to ABT-737, cells were stained with either

10 mmol/L of the dye dihydroethidium (HE, Invitrogen) tomeasure intracellular superoxide or 10 mmol/L of the dye dihy-drodichlorofluorescein-diacetate (H2DCFDA, Invitrogen) in 1-mLmedia to measure intracellular hydrogen peroxide. Cells werestained for 30 minutes at 37�C, and fluorescence was measuredusing an EnVision 2103 Multilabel Reader (PerkinElmer).

Whole-cell lysatesCells were washed once in cold PBS and recovered by centri-

fugation. Briefly, cell pellets were dislodged into cold lysis buffer(20 mmol/L Tris-acetate, pH 7.0, 0.27 mol/L sucrose, 1 mmol/LEDTA, 1mmol/L EGTA, 50mmol/L sodiumfluoride, 1%TritonX-100, 10 mmol/L b-glycero-phosphate, 1 mmol/L dithiothreitol,10 mmol/L p-nitrophenyl-phosphate, and antiproteases) andcentrifuged at 15,000 rpm for 20 minutes. Supernatants were

Cell Death Induction by BIRO1

www.aacrjournals.org Mol Cancer Res; 13(1) January 2015 87




recovered and stored at �70�C until use. Total protein in celllysates was quantified by Bradford method.

Cell lysates used for immunoprecipitation were prepared in0.1% Triton X-100 lysis buffer (standard lysis buffer with only0.1% Triton X-100).

Isolation of mitochondrial extractsCells were lysed by mechanical homogenization using a small

pestle, and mitochondrial extraction was performed using themitochondrial extraction kit from Pierce according to the man-ufacturer's instructions.

Western blotting experimentsEqual quantities of total protein lysates were resolved by 8% to

15% SDS-PAGE and electrotransferred onto polyvinylidenedifluoride membranes. Nonspecific protein binding was blockedby incubating the membrane with a blocking solution (1� TBS,0.1%Tween 20, 5%nonfat driedmilk powder) for 1 hour at roomtemperature. The blots were then probed with primary antibodiesovernight. The immune complex was detected by using a perox-idase-conjugated secondary antibody and the chemiluminescentdetection kit according to the manufacturer's specifications(Millipore).

For non-denaturing SDS-PAGE, b-mercaptoethanol was notadded to samples.

Pull-down and immunoprecipitation experimentsY79 protein extracts (1 mg) were first incubated with a GST

protein bound to glutathione–agarose beads for 3 hours at 4�C.Following this preclearing step, the supernatant was incubatedwith a GST/BIRO1 fusion protein bound to glutathione–agarosebeads for 16 hours at 4�C. After 4 washes, the proteins pulleddown with the GST/BIRO1 were subjected to SDS-PAGE andimmunoblotting.

Immunoprecipitations were performed with the PIERCE Co-IPKit as mentioned by the manufacturer.

Immunofluorescence microscopyCells were grown on coverslips at 60% density and fixed in ice-

cold acetone: methanol (1:1) for 5 minutes at room temperature,followed by permeabilization with 0.2% Triton X-100/PBS for 5minutes. Cells were then blocked using 2% BSA in PBS for 1 hourand incubated with the monoclonal anti-ezrin antibody at thedilution of 1:100. Bound antibodies were detected with AlexaFluor-594 (1:4,000; Molecular Probes, Inc.) secondary anti-mouse antibody.

Mitochondria were detected by staining with MitoTracker RedCM-H2Xros (Molecular Probes) at 200 nmol/L for 30 minutes.Nuclei were counterstained with the DNA dye Hoechst 33285(Sigma). Coverslips were mounted with Vectorshield aqueousmount (Vector Laboratories) andobserved andphotographedusingan Olympus BL51 fluorescence microscope at 400�magnification.

ResultsExpression of BCL2 family members in tumor cell lines

Analyzing apoptosis induced by BH3 domains, we investi-gated the expression of the different partners involved inmitochondrial outer membrane permeabilization, that is, anti-apoptotic (BCL2, BCL2L1 and MCL1), proapoptotic (BAX andBAK1), and BH3-only proapoptotic proteins (BCL2L1 and

BBC2) in human retinoblastoma tumoral cells Y79, WERI-Rb,in primary mouse retinoblastoma (PR), in Rb4820 andRb6226, 2 primary mouse tumoral cell lines derived fromprimary retinoblastoma, in 661W photoreceptor cell line, andin nonocular tumoral cells. While BAX was highly expressed inWERI-Rb, Rb4820, Rb6226, and in 4 different primary retino-blastoma tumors and in 661W cells, it was completely absent inY79 cells (Fig. 1A). BAK1 was lacking in Rb4820 and 661W.BCL2L1 was abundantly expressed in Y79 and WERI-Rb, where-as MCL1 was detected in both these human cells. These datasuggest that WERI-Rb, the 4 PR, and Rb6226 should be moresensitive to cell death induction than Y79 and Rb4820.

Induction of cell death in retinoblastoma cell lines using BH3peptidic compounds

Cellular delivery of peptides derived from the BH3 domainwasachieved by linking BH3 sequences to the peptidic transporterTAT48-57 from HIV (10 amino acids, GRKKRRQRRR), known tohave a good efficacy in crossing cellmembranes fromdifferent celltypes (25) and in transducing healthy cells from various compart-ments of the eye following injection into the vitreous body orsubretinal space of adult mice (26). Various peptides weredesigned on the basis of the chemical and conformational prop-erties of BH3 domains (Table 1). To enhance protease resistanceand improve stability, peptides were synthesized in D-retroinverso conformation (25, 26). To assess the capacity of D-TATpeptides to cross the cellmembrane, cell lineswere incubatedwiththe D-TAT/FITC–labeled peptide. As observed in Fig. 1B, thefluorescent peptide was able to very rapidly (1 hour) enter tumorcells, as well as 661W cells. The entry of BIRO1/FITC, the BH3-labeled peptide derived from BCL2L11, was also tested and itslocalization was determined under a fluorescent microscope,showing a diffuse repartition in the whole cell without accumu-lation in any specific compartment (Fig. 1C).

We next examined whether cell-permeable BH3 peptides couldinduce cell death by exposing Y79 and WERI-Rb to peptidescorresponding to the BH3 domains of BCL2L11, BAD, BIK, andBOK. BIRO1 exhibited a massive killing activity (80%; Fig. 2A),whereas BAD displayed a moderate cell death induction, and theother peptides showed little, if any, killing effect. As a control, cellswere exposed to a mutated form of BIRO1, BIRO1 MT, in whichthe conserved leucine and aspartate residues were mutated toalanine. The mutation of these 2 critical amino acids has beenshown to disrupt interaction with other members of the BCL2protein family (19, 22). The prodeath effect of BIRO1 MT wasdecreased comparedwithBIRO1,withonly20%of cell death (Fig.2A). SN50 peptide was used as a negative control, as a peptideunrelated to BCL2 proteins family.

To assess the kinetics of ATP content decrease in Y79 andWERI-Rb, cells were exposed to BIRO1 for 48 hours and ATP content wasmeasured at 4, 8, 24, and 48 hours. As shown in Fig. 2B, ATPcontent was already massively decreased 4 hours posttreatment inWERI-Rb (70%) and this downregulation was already well pro-moted inY79 (40%). A rapid reduction in cellular ATP contentwasdetected in cells undergoing necrosis. To discriminate betweennecrosis and apoptosis, Annexin V-FITC/PI double staining anal-ysis was performed by flow cytometry on Y79 cells treated withBIRO1. After only 2 hours of BIRO1 treatment, Y79 culture showed25% of PI-positive cells and 52% of cells positive for doublestaining (Annexin V-FITC and PI), which suggests the activationof a necrotic death pathway rather than apoptosis (Fig. 2C).

Allaman-Pillet et al.

Mol Cancer Res; 13(1) January 2015 Molecular Cancer Research88




In necrosis cell death, cell swelling occurs with membraneblebbing leading to membrane rupture. To observe the impactof BIRO1 on cell membrane behavior, we determined the local-ization of Ezrin (EZR) by immunofluorescence. EZR is a proteinlocalized just beneath the plasma membrane, playing the role oflinker between the plasma membrane and actin cytoskeleton. Asshown in Fig. 2D, the treatment of Y79 with BIRO1 quicklyinduced outgrowths followed by cell membrane breakdown.

To investigatewhether BIRO1killing effectwas restricted toY79and WERI-Rb cell lines, we tested cell death induction in 661Wcells, as well as in nonocular tumoral cells. MDA-465 cellswere the only one to display cell death (60%) after exposureto 10 mmol/L BIRO1 (Fig. 2E). At higher peptide concentration

(20 mmol/L), U2OS and HCT-116 showed 80% and 50%cell death. As a control for toxicity of high peptidic concentration(>10 mmol/L), cells were exposed to 20 mmol/L of the mutatedform of BIRO1. No toxic effect was observed.

Effect of BIRO1 on primary mouse retinoblastomaAs the killing activity of BIRO1 was highly variable from one

cell line to another, the effect of the peptide on primary retino-blastoma tumors was difficult to predict. Primary mouse retino-blastoma were isolated from the SV40-LT transgenic mice (24)and their sensitivity to BIRO1 was tested. Retinoblastoma tumorswere very sensitive to BIRO1, as observed by Hoechst/PI staining,whereas BIRO1 MT had a weak effect (Fig. 2F). These resultsinstigate us to further examine BIRO1 efficacy in vivo in mouseretinoblastoma models.

Cell death induced by BIRO1 is caspase-independentThe rapid ATP content decrease and the early PI staining

observed in Y79 and WERI-Rb treated with BIRO1 suggested thata mechanism other than apoptosis was implicated in the cell

Table 1. Sequence of the different BH3 peptides

BIRO1 W I A Q E L R R I G D E F N A Y Y ABIRO1 MT W I A Q E A R R I G A E F N A Y Y ABAD R Y G R E L R R M S D E F V D S F K G GBOK E V C T V L L R L G D E L E Q I R PBIK A L A L R L A C I G D E M D V S L R

Figure 1.BCL2 family members content andcellular import of the D-TAT/FITC–labeled peptide. A, protein expressionof BCL2 protein family members wasdetermined by Western blot analysisin Y79, WERI-Rb, 661W, primarymouse retinoblastoma (PR1–4), aswellas in cell lines derived from theseprimary tumors (Rb4820, Rb6226)and in nonocular tumoral cell linesBT-549 and MDA-465 (human breastcancer cells), HCT116 (human coloncancer cells), U2OS and SaOS (humanosteosarcoma). B, various cell lineswere incubated with the D-TAT/FITC-labeled peptide (5 mmol/L) for 1 hourand were visualized using live cellfluorescent microscopy. C, Y79 cellswere exposed for 1 hour to 10 mmol/LBIRO1/FITC, and localization of thegreen fluorescence was evaluatedusing live cell fluorescent microscopy.Mitochondria were visualized afterstaining with a MitoTracker (red).






Figure 2.Effect of BH3 peptides onWERI-Rb andY79 andon primarymouse retinoblastoma. A, the BIRO1 peptide induces cell death inWERI-Rb andY79 retinoblastoma cells.Y79 andWERI-Rb cellswere exposed todifferentBH3 cell-permeable peptides (3 and 10mmol/L) and to apeptide unrelated toBCL2proteins family (SN50) aswell asto cisplatin (100 mmol/L) and etoposide (10 mmol/L) for 24 hours. Cell viability was determined by measuring ATP content of the cells using a luminescenceassay. Independent experimentswere repeated 5 times. Dark grey column, Y79; grey column,WERI-Rb; B-MT,mutated BIRO1. B, the cell death promoted byBIRO1 isa very rapid process. Y79 and WERI-Rb cells were exposed to various concentrations of BIRO1, and cell viability was determined at different time point bymeasuring theATP content of the cells using a luminescence assay. Independent experiments havebeen repeated 5 times. Grey lines, Y79; dark lines,WERI-Rb; circle,10mmol/LBIRO1; triangle, 3mmol/LBIRO1; square, noBIRO1. C,flowcytometric analysis of plasmamembraneswithAnnexinV-FITC/PI double staining. Y79 cellswereincubated either in the absence of BIRO1 for 5 hours (control) or in the presence of 10 mmol/L BIRO1 for 2 and 5 hours. Undamaged cells were stained withnegative Annexin V-FITC/PI (bottom left quadrant). After incubation with BIRO1 for 2 hours, a significant number of cells were stained with positive Annexin V-FITCand positive PI (top right quadrant). D, Y79 cells were left untreated (control) and treated with 10 mmol/L BIRO1 for 45 and 90 minutes. Cells were thenimmunostained with Ezrin monoclonal primary antibody and Alexa Fluor anti-mouse secondary (595-red) antibody. �, membrane swelling; dark arrow, membraneblowout. E, differential death response of various tumoral and nontumoral cell lines to BIRO1 peptide. In addition to WERI-Rb and Y79 cells, other tumoralcell lines and 661Wwere exposed to increasing concentrations of BIRO1 and to 20 mmol/L of the inactive BIRO1 MT for 24 hours, and cell viability was determined atdifferent time point bymeasuring the ATP content of the cells using a luminescence assay. Independent experiments have been repeated 3 times. F, retinoblastomathat developed in SV40-LT transgenic mice were isolated at 3 months of age. Primary tumoral cells were left untreated for 48 hours before treatment for 16 hourswith 10 mmol/L BIRO1 or with the mutated form of the peptide (BIRO1 MT). Photographs showing Hoechst 33342 (blue)/propidium iodide (red, dead cells)staining is representative of results obtained in 3 different primary retinoblastoma.






death program. To assess whether caspases were involved, thebroad-spectrum caspase inhibitor Z-VAD-FMK was added to Y79and WERI-Rb exposed to BIRO1. The BH3 peptide induced celldeath in the presence of Z-VAD-FMK (Fig. 3A). In addition, wewere unable to measure any caspase activation by Western blot-ting (data not shown) or fluorescent assay in Y79 and WERI-Rbexposed to BIRO1 (Fig. 3B), whereas treatment with etoposide orcisplatin triggered caspase activation. Caspase activation can alsobe visualized by the cleavage of caspase substrate proteins. Asshown in Fig. 3C, Y79 and WERI-Rb exposed to BIRO1 did notlead to PARP cleavage whereas cisplatin did. In agreement with anonapoptotic death program, we were unable to detect any BAXtranslocation (data not shown), cytochrome c, or AIF release inboth human retinoblastoma cell lines (data not shown). Check-ing for DNA fragmentation, we noticed chromatin cleavagein WERI-Rb exposed to BIRO1 as well as to cisplatin (data notshown), whereas no DNA fragmentation was detected in Y79.These results suggest that a small fraction ofWERI-Rb cells enteredinto apoptosis, whereas the majority of the cells died throughnecrosis. Relating to Y79, the absence of BAXmost likely repressedany apoptotic program through the mitochondria, leading to anecrotic death process.

Cell death induced by BIRO1 did not implicate autophagy ornecroptosis

To determine whether the autophagy or the necroptoticmachinery were modulated in Y79 and WERI-Rb exposed toBIRO1, cells were pretreated with 3-MA and necrostatin (Nec-1), inhibitors of autophagy and necroptosis, respectively. Neither3-MA nor Nec-1 protected Y79 and WERI-Rb against cell deathinduced by BIRO1 (Fig. 3A). To further confirm an autophagy-independent killing pathway, the absence of conversion of LC3Ito the lipidated form LC3II was assessed (refs. 27, 28; Fig. 3D).Our data suggested therefore that BIRO1-induced cell death inY79 andWERI-Rbwas notmediated by autophagy or necroptosis.

Cell death induced by BIRO1 did not induce calpains activationor ROS production

Calpains are important players in programmed necrosis. Thesecalcium-activated proteases initiate lysosomal disruption (29)followed by the release of cathepsins and the rearrangement ofthe actin cytoskeleton (30). To clarify whether BIRO1 inducedmodulation of calpains activity and downstream release of lyso-somal proteases, we tested whether calpains inhibitors (calpas-tatin, ALLN) as well as cathepsin B and L inhibitors (ALLN)decreased BIRO1 effect on cell viability. Both calpastatin andALLN failed to affect cell death induced by BIRO1 (Fig. 3A).

ROS is another effector in cell death signaling pathways,including apoptosis and necrosis. To explore its potential rolein BIRO1-induced cell death, Y79 and WERI-Rb were pretreatedwith the antioxidant BHA before exposure to BIRO1. As shownin Fig. 4A, BHA did not attenuate retinoblastoma cell death. Wealso evaluated BIRO1 effect on intracellular ROS production,that is, hydrogen peroxide and superoxide, using 2 differentfluorescent probes (HE or H2DCFDA). Exposure of WERI-Rb toBIRO1 did not result in an increase in intracellular ROS pro-duction, neither hydrogen peroxide (Fig. 4B) nor superoxide(data not shown), as measured by fluorescence emission. Atthe same time, the BH3 mimetics ABT-737, known to induceWERI-Rb apoptosis (21), induced hydrogen peroxide produc-tion (Fig. 4B).

BIRO1 promoted mitochondrial fragmentationin Y79

Under physiologic conditions, mitochondria are elongatedand filamentous. Upon stress conditions or apoptotic stimuli,mitochondria become fragmented. Mitochondrial fissionimplicates the constriction and cleavage of mitochondria byfission proteins that mediate the remodeling of the outer andinner mitochondrial membranes (31). It has previously beenshown that Bax�/�/Bak�/� MEF cells exposed to a high con-centration of a BID BH3 peptide (100 mmol/L) displayedmitochondrial fragmentation characterized by spherical mito-chondria (32). To assess whether any morphologic changes inmitochondria were generated in Y79 following exposure toBIRO1, cells were stained with the fluorescent dye MitotrackerRed CM-H2Xros, and mitochondrial morphology was exam-ined (Fig. 5A). Untreated cells exhibited elongated tubularmitochondria, whereas mitochondria of Y79 treated withBIRO1 displayed a conversion of tubular-fused mitochondriainto isolated small punctuated organelles after 1 hour only (Fig.5A), indicative of mitochondrial fragmentation.

Mitochondrial fragmentation can result either from anincrease in fission activity or an inhibition of fusion(33–36). To address whether BIRO1 induced fragmentationby recruiting the key fission protein DNM1L, we investigatedthe expression and localization of DNM1L in Y79. Undernormal conditions, the main fraction of DNM1L was cytosolic(Fig. 5B), with a weak portion localized at mitochondria.Following exposure to BIRO1, we were unable to detect anytranslocation of DNM1L to mitochondria or DNM1L dimer-ization (Fig. 5C). Furthermore, the treatment of Y79 cells withthe DNM1L inhibitor mdivi1, which attenuates DNM1L self-assembly, showed no protective effect following BIRO1 treat-ment (Fig. 5D). These results suggested that mitochondrialfission induced by BIRO1 in Y79 was DNM1L-independentand might implicate unidentified regulators.

Another possible pathway leading to mitochondrial fragmen-tation is the proteolytic processing and inactivation of OPA1.BIRO1 had no effect on OPA1 (Fig. 5E), indicating that BIRO1-induced mitochondrial fragmentation was not the side effect ofthe inhibition of mitochondrial membrane fusion.

The impact of BIRO1 on factors involved in mitochondriadynamics

While Y79 mitochondria fragmentation and cell membraneblowout were very early events following BIRO1 treatment (60–90 minutes; Figs. 5A and 2D), we measured a stable content invarious proteins, including proteins implicated in mitochondriadynamics and mitochondria-dependent cell death, that is, volt-age-dependent anion channel (VDAC1), hexokinase-2 (HK2),OPA1, and BCL2 proteins (Fig. 6A). A downregulation of BID,BBC2, andDNM1L as well as the cleavage ofMCL1were observed8 hours posttreatment. An increase in the cleaved form of phos-phoglycerate mutase family member 5 (PGAM5) isoform 1 wasobserved 8 hours posttreatment.

BIRO1 interacted with BCL2L1BIRO1 derived from the BH3 domain of BH3-only proteins

induced necrosis in Y79 and WERI-Rb which implicated mito-chondria. To determinewhichmechanismwas responsible for thehigh Y79 and WERI-Rb sensitivity to BIRO1, we investigated thepotential interaction between BIRO1 and antiapoptotic members






Figure 3.BIRO1-induced cell death is notapoptosis and is not related toautophagy activation. A, BIRO1-induced cell death is caspase-independent and does not rely onautophagy or necroptosis activation.Y79 andWERI-Rbwere treatedwith 10mmol/L BIRO1 for 24 hours. BeforeBIRO1 treatment, cells were exposedfor 1 hour to 10 mmol/L zVAD (a broad-range caspase inhibitor) as well as to0.5 mmol/L 3-MA (autophagyinhibitor), 50 mmol/L Nec-1 (RIPK1inhibitor), and 10 mmol/L calpastatinand ALLN (calpains inhibitors). Cellviability was determined bymeasuringATP content of the cells using aluminescence assay. Independentexperiments were repeated 4 times.Dark grey column, Y79; grey column,WERI-Rb. B, Y79 and WERI-Rb cellswere treatedwith 10 mmol/L BIRO1 andcaspase-3/7 activity was measuredusing ahomogeneous and luminescentassay. Independent experiments havebeen repeated 3 times. C, followingexposure of Y79 and WERI-Rb toBIRO1 and cisplatin, the cleavage ofPARP, a downstream substrate ofcaspase-3, was measured by Westernblotting experiment using an antibodythat recognize the full-length protein(105 kDa), as well as the cleavedprotein (85 kDa). Independentexperiments have been repeated 3times. D, Y79 and WERI-Rb wereexposed toBIRO1 for 3 hours, and LC3IIformation was assessed by Westernblotting during this period of time.






of the BCL2 proteins family. We detected an interaction ofBIRO1 with BCL2L1 in Y79 by pull-down experiments, whereasBIRO1 was unable to bind BCL2 and MCL1 (Fig. 6B). We alsoverified whether other factors playing a role in mitochondriadynamics were able to interact with BIRO1 (Fig. 6B). None ofthese factors were found to bind to BIRO1 in Y79. The mutatedform of BIRO1 was unable to interact with BCL2L1 (data notshown).

Our hypothesis was therefore that BIRO1/BCL2L1 interac-tion induced necrosis by releasing a deleterious factor fromBCL2L1. BCL2L1 is known to bind different factors and wefocused on the interaction of BCL2L1 with BID, BBC2, DNM1L,HK2, VDAC1, and PGAM5 in Y79 (Fig. 6C). BCL2L1 was foundto interact with BID, HK2, and VDAC1. Following treatment ofY79 with BIRO1, the BCL2L1/BID interaction was decreased,whereas the BCL2L1/HK2 and BCL2L1/VDAC1 interactionswere increased (Fig. 6D). A similar experiment was performedwith BIRO1 MT and no effect was observed (data not shown).

DiscussionSpecific modulations of various pathways, that is, PI3K/AKT,

JAK-STAT, p53, progressively lead to resistance of retinoblas-toma cells to death (2–5). The emerging small anticancer drugswhich target BCL2 family proteins offer a way to overcome thedeath resistance of cancer cells (13–21).

In this report, we investigated the minimal death domain ofdifferent BH3-only proteins for their capacity to induce celldeath in human retinoblastoma cell lines Y79 and WERI-Rb, aswell as in primary mouse retinoblastoma. Among the differentcell penetrating peptides tested, BIRO1 derived from the BH3domain of BCL2L11 was the most effective in killing both Y79and WERI-Rb, whereas it was nontoxic for the photoreceptors661W cell line. Among the other nonocular tumoral cell linesexposed to BIRO1, cell response was heterogeneous, with MDAbeing relative sensitive and all others being more resistant.Interestingly, primary mouse retinoblastoma was shown to besensitive to BIRO1. The high killing effect of BIRO1 comparedwith the other peptides suggests that not only BCL2 andBCL2L1must be inhibited but also other antiapoptotic proteinssuch as MCL1. Indeed, while the BH3 domain of BCL2L11 isable to interact and inhibit all the antiapoptotic members ofBCL2 family proteins, the BH3 domains of BAD and BIK inhibitonly BCL2, BCL2L1, BCL2L2, and MCL1, BCL2A1 respectively.

The high sensitivity of Y79 to BIRO1 despite the absence ofBAX questioned us about the molecular mechanism leading toY79 cell death. Indeed, BAX is known to be necessary to triggerapoptosis. Investigating markers of apoptosis after exposure toBIRO1 demonstrated that the apoptotic cell death was notinvolved in Y79 and weakly triggered in WERI-Rb. BIRO1 didnot activate caspase-3/7 in both Y79 and WERI-Rb, as observedby direct assays or indirectly by using caspase inhibitors orexamining caspase-3 substrate cleavage (PARP). BIRO1 did notinduce DNA fragmentation in Y79 whereas it did in WERI-Rb.This difference in one of the last step of the apoptotic processsuggests that a small part of the WERI-Rb cells is able to initiateapoptosis. We also examined the effect of BIRO1 on mitochon-dria. BIRO1 promoted mitochondria fragmentation very rap-idly (1 hour), suggesting an early mitochondrial function inBIRO1-induced cell death.

As BIRO1 did not kill retinoblastoma cells by apoptosis, weassessed the potential roles of necrosis and autophagy. Whileautophagy can participate to cell death, its activation may alsoestablish a cellular protection against certain stress, in partic-ular following deprivation of nutrients, and therefore againstcell death (37). The BCL2 protein family are implicated incontrolling autophagy, although autophagic mechanisms arequite different from apoptotic ones, and require distinct reg-ulator proteins (38–40), including Beclin-1. We thereforetested whether BIRO1 was able to modulate the interactionof Beclin-1 with BCL2 and/or BCL2L1 to induce autophagy.BIRO1 had no effect on the autophagic process in Y79 andWERI-Rb as observed by LC3II absence following peptidetreatment.

In necrosis, studies have shown that mitochondria mayconstitute a regulatory element. Necrosis has long been con-sidered as a passive process devoid of controlled signalingevents. However, evidences for a regulated necrosis mechanismare accumulating especially from studies of death receptorsleading to TNF-induced necrosis, called necroptosis. In

Figure 4.The cell death process induced by BIRO1 is not combined with ROSproduction in retinoblastoma cell lines. A, Y79 and WERI-Rb were treatedwith 10 mmol/L BIRO1 for 24 hours. Before BIRO1 treatment, cells wereexposed for 1 hour to 100 mmol/L BHA (antioxidant). Cell viability wasdetermined by measuring ATP content of the cells using a luminescenceassay. Independent experiments have been repeated 3 times. Dark greycolumn, Y79; grey column, WERI-Rb. B, WERI-Rb cells were treated with 10mmol/LBIRO1 during 1 hour and further stained for 30minuteswithH2DCFDA,a cell-permeable indicator for reactive oxygen species (ROS). Changes inoverall fluorescent intensity observed after exposure to BIRO1 was measuredand quantified using an alpha-screen apparatus. As a positive control, cellswere exposed to 1 mmol/L ABT-737 and 100 mmol/L H2O2, which caused arapid 50% and 100% increase of ROS, respectively. Independent experimentshave been repeated 3 times.






addition to the features of necrosis in BIRO1-treated Y79 andWERI-Rb, that is, absence of caspase activation and DNAfragmentation, other observations support the necrotic celldeath pathway. ATP content in both Y79 and WERI-Rb droppedrapidly and cell membranes were quickly permeabilized. How-ever, other events involved in necrotic cell death were not seenin BIRO1-induced cell death, such as calpain activation or ROSproduction. In necroptosis, the kinase RIPK1 disrupts theinteraction of the adenine nucleotide translocase (ANT) withcyclophilin D at the mitochondrial inner membrane, causing

mitochondrial dysfunction and cell death (41). We have exam-ined whether RIPK1 was involved in BIRO1 cell death usingNec-1, an allosteric inhibitor of RIPK1 (42) able to protect cellsfrom necroptosis. This inhibitor had no protective effect, sug-gesting therefore that BIRO1 induces a programmed necrosisdifferent from necroptosis.

BIRO1-induced programmed necrosis directly implicatedmitochondria. Indeed, the Y79 mitochondria, which displayedan elongated filamentous morphology under basal conditions,showed an altered fragmented shape and size following BIRO1

Figure 5.BIRO1 induces mitochondrial fragmentation in Y79. A, Y79 were left untreated or treated with 10 mmol/L BIRO1 for 1 to 3 hours. Mitochondria were stainedwith MitoTracker Red CM-H2Xros (200 nmol/L), and fluorescence microscopy was performed. B, mitochondrial and cytosolic fractions were isolated from Y79cells. DNM1L level in these 2 fractions was analyzed by Western blotting using corresponding antibodies. The mitochondrial fractionation was verified withthe mitochondrial marker VDAC1. C, Y79 cell extracts were separated by SDS-gel electrophoresis in the presence (denaturing conditions) or absence (nondenaturingconditions) of b-mercaptoethanol followed by Western blotting with anti-DNM1L antibody. Under nonreducing conditions, a DNM1L oligomer (�) was observedinaddition to themonomer form.D, Y79werepretreatedwith themitochondrialfission inhibitormdivi-1 for 60minutes followedby treatmentwithofBIRO1 (10mmol/L)for 16 hours. Cell viability was determined by measuring ATP content of the cells using a luminescence assay. Dark grey column, Y79; grey column, WERI-Rb.E, the content of OPA1 playing a role in mitochondrial fusion was determined by Western blot analysis in Y79 following BIRO1 treatment (10 mmol/L).






exposure. Mitochondrial fission is a process associated withapoptosis, as well as with programmed necrosis or mitophagy.DNM1L has been shown to be involved in mitochondrial fission(33, 36), its translocation from cytosol to mitochondria inducingmitochondria constriction. In Y79, we were unable to detect anyDNM1L translocation to mitochondria, suggesting that uniden-tified regulators of fission machinery might contribute to celldeath induced by BIRO1.

PGAM5 seems also to play an important role in mitochondrialdynamics at the interphase of apoptosis, necrosis, and autophagyas reported (43). The PGAM5 gene encodes 2 protein isoforms,PGAM5-L (32 kDa) andPGAM5-S (28 kDa), both expressed at thelevel of the outer mitochondrial membrane (OMM). The accu-mulation of the PGAM5-S results in the formation of fragmentedmitochondria (43). PGAM5 has been shown to be activated innecroptosis triggering DNM1L dephosphorylation and translo-cation to mitochondria leading to mitochondria fragmentation(44). In addition, upon mitochondrial membrane potential loss,PGAM5 is cleaved (45), being therefore a good marker for this

event. We observed that BIRO1 induced an accumulation of thecleaved PGAM-L isoform. As PGAM5 has been shown to interactwith BCL2L1 (46), we verified whether BIRO1 acted in Y79 bymodulating PGAM5/BCL2L1 interaction. This was not the case(Fig. 6D). We were then unable to determine whether BIRO1 hada direct effect on PGAM5 or whether PGAM5 cleavage was asecondary effect of BIRO1 on mitochondria. This point needsadditional clarification.

VDAC1 is another abundant protein of theOMMplaying a rolein mitochondria-mediated cell death (47, 48) when it sets upchannels in the OMM. The conductance throughout these chan-nels is modulated by various factors, including VDAC1/BCL2L1interaction, which allows the channel to be in an open state,maintaining adenine nucleotide flux and the outer membranepermeability. The disruption of the VDAC1/BCL2L1 interactiontriggers mitochondrial permeability modulation and cell death(49, 50). InY79, BIRO1 seemed to strengthen theVDAC1/BCL2L1binding (Fig. 6D), which is in discrepancy with the protectiveeffect of the VDAC1/BCL2L1 interaction.

Figure 6.Effect of BIRO1 on proteins involved inmitochondria dynamics. A, thecontent of factors playing a role at themitochondrial level was determinedby Western blot analysis in Y79following BIRO1 treatment (10mmol/L). B, the interaction of theGST/BIRO1 protein with potentialpartners was determined by pull-down experiments using Y79 proteinsextracts. The GST alone was used as anegative control. PD, pull-down; SN,supernatant; WE, whole extracts.C, the proteins able to interact withBCL2L1 in Y79 cells were estimated byco-immunoprecipitation using anantibody against BCL2L1. IP,immunoprecipitated; SN, supernatant;WE, whole extracts. D, the modulationof the interaction of BCL2L1 with itspartners following BIRO1 exposure (10mmol/L, 4 hours)was estimated by co-immunoprecipitation in Y79 andcompared with untreated cells (C). IP,immunoprecipitated; WE, wholeextracts.






Our study demonstrated that BH3 peptides enhanced celldeath in retinoblastoma cells. Our finding is supported by arecent report (21) showing that retinoblastoma cells are sen-sitive to another BH3 mimetics, ABT-737. Further investiga-tions are however necessary to clarify the exact mechanismsinvolved in the cell death process induced by BIRO1. From ourstudy, it appears that an unknown programmed necrotic mech-anism is implicated.

Disclosure of Potential Conflicts of InterestN. Allaman-Pillet and D.F. Schorderet have ownership interests (including

patents) as EPFL and IRO are seeking a patent for the use of BIRO1. No potentialconflicts of interest were disclosed by the other author.

Authors' ContributionsConception and design: N. Allaman-Pillet, D.F. SchorderetDevelopment of methodology: N. Allaman-Pillet, D.F. SchorderetAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): N. Allaman-Pillet, A. Oberson, D.F. Schorderet

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): N. Allaman-Pillet, D.F. SchorderetWriting, review, and/or revision of the manuscript: N. Allaman-Pillet, D.F.SchorderetAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): N. Allaman-Pillet, A. Oberson, D.F. SchorderetStudy supervision: N. Allaman-Pillet, D.F. Schorderet

AcknowledgmentsThe authors thank Carole Herkenne, Ang�elique Schmid, and C�eline

Agosti for precious technical assistance and Dr. M. Al-Ubaidi for providingthe 661W cell line. The SV40-LT transgenic mice are a gift of J.M. O'Brien.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received May 8, 2014; revised July 14, 2014; accepted August 2, 2014;published OnlineFirst August 20, 2014.

References1. KnudsonAG.Mutation and cancer: statistical studyof retinoblastoma. Proc

Natl Acad Sci U S A 1971;68:820–3.2. Corson TW,Gallie BL.Onehit, twohits, three hits,more?Genomic changes

in the development of retinoblastoma. Genes Chomosomes Cancer2007;46:617–34.

3. Sampieri K, Mencarelli MA, Carmela EM, Toti P, Lazzi S, Bruttini M, et al.Genomic differences between retinoma and retinoblastoma. Acta Oncol2008;47:1483–92.

4. Dimaras H, Gallie BL. The p75 NTR neurotrophin receptor is a tumorsuppressor in human and murine retinoblastoma development. Int JCancer 2008;122:2023–9.

5. Dimaras H, Khetan V, Halliday W, Orlic M, Prigoda NL, Piovesan B, et al.Loss of RB1 induces non-proliferative retinoma: increasing genomic insta-bility correlates with progression to retinoblastoma. Hum Mol Genet2008;17:1363–72.

6. Chantada G, Fadino A, Manzitti J. Late diagnosis of retinoblastomas in adeveloping country. Arch Dis Child 1999;80:171–4.

7. Chen L,Willis SN,Wei A, Smith BJ, Fletcher JI, HindsMG, et al. Differentialtargeting of prosurvival Bcl-2 proteins by their BH3-only ligands allowscomplementary apoptotic function. Mol Cell 2005;17:393–403.

8. Adams JM, Cory S. Life-or-death decisions by the Bcl-2 protein family.Trends Biochem Sci 2001;26:61–6.

9. Youle RJ, Strasser A. The BCL-2 protein family: opposing activities thatmediate cell death. Nat Rev Mol Cell Biol 2008;9:47–59.

10. Moldoveanu T, Liu Q, Tocilj A, Watson M, Shore G, Gehring K. The X-raystructure of a BAK homodimer reveals an inhibitory zinc binding site. MolCell 2006;24:677–88.

11. Willis SN, Adams JM. Life in the balance: how BH3-only proteins induceapoptosis. Curr Opin Cell Biol 2005;17:617–25.

12. Zong WX, Lindsten T, Ross AJ, MacGregor GR, Thompson CB. BH3-only proteins that bind pro-survival Bcl-2 family members fail toinduce apoptosis in the absence of Bax and Bak. Genes Dev 2001;15:1481–6.

13. Moreau C, Cartron PF, Hunt A,Meflah K, GreenDR, Evan G, et al. MinimalBH3 peptides promote cell death by antagonizing anti-apoptotic proteins.J Biol Chem 2003;278:19426–35.

14. Shangary S, Oliver CL, Tillman TS, Cascio M, Johnson DE. Sequence andhelicity requirements for the proapoptotic activity of Bax BH3 peptides.Mol Cancer Ther 2004;3:1343–54.

15. Oltersdorf T, Elmore SW, Shoemaker AR, Armstrong RC, Augeri DJ, BelliBA, et al. An inhibitor of Bcl-2 family proteins induces regression of solidtumours. Nature 2005;435:677–81.

16. Cory S, Adams JM. Killing cancer cells by flipping the Bcl-2/Bax switch.Cancer Cell 2005;8:5–6.

17. Reed JC, Pellecchia M. Apoptosis-based therapies for hematologic malig-nancies. Blood 2005;106:408–18.

18. WalenskyLD,PitterK,Morash J,OhKJ,BarbutoS, Fisher J, et al. A stapledBIDBH3 helix directly binds and activates BAX. Mol Cell 2006;24:199–210.

19. Certo M, Del Gaizo Moore V, Nishino M, Wei G, Korsmeyer S, ArmstrongSA, et al. Mitochondria primed by death signals determine cellular addic-tion to antiapoptotic BCL-2 family members. Cancer Cell 2006;9:351–65.

20. Chonghaile TN, Letai A. Mimicking the BH3 domain to kill cancer cells.Oncogene 2008;27 Suppl 1:S149–57.

21. Allaman-Pillet N, Oberson A, Munier F, Schorderet DF. The Bcl-2/Bcl-XLinhibitor ABT-737 promotes death of retinoblastoma cancer cells. Oph-thalmic Genet 2013;34:1–13.

22. Letai A, Bassik MC, Walensky LD, Sorcinelli MD, Weiler S, Korsmeyer SJ.Distinct BH3 domains either sensitize or activatemitochondrial apoptosis,serving as prototype cancer therapeutics. Cancer Cell 2002;2:183–92.

23. Uren RT, Dewson G, Chen L, Coyne SC, Huang DC, Adams JM, et al.Mitochondrial permeabilization relies on BH3 ligands engaging mul-tiple prosurvival Bcl-2 relatives, not Bak. J Cell Biol 2007;177:277–87.

24. Windle JJ, Albert DM, O'Brien JM, Marcus DM, Disteche CM, BernardsR, et al. Retinoblastoma in transgenic mice. Nature 1990;343:665–9.

25. BonnyC,ObersonA,Negri S, SauserC,SchorderetDF.Cell-permeablepeptideinhibitors of JNK: novel blockers of beta-cell death. Diabetes 2001;50:77–82.

26. Schorderet DF,Manzi V, Canola K, Bonny C, Arsenijevic Y,Munier FL, et al.D-TAT transporter as an ocular peptide delivery system. Clin ExperimentOphthalmol 2005;33:628–35.

27. Kabeya Y, Mizushima N, Ueno T, Yamamoto A, Kirisako T, Noda T, et al.LC3, a mammalian homologue of yeast Apg8p, is localized in autophago-some membranes after processing. EMBO J 2000;19:5720–8.

28. Kabeya Y, Mizushima N, Yamamoto A, Oshitani-Okamoto S, Ohsumi Y,Yoshimori T. LC3, GABARAP and GATE16 localize to autophagosomalmembrane depending on form-II formation. J Cell Sci 2004;117:2805–12.

29. Yamashima T, Tonchev AB, Tsukada T, Saido TC, Imajoh-Ohmi S, MomoiT, et al. Sustained calpain activation associated with lysosomal ruptureexecutes necrosis of the postischemic CA1 neurons in primates. Hippo-campus 2003;13:791–800.

30. Syntichaki P, Tavernarakis N. Death by necrosis. Uncontrollable catastro-phe, or is there order behind the chaos? EMBO Rep 2002;3:604–9.

31. Okamoto K, Shaw JM. Mitochondrial morphology and dynamics inyeast and multicellular eukaryotes. Annu Rev Genet 2005;39:503–536.

32. Shroff EH, Snyder CM, Budinger GR, Jain M, Chew TL, Khuon S, et al. BH3peptides inducemitochondrial fission and cell death independent of BAX/BAK. PLoS One 2009;4:e5646.

33. James DI, Parone PA, Mattenberger Y, Martinou JC. hFis1, a novel com-ponent of the mammalian mitochondrial fission machinery. J Biol Chem2003;278:36373–9.

34. Schauss AC, Bewersdorf J, Jakobs S. Fis1p and Caf4p, but not Mdv1p,determine the polar localization of Dnm1p clusters on the mitochondrialsurface. J Cell Sci 2006;119:3098–3106.






35. Griparic L, Kanazawa T, van der Bliek AM. Regulation of themitochondrialdynamin-like protein Opa1 by proteolytic cleavage. J Cell Biol 2007;178:757–64

36. Detmer SA, Chan DC. Functions and dysfunctions of mitochondrialdynamics. Nat Rev Mol Cell Biol 2007;8:870–9.

37. Levine B, Kroemer G. Autophagy in the pathogenesis of disease. Cell2008;132:27–42.

38. Maiuri MC, Le Toumelin G, Criollo A, Rain JC, Gautier F, Juin P, et al.Functional and physical interaction between Bcl-X(L) and a BH3-likedomain in Beclin-1. EMBO J 2007;26:2527–39.

39. Maiuri MC, Zalckvar E, Kimchi A, Kroemer G. Self-eating and self-killing:crosstalk between autophagy and apoptosis. Nat Rev Mol Cell Biol 2007;8:741–52.

40. Rashmi R, Pillai SG, Vijayalingam S, Ryerse J, Chinnadurai G. BH3-onlyprotein BIK induces caspase-independent cell death with autophagic fea-tures in Bcl-2 null cells. Oncogene 2008;27:1366–75.

41. Temkin V, Huang Q, Liu H, Osada H, Pope RM. Inhibition of ADP/ATPexchange in receptor-interacting protein-mediated necrosis. Mol Cell Biol2006;26:2215–25.

42. Degterev A, Hitomi J, Germscheid M, Ch'en IL, Korkina O, Teng X, et al.Identification of RIP1 kinase as a specific cellular target of necrostatins. NatChem Biol 2008;4:313–21.

43. Lo SC, Hannink M. PGAM5 tethers a ternary complex containing Keap1and Nrf2 to mitochondria. Exp Cell Res 2008;314:1789–803.

44. Wang Z, Jiang H, Chen S, Du F, Wang X. The mitochondrial phosphatasePGAM5 functions at the convergence point of multiple necrotic deathpathways. Cell 2012;148:1–2.

45. Sekine S, Kanamaru Y, Koike M, Nishihara A, Okada M, Kinoshita H, et al.Rhomboid protease PARL mediates the mitochondrial membrane poten-tial loss-induced cleavage of PGAM5. J Biol Chem 2012;287:34635–45.

46. Lo SC, Hannink M. PGAM5, a Bcl-XL-interacting protein, is a novelsubstrate for the redox-regulated Keap1-dependent ubiquitin ligase com-plex. J Biol Chem 2006;281:37893–903.

47. Shimizu S,Matsuoka Y, Shinohara Y, Yoneda Y, Tsujimoto Y. Essential roleof voltage dependent anion channel in various forms of apoptosis inmammalian cells. J Cell Biol 2001;152:237–50.

48. McCommis KS, Baines CP. The role of VDAC in cell death: friend or foe?Biochim Biophys Acta 2012;1818:1444–50.

49. Arbel N, Shoshan-Barmatz V. Voltage-dependent anion channel 1-basedpeptides interact with Bcl-2 to prevent antiapoptotic activity. J Biol Chem2010;285:6053–62.

50. Arbel N, Ben-Hail D, Shoshan-Barmatz V. Mediation of the antiapoptoticactivity of Bcl-xL protein upon interaction with VDAC1 protein. J BiolChem 2012;287:23152–61.






2015;13:86-97. Published OnlineFirst August 20, 2014.Mol Cancer Res Nathalie Allaman-Pillet, Anne Oberson and Daniel F. Schorderet Fragmentation and Death of Retinoblastoma CellsBIRO1, a Cell-Permeable BH3 Peptide, Promotes Mitochondrial

Updated version

10.1158/1541-7786.MCR-14-0253doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/13/1/86.full#ref-list-1

This article cites 50 articles, 21 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/1/86.full#related-urls

This article has been cited by 1 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://mcr.aacrjournals.org/content/13/1/86To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-14-0253

http://mcr.aacrjournals.org/content/13/1/86.full#ref-list-1

http://mcr.aacrjournals.org/content/13/1/86.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/1/86


biro1, a cell-permeable bh3 peptide, promotes ... · bh3 peptide–based approach uses the prodeath...

Documents