PK11195 potently sensitizes to apoptosis induction ... - CiteSeerX

Oncogene (2005) 24, 7503–7513

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

ORIGINAL PAPERS

PK11195 potently sensitizes to apoptosis induction independently from the peripheral benzodiazepin receptor Rosa-Ana Gonzalez-Polo1, Gabrielle Carvalho1, Thorsten Braun1, Didier Decaudin2, Claire Fabre1, Nathanael Larochette1, Jean-Luc Perfettini1, Mojgan Djavaheri-Mergny3, Ibtissam Youlyouz-Marfak4, Patrice Codogno3, Martine Raphael5, Jean Feuillard4 and Guido Kroemer*,1 1 Centre National de la Recherche Scientifique, UMR8125, Institut Gustave Roussy, Pavillon de Recherche 1, 39 rue Camille-Desmoulins, 94805 Villejuif, France; 2Department of Hematology, Institut Curie, 75005 Paris, France; 3Institut Andre´ Lwoff , INSERM U504, 16 avenue Paul-Vaillant-Couturier, 94807 Villejuif Cedex, France; 4Centre National de la Recherche Scientifique, UMR CNRS 6101 et Laboratoire d’He´matologie, faculte´ de Me´dicine et CHU Dupuytren, Limoges, France; 5Service d’He´matologie Biologique, U473, CHU Kremlin Biceˆtre, France

1-(2-Chlorophenyl-N-methylpropyl)-3-isoquinolinecarboxamide (PK11195) is a prototypic ligand of the peripheral benzodiazepine receptor (PBR), a mitochondrial outer membrane protein. PK11195 can be used to chemosensitize tumor cells to a variety of chemotherapeutic agents, both in vitro and in vivo. PK11195 has been suggested to exert this effect via inhibition of the multiple drug resistance (MDR) pump and by direct mitochondrial effects which could be mediated by the PBR. Here, we established a model system in which PK11195 and another PBR ligand, 7-chloro-5-(4chlorophenyl)-1,3-dihydro-1-methyl-2H-1,4-benzodiazepin2-one (Ro5-4864), sensitize to nutrient depletion-induced cell death. In this MDR-independent model, PK11195 and Ro5-4864 are fully active even when the PBR is knocked down by small interfering RNA. Cells that lack PBR possess low-affinity binding sites for PK11195 and Ro54864. The starvation-sensitizing effects of PK11195 are not due to a modulation of the adaptive response of starved cells, namely autophagy and NF-jB activation. Rather, it appears that the combination of PK11195 with autophagy or NF-jB inhibitors has a potent synergistic death-inducing effect. Starved cells treated with PK11195 exhibit characteristics of apoptosis, including loss of the mitochondrial transmembrane potential, mitochondrial cytochrome c release, caspase activation and chromatin condensation. Accordingly, stabilization of mitochondria by overexpression of Bcl-2 or expression of the viral mitochondrial inhibitor (vMIA) from cytomegalovirus inhibits cell death induced by PK11195 plus starvation. Thus, PK11195 potently sensitizes to apoptosis via a pathway that involves mitochondria, yet does not involve the PBR. Oncogene (2005) 24, 7503–7513. doi:10.1038/sj.onc.1208907; published online 8 August 2005 Keywords: autophagy; caspase; Bcl-2; NF-kappaB

*Correspondence: G Kroemer; E-mail: [email protected] Received 20 April 2005; revised 31 May 2005; accepted 7 June 2005; published online 8 August 2005

Introduction The peripheral benzodiazepine receptor (PBR) has been suggested as a putative target for therapeutic cell death induction (Casellas et al. , 2002; Castedo et al., 2002). The PBR protein is overexpressed in some tumors, and this overexpression has a negative prognostic impact on breast cancer (Galiegue et al., 2004) and colorectal cancer (Maaser et al., 2005). In contrast, knockdown of PBR expression by an antisense construct enhances the tumorigenicity of murine cancer cell lines (Weisinger et al., 2004). Ligation of PBR with synthetic ligands such as 1-(2-chlorophenyl-N-methylpropyl)-3-isoquinolinecarboxamide (PK11195) or 7-chloro-5-(4-chlorophenyl)-1,3-dihydro-1-methyl-2H-1,4-benzodiazepin-2-one (Ro5-4864) facilitates apoptosis induction in human tumor cells by a variety of chemotherapeutic agents, including doxorubicin (Hirsch et al., 1998; Sutter et al., 2004), daunorubicin (Jakubikova et al., 2002), etoposide (Decaudin et al., 2002; Okaro et al., 2002), 5-fluorouracil, UV and g-irradiation (Okaro et al., 2002), ifosfamide (Decaudin et al., 2002), paclitaxel, docetaxel (Hirsch et al., 1998; Sutter et al., 2004), colchicine (Jorda et al., 2005), arsenicals (Larochette et al., 1999; Muscarella et al., 2003), lonidamine (Ravagnan et al., 1999), and bortezomib (Chauhan et al., 2004). In addition, PK11195 can sensitize cells to apoptosis induction by anti-CD95 (Decaudin et al., 2002), the cytokine mda-7/interleukin-24 (Lebedeva et al., 2003), the proapoptotic second messenger ceramide (Hirsch et al., 1998) and the Bcl-2 antagonist HA14.1 (Chen et al., 2002; Sutter et al., 2004). Finally, PK11195 sensitizes acute myeloid leukemia cells to an anti-CD33 antibody conjugated to a calicheamicin toxin (gemtuzumab ozogamicin) (Walter et al., 2004). Some of these effects have been obtained in vivo, in mice bearing human lung cancer xenografts treated with etoposide or ifosfamide (Decaudin et al., 2002), in mice grafted with human cholangiocarcinoma carcinoma cells treated with etoposide (Okaro et al., 2002) and in mice injected

Apoptosis sensitization by PK11195 R-A Gonzalez-Polo et al

7504

with human acute myeloid leukemia cells (Walter et al., 2004). PK11195 has been administrated safely to patients (Ferry et al., 1989; Ansseau et al., 1991), and has been suggested to be included in clinical trials as a chemosensitizing agent (Decaudin et al., 2002; Okaro et al., 2002; Walter et al., 2004). However, alternative PBR ligands with proapoptotic anticancer effects are being developed (Casellas et al., 2002; Castedo et al., 2002; Boitano et al., 2003; Decaudin, 2004). The mechanism of action of PK11195 is controversial. PK11195 binds with nanomolar affinity to the p18 PBR protein (Snyder et al., 1990; Casellas et al., 2002), yet facilitates apoptosis induction at doses in the higher micromolar range (Hirsch et al., 1998; Larochette et al., 1999; Ravagnan et al., 1999; Chen et al., 2002; Decaudin et al., 2002; Jakubikova et al., 2002; Okaro et al., 2002; Lebedeva et al., 2003; Muscarella et al., 2003; Chauhan et al., 2004; Sutter et al., 2004; Walter et al., 2004; Jorda et al., 2005). PBR is preponderantly localized in the outer mitochondrial membrane and forms a molecular complex with two putative regulators of apoptosis (McEnery et al., 1992), the voltage-dependent anion channel (Shimizu et al., 2001) and the adenine nucleotide translocase (Vieira et al., 2000), which in turn are regulated by proteins from the Bcl-2 family (Shimizu et al., 2000; Belzacq et al., 2003). These proteins participate in the regulation of mitochondrial membrane permeabilization (MMP), one of the decisive steps of the apoptotic process (Green and Kroemer, 2004). Indeed, micromolar concentrations of PK11195 can facilitate MMP when added to isolated mitochondria (Hirsch et al., 1998). In a variety of systems, PBR ligand can reduce or abrogate the MMP-inhibitory effect of Bcl-2like proteins, including Bcl-2 itself (Hirsch et al., 1998; Decaudin et al., 2002; Muscarella et al., 2003; Chauhan et al., 2004; Walter et al., 2004), Bcl-XL (Okaro et al., 2002; Walter et al., 2004) and Mcl-1 (Okaro et al., 2002), leading to the suggestion that PBR might exert local effects on mitochondrial membranes that nullify the cytoprotective action of Bcl-2 and its homologues (Hirsch et al., 1998). Beyond its effects on mitochondria, PK11195 has been suggested to have a series of additional effects, namely inhibition of the multiple drug resistance (MDR) pumps (Banker et al., 2002; Jakubikova et al., 2002), induction of reactive oxygen species (Fennell et al., 2001), activation of c-Jun NH2 terminal kinase (JNK) (Chauhan et al., 2004) or activation of the p38-mitogenactivated protein kinase (MAPK) pathway (Sutter et al., 2003). Based on these uncertainties and incognita, we decided to re-examine the mode of mechanisms through which PK11195 sensitizes to cell death induction. Here we show that PK11195 can sensitize to starvationinduced cell death, that is, a modality of cell death that should not be modulated by the MDR. Using this model, we demonstrate that PK11195 acts in cells lacking PBR and explore the mode of action of PK11195, which potentiates the mitochondrial pathway of apoptosis induction. Oncogene

Results and discussion PK11195 sensitizes to starvation-induced death through a PBR-independent pathway HeLa cells survive for a limited period in serum-free or serum- plus nutrient-free conditions. However, in the presence of PK11195 (75–100 mM), which has no or little effects on cell survival by itself, cells undergo massive cell death, as indicated by a DCm loss (as determined with the DCm-sensitive dye 3,30 dihexyloxacarbocyanine iodide (DiOC6(3))) and an increased plasma membrane permeability (as determined with the vital dye propidium iodine (PI)) (Figure 1a, b). In contrast with PK11195, MDR inhibitors such as verapamil did not enhance starvation-induced apoptosis (not shown). We developed two distinct small interfering RNA heteroduplexes specific for PBR that both effectively downregulated PBR protein expression within 48 h after transfection into HeLa cells (Figure 2a). While knockdown of the PBR gene had a minor sensitizing effect on nutrient-depleted HeLa cells, this manipulation did not abolish the effects of PK11195, which continued to sensitize HeLa cells to starvation-induced death (Figure 2b, c). PK11195 enhanced the production of reactive oxygen species in nutrient-depleted HeLa cells (as measured by the oxidation of the nonfluorescent dye hydroethidine (HE) into its fluorescent product ethidium (Eth)). Again, this effect was not counteracted by PBR knockdown (Figure 2d, e). Another ligand of PBR, RO5-4864, also sensitized to cell death induction by starvation, in a PBR-independent fashion (Figure 2f). Finally, Jurkat cells, which reportedly lack PBR expression (Carayon et al., 1996; Rochard et al., 2004), also exhibited enhanced cell death when treated with a combination of nutrient depletion and PK11195 (Figure 3) or RO5-4864 (not shown). Thus, PK11195 and RO5-4864 mediate their death-sensitizing effect in a PBR-independent fashion. Low-affinity binding sites for PK11195 and Ro5-4864 in cell lines lacking PBR The cell death-sensitizing effects of PK11195 and Ro54864 detailed above (Figures 1 and 3) and described in the literature are all obtained at doses (10–100 mM) which are well beyond the affinity for the PBR (that is, lower than 1 nM) (see Introduction). This prompted us to confirm the existence of an alternative low-affinity PBR-binding site. At a concentration of 10 nM [3H]PK11195, we observed a high specific binding in the SNB19 cell line and low binding in SHEP and Jurkat cells (Figure 4a). In contrast, at a concentration of 1 mM [3H]PK11195 (100-fold higher) in SHEP-control and Jurkat cell lines, we observed an increase in the specific binding, suggesting the existence of low-affinity mPBR. Pre-exposure of Jurkat cells to Ro5-4864 or diazepam before addition of the radioactive ligand abolished the specific binding of [3H]PK11195 (1 mM) to the mPBR (Figure 4b). Scatchard analyses of the specific binding of [3H]RO5-4864 in Jurkat cells revealed a single class of binding sites with a dissociation constant (Kd) of


7505

a

104 10

CM 4

1

3

CM-serum

104 4

3

3

104

NF+serum 3

2

3

104

10

102

102

102

102

101

101

101

101

8 100

87 101

102

103

104

PI

104

8

100

1

36 100

104

57 101

102

103

61

100

104

7

8

87 101

102

103

10

104

2

104

103

103

102

102

102

102

101

101

101

101

100

61 101

102

103

104

100

31 100

1 101

102

103

100

104

33 100

55 101

102

103

104

32 100

103

29

Co

100

103

100

3

10

100 104

5

3

10

100

NF

60 101

102

103

68

104

6

PK11195

25

100

100

1 101

102

103

104

DiOC6(3)

b DiOC6(3)low 100

PIhigh

% Cells

80 60 40 20 0 0

50 CM

100

0

50

100

CM-serum

0

50

100

NF+serum

0

50

100 PK11195 µM

NF

Figure 1 PK11195-mediated sensitization of HeLa cells to starvation-induced cell death. HeLa cells were cultured in the indicated conditions (CM: complete medium; CM-serum: without FCS; NF: nutrient- and serum-free medium; NF þ serum: nutrient-free medium supplemented with FCS, in the absence (Co) or presence of the PBR ligand PK11195 (100 mM) for 15 h, followed by simultaneous staining of the DCm with DiOC6(3) and viability with PI. Representative FACS diagrams are shown in (a) and quantitative data (means of three independent experiments7s.d.) are shown in (b) for different PK11195 concentrations

11.3 nM (Figure 4c). Using the 143N2 cell line as a positive control for PBR expression (Miccoli et al., 1998a), we observed an additional low-affinity bindingsite type (Kd 0.6 nM) after 5 min of incubation with [3H]RO5-4864 (data not shown). Incubation with [3H]RO5-4864 for 30 min showed two classes of binding sites with Kd’s of 0.5 and 25 nM (Figure 4d), respectively. These data indicate the existence of a second PBRbinding site that binds PK11195 and Ro5-4864 with an affinity higher than 10 nM. Mechanistic aspects of PK11195-facilitated cell death induction Starvation-induced cell death can be exacerbated by inhibitors of autophagy such as hydroxychloroquine

(HCQ), an agent that inhibits the fusion of autophagic vacuoles (AV) and lysosomes (Boya et al., 2003, 2005) and hence causes the AV accumulation (as detected by two methods: vacuolization of the cytoplasma visible as ‘holes’ in CMFDA-stained cells, and recruitment of an LC3–GFP fusion protein to AV) (Figure 5a, b). In contrast to HCQ, PK11195 did not induce any AV accumulation (Figure 5a, b). Rather, PK11195 inhibited the HCQ-induced autophagic vacuolization (Figure 5c). However, as compared to the prototypic autophagy inhibitor 3-methyladenine (3-MA), PK11195 did not inhibit autophagy (as determined by quantifying the turnover of proteins in starved cells) (Figure 5d). Methylpyruvate, a cell-permeable pyruvate derivative, reduced cell death induced by the combination of HCQ and starvation (Figure 5e), presumably because it Oncogene


7506

substitutes for the reduced generation of endogenous nutrients via autophagic catabolism (Lum et al., 2005). In contrast, methylpyruvate did not rescue cells exposed

a

b

Emerin

PBR04

CM

104

1

104

1

PBR10

103

102

102

102

101

7 0

91 1

2

100

3

10 10 10 10 10

4

NF+PK11195

104

1

103

100

PBR

NF 3

103

101

siRNA Sham

to PK11195 plus starvation (Figure 5e). In conclusion, PK11195 does not act on the autophagic regulation of cell death.

46

7

siRNA Emerin

101

14

82

0

1

2

3

10 10 10 10 10

30

100 4

0

17 1

2

3

10 10 10 10 104

PI 104

GAPDH

4 1 10

1

3

104

1

103

103

103

102

102

102

101 10

0

88 1

2

3

10 4

10 10 10 10 10

0

14

siRNA PBR10

101

101

10

0

40

17

89

31

100

0

15

10 10 10 10 104

100 101 102 103 104

1

2

3

DiOC6(3)

c DiOC6(3)low

100 80 60

PIhigh

Sham siRNA Emerin siRNA PBR04 siRNA PBR10

40 20

0 PK11195 µM

0

50

75

100

0

50

CM CM 6

NF

800

800

12

600

600

600

400

400

400

200

200

200

0

0 100 101 102 103 104

FSC 800

32

siRNA Emerin

15

34

600

600

600

400

400

400

200

200

200

0

0

0 100 101 102 103 104

HE

f

30 25

15 10 5

siRNA PBR10

0

Co

PK11195

CM 100 101 102 103 104

Eth 90 80 70

DiOC6(3)low


20 100 101 102 103 104

800

7

60


PI+

50 40 30 20 10 0 Co

Oncogene

100

40 35

0 100 101 102 103 104

800

100 101 102 103 104

e

NF+PK11195

Eth)high

800

%(HE

d

75

NF

Ro5-4864 CM

Co

Ro5-4864

NF

Co

PK11195

NF


7507

starvation-induced apoptosis, since transfection with a dominant-negative p53 mutant or treatment of the cells with cyclic pifithrin-a failed to modulate cell death in these conditions (not shown). Starved HeLa cells manifested the activation of the NF-kB pathway (as determined by electrophoretic mobility shift assays (EMSA)). While the IKK inhibitor BAY 11-7085 prevented NF-kB activation, no major inhibitory effect was observed for PK11195 in EMSA (Figure 8a). BAY 11-7085 strongly sensitized to the induction of induced apoptosis by the combination of starvation, as well as by the combination of PK11195 and starvation, both in HeLa cells (Figure 8b) and in PRI B lymphoma cells (Figure 8c). A PRI cell line engineered to overexpress a

5000

b 2000

4000

1600

[ H] PK11195

3000

3

2000 1000

PK11195 and NF-kB inhibition cooperate to induce starvation-induced apoptosis

0

c

PK* 1µM

6

Bound/free

3 2

% Cells

80 60 40 20 0

Co

PK11195 CM

Co

6h

PK11195 NF

Co

PK11195 CM

Co

PK11195 NF

15 h

Figure 3 Sensitization of Jurkat cells to cell death induction by a combination of nutrient depletion and PK11195. Jurkat cells were cultured in CM or NF, for 6 or 15 h, in the absence (Co) or presence of PK11195, and the frequency of cells with low DiOC6(3) or high PI incorporation was measured by FACS. This experiment has been performed three times, with similar results

Ro5-4864 Diazepam Jurkat

Kd=0.47 nM

8 7

4

Co

9

Kd=11.3 nM

5

Co FOG

d

7

6 5 4 3

Kd=25 nM

2 1

1 0

100

0

SNB19 SHEP-c Jurkat SHEP-c Jurkat

0

DiOC6(3)low PIhigh

800 400

PK* 10nM

Bound/free

We explored the possibility that PK11195 might modulate prominent apoptosis regulators at the transcriptional level. PK11195 did not activate the p53 pathway, as determined by adding PK11195 to HeLa cells transfected with a p53-inducible reporter gene (not shown). Moreover, p53 appeared to be irrelevant to

1200

3

a [ H] PK11195

Morphological analysis of cells subjected to the combined treatment of PK11195 and starvation indicate a typical apoptotic chromatin condensation, associated with caspase-3 activation and cytochrome c (Cyt c) release from mitochondria (Figure 6a, b). This combined treatment also induced the release of AIF from mitochondria, which was released at the same time as Cyt c (not shown). The broad-spectrum caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (Z-VAD-fmk) prevented chromatin condensation (not shown) and retarded cell death (but not the DCm loss) of PK11195-treated, starved HeLa cells (Figure 6c). Transfection-enforced overexpression of two proteins that stabilize mitochondria and prevent Cyt c release, namely Bcl-2 or expression of the cytomegalovirus-encoded vMIA protein (Kroemer, 1997; Poncet et al., 2004), also suppressed cell death induction by PK11195 (Figure 7). Thus, PK11195 facilitates starvation-induced cell death via the intrinsic (mitochondrion-dependent) pathway of caspase activation.

20

40 60 Bound

80

100

0

0

10

20 30 Bound

40

50

Figure 4 [3H]PK11195- and [3H]Ro5-4864-binding studies. (a) The human glioma cell line SNB19/FOG, the neuroblastoma cell line SHEP-control and the T-cell line Jurkat were cultured for 30 min in the presence of [3H]PK11195 at a concentration of 10 or 1000 nM. (b) Cells were previously exposed for 4 h to Ro5-4864 or diazepam, before addition of 1 mM [3H]PK11195. Specific affinity of [3H]PK11195 to the mPBR was measured using liquid scintillation spectroscopy in (a) and (b). Scatchard plots of [3H]Ro5-4864 were performed with Jurkat (c) and 143N2/FOG (d) cell lines, as described in ‘Materials and methods’ section. Scatchard plot of the specific binding data was determined using a least-square fit to determine the regression line. Scatchard analyses of the specific binding of [3H]Ro5-4864 on Jurkat cell line showed a single class of binding sites (c). In contrast, for the 143N2 cell line, Scatchard analyses revealed two classes of binding sites with different affinities (d). The existence of two distinct binding sites is indicated by the fact that the best fit of the data is obtained with two (rather than with one single) linear functions, as depicted graphically by two intersecting lines

Figure 2 PK11195-mediated death sensitization of HeLa cells is independent of the PBR. (a) Knockdown of PBR using siRNA. HeLa cells were sham transfected or transfected with siRNA specific for Emerin or two specific siRNAs specific for PBR, and the downregulation of PBR was corroborated by immunoblot analysis, 48 h after transfection. (b, c) Sensitization of HeLa cells to PK11195 in the absence of PBR. Cells were subjected to the indicated siRNA transfection and, 48 h later, cells were cultured in rich medium (CM) or nutrient-depleted medium (NF), in the absence or presence of PK11195 for 15 h, stained with DiOC6(3) and viability with PI, and analysed by FACS. Representative FACS pictograms are shown in (b) and mean results of triplicates (X7s.d.) are shown in (c). (d, e) PBR-independent, PK11195-elicited formation of superoxide anion. HeLa cells subjected to Emerin or PBR siRNA were cultured in the indicated medium, in the absence or presence of PK11195 (as in b, c), followed by quantitation of the frequency of cells that exhibit enhanced oxidation of hydroethidin (HE) into the fluorescent product ethidium (Eth). Typical FACS data are shown in (d), and the mean values of triplicate determinations are shown in (e). (f) PBR-independent sensitization to Ro5-4864. HeLa cells treated as in (a)–(e) were cultured in the absence or presence of nutrients and/or Ro5-4864 (100 mM) for 15 h, followed by determination of the DCm and viability. All results are representative of at least three independent determinations Oncogene


7508

a

CM

NF

NF+ PK11195

NF+ HCQ

10 µM

CFMDA CM

b

LC3-GFP

d

70

e PI high

Degradation (% h -1)

50 40 30 20

100 1.5

1

80 60 40 20

10 0 HCQ PK11195

NF+ HCQ

Ho

60 % Cells LC3-GFPVac

NF+ PK11195

% DiOC6(3)low

c

NF

-

+

+ -

+ +

0.5 3-MA PK11195

- - CM

+ - + NF

0 HCQ PK11195 Methyl-pyruvate -

+ - - + - + - - + - - + + + CM

-

+ - - + - + - - + - - + + + NF

Figure 5 Effect of PK11195 on the autophagic system. (a)–(c) Failure of PK11195 to elicit AV. HeLa cells were cultured in the presence (CM) or absence (NF) of nutrients, and in the presence of PK11195 or HCQ, which serves as a positive control. The presence of cytoplasmic vacuoles was determined by staining with CMFDA (a), and the presence of autophagic vacuoles was confirmed by transfection with an LC3–GFP fusion construct that relocalizes to them (b), quantified in (c), X7s.d., X ¼ 3). (d) Failure of PK11195 to inhibit protein turnover. The protein half-life was determined by pulse chasing with radioactive valine nutrient-depleted HeLa cell cultured in the absence or presence of 3-MA (a positive control of autophagy inhibition) or PK11195. Results are mean values of triplicates7s.d. of one out of three experiments. (e) Failure of methylpyruvate to antagonize PK11195-induced cell killing. Cells were cultured in the presence of the indicated combinations of agents (30 mg/ml HCQ, 75 mM PK11195, 10 mM methylpyruvate), and the indicated parameters of cell death were measured in triplicate

dominant-negative inhibitor of NF-kB (IkB), the IkB super-repressor (IKSR), in a tetracyclin-inducible (Teton) fashion, underwent BAY 11-7085 plus PK11195 induced DCm loss. IKSR induction by tetracyclin could be monitored because IKSR and the NGF receptor (NGFR) were coexpressed in a polycistronic vector, and surface expression of NGFR was measured by immunofluorescence in live cells (which exclude the vital dye 40 ,6-diamidino-2-phenylindole dihydrochloride (DAPI)). NGFR þ cells exhibited a higher DCm loss than NGFR Oncogene

cells, when exposed to PK11195 plus BAY 11-7085, confirming that NF-kB inhibition has a sensitizing effect to PK11195 (Figure 8c). The cell-permeable antioxidant glutathione ester inhibited the BAY 11-7085-mediated sensitization to starvation-induced apoptosis, yet had no effect on PK11195-induced apoptosis (Figure 8d), underscoring that the mechanisms of proapoptotic sensitization between BAY 11-7085 and PK11195 differ and that PK11195 does not act as a pro-oxidant in this particular system.


7509

a

NF

NF+PK11195 100

DiOC6(3)low PI high

% Cells

80 60 40 20 0

Cyt c Casp-3a Hoechst 33342

0 50 75 100

Neo

10 µM

0 50 75 100

Bcl-2

CM

b

60 50

vMIA

0 50 75 100

Neo

0 50 75 100

Bcl-2

0 50 75 100 PK11195 µM

vMIA

NF

Figure 7 Contribution of mitochondria to cell death induction by PK11195. HeLa cells stably transfected with vector only (Neo) or with either of the two mitochondrion-stabilizing proteins Bcl-2 or vMIA were subjected to nutrient depletion in the presence or absence of PK11195, and the loss of the DCm or of the viability was measured (X7s.d., n ¼ 3)

Diffuse Cyt c Active caspase 3 Nuclear apoptosis

40

% Cells

0 50 75 100

30 20 10 0

c 100

0

50

100

0

CM

50

100

PK11195 µM

50 100

PK11195 µM

NF

PI high

% DiOC6(3)low

80 60 40 20 0

0

50 100

0

50 100

+Z-VAD-fmk

CM

0

50 100

0

+Z-VAD-fmk

NF

Figure 6 Contribution of caspases to cell death induction by PK11195. (a, b) Caspase activation through the mitochondrial pathway. HeLa cells were cultured in the absence or presence of nutrients and/or PK11195 (100 mM in (a)) for 15 h, followed by fixation, permeabilization and immunofluorescence staining of Cyt c, Casp3a and nuclear chromatin condensation. Representative microphotographs are depicted in (a) and the frequency of the measured changes is analysed in (b) (X7s.d., n ¼ 3). (c) Effect of the caspase inhibitor Z-VAD-fmk on cell killing. Cells were cultured in the absence or presence of nutrients, PK11195, and/or Z-VAD-fmk for 15 h, and the frequency of cells with low DiOC6(3) or high PI incorporation was determined as in Figure 1a

Concluding remarks The data contained in this paper indicate that the potential anticancer agents PK11195 and Ro5-4864 sensitize to cell death induction via a novel, unexpected mechanism. First, in the model that we have chosen (starvation-induced cell death), PK11195 and Ro5-4864 cannot act through an inhibitory effect of the MDR pumps (Banker et al., 2002; Jakubikova et al., 2002) (Figure 1). Second, PK11195 and Ro5-4864 do not act through the PBR, because knockdown of PBR (Figure 2) or spontaneous loss of PBR (Figure 3) does not curtail the death-sensitizing effect of PK11195 or Ro5-4864. Accordingly, cells that possess the PBR (which binds to PK11195 with an affinity of approximately 1 nM) have a second, low-affinity PK11195 or Ro5-4864-binding site

(KdB10 nM), and such a binding site is also found in Jurkat cells (Figure 4). Whether this binding site is relevant to apoptosis sensitization, however, is not clear, because apoptosis sensitization is obtained at concentrations (10–100 mM) that are much higher than the lowaffinity binding site, and it is formally possible that PK11195 and Ro5-4864 sensitize to apoptosis through ‘nonspecific’ effects, for instance, physicochemical effects on membranes. PK11195 sensitizes to starvation through an unknown mechanism, because it does not subvert the known defense systems acting to rescue cells from starvation, namely autophagy and NF-kB activation. First, PK11195 fails to inhibit autophagy (Figure 5). Rather, PK11195 precipitates a direct activation of the mitochondrial apoptotic program, as exemplified by Cyt c release, caspase-3 activation (Figure 6a, b), as well as the inhibitory effect of Z-VAD-fmk (Figure 6c) and Bcl-2 or vMIA (Figure 7) on cell death. Clearly, it appears that caspase activation is a postmitochondrial event in PK11195-elicited apoptosis, because Z-VAD-fmk fails to maintain the DCm in the cells (Figure 6c), while Bcl-2 and vMIA both stabilize DCm and maintain viability (Figure 7). p38 MAPK and JNK inhibition by PK11195 (Sutter et al., 2003; Chauhan et al., 2004) are irrelevant to starvation-induced cell death because specific inhibitors of these stress kinases did not sensitize to nutrient depletion-induced apoptosis (not shown). Moreover, although PK11195 can cause the production of superoxide anion (Figure 2d, e), antioxidants such as glutathione ester (Figure 8d) or the superoxide dismutase mimic MnTBAP (not shown) fail to antagonize PK11195-mediated killing. In addition, PK11195 fails to interfere with the NF-kB rescue pathway (Figure 8). Accordingly, some data indicate a cooperative cell death-sensitizing effect when PK11195 and NF-kB inhibition are combined (Figure 8b, c). Thus, the death-sensitizing effect of PK11195 does not involve any of the before-mentioned pathways. This is in accord with the previous observation that PK11195 can act on isolated mitochondria, in a cell-free system, to facilitate apoptosis-related MMP (Hirsch et al., 1998). Altogether, our results indicate that PK11195 sensitizes to apoptosis induction through a unique, novel Oncogene


7510

pathway. This pathway, which remains to be explored in its details, may be particularly relevant to cancer cell biology because PK11195 has a unique capacity to

a

sensitize transformed cells to apoptosis, yet has little effects on normal cells, as indicated by its low level of intrinsic toxicity.

CM

PK11195 BAY 11-7085 HCQ

-

+ -

+ -

NF + + -

+

-

+ -

+ -

+ + -

+

NF-kB

b

DiOC6 (3)low 100

PI high

% Cells

80 60 40 20 0

0

50 75 100

0 50 75 100 +Bay 11-7085

0

CM

c

0 50 75 100 PK11195 µM +Bay 11-7085

50 75 100

NF

100

Without doxycyclin, NGFR With doxycyclin, NGFR+

% DiOC6(3)low

80 60 40 20 0

50

0

100

0 50 100 +Bay 11-7085

0

50

CM

d

0 50 100 PK11195 µM +Bay 11-7085

100

NF

100

DiOC6 (3)low PIhigh

% Cells

80 60 40 20 0

-

+ -

+ -

+ CM

Oncogene

+ +

+ +

-

+ -

+ -

+ NF

+ +

+ +

PK11195 Bay 11-7085 Glut-ester


7511 Materials and methods

fmk (Bachem) was added at the same time as cell death inducers at 50 mM (Boya et al., 2005).

Cell lines and culture conditions HeLa cells were stably transfected with pcDNA3.1 control vector (Neo), with human Bcl-2 (Bcl-2) or the cytomegalovirus UL37 exon 1 gene coding for vMIA (vMIA, kindly provided by Dr V Goldmacher) (Goldmacher et al., 1999; Belzacq et al., 2001). Cells were cultured in DMEM medium supplemented with 10% FCS, 1 mM pyruvate and 10 mM Hepes. Jurkat human T leukemia cells were grown in RPMI 1640 medium supplemented with 10% FCS. Lymphocytes B immortalized with the Epstein–Barr virus (PRI cells) were transfected with an episomal Tet on a double inducible construct (BaranMarszak et al., 2004) containing the cassette encoding an inactive truncated form of nerve growth factor receptor (NGFR), and the cassette encoding the mutated dominantnegative I-kB alpha cDNA (Feuillard et al., 2000) under the control of the bidirectional tetO7 tetracycline-responsive promoter (Baron et al., 1995). These cells were cultured in RPMI 1640 (Biomedia) supplemented with 10% fetal calf serum, 100 U/ml penicillin, 10 mg/ml streptomycin (GibcoBRL, Life Technologies), 2 mM L-glutamine (Eurobio) and, optionally, doxycyclin (1 mg/ml, Sigma). Cultures were maintained in a humidified atmosphere containing 5% CO2. Amino-acid starvation of cells was performed using serum-free EBSS medium (NF) (Sigma). Transfection and RNA interference siRNAs were synthesized by Proligo France SAS. For PBR (National Center for Biotechnology Information, accession number U12421), RNA sequences started at positions 226 (GAGAAGGCUGUGGUUCCCC) and 446 (CACUCAA CUACUGCGUAUG) from PBR04 and PBR10, respectively. As a control, we targeted the nuclear envelope protein Emerin, whose knockdown does not induce any known phenotype (Harborth et al., 2001). Cells were cultured in six-well plates and transfected at 80% confluence with Oligofectamine reagent (InVitrogen) following the manufacturer’s instructions. After 3 h 10% FCS was added, and cells were left for another 24–48 h before they were trypsinized and used for experiments. Transient transfection was performed with Lipofectamine 2000t reagent (Invitrogen), and cells were used 24 h after transfection. The formation of vacuoles was followed by means of a LC3-GFP plasmid (Kabeya et al., 2000). Reagents and cell death induction HCQ (Sanofi-Synthelabo) was used from a 30 mg/ml stock solution at 30 mg/ml unless otherwise specified. 3-MA (Fluka) was used at 10 mM, methyl pyruvate (Sigma) was used at 10 mM, glutathione monoethyl ester from Calbiochem was used at 1 mM. Bay 11-70585 (10 mM), Rho5-4864 (100 mM) and PK11195 were from Sigma. For caspase inhibition, z-VAD-

Flow cytometry The following fluorochromes were employed to determine apoptosis-associated changes by cytofluorometry: DiOC6(3) (40 nM) for DCm quantification, propidium iodide (PI, 5 mg/ml) and DAPI (10 mg/ml) for determination of cell viability (all of them from Molecular Probes). For the determination of superoxide anion generation, HE (10 mM) was used. Cells were trypsinized and labeled with the fluorochromes at 371C, followed by cytofluorometric analysis with a FACS Scan (Becton Dickinson). Immunofluorescence Cells cultured on coverslips were stained with Cell Tracker Green 5-chloromethylfluorescein diacetate (CMFDA, Molecular Probes, 1 mM) and Hoechst 33342 (Sigma, 2 mM), followed by fluorescence microscopic assessment with a Leica IRE2 equipped with a Leica DC300F camera. Alternatively, cells were fixed with paraformaldehyde (4% w:v) and picric acid (0.19% v:v) for LC3-GFP and immunofluorescence assays. Cells were stained for the detection of Cyt c (mAb 6H2.B4 from Pharmingen), or Casp3a (polyclonal antibody from Cell Signaling Technology), developed by goat antimouse or anti-rabbit immunoglobulin Alexas fluor conjugates (Molecular Probes) (Cregan et al., 2002). Western blot analysis Cells were washed by cold PBS at 41C and lysed in a buffer containing 50 mM Tris-HCl, pH 6.8, glycerol 10%, 2% SDS, 10 mM DTT and 0.005% blue bromophenol. In all, 40 mg of protein was loaded on a 15% SDS–PAGE and transferred to nitrocellulose. The membrane was incubated for 1 h in PBSTween 20 (0.05%) containing 5% nonfat milk. Primary antiPBR antibody (Trevigen, Gaithersburg, MD, USA) was added and revealed with the appropriate horseradish peroxidaselabeled secondary antibodies (Southern Biotechnologies Associates) and detected by SuperSignal West Pico chemiluminiscent substrate (Pierce). Anti-GAPDH (Chemicon) was used to control equal loading. [3H]PK11195- and [3H]RO5-4864-binding studies The specific affinity of [3H]PK11195 (Dupond/NEN, Les Ulis, France) to the mPBR was defined by binding assays using SHEP-control, 143N2 and Fas-resistant Jurkat cell lines, as described previously (Miccoli et al., 1998b). A human glioma cell line SNB19 (Decaudin et al., 2002) was used as positive control (Miccoli et al., 1998a), and cells were exposed to [3H]PK11195 in the presence or absence of RO5-4864, PK11195, diazepam or anti-CD95. Briefly, the culture medium of the cells in the experimental growth phase cultured in 60 mm

Figure 8 PK11195 enhances cell death independently from the NF-kB pathway. (a) Failure of PK11195 to inhibit NF-kB activation. HeLa cells were incubated for 6 h in the presence (CM) or absence (NF) of nutrients and/or the indicated drugs (75 mM PK11195, 30 mg/ ml HCQ, 10 mM BAY 11-7085), followed by nuclear extraction and determination of the NF-kB binding to nuclear DNA using EMSA. (b, c) Joint effect of PK11195 and BAY 11-7085 on cell death induction. HeLa cells were subjected to the indicated joint treatment and the frequency of dead and dying cells was determined with DiOC6(3) and PI (b). Alternatively, PRI cells manipulated to express a doxycyclin-inducible NGFR plus IKSR (see Materials and methods) were analysed for their capacity to dye in response to nutrient depletion, PK11195 and/or BAY 11-7085. The frequency of DiOC6(3)low cells was determined among the viable population staining positively for NFGR, after doxycyclin induction (c). (d) Failure of glutathione ester to antagonize PK11195-induced killing. Cells were cultured in the presence of the indicated agents (75 mM PK11195, 10 mM BAY 11-7085, 1 mM glutathione ester) for 15 h and the DCm or viability was quantitated (X7s.d., n ¼ 3) Oncogene


7512 Petri dishes was replaced by 140 mM NaCl buffered with 10 mM NaH2PO4 (pH 7.4, 371C), 2.6 mM KCl and 1.4 mM KH2PO4 for 10 min to remove any binding inhibitors present in the medium. The washing buffer was replaced with a new buffer containing 1 or 1000 nM [3H]PK11195. Nonspecific binding was determined in the presence of unlabeled PK11195 at a concentration 1000-fold higher. After incubation (30 min at 371C), cells were extracted in 2 ml of 0.2 M NaOH and placed into vials with 5 ml of scintillation liquid (ICN Pharmaceuticals, Orsay, France) and assayed for tritium by liquid scintillation spectroscopy at 60% efficiency. To define the affinity of [3H]RO5-4864 for the mPBR, binding assays were performed using Jurkat and 143N2 cell lines. The washing buffer was replaced by a buffer containing increasing concentrations of [3H]RO5-4864 (range from 5 to 500 nM). Nonspecific binding was determined in the presence of unlabeled RO5-4864 at a 1000-fold excess. After incubation, the protocol of cell extraction was that described above. Scatchard plot of the specific binding data was determined using a least-square fit to calculate the regression line. Analysis of protein degradation HeLa cells were incubated with 0.2 mCi/ml. L-[14C]valine (Perkin-Elmer Life Science) in complete medium for 24 h at 371C. Unincorporated radioisotope was removed by washing the cells three times with PBS (pH 7.4). Cells were then incubated in HBSS buffer plus 0.1% BSA (nutrient-free medium) in the presence of 10 mM cold valine, for 1 h (prechase period). After this time, the medium was replaced by the appropriate fresh medium (nutrient-free medium or complete medium) plus cold valine 10 mM in the presence or absence of 10 mM 3-MA and 75 mM PK11195 for 4 h (chase period). Cells and radiolabeled proteins from the medium were precipitated in trichloroacetic acid at the final concentration of 10% (v/v). The precipitated proteins were separated from the soluble radioactivity by centrifugation at 600 g for 20 min, and then dissolved in 1 ml of 0.2 N NAOH. The rate of protein degradation was calculated by determining the ratio of acidsoluble radioactivity recovered from both cells and medium to the ratio of radioactivity in TCA-precipitated proteins obtained from both cells and medium (Pattingre et al., 2003).

Nuclear protein extraction and EMSA Nuclear extracts from HeLa cells were prepared in lysis buffer (20 mM HEPES (pH 7.9), 350 mM NaCl, 1 mM EGTA, 1 mM dithiothreitol and complete protease inhibitors (Roche)). Extracts were coincubated in binding buffer (25 mM Tris-Hcl, pH 8, 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, 0.5 mM dithiothreitol, 10% glycerol) and 1 mg/ml poly[dI-dC]) with g-32P-labeled NF-kB (50 -ACAAGGGACTTTCCGCTGGG GACTTTCCAG-30 ) probe and T4 polynucleotide kinase (Roche) for 30 min at room temperature. Specificity was assessed by incubating nuclear extracts with the non-radiolabeled NF-kB probe or a mutated NF-kB oligonucleotide probe (50 -ACAACTCACTTTCCGCTGCTCACTTTCCAG-30 ) obtained from Invitrogen. Abbreviations AV, autophagic vacuoles; Casp3a, activated caspase-3; Co, control; Cyt c, cytochrome c; DCm, mitochondrial transmembrane potential; DAPI, 40 ,6-diamidino-2-phenylindole dihydrochloride; DiOC6(3), 3,30 dihexyloxacarbocyanine iodide; EMSA, electrophoretic mobility shift assays; FSC, forward scatter channel; HCQ, hydroxychloroquine; Ho, Hoechst 33342; 3-MA, 3-methyladenine; IKSR, inhibitor of NF-kB super-repressor; MDR, multiple drug resistance; MMP, mitochondrial membrane permeabilization; NGFR, nerve cell growth factor receptor; PBR, peripheral benzodiazepine receptor; PI, propidium iodine; PK11195, 1-(2-chlorophenyl-N-methylpropyl)3-isoquinolinecarboxamide; Ro5-4864, 7-chloro-5-(4-chlorophenyl)-1,3-dihydro-1-methyl-2H-1,4-benzodiazepin-2-one; Z-VADfmk, N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone. Acknowledgements We thank Dr T Yoshimori (National Institute for Basic Biology, Okazaki, Japan) for the LC3-GFP construct. Dr GK is supported by Cance´ropoˆle Ile-de-France, Ligue Nationale contre le cancer, European Community (Active p53, RIGHT), and Sidaction. R-A G-P received a fellowship from the European Union (FP6-2002-5000698). IYM and JF are supported by Cance´ropoˆle Grand-Sud-Ouest, Ligue Nationale contre le Cancer and Conseil Re´gional du Limousin.

References Ansseau M, von Frenckell R, Cerfontaine JL and Pa-part P. (1991). Pharmacopsychiatry, 24, 8–12. Banker DE, Cooper JJ, Fennell DA, Willman CL, Appelbaum FR and Cotter FE. (2002). Leuk. Res., 26, 91–106. Baran-Marszak F, Feuillard J, Najjar I, Le Clorennec C, Bechet JM, Dusanter-Fourt I, Bornkamm GW, Raphael M and Fagard R. (2004). Blood, 104, 2475–2483. Baron U, Freundlieb S, Gossen M and Bujard H. (1995). Nucleic Acids Res., 11, 3605–3606. Belzacq AS, El Hamel C, Vieira HLA, Cohen I, Haouzi D, Metivier D, Marchetti P, Goldmacher V, Brenner C and Kroemer G. (2001). Oncogene, 20, 7579–7587. Belzacq AS, Vieira HL, Verrier F, Vandecasteele G, Cohen I, Prevost MC, Petit PX, Kahn A, Roques B, Rizzuto R, Brenner C and Kroemer G. (2003). Cancer Res., 63, 541–546. Boitano A, Ellman JA, Glick GD and Opipari AJ. (2003). Cancer Res., 63, 6870–6876. Boya P, Gonzalez-Polo R-A, Casares N, Perfettini J-L, Dessen P, Larochette N, Metivier D, Meley D, Souquere S, Pierron G, Ohsumi Y, Codogno P and Kroemer G. (2005). Mol. Cell. Biol., 25, 1025–1040. Oncogene

Boya P, Gonzalez-Polo R-A, Poncet D, Andreau K, Roumier T, Perfettini J-L and Kroemer G. (2003). Oncogene, 22, 3927–3936. Carayon P, Portier M, Dussossoy D, Bord A, Petitpretre G, Canat X, Le Fur G and Casellas P. (1996). Blood, 87, 3170–3178. Casellas P, Galiegue S and Basile AS. (2002). Neurochem. Int., 40, 475–486. Castedo M, Perfettini J-L and Kroemer G. (2002). J. Exp. Med., 196, 1121–1126. Chauhan D, Li G, Podar K, Hideshima T, Mitsiades C, Schlossman R, Munshi N, Richardson P, Cotter FE and Anderson K. (2004). Blood, 104, 2458–2466. Chen J, Freeman A, Liu J, Dai Q and Lee RM. (2002). Mol. Cancer Ther., 1, 961–967. Cregan SP, Fortin A, MacLaurin JG, Callaghan SM, Cecconi F, Park DS, Dawson TM, Kroemer G and Slack RS. (2002). J. Cell Biol., 158, 507–517. Decaudin D. (2004). Anticancer Drugs, 15, 737–745. Decaudin D, Castedo M, Beurdeley-Thomas A, Nemati F, de Pinieux G, Caron A, Pouillart P, Wijdenes J, Rouillard D,


7513 Kroemer G and Poupon M-F. (2002). Cancer Res., 62, 1388–1393. Fennell DA, Corbo M, Pallaska A and Cotter FE. (2001). Br. J. Cancer, 84, 1397–1404. Ferry A, Jaillon P, Lecocq B, Lecocq V and Jozefczak C. (1989). Fundam. Clin. Pharmacol., 3, 383–392. Feuillard J, Schuhmacher M, Kohanna S, Asso-Bonnet M, Ledeur F, Joubert-Caron R, Bissieres P, Polack A, Bornkamm GW and Raphael M. (2000). Blood, 95, 2068–2075. Galiegue S, Casellas P, Kramar A, Tinel N and SimonyLafontaine J. (2004). Clin. Cancer Res., 10, 2058–2064. Goldmacher VS, Bartle LM, Skletskaya S, Dionne CA, Kedersha NL, Vater CA, Han JW, Lutz RJ, Watanabe S, McFarland EDC, Kieff ED, Mocarski ES and Chittenden T. (1999). Proc. Natl. Acad. Sci. USA, 96, 12536–12541. Green DR and Kroemer G. (2004). Science, 305, 626–629. Harborth J, Elbashir SM, Bechert K, Tuschl T and Weber K. (2001). J. Cell Sci., 114, 4557–4565. Hirsch T, Decaudin D, Susin SA, Marchetti P, Larochette N, Resche-Rigon M and Kroemer G. (1998). Exp. Cell Res., 241, 426–434. Jakubikova J, Duraj J, Hunakova L, Chorvath B and Sedlak J. (2002). Neoplasma, 49, 231–236. Jorda EG, Jimenez A, Verdaguer E, Canudas AM, Folch J, Sureda FX, Camins A and Pallas M. (2005). Apoptosis, 10, 91–104. Kabeya Y, Mizushima N, Ueno T, Yamamoto A, Kirisako T, Noda T, Kominami E, Ohsumi Y and Yoshimori T. (2000). EMBO J., 19, 5720–5728. Kroemer G. (1997). Nat. Med., 3, 614–620. Larochette N, Decaudin D, Jacotot E, Brenner C, Marzo I, Susin SA, Zamzami N, Xie Z, Reed JC and Kroemer G. (1999). Exp. Cell Res., 249, 413–421. Lebedeva IV, Su ZZ, Sarkar D, Kitada D, Dent P, Waxman S, Reed JC and Fisher PB. (2003). Cancer Res., 63, 8138–8144. Lum JJ, Bauer DE, Kong M, Harris HM, Li C and Lindsten T. (2005). Cell, 120, 237–248. Maaser K, Grabowski P, Oezdem Y, Krahn A, Heine B, Stein H, Buhr H, Zeitz M and Scherubl H. (2005). Clin. Cancer Res., 11, 1751–1756.

McEnery MW, Snowman AM, Trifiletti RR and Snyder SH. (1992). Proc. Natl. Acad. Sci. USA, 89, 3170–3174. Miccoli L, Beurdeley-Thomas A, De Pinieux G, Sureau F, Oudard S, Dutrillaux B and Poupon MF. (1998a). Cancer Res., 58, 5777–5786. Miccoli L, Poirson-Bichat F, Sureau F, Gonçalves RB, Bourgeois Y, Dutrillaux B, Poupon M-F and Oudard S. (1998b). J. Natl. Cancer Inst., 90, 1400–1406. Muscarella DE, O’Brien KA, Lemley AT and Bloom SE. (2003). Toxicol Sci., 74, 66–73. Okaro AC, Fennell DA, Corbo M, Davidson BR and Cotter FE. (2002). Gut., 51, 556–561. Pattingre S, Bauvy C and Codogno P. (2003). J. Biol. Chem., 278, 16667–16674. Poncet D, Boya P, Larochette N, Jalil A-A, Paulau A-L, Cartron P-F, Vallette FM, Schnebelen C, Bartle LM, Skaleskaya A, Martinou J-C, Goldmacher VS, Kroemer G and Zamzami N. (2004). J. Biol. Chem., 279, 22605–22614. Ravagnan L, Marzo I, Costantini P, Susin SA, Zamzami N, Petit PX, Hirsch F, Poupon M-F, Miccoli L, Xie Z, Reed JC and Kroemer G. (1999). Oncogene, 18, 2537–2546. Rochard P, Galiegue S, Tinel N, Peleraux A, Bord A, Jbilo O and Casellas P. (2004). Gene Expr., 12, 13–27. Shimizu S, Konishi A, Kodama T and Tsujimoto Y. (2000). Proc. Natl. Acad. Sci. USA, 97, 3100–3105. Shimizu S, Matsuoka Y, Shinohara Y, Yoneda Y and Tsujimoto Y. (2001). J. Cell Biol., 152, 237–250. Snyder SH, McEnery MW and Verma A. (1990). Neurochem. Res., 15, 119–123. Sutter AP, Maaser K, Barthel B and Scherubl H. (2003). Br. J. Cancer, 89, 564–572. Sutter AP, Maaser K, Grabowski P, Bradacs G, Vormbrock K, Hopfner M, Krahn A, Heine B, Stein H, Somasundaram R, Schuppan D, Zeitz M and Scherubl H. (2004). J. Hepatol., 41, 799–807. Vieira HL, Haouzi D, El Hamel C, Belzacq AS, Brenner C and Kroemer G. (2000). Cell Death Differ., 7, 1146–1154. Walter RB, Raden BW, Cronk MR, Bernstein ID, Appelbaum FR and Banker DE. (2004). Blood, 103, 4276–4284. Weisinger G, Kelly-Hershkovitz E, Veenman L and Spanier I. (2004). Biochemistry, 43, 12315–12321.

Oncogene