J. Biol. Chem.-2005-Lin-23758-65

Bcl-2 Rescues Ceramide- and Etoposide-induced MitochondrialApoptosis through Blockage of Caspase-2 Activation*S

Received for publication, October 29, 2004, and in revised form, February 17, 2005Published, JBC Papers in Press, April 6, 2005, DOI 10.1074/jbc.M412292200

Chiou-Feng Lin, Chia-Ling Chen, Wen-Tsan Chang, Ming-Shiou Jan**, Li-Jin Hsu,Ren-Huang Wu, Yi-Ting Fang, Ming-Jer Tang, Wen-Chang Chang, and Yee-Shin Lin

From the Departments of Microbiology and Immunology, Biochemistry, Physiology, and Pharmacology and theInstitute of Basic Medical Sciences, National Cheng Kung University Medical College, Tainan 701, Taiwan and the**Department of Microbiology and Immunology, Chung-Shan Medical University, Taichung 402, Taiwan

Recent studies indicate that caspase-2 is involved inthe early stage of apoptosis before mitochondrial dam-age. Although the activation of caspase-2 has beenshown to occur in a large protein complex, the mecha-nisms of caspase-2 activation remain unclear. Here wereport a regulatory role of Bcl-2 on caspase-2 upstreamof mitochondria. Stress stimuli, including ceramide andetoposide, caused caspase-2 activation, mitochondrialdamage followed by downstream caspase-9 and -3 acti-vation, and cell apoptosis in human lung epithelial cellline A549. When A549 cells were pretreated with thecaspase-2 inhibitor benzyloxycarbonyl-Val-Asp(-OMe)-Val-Ala-Asp(-OMe)-fluoromethyl ketone or transfectedwith caspase-2 short interfering RNA, both ceramide-and etoposide-induced mitochondrial damage and apo-ptosis were blocked. Overexpression of Bcl-2 preventedceramide- and etoposide-induced caspase-2 activationand mitochondrial apoptosis. Furthermore, caspase-2was activated when A549 cells were introduced withBcl-2 short interfering RNA or were treated with Bcl-2inhibitor, which provided direct evidence of a negativeregulatory effect of Bcl-2 on caspase-2. Cell survival wasobserved when caspase-2 was inhibited in Bcl-2-silenc-ing cells. Blockage of the mitochondrial permeabilitytransition pore and caspase-9 demonstrated that Bcl-2-modulated caspase-2 activity occurred upstream of mi-tochondria. Further studies showed that Bcl-2 was de-phosphorylated at serine 70 after ceramide andetoposide treatment. A protein phosphatase inhibitor,okadaic acid, rescued Bcl-2 dephosphorylation andblocked caspase-2 activation, mitochondrial damage,and cell death. Taken together, ceramide and etoposideinduced mitochondria-mediated apoptosis by initiatingcaspase-2 activation, which was, at least in part, regu-lated by Bcl-2.

During the process of apoptosis, there is, in general, a reduc-tion of mitochondrial transmembrane potential (m) followedby the release of cytochrome c, which binds to Apaf-1 and

promotes caspase-9 and -3 activation (13). Bcl-2 family pro-teins serve as critical regulators of mitochondrial apoptosis,functioning as either inhibitors or promoters of cell death (4).Bcl-2 inhibits apoptosis by blocking cytochrome c release frommitochondria (5) through prevention of channel formation,which is mediated by proapoptotic Bax and Bid (68). A recentstudy (9) indicated that, in healthy cells, Bcl-2 adopts a typicaltail-anchored topology. Induction of apoptosis by ceramide oretoposide triggered a change of Bcl-2 to the multispanningtransmembrane topology. In addition to membrane topology,Bcl-2 phosphorylation is required for its full anti-apoptoticfunction (10, 11).Caspase-2 acts upstream of mitochondria to promote cyto-

chrome c release and apoptosis (1218), although caspase-2may also act downstream of mitochondria (19, 20). One study inthe mechanisms of caspase-2 activation showed that caspase-2complex formation occurs independently of an Apaf-1/apopto-some pathway and that the recruitment of caspase-2 into thiscomplex is sufficient to mediate its activation upstream ofmitochondria (21). A recent report (22) demonstrated that ac-tivation of caspase-2 occurs in the complex that contains thep53-induced death-domain-containing protein and the adapterprotein RAIDD (RIP (ribosome-inactivating protein)-associatedICH-1/CED-homologous protein with death domain). Increasedp53-induced death-domain-containing protein may result incaspase-2 activation to regulate apoptosis induced by genotoxicstress. Interestingly, Bcl-2 suppresses p53-dependent apopto-sis that requires Bax and caspase-2 as essential apoptoticmediators (23).We previously demonstrated sequential caspase-2 and -8

activation upstream of mitochondria during ceramide- and eto-poside-induced apoptosis (24). In the present study, the rela-tionship between Bcl-2 and caspase-2 in mitochondrial apopto-sis induced by ceramide and etoposide was investigated. Bcl-2overexpression rescued ceramide- and etoposide-induced apo-ptosis (2532). Furthermore, ceramide caused Bcl-2 dysfunc-tion through its dephosphorylation at serine 70 mediated byprotein phosphatase 2A (33). Chemotherapeutic etoposidecaused Bcl-2 cleavage, which led to cell apoptosis (34). In ad-dition, protein phosphatase facilitated cells undergoing etopo-side-induced apoptosis (35, 36). Using Bcl-2 short interferingRNA or Bcl-2 inhibitor ethyl 2-amino-6-bromo-4-(1-cyano-2-ethoxy-2-oxoethyl)-4H-chromene-3-carboxylate (HA14-1),1 we

* This work was supported by Grant 91-B-FA09-1-4 from the Minis-try of Education (MOE) Program for Promoting Academic Excellence ofUniversity (Taiwan). The costs of publication of this article were de-frayed in part by the payment of page charges. This article musttherefore be hereby marked advertisement in accordance with 18U.S.C. Section 1734 solely to indicate this fact.S The on-line version of this article (available at http://www.jbc.org)

contains supplemental Figs. S1S3. A postdoctoral fellow supported by the National Health Research

Institutes, Taiwan, ROC (PD9403). To whom correspondence should be addressed. Tel.: 886-6-235-3535

(ext. 5646); Fax: 886-6-208-2705; E-mail: [email protected].

1 The abbreviations used are: HA14-1, ethyl 2-amino-6-bromo-4-(1-cyano-2-ethoxy-2-oxoethyl)-4H-chromene-3-carboxylate; MDCK,Madin-Darby canine kidney; z, benzyloxycarbonyl; fmk, fluoromethylketone; PI, propidium iodide; GFP, green fluorescent protein; EGFP,enhanced GFP; siRNA, short interfering RNA; OA, okadaic acid; DAPI,4,6-diamidino-2-phenylindole; OD, optical density.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 25, Issue of June 24, pp. 2375823765, 2005 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org23758

by guest on June 21, 2015http://w

ww

.jbc.org/D

ownloaded from

showed directly that Bcl-2 negatively regulated caspase-2.Upon ceramide and etoposide stimulation, protein phospha-tase-mediated Bcl-2 dephosphorylation led to activation ofcaspase-2, mitochondrial damage, and apoptosis.

EXPERIMENTAL PROCEDURES

Cell Cultures and ReagentsThe human lung epithelial cell lineA549, Madin-Darby canine kidney (MDCK) cells, and their bcl-2 trans-fectants (A549-B2 and MDCK-B6) and vector controls (A549-P2 andMDCK-C1) were provided by Dr. M. T. Lin, Department of Biochemis-try, and Dr. M. J. Tang, Department of Physiology, National ChengKung University, Taiwan. A549 and MDCK cells were cultured inDulbeccos modified Eagles medium supplemented with 10% fetal bo-vine serum, 50 units/ml penicillin, and 0.05 mg/ml streptomycin. Theywere maintained at 37 C in 5% CO2. Human prostate cancer DU145cells were cultured in minimum essential medium supplemented with10% fetal bovine serum. A549-B2, A549-P2, MDCK-B6, and MDCK-C1cells were constructed using lipofection with a transfection reagent(Lipofectamine 2000, Invitrogen) and then cultured in Dulbeccos mod-ified Eagles medium supplemented with 10% fetal bovine serum and500 g/ml G418 (Calbiochem). Bcl-2 cDNA constructed in expressionvector, pUSEamp, was purchased from Upstate Biotechnology. Cera-mide analogue C2-ceramide (BioMol) and etoposide (BioVision) weredissolved in Me2SO. The broad-spectrum caspase inhibitor benzyloxy-carbonyl-Val-Ala-Asp(-OMe)-fluoromethyl ketone (z-VAD-fmk) andcaspase-9, -3, and -2 inhibitors benzyloxycarbonyl-Leu-Glu(-OMe)-His-Asp(-OMe)-fluoromethyl ketone (z-LEHD-fmk), benzyloxycarbonyl-Asp(-OMe)-Glu(-OMe)-Val-Asp(-OMe)-fluoromethyl ketone (z-DEVD-fmk), and benzyloxycarbonyl-Val-Asp(-OMe)-Val-Ala-Asp(-OMe)-fluoromethyl ketone (z-VDVAD-fmk), respectively, were purchasedfrom Sigma and dissolved in Me2SO. Bcl-2 inhibitor HA14-1 (Tocris),bongkrekic acid (Calbiochem), and okadaic acid (OA, Sigma) were dis-solved in Me2SO.

Analysis of Cell ApoptosisFor apoptosis detection characterized byDNA fragmentation, cells were fixed with 70% ethanol in phosphate-buffered saline for propidium iodide (PI, Sigma) staining and then wereanalyzed using flow cytometry (FACSCalibur; BD Biosciences). DAPI(Sigma) was also used for apoptotic cell staining in 5 g/ml for 30 minat room temperature and was followed by microscopy detection. Apo-

ptotic cell membrane disruption characterized by the presence of phos-phatidylserine was performed using the annexin V-phycoerythrin de-tection kit (BioVision).

Detection of Caspase ActivationCellular caspase activation wasdetermined using the ApoAlert caspase colorimetric assay kit (Clon-tech) for caspase-3 and an ApoAlert caspase fluorescent assay kit forcaspase-9 according to the manufacturers instructions. Caspase-2 ac-tivity was detected using a caspase-2 assay kit (Calbiochem). Opticaldensity (OD) measurements were performed using a microplate reader,and the substrate activities shown as p-nitroanilide (nmol) were calcu-lated for caspase-3 and -9. For caspase-2, the relative substrate activitywas shown by the OD values. Caspase-3 activation monitored in cellswas performed using PhiPhiLux-G2D2 staining (OncoImmunol) anddetected with fluorescent microscopy. The activation of caspases wasalso detected using Western blot analysis as described below.

Western Blot AnalysisTo detect cytochrome c release, cytosolic ex-tract without the mitochondrial fraction was separated using an ApoA-lert cell fractionation kit (Clontech) according to the manufacturersinstructions. To detect other proteins, total cell lysate was used. West-ern blotting was then performed (BD Biosciences). Briefly, cell extractwas separated by SDS-PAGE and then transferred to a polyvinylidenedifluoride membrane (Millipore). After blocking, blots were developedwith a series of antibodies as indicated. Rabbit antibodies specific forhuman caspase-9 and -3 (Cell Signaling Technology), cytochrome c(Santa Cruz Biotechnology), green fluorescent protein (GFP) (SantaCruz Biotechnology), Bcl-2, and phospho-Bcl-2 serine 70 (R&D) wereused. Monoclonal antibodies against human caspase-2 (R&D) and -ac-tin (Sigma) were used. Finally, blots were hybridized with horseradishperoxidase-conjugated goat anti-rabbit IgG or anti-mouse IgG (Calbio-chem) and developed using an AEC substrate kit (Zymed LaboratoriesInc.).

Mitochondrial Functional AssayThe loss of mitochondrial trans-membrane potential (m) value was determined using rhodamine 123(Sigma). Cells were incubated with rhodamine 123 (50 M) for 30 min incultured medium. After being washed with phosphate-buffered saline,cells were resuspended in cold phosphate-buffered saline and immedi-ately underwent flow cytometric analysis. Mitochondrial dehydrogen-ase activity was determined using a WST-8 assay kit (Dojindo Labora-tories, Kumamoto, Japan).

FIG. 1. Ceramide and etoposide induce mitochondria-mediated apoptosis in A549 cells. A, human A549 epithelial cells were treatedwith or without 50 M C2-ceramide or 25 M etoposide for 48 h. The changes in cell morphology (characterized by cell rounding up) and in nuclearmorphology (characterized by nuclear fragmentation) are shown by phase-contrast microscopic observation and DAPI staining, respectively. B,time- and dose-dependent (as indicated) ceramide- or etoposide-induced A549 cell apoptosis was detected using PI staining followed by flowcytometric analysis. The percentages of apoptotic cells are shown (means S.D. of triplicate cultures). C, using rhodamine 123 followed by flowcytometric analysis, the percentages of ceramide- or etoposide-treated cells with m (MTP) reduction at different times are shown (means S.D.of triplicate cultures). D, the activation of caspase-9 and -3 induced by ceramide or etoposide time dependently as determined by caspase activityassay kits are shown (means S.D. of triplicate cultures). pNA, p-nitroanilide. E, with or without caspase inhibitors as indicated, ceramide- oretoposide-induced cell apoptosis at 48 h was detected using PI staining. The percentages of apoptotic cells are shown (means S.D. of triplicatecultures).

Bcl-2 Modulates Caspase-2 Activation 23759


ww

.jbc.org/D

ownloaded from

ImmunostainingFor intracellular immunostaining, cells werefixed with 1% formaldehyde in phosphate-buffered saline and perme-abilized with 0.01% saponin in phosphate-buffered saline. A series ofantibodies was used as indicated, followed by fluorescein isothiocya-nate-conjugated goat anti-rabbit IgG (Calbiochem) staining. Rabbitanti-Bcl-2 and anti-phospho-Bcl-2 serine 70 (R&D) antibodies wereused for flow cytometric analysis. For confocal microscopy, rabbitanti-truncated Bid (tBid) (Calbiochem) antibodies were used. MitoTracker Red CMXRos (Molecular Probes) was used for mitochondrialstaining.

Short Interfering RNA (siRNA) PreparationPlasmids expressingshort hairpin RNA were constructed using standard techniques. ThepSUPER/enhanced GFP (EGFP) contained the GFP gene frompEGFP-N1 (Clontech) inserted into pSUPER vector (37) (kindly pro-vided by Dr. R. Agami, The Netherlands Cancer Institute, Amsterdam,The Netherlands). To generate pSUPER-Casp2/EGFP and pSUPER-Bcl2/EGFP, pSUPER/EGFP was digested with BglII and HindIII, andthe annealed targeting oligonucleotides ACAGCTGTTGTTGAGCGAAfor caspase-2 (15) and GCTGCACCTGACGCCCTTC for Bcl-2 (23) were

ligated into the vector. To generate the double knockdown constructpSUPER-Bcl2/Casp2/EGFP, the Casp2 short hairpin RNA expressioncassette from pSUPER-Casp2/EGFP was inserted into pSUPER-Bcl2/EGFP. The pSUPER-Casp2/EGFP/Neo, pSUPER-Bcl2/EGFP/Neo, andpSUPER-Bcl2/Casp2/EGFP/Neo were generated by inserting the Neor

gene from the pIRESneo2 (Clontech) into the pSUPER-Casp2/EGFP,pSUPER-Bcl2/EGFP, and pSUPER-Bcl2/Casp2/EGFP vectors, respec-tively (supplemental Fig. S1).

A549 and DU145 cells were cultured in 6-well plastic plates inDulbeccos modified Eagles medium and minimum essential medium,respectively, supplemented with 10% fetal bovine serum (5 105/well). Before short hairpin RNA-expression vector transfection, cellswere washed with serum-free medium and cultured with 2 l ofLipofectamine 2000 and 1 g of DNA. After 6 h of incubation, cellswere maintained in cultured medium containing 10% fetal bovineserum for an additional 24 h before experiments. A FACSAria cellsorter (BD Biosciences) was used to sort EGFP-positive cells forvector control and siRNA-expressing cells in some experiments asindicated.

FIG. 2. Requirement of caspase-2 activation during ceramide- or etoposide-induced mitochondrial apoptosis. A, A549 cells weretreated with 50 M C2-ceramide or 25 M etoposide for various time periods followed by a caspase-2 activity assay; the substrate relative activities(OD) are shown (top, means S.D. of triplicate cultures). Also, ceramide (C)- or etoposide (E)-induced caspase-2 processing was determined usingWestern blot analysis (bottom). Protein expression of -actin was used as an internal control. B, ceramide or etoposide induced caspase-2-mediatedmitochondrial disruption. With or without caspase-2 inhibitor z-VDVAD-fmk or broad spectrum caspase inhibitor z-VAD-fmk, m (MTP)reduction and cell apoptosis induced by 50 M C2-ceramide or 25 M etoposide for 48 h were determined using rhodamine 123 and PI staining,respectively, followed by flow cytometric analysis (means S.D. of triplicate cultures). The activities of caspase-9 and -3 (means S.D. of triplicatecultures) were determined using caspase activity assay kits (top). Ceramide- or etoposide-induced cytosolic cytochrome c expression with or withoutcaspase inhibitors was determined using Western blot analysis (bottom). Protein expression of -actin was used as an internal control. pNA,p-nitroanilide. C, blockage of ceramide- or etoposide-induced cell apoptosis by caspase-2 siRNA. A549 cells were transfected with siRNA tocaspase-2 as described under Experimental Procedures. The expression of procaspase-2 (45 kDa) in vector-transfected cells (Vector) and caspase-2siRNA-transfected cells (Casp-2 siRNA) was detected using Western blot analysis (top). Protein expression of EGFP was used as an internalcontrol. The changes in morphology of transfected cells with 50 M C2-ceramide or 25 M etoposide treatment for 48 h were determined usingfluorescent microscopy (EGFP) plus phase-contrast microscopy (Merge). The apoptotic cells are shown in transfected cells (filled arrowheads) oruntransfected cells (open arrowheads). D, ceramide- or etoposide-treated transfected cells were stained with annexin V and caspase-3-specificfluorescence substrate PhiPhiLux-G2D2 as described under Experimental Procedures. The percentages of annexin V-positive and active-caspase-3-positive cells are shown (means S.D. of triplicate cultures).

Bcl-2 Modulates Caspase-2 Activation23760


ww

.jbc.org/D

ownloaded from

RESULTS AND DISCUSSION

Ceramide and Etoposide Induce Mitochondria-mediated Apo-ptosis in A549 CellsApoptotic cell death induced by ceramideand etoposide has been widely reported (2832, 38, 39). Toinvestigate the involvement of mitochondrial dysfunction andcaspase activation, C2-ceramide and etoposide were used toinduce apoptosis in A549 human lung epithelial cells. First, cellgrowth inhibition after ceramide and etoposide treatment wasdemonstrated using a WST-8 assay to detect mitochondrialdehydrogenase activity (data not shown). A549 cells that hadbeen exposed to ceramide and etoposide for 48 h were fixed andstained with DAPI. In ceramide- and etoposide-treated cul-tures, but not in untreated cultures, cells exhibited apoptoticmorphology (Fig. 1A). The dose and time kinetics of ceramide-and etoposide-induced cell apoptosis were shown using PIstaining and then flow cytometric analysis (Fig. 1B). To furtherinvestigate the involvement of mitochondrial damage, the lossof mitochondrial transmembrane potential (m) was deter-mined. Using lipophilic cationic fluorochrome rhodamine 123staining, we found that ceramide and etoposide time depend-ently induced m reduction in A549 cells (Fig. 1C). Using acaspase activity assay andWestern blot analysis, the activity ofcaspase-9 and -3 (Fig. 1D) and the processing of caspase-3 fromproform (35 kDa) to active form (17 kDa) (data not shown) wereobserved time dependently after exposure to ceramide andetoposide. An increase in cytosolic cytochrome c expression wasalso observed (data not shown). Furthermore, pretreatmentwith the irreversible caspase-9 and -3 inhibitors z-LEHD-fmkand z-DEVD-fmk, respectively, blocked cell death detected us-ing PI staining (Fig. 1E). We therefore ascertained the involve-ment of mitochondrial damage, cytochrome c release, and

caspase-9 and -3 activation in ceramide- and etoposide-inducedA549 cell apoptosis.Requirement of Caspase-2 Activation before Mitochondrial

DamageOur previous study (24) demonstrated that activa-tion of caspase-2 was required for ceramide- and etoposide-induced T cell apoptosis before mitochondrial damage. We thusconfirmed the involvement of caspase-2 in response to ceramideand etoposide stimulation in A549 cells. Time-dependent acti-vation of caspase-2 was shown by the activity assay (Fig. 2A,top). The caspase-2 cleavage was also detected using Westernblot analysis (Fig. 2A, bottom). To examine whether caspase-2was required for the mitochondrial intrinsic pathway of apo-ptosis, we inactivated caspase-2 in A549 cells by pretreatmentwith the inhibitor z-VDVAD-fmk. The results showed the inhi-bition of cell apoptosis, m reduction, and caspase-9 and -3activation after z-VDVAD-fmk pretreatment in response toceramide and etoposide stimulation (Fig. 2B, top). Accordingly,cytosolic cytochrome c (Fig. 2B, bottom) and cleavage fragmentsof active caspase-9 and -3 (data not shown) were shown in theceramide- and etoposide-treated cultures, but not in the cul-tures pretreated with z-VDVAD-fmk. The broad-spectrumcaspase inhibitor z-VAD-fmk caused an effect similar to that ofz-VDVAD-fmk (Fig. 2B). We further introduced the short hair-pin RNA specific for caspase-2 into A549 cells for interferencewith caspase-2 expression. The mean transfection efficiency ofcaspase-2 siRNA was 52.3% and of vector control was 60.8%according to a flow cytometric analysis of EGFP-positive cellsin a representative experiment. An 50% inhibition ofcaspase-2 expression was observed in the caspase-2 siRNA-transfected cells compared with the vector control (Fig. 2C,top). After ceramide or etoposide treatment, apoptotic cells

FIG. 3. Overexpression of Bcl-2 inactivates caspase-2, which rescues ceramide- or etoposide-induced apoptosis. A, A549 cells withBcl-2 overexpression were established as described under Experimental Procedures. The protein expression of Bcl-2 in wild-type (WT),vector-transfected (P2), and Bcl-2-transfected (B2) was detected using Western blot analysis (top) and immunostaining followed by flow cytometricanalysis (bottom). The relative mean fluorescence intensity (MFI) is shown. For Western blotting, protein expression of -actin was used as aninternal control. B, cell apoptosis induced by 50 M C2-ceramide or 25 M etoposide at 48 h was detected using DAPI staining (left) or PI staining(middle) followed by fluorescent microscopy or flow cytometric analysis, respectively. The percentages of apoptotic cells are shown. For mreduction detection, ceramide- or etoposide-treated cells were detected using rhodamine 123 followed by flow cytometric analysis (right). Thepercentages of cells with m reduction are shown. C, caspase activation in A549-P2 and A549-B2 cells induced by 50 M C2-ceramide (top) or 25M etoposide (bottom) for various time periods was determined using activity assay kits. The substrate activities for caspase-9 and -3 (pNA,p-nitroanilide) (left, means S.D. of triplicate cultures) and the relative activity for caspase-2 using OD (right) are shown. D, the processing ofcaspase-2 in these cells after ceramide (top) or etoposide (bottom) treatment was detected using Western blot analysis.



ww

.jbc.org/D

ownloaded from

characterized by destructive morphology (Fig. 2C, all arrow-heads) were seen in both EGFP-positive vector control (filledarrowheads) and EGFP-negative cells (open arrowheads). Inthe caspase-2 siRNA-transfected group, however, only EGFP-negative cells underwent apoptosis (open arrowheads),whereas EGFP-positive caspase-2 siRNA-expressing cellsshowed resistance to ceramide and etoposide stimulation withintact morphology. Similarly, ceramide- and etoposide-inducedapoptosis was blocked in caspase-2 siRNA-expressing cells,detected using annexin V staining and caspase-3 activity inEGFP-positive cells (Fig. 2D). These results showed that inter-ference with the expression of caspase-2 blocked the mitochon-dria-dependent pathway of apoptosis induced by ceramide andetoposide.Bcl-2 Blocks Caspase-2-mediated Ceramide- and Etoposide-

induced ApoptosisBcl-2 acts as an anti-apoptotic factor toblock death signals, including those from ceramide and etopo-side (2532). To investigate the relation between Bcl-2 andcaspase-2 activity during ceramide- and etoposide-induced apo-ptosis, we used Bcl-2-overexpressing cells. First, an elevatedexpression of Bcl-2 in A549 cells transfected with bcl-2 (A549-B2), compared with the wild-type (WT) and vector control(A549-P2), was confirmed using Western blot analysis (Fig. 3A,top), intracellular immunostaining, and then flow cytometricanalysis (Fig. 3A, bottom). Cell apoptosis, characterized byDNA fragmentation, was detected using DAPI staining (Fig.3B, left) or PI staining followed by flow cytometric analysis(Fig. 3B, middle). Compared with A549-P2 cells, which had45.1 and 35.7% apoptotic cells after ceramide and etoposidestimulation, respectively, apoptotic cells were reduced to 15.1and 8.3% in A549-B2 cells. Using rhodamine 123 staining, the

m reductions in A549-P2 and A549-B2 cells were 39.2 and7.1%, respectively, after ceramide treatment and 45.6 and 4.8%after etoposide treatment (Fig. 3B, right). Ceramide-inducedtBid expression was detected using tBid-specific antibody incombination with Mito Tracker Red dye. Results showed Bidcleavage and translocation to mitochondria in ceramide-treated A549-P2 but not in A549-B2 cells (data not shown). Acaspase substrate activity assay revealed a time-dependentincrease in caspase-9, -3, and -2 activities in A549-P2 but not inA549-B2 cells (Fig. 3C). Also, cleavage of procaspase-2 was seenin A549-P2 but not in A549-B2 cells after ceramide and etopo-side treatment (Fig. 3D). MDCK cells with or without Bcl-2overexpression after ceramide and etoposide stimulationshowed similar results (supplemental Fig. S2).Role of Bcl-2 in Regulating Caspase-2 ActivationBcl-2 has

blocked caspase-3 and -2 sequential activation after mitochon-drial damage (19). In that case, caspase-2 acted downstream ofmitochondria and caspase-3, and Bcl-2 acted at a point down-stream from the release of mitochondrial cytochrome c. Bcl-2had also acted upstream or downstream of caspase-8 in Fas- orstaurosporine-induced apoptosis (4, 40, 41). Our previous studydemonstrated caspase-2 and -8 sequential activation in ceram-ide- and etoposide-induced mitochondrial damage in concomi-tant with the production of tBid and the release of cytochromec (24). To directly investigate the regulation of Bcl-2 oncaspase-2, we used the gene-silencing technique and Bcl-2 in-hibitor. siRNA transfection in A549 cells induced Bcl-2 orcaspase-2 knockdown, or both. After cell sorting, flow cytomet-ric analysis showed, and Western blot analysis confirmed, thatthe EGFP-positive cells with Bcl-2 siRNA, caspase-2 siRNA,Bcl-2 plus caspase-2 siRNA, and vector control were 82.2, 84,

FIG. 4. Down-regulation of Bcl-2 results in caspase-2 activation. A, A549 cells were transfected with Bcl-2 siRNA, caspase-2 siRNA, orBcl-2 plus caspase-2 siRNA as described under Experimental Procedures. After cell sorting, the expression of Bcl-2 (28 kDa) and caspase-2 (45kDa) in vector-transfected cells (Vector) and siRNA-transfected cells was detected using Western blotting (top). Protein expression of EGFP wasused as an internal control. In the bottom panel, EGFP-positive cells are indicated as successfully transfected cells (all arrowheads). With orwithout z-VDVAD-fmk, the morphology changes of 48-h posttransfected cells were detected using phase-contrast plus fluorescent microscopy. Theapoptotic cells (open arrowheads) are shown. B, for apoptosis and caspase-activation analysis, 24 and 48 h posttransfection, cells were stained withannexin V or caspase-3-specific fluorescence substrate using PhiPhiLux-G2D2 staining, respectively. The percentages of positive cells in vector-transfected and Bcl-2 siRNA-transfected groups are shown (means of duplicate cultures). After cell sorting, the caspase-2 activity was detectedusing a caspase-2 activity assay kit. The substrate relative activity is shown using OD (means of duplicate cultures). C, A549 cells were treatedwith 100 M Bcl-2 inhibitor HA14-1 for 18 and 36 h, and cell apoptosis was detected by PI staining followed by flow cytometric analysis. Also,activation of caspase-3 and -2 was determined using caspase activity assay kits. The substrate activity for caspase-3 (pNA, p-nitroanilide) and therelative activity for caspase-2 using OD are shown (means of duplicate cultures).



ww

.jbc.org/D

ownloaded from

90, and 89.8%, respectively, in one representative experiment(Fig. 4A, top). We next showed that the hallmarks of apoptosiswere present in EGFP-positive cells with Bcl-2 siRNA but notin those with vector control. Cells with Bcl-2 siRNA transfec-tion showed apoptotic morphology compared with untrans-fected cells or vector control cells (Fig. 4A, open arrowheads).Bcl-2-silencing cells co-transfected with caspase-2 siRNA orpretreated with z-VDVAD-fmk rescued apoptosis caused byBcl-2 inhibition. Bcl-2 siRNA-transfected cells showed a time-dependent increase of apoptotic cell death with annexin Vbinding and caspase-3 activation compared with vector controlcells (Fig. 4B). To further investigate caspase-2 activation afterBcl-2 knockdown, we sorted EGFP-positive cells and deter-mined the activation of caspase-2. A substrate activity detec-tion assay suggested that caspase-2 was activated in Bcl-2siRNA-transfected cells but not in vector control cells (Fig. 4B).HA14-1, a small molecule inhibitor of Bcl-2, induced cell apo-ptosis via the mitochondrial intrinsic pathway through Bcl-2structural dysfunction (42, 43). HA14-1 caused Bax transloca-tion to mitochondria and cytochrome c release (43). Our resultsshowed that treatment with HA14-1 induced A549 cells toundergo apoptosis. The m reduction (data not shown) andcaspase-3 and -2 activation (Fig. 4C) were also observed. HA14-1-induced cell apoptosis and caspase activation were detectedin Bcl-2-overexpressing cells (data not shown). To verify thespecificity of HA14-1 on Bcl-2 dysfunction, we tested a Bcl-2-independent apoptotic pathway in DU145 cells, human pros-

tate cancer cells. A previous study (44) showed that Bcl-2down-regulation by antisense RNA and siRNA strategies wasunable to facilitate cell apoptosis in DU145 cells. Our resultsconfirmed that DU145 cells resisted a serial dose of Bcl-2siRNA (supplemental Fig. S3A). DU145 cells treated with 100M HA14-1 did not undergo apoptosis, m reduction, orcaspase-3 and -2 activation (supplemental Fig. S3B). However,HA14-1 may have caused DU145 cell death at higher concen-trations (data not shown). These results corresponded with aprevious report (23) demonstrating the involvement ofcaspase-2 in Bcl-2-silencing cell apoptosis.Based on our previous findings (24), ceramide- and etopo-

side-induced caspase-2 activation acted upstream of caspase-8activation, Bid cleavage, and m reduction. To further con-firm that Bcl-2-modulated caspase-2 activation acted upstreamof mitochondrial damage, we used the mitochondrial perme-ability transition pore inhibitor bongkrekic acid and caspase-9inhibitor z-LEHD-fmk. After pretreatment with bongkrekicacid or z-LEHD-fmk, apoptosis (Fig. 5, top) and caspase-3 ac-tivation (middle) caused by Bcl-2 knockdown or HA14-1 wascompletely blocked. Most importantly, caspase-2 activation in-duced by Bcl-2 knockdown or HA14-1 was not inhibited in thepresence of bongkrekic acid or z-LEHD-fmk (bottom). We con-cluded that Bcl-2-modulated caspase-2 activation functionedupstream of mitochondria.Bcl-2 Dysfunction Caused by Phosphatase during Ceramide-

and Etoposide-induced ApoptosisIt had been reported thatceramide caused Bcl-2 dephosphorylation at serine 70 and re-sulted in Bcl-2 functional destruction through protein phospha-

FIG. 5. Caspase-2 activation acts upstream of mitochondrialdamage after Bcl-2 dysfunction. With or without bongkrekic acid(BA) or caspase-9 inhibitor z-LEHD-fmk, A549 cells were treated with50 M C2-ceramide, 25 M etoposide, or 100 M HA14-1, or transfectedwith Bcl-2 siRNA. Cell apoptosis and caspase activation were deter-mined at 48 h. Cell apoptosis was detected using PI (filled bar) orannexin V (open bar) staining followed by flow cytometric analysis orfluorescent microscopy. The percentages of apoptotic cells and annexinV-positive cells are shown (means of duplicate cultures). The activationof caspase-3 and -2 was detected by substrate cleavage using PhiPhi-Lux-G2D2 staining or an activity assay kit. The quantified concentra-tions for caspase-3 (filled bar) and the percentages of active caspase-3-positive cells (open bar) are shown. Bcl-2 siRNA-transfected cells weresorted and used for caspase-2 activity detection. The substrate-relativeactivity is shown using OD (means of duplicate cultures).

FIG. 6. Phosphatase causes Bcl-2 dysfunction, which facili-tates ceramide- and etoposide-induced mitochondrial apopto-sis. A, Western blot analysis of phospho-Bcl-2 serine 70 (pBcl-2) andBcl-2 protein expression (28 kDa) in total A549 cell extracts at varioustime points after treatment with 50 M C2-ceramide (top) or 25 Metoposide (bottom). Protein expression of -actin was used as an inter-nal control. B, using immunofluorescence staining followed by flowcytometric analysis, the expression of pBcl-2 in A549 cells after cera-mide treatment with or without 100 nM protein phosphatase inhibitorOA was detected at 24 h. The histogram of protein expression is shownas a representative of two individual experiments. The mean fluores-cence intensity (MFI) of each protein expression is shown. C, A549 cellswere treated with ceramide (solid line) or etoposide (dotted line) with orwithout 100 nM OA for 24 and 48 h. Cell apoptosis and m (MTP)reduction were detected using PI and rhodamine 123 staining, respec-tively, followed by flow cytometric analysis. The percentages of apo-ptotic cells and positive cells with m reduction are shown (means ofduplicate cultures). Also, activation of caspase-3 and -2 were deter-mined using caspase activity assay kits. The activity as relative OD isshown (means of duplicate cultures).



ww

.jbc.org/D

ownloaded from

tase 2A (33). We therefore investigated whether the dephos-phorylation of Bcl-2 at serine 70 also occurred in A549 cells andrelated to caspase-2 activation. Indeed, Bcl-2 was dephospho-rylated at serine 70 by 12 and 24 h after ceramide and etopo-side treatment, respectively, and the expression level of Bcl-2decreased at 48 and 24 h posttreatment (Fig. 6A). Similarresults with Bcl-2 serine 70 dephosphorylation were seen usingintracellular immunostaining followed by flow cytometric anal-ysis. Moreover, dephosphorylation of Bcl-2 caused by ceramidewas rescued by pretreatment with the phosphatase inhibitorOA (Fig. 6B). Furthermore, ceramide- and etoposide-inducedcell apoptosis, m reduction, and caspase-3 and -2 activationwere all repressed by OA (Fig. 6C). OA caused maximal inhi-bition at a dose of 100 nM, but a higher dose was cytotoxic (datanot shown). Based on these results, ceramide- and etoposide-induced mitochondrial damage was initiated by caspase-2 ac-tivation, caspase-2 was regulated by Bcl-2, and Bcl-2 was, atleast in part, regulated by protein phosphatase. Nevertheless,although the phosphorylated status of Bcl-2 appears involved,the modulatory role of Bcl-2 on caspase-2 activation remains tobe defined.Concluding RemarksIn the present study, the interrela-

tionship between Bcl-2 and caspase-2 was revealed. Bcl-2knockdown by siRNA or Bcl-2 inhibition by an inhibitor re-sulted in an autonomic activation of caspase-2, providing directevidence that caspase-2 is negatively regulated by Bcl-2. Uponapoptotic stimulation by ceramide and etoposide, Bcl-2 wasdephosphorylated by protein phosphatase and lost its regula-tory control on caspase-2. This study, therefore, addresses anovel anti-apoptotic mechanism of Bcl-2. In its phosphorylatedstate, it blocks caspase-2 activation, although the underlyingmechanism remains unclear. However, Bcl-2-modulatedcaspase-2 activation and cell apoptosis were not involved inBcl-2-insensitive cells such as DU145. Recently, the silencing ofBcl-2 by siRNA or inhibitor has been used in tumor therapy(23, 42, 43). Cell apoptosis occurred in Bcl-2-silencing cellsthrough caspase-2- and p53-regulated pathways (23). In addi-tion, Bcl-2 had been shown to act upstream of caspase-2 acti-vation in PC12 cell apoptosis induced by growth factor depri-vation (45). Based on our results, ceramide- and etoposide-induced caspase-2 activation before mitochondrial damage wasinitiated because of Bcl-2 dysfunction. In ceramide- and etopo-side-induced apoptosis, Bcl-2 dysfunction occurred because ofits dephosphorylation at serine 70. Furthermore, there was adecrease in Bcl-2 expression after ceramide and etoposidetreatment. Involvement of Bcl-2 cleavage in the accelerationof etoposide-induced U937 cell apoptosis had previously beenreported (34).Apoptotic signaling mediated by ceramide has provided new

insights into the mechanism of action of chemotherapy andradiotherapy in antitumor activity (4648). The up-regulationof the endogenous ceramide level induced by etoposide wasdemonstrated (30, 49, 50). Ceramide-induced apoptosis hasbeen associated with dephosphorylation of various kinasessuch as Akt, Bcl-2-associated death promoter (BAD), forkheadtranscription factor (FKHR), and GSK-3 (5154). Whetherthese kinases are involved in the regulation of protein phos-phatase on Bcl-2 and in the regulation of Bcl-2 on caspase-2remains to be investigated. Our preliminary results show thatBcl-2 and caspase-2 can be co-precipitated. Whether Bcl-2 andcaspase-2 can directly interact with each other or whetheradaptor proteins are necessary for Bcl-2 and caspase-2 bindingrequires further investigation. The possible regulatory mecha-nisms between protein phosphatase, protein kinase, Bcl-2, andcaspase-2 need to be further explored. Taken together, these

findings shed light on the role of Bcl-2 in the inhibition ofcaspase-2 before mitochondrial damage.

AcknowledgmentsWe thank Wan-Hua Tsai from the Institute ofBasic Medical Sciences (NCKU, Taiwan) for conducting the Bcl-2 over-expression system and Ming-Chen Yang and Wen-Wei Chang from theDepartment of Microbiology and Immunology (NCKU, Taiwan) for as-sistance with cell sorting. We also thank Dr. C. W. Chiang for commentson the manuscript and Bill Franke for editorial assistance.

REFERENCES

1. Green, D. R., and Reed, J. C. (1998) Science 281, 130913122. Kroemer, G., and Reed, J. C. (2000) Nat. Med. 6, 5135193. Ravagnan, L., Roumier, T., and Kroemer, G. (2002) J. Cell. Physiol. 192,

1311374. Reed, J. C. (1998) Oncogene 17, 322532365. Yang, J., Liu, X., Bhalla, K., Kim, C. N., Ibrado, A. M., Cai, J., Peng, T. I.,

Jones, D. P., and Wang, X. (1997) Science 275, 112911326. Brenner, C., Cadiou, H., Vieira, H. L., Zamzami, N., Marzo, I., Xie, Z., Leber,

B., Andrews, D., Duclohier, H., Reed, J. C., and Kroemer, G. (2000) Onco-gene 19, 329336

7. Luo, X., Budihardjo, I., Zou, H., Slaughter, C., and Wang, X. (1998) Cell 94,481490

8. Desagher, S., Osen-Sand, A., Nichols, A., Eskes, R., Montessuit, S., Lauper, S.,Maundrell, K., Antonsson, B., and Martinou, J. C. (1999) J. Cell Biol. 144,891901

9. Kim, P. K., Annis, M. G., Dlugosz, P. J., Leber, B., and Andrews, D. W. (2004)Mol. Cell 14, 523529

10. Ito, T., Deng, X., Carr, B., and May, W. S. (1997) J. Biol. Chem. 272,1167111673

11. Ruvolo, P. P., Deng, X., Carr, B. K., and May, W. S. (1998) J. Biol. Chem. 273,2543625442

12. Guo, Y., Srinivasula, S. M., Druilhe, A., Fernandes-Alnemri, T., and Alnemri,E. S. (2002) J. Biol. Chem. 277, 1343013437

13. Paroni, G., Henderson, C., Schneider, C., and Brancolini, C. (2002) J. Biol.Chem. 277, 1514715161

14. Robertson, J. D., Enoksson, M., Suomela, M., Zhivotovsky, B., and Orrenius, S.(2002) J. Biol. Chem. 277, 2980329809

15. Lassus, P., Opitz-Araya, X., and Lazebnik, Y. (2002) Science 297, 1352135416. Kumar, S., and Vaux, D. L. (2002) Science 297, 1290129117. Schweizer, A., Briand, C., and Grutter, M. G. (2003) J. Biol. Chem. 278,

424414244718. Troy, C. M., and Shelanski, M. L. (2003) Cell Death Differ. 10, 10110719. Swanton, E., Savory, P., Cosulich, S., Clarke, P., and Woodman, P. (1999)

Oncogene 18, 1781178720. OReilly, L. A., Ekert, P., Harvey, N., Marsden, V., Cullen, L., Vaux, D. L.,

Hacker, G., Magnusson, C., Pakusch, M., Cecconi, F., Kuida, K., Strasser,A., Huang, D. C., and Kumar, S. (2002) Cell Death Differ. 9, 832841

21. Read, S. H., Baliga, B. C., Ekert, P. G., Vaux, D. L., and Kumar, S. (2002)J. Cell Biol. 159, 739745

22. Tinel, A., and Tschopp, J. (2004) Science 304, 84384623. Jiang, M., and Milner, J. (2003) Genes Dev. 17, 83283724. Lin, C. F., Chen, C. L., Chang, W. T., Jan, M. S., Hsu, L. J., Wu, R. H., Tang,

M. J., Chang, W. C., and Lin, Y. S. (2004) J. Biol. Chem. 279, 407554076125. Zamzami, N., Marchetti, P., Castedo, M., Decaudin, D., Macho, A., Hirsch, T.,

Susin, S. A., Petit, P. X., Mignotte, B., and Kroemer, G. (1995) J. Exp. Med.182, 367377

26. Zhang, J., Alter, N., Reed, J. C., Borner, C., Obeid, L. M., and Hannun, Y. A.(1996) Proc. Natl. Acad. Sci. U. S. A. 93, 53255328

27. Smyth, M. J., Perry, D. K., Zhang, J., Poirier, G. G., Hannun, Y. A., and Obeid,L. M. (1996) Biochem. J. 316, 2528

28. Elliott, M. J., Murphy, K. M., Stribinskiene, L., Ranganathan, V., Sturges, E.,Farnsworth, M. L., and Lock, R. B. (1999) Cancer Chemother. Pharmacol.44, 111

29. Zhang, J., Reedy, M. C., Hannun, Y. A., and Obeid, L. M. (1999) J. Cell Biol.145, 99108

30. Tepper, A. D., de Vries, E., van Blitterswijk, W. J., and Borst, J. (1999) J. Clin.Investig. 103, 971978

31. Murphy, K. M., Ranganathan, V., Farnsworth, M. L., Kavallaris, M., and Lock,R. B. (2000) Cell Death Differ. 7, 102111

32. Eliseev, R. A., Gunter, K. K., and Gunter T. E. (2003) Exp. Cell Res. 289,275281

33. Ruvolo, P. P., Deng, X., Ito, T., Car, B. K., and May, W. S. (1999) J. Biol. Chem.274, 2029620300

34. Fujita, N., and Tsuruo, T. (1998) Biochem. Biophys. Res. Commun. 246,484488

35. Wolf, C. M., Reynolds, J. E., Morana, S. J., and Eastman, A. (1997) Exp. CellRes. 230, 2227

36. Morana, S. J., Wolf, C. M., Li, J., Reynolds, J. E., Brown, M. K., and Eastman,A. (1996) J. Biol. Chem. 271, 1826318271

37. Brummelkamp, T. R., Bernards, R., and Agami, R. (2002) Science 296,550553

38. Mathias, S., Pena, L. A., and Kolesnick, R. N. (1998) Biochem. J. 335, 46548039. Hannun, Y. A., and Ruberto, C. (2000) Trends Cell Biol. 10, 738040. Johnson, V. L., Cooper, I. R., Jenkins, J. R., and Chow, S. C. (1999) Exp. Cell

Res. 251, 17518441. Kawahara, A., Kobayashi, T., and Nagata, S. (1998) Oncogene 17, 2549255442. Wang, J. L., Liu, D., Zhang, Z. J., Shan, S., Han, X., Srinivasula, S. M., Croce,

C. M., Alnemri, E. S., and Huang, Z. (2000) Proc. Natl. Acad. Sci. U. S. A.97, 71247129

43. An, J., Chen, Y., and Huang, Z. (2004) J. Biol. Chem. 279, 1913319140



ww

.jbc.org/D

ownloaded from

44. Raffo, A., Lai, J. C., Stein, C. A., Miller, P., Scaringe, S., Khvorova, A., andBenimetskaya, L. (2004) Clin. Cancer Res. 10, 31953206

45. Haviv, R., Lindenboim, L., Yuan, J., and Stein, R. (1998) J. Neurosci. Res. 52,491497

46. Gulbins, E., and Kolesnick, R. (2003) Oncogene 22, 7070707747. Jaffrezou, J. P., and Laurent, G. (2004) Bull. Cancer (Paris) 91, 13316148. Ogretmen, B., and Hannun, Y. A. (2004) Nat. Rev. Cancer 4, 60461649. Boesen-de Cock, J. G. R., Tepper, A. D., de Vries, E., van Blitterswijk, W. J.,

and Borst, J. (1999) J. Biol. Chem. 274, 1425514261

50. Toman, R. E., Movsesyan, V., Murthy, S. K., Milstien, S., Spiegel, S., andFaden, A. I. (2002) J. Neurosci. Res. 68, 323330

51. Stoica, B. A., Movsesyan, V. A., Lea, P. M., IV, and Faden, A. I. (2003) Mol.Cell. Neurosci. 22, 365382

52. Zhou, H., Summers, S. A., Birnbaum, M. J., and Pittman, R. N. (1998) J. Biol.Chem. 273, 1656816575

53. Schubert, K. M., Scheid, M. P., and Duronio, V. (2000) J. Biol. Chem. 275,1333013335

54. Ruvolo, P. P. (2003) Pharmacol. Res. 47, 383392



ww

.jbc.org/D

ownloaded from

Tang, Wen-Chang Chang and Yee-Shin LinRen-Huang Wu, Yi-Ting Fang, Ming-JerChang, Ming-Shiou Jan, Li-Jin Hsu, Chiou-Feng Lin, Chia-Ling Chen, Wen-Tsan

ActivationApoptosis through Blockage of Caspase-2Etoposide-induced Mitochondrial Bcl-2 Rescues Ceramide- andMechanisms of Signal Transduction:

doi: 10.1074/jbc.M412292200 originally published online April 6, 20052005, 280:23758-23765.J. Biol. Chem.

10.1074/jbc.M412292200Access the most updated version of this article at doi:

.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the

Alerts:

When a correction for this article is posted When this article is cited

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2005/05/11/M412292200.DC1.html

http://www.jbc.org/content/280/25/23758.full.html#ref-list-1This article cites 54 references, 29 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

Documents

J. Biol. Chem.-2005-Lin-23758-65