Synthetic glycosidated phospholipids induce apoptosis through ...

4 downloads 50 Views 641KB Size Report
Mar 25, 2011 - Gajate C, Mollinedo F (2007) Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and ...
Apoptosis (2011) 16:636–651 DOI 10.1007/s10495-011-0592-2

ORIGINAL PAPER

Synthetic glycosidated phospholipids induce apoptosis through activation of FADD, caspase-8 and the mitochondrial death pathway Clarissa von Haefen • Jana Wendt • Geo Semini • Marco Sifringer • Claus Belka • Silke Radetzki • Werner Reutter • Peter T. Daniel • Kerstin Danker

Published online: 25 March 2011 Ó Springer Science+Business Media, LLC 2011

Abstract Apoptosis is modulated by extrinsic and intrinsic signaling pathways through the formation of the death receptor-mediated death-inducing signaling complex (DISC) and the mitochondrial-derived apoptosome, respectively. Ino-C2-PAF, a novel synthetic phospholipid shows impressive antiproliferative and apoptosis-inducing activity. Little is known about the signaling pathway through which it stimulates apoptosis. Here, we show that this drug induces apoptosis through proteins of the death receptor pathway, which leads to an activation of the intrinsic apoptotic pathway. Apoptosis induced by Ino-C2PAF and its glucosidated derivate, Glc-PAF, was dependent on the DISC components FADD and caspase-8. This can be inhibited in FADD-/- and caspase-8-/- cells, in Peter T. Daniel and Kerstin Danker—shared senior authorship. C. von Haefen (&)  M. Sifringer Department of Anaesthesiology and Intensive Care Medicine, Charite´-Universita¨tsmedizin Berlin, Campus Virchow-Klinikum, 13353 Berlin, Germany e-mail: [email protected] J. Wendt  S. Radetzki  P. T. Daniel Department of Haematology, Oncology and Tumor Immunology, University Medical Center Charite´, Campus Buch, Berlin, Germany C. Belka Department of Radiation Oncology, University of Munich, Munich, Germany G. Semini  K. Danker Institute of Biochemistry, Campus Virchow Klinikum, University Medical Center Charite´, Berlin, Germany W. Reutter Institute of Biochemistry and Molecular Biology, Campus Benjamin Franklin, Berlin, Germany

123

which the breakdown of the mitochondrial membrane potential, release of cytochrome c and activation of caspase-9, -8 and -3 do not occur. In addition, the overexpression of crmA, c-Flip or dominant negative FADD as well as treatment with the caspase-8 inhibitor z-IETD-fmk protected against Ino-C2-PAF-induced apoptosis. Apoptosis proceeds in the absence of CD95/Fas-ligand expression and is independent of blockade of a putative death-ligand/ receptor interaction. Furthermore, apoptosis cannot be inhibited in CD95/Fas-/- Jurkat cells. Expression of Bcl2 in either the mitochondria or the endoplasmic reticulum (ER) strongly inhibited Ino-C2-PAF- and Glc-PAF-induced apoptosis. In conclusion, Ino-C2-PAF and Glc-PAF trigger a CD95/Fas ligand- and receptor-independent atypical DISC that relies on the intrinsic apoptotic pathway via the ER and the mitochondria. Keywords Synthetic phospholipid analogs  DISC  Apoptosis  Mitochondrial pathway

Introduction It has become evident that apoptosis contributes to the cytotoxic effects of genotoxic therapies such as hypoxia, irradiation or chemotherapy, implying that apoptosis resistance may limit treatment efficacy. Thus, the aberrant apoptosis pathways of tumor cells present an attractive target for the modulation of therapy response [1, 2]. The extrinsic (membrane receptor-mediated) pathway involves the death-inducing signaling complex (DISC), which forms upon ligand-receptor binding (e.g., CD95/Fas ligand binding to the corresponding CD95/Fas receptor). The accumulation of Fas-Associated protein with Death Domain (FADD) and procaspase-8 in the DISC generates

Apoptosis (2011) 16:636–651

active caspase-8, which is released from the complex to activate the executioner caspases. The intrinsic (mitochondrial) pathway involves alterations of mitochondrial function, including disruption of the mitochondrial membrane potential and release of pro-apoptotic proteins such as cytochrome c from the mitochondrial intermembrane space into the cytosol. This activates the formation of the apoptosome, a complex composed of cytochrome c, Apaf-1 and procaspase-9. Formation of the apoptosome is followed by the activation of caspase-9, which promotes the activation of the executioner caspases. The extrinsic pathway feeds into the intrinsic pathway via activation of Bid by caspase-8. Bid then induces the release of cytochrome c from the mitochondria. Synthetic alkyl-phospholipid analogs (APLs) are a new class of antitumor agents which target cell membranes [3–10]. These compounds differ from conventional anticancer therapeutics that act either via a DNA damage-induced stress response or, in the case of the new targeted therapeutics, via interference with kinase pathways. Although the potency of phospholipid analogs such as edelfosine, perifosine and hexadecylphosphocholine (He-PC) has been established in cancer therapy, the signaling mechanism involved in the initiation and execution of cell death is still unclear. Further, significant efforts have been made to synthesize phospholipid analogs with high antiproliferative capacity but few side effects. The introduction of glucose into the sn-2 position of the glycerol backbone of phosphatidylcholine or the platelet-activating factor (PAF) instead of an ordinary fatty acid generated 2-glucosephosphatidylcholine (Glc-PC) [11] and 1-O-octadecyl-2-Oa-D-glucopyranosyl-sn-glycero-3-phosphatidylcholine (Glc-PAF) [12], respectively. In particular, Glc-PAF showed increased antiproliferative capacity compared to existing APLs, but this compound is still rather cytotoxic. The related 1-O-octadecyl-2-O-(2-(myo-inositolyl)-ethyl)sn-glycero-3-(R/S)-phosphatidylcholine (Ino-C2-PAF) contains the cyclic alcohol inositol in the sn-2 position of the glycerol backbone and was shown to inhibit cell proliferation more powerfully than its glucose-containing analogs or the structurally related glycerol-free compound He-PC. In immortalized keratinocytes, Ino-C2-PAF induced differentiation and low levels of apoptosis [13]. In light of this cancer-specific cytotoxicity profile, we aimed to dissect the signaling events in its regulation of tumor cell apoptosis. To this end, we studied the comparative contributions of the intrinsic, mitochondrial cell death pathway and the extrinsic, death-receptor-activated pathway of apoptosis in two independent cell-line systems: a T-lineage acute lymphoblastic leukemia (T-ALL) cell line (Jurkat) and a Burkitt-like B cell lymphoma cell line (BJAB). In contrast to earlier studies showing that conventional anticancer drugs trigger a CD95/Fas- and FADD-independent

637

mode of apoptosis [14–20], experiments in caspase-8- or FADD-deficient or inhibited cells established that this novel group of alkyl-ether lipids triggers cell death through a caspase-8/FADD-dependent mechanism. However, this mechanism is CD95/Fas ligand- and receptor-independent and coincides with low-level induction of FADD/caspase-8 complexes. In this case, cell death could be inhibited by ectopic expression of Bcl-2 protein, Bcl-2 localized at the mitochondria and Bcl-2 located at the endoplasmic reticulum (ER), indicating an ER- and mitochondrial-dependence of cell death execution.

Results Ino-C2-PAF and Glc-PAF induce apoptosis in Jurkat lymphoma cells MTT proliferation assays revealed that all compounds reduced cell proliferation in malignant lymphoid cells. InoC2-PAF, Glc-PAF, Glc-PC or the alkyl-phosphocholine He-PC significantly inhibited proliferation of Jurkat T-lineage acute T-ALL cells at concentrations of 2.5–20 lM (Fig. 1a). Furthermore, we studied cell death induction in Jurkat cells in order to investigate the mode of growth inhibition induced by different phospholipid analogs. Jurkat cells were treated with Ino-C2-PAF, Glc-PAF, Glc-PC and He-PC in doses ranging from 2.5 to 20 lM. Apoptosis was determined on the single-cell level by flow cytometric measurement of genomic DNA fragmentation. The dose–response curve confirmed a high rate of apoptosis induction by Ino-C2-PAF, with 79% of cells exhibiting apoptotic DNA fragmentation after exposure to 10 lM Ino-C2-PAF for 72 h (Fig. 1b). Glc-PAF also induced apoptotic cell death in Jurkat cells, but was less effective than Ino-C2-PAF at equimolar concentrations. After exposure of Jurkat cells to 10 lM Glc-PAF, 20% of cells exhibited apoptotic DNA fragmentation after 72 h (Fig. 1b). In contrast to Ino-C2-PAF and Glc-PAF, the phosphatidylcholine derivative Glc-PC and the alkylphosphocholine He-PC failed to induce apoptosis in wild type Jurkat cells at doses up to 20 lM. After incubation with 20 lM Glc-PC, only 7.3% of cells exhibited DNA fragmentation, while 9.0% of He-PC-treated cells were apoptotic (Fig. 1b). Thus, the PAF component of the molecule appears to be critical for apoptosis induction. Further, Glc-PC and He-PC seem to exert their antiproliferative effects via non-apoptotic routes in Jurkat cells. To identify the time point when Ino-C2-PAF, the most potent apoptotic substance, starts to yield apoptosis, we treated Jurkat cells with Ino-C2-PAF in doses ranging from 2.5 to 20 lM at 24 and 48 h. The level of apoptotic cell death at 24 and 48 h was rather low compared to 72 h (Fig. 1c).

123

638

B

Ino-C2-PAF He-PC Glc-PAF Glc-PC

A

Ino-C2-PAF He-PC Glc-PAF Glc-PC

100 100

hypodiploid cells (%)

80

cell proliferation (%)

80

60

40

60

40

20

20

0

0 5

0

10

0

20

15

C

5

10

15

20

[µM]

[µM] 100

apoptotic necrotic

80

cell death (%)

Fig. 1 The phospholipid analogs Ino-C2-PAF and GlcPAF induce apoptosis in Jurkat cells. a Jurkat cells were cultured in the presence or absence of Ino-C2-PAF, GlcPAF, He-PC or Glc-PC for 72 h. Cell proliferation was measured using the MTT assay. Mean ± standard deviation from 12 independent experiments are shown. b Cells were cultured in the presence or absence of Ino-C2-PAF, GlcPAF, He-PC or Glc-PC for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages of cells displaying a sub-G1, hypodiploid DNA content are shown. Data shown is the mean ± standard deviations from triplicate experiments. c Cells were cultured in the presence or absence of Ino-C2-PAF for 24 and 48 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages of cells displaying a sub-G1, hypodiploid DNA content are shown. Data shown is the mean ± standard deviations from triplicate experiments

Apoptosis (2011) 16:636–651

60

40

20

.0 20

0

.0 10

5

5.

2.

0 0.

.0 20

0

.0

5.

10

5 2.

0.

0

0

[µM] 24 h

The level of necrotic cell death was around 30% after treatment with 20 lM Ino-C2-PAF for 24 h and remained unchanged after 48 h (Fig. 1c). Ino-C2-PAF-induced apoptotic cell death is dependent on FADD and caspase-8 To address the functional involvement of the death receptor pathway components FADD and caspase-8, we analyzed the induction of apoptotic cell death by Ino-C2-PAF in FADD- or caspase-8-deficient Jurkat cells. Treatment with 20 lM Ino-C2-PAF for 72 h induced apoptotic DNA

123

48 h

fragmentation in 83% of wild type Jurkat cells, but in only 42% of FADD-deficient cells. Similar results were obtained in caspase-8-negative Jurkat cells; loss of caspase-8 reduced the number of cells undergoing genomic DNA fragmentation to 46% (Fig. 2a). To distinguish between viable and apoptotic cells after treatment for 72 h with increasing concentrations of Ino-C2-PAF, we stained the cells with propidiumiodide (PI) to determine permeabilization of the plasma membrane which is a feature of necrotic cell death. The level of necrotic cell death was rather low compared to the induction of apoptotic cells. Treatment with 20 lM Ino-C2-PAF for 24 h induced PI

Apoptosis (2011) 16:636–651 Jurkat wt Jurkat FADD -/Jurkat caspase-8 -/-

A

Jurkat A3 Jurkat FADD -/Jurkat caspase-8 -/-

B

100

100

80

PI positive cells (%)

hypodiploid cells (%)

60

40

20

80

60

40

20

0

0 0

5

10

15

20

0

5

10

15

[µM]

Jurkat

D

c-Flip

58

ß-actin

42

E

100 80 60 40

80

60

40

20

0 0

5

10

15

20

[µM]

20

iu

k

µ

µ

TD

-fm

m

0

ed

IE zF+ A 2P In

oC

hypodiploid cells (%)

100

kDa

c-Flip

mock

hypodiploid cells (%)

C

20

[µM] Jurkat mock Jurkat c-Flip

m

Fig. 2 FADD and caspase-8 are required in Ino-C2-PAFinduced apoptosis. a FADDdeficient (Jurkat-FADD-/-), caspase-8-deficient Jurkat cells (Jurkat-caspase-8-/-) or the parental cell line Jurkat A3 (Jurkat wt) were cultured for 72 h in the presence or absence of Ino-C2-PAF. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. The mean values of triplicate cultures are shown. b FADD-deficient (JurkatFADD-/-), caspase-8deficient Jurkat cells (Jurkatcaspase-8-/-) or the parental cell line Jurkat A3 (Jurkat wt) were cultured in the presence or absence of Ino-C2-PAF for 72 h. Cell viability was measured by staining with PI. Cells which are permeable for PI previous permeabilization are defined as necrotic cells and termed PI positive cells. Mean ± standard deviations of three independent experiments are shown. c Immunoblot for c-Flip of mock- and Jurkat cells overexpressing c-Flip. Similar results were obtained in a second trial of the experiment. ß-actin served as a loading control. d Jurkat cells overexpressing c-Flip (or mockexpressing) were cultured in the presence or absence of Ino-C2PAF for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. e Wild type Jurkat cells were cultured in the presence or absence of the caspase-8 inhibitor z-IETD-fmk (20 lM) for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. The mean values of three different cultures are shown

639

positivity in 42% of wild type cells, in 22% of FADDdeficient cells and in 30% of caspase-8-negative Jurkat cells (Fig. 2b). The overexpression of cellular Flip (c-Flip) (Fig. 2c), a physiological apoptosis inhibitor that interacts with FADD and caspase-8, suppressed Ino-C2-PAFinduced DNA fragmentation, and 83% of the mock-transfected Jurkat cells became apoptotic upon treatment with

20 lM Ino-C2-PAF for 72 h compared to 42% of Jurkat cells overexpressing c-Flip (Fig. 2d). This confirmed the involvement of FADD and caspase-8 in Ino-C2-PAFinduced cell death. Furthermore, Ino-C2-PAF-induced DNA fragmentation could be reduced from 71 to 18% by pretreating wild type Jurkat cells with the caspase-8 inhibitor z-IETD-fmk (Fig. 2e).

123

640

Ino-C2-PAF and Glc-PAF induce CD95/Fas ligand- and receptor-independent formation of a FADD/caspase-8 DISC The functional experiments described above established that cell death induced by Ino-C2-PAF or Glc-PAF is largely dependent on components of the death-receptor pathway, namely FADD and caspase-8. Thus, we further asked whether glycosidated phospholipids interfere with the formation of the DISC, which is physiologically induced by ligand receptor interaction (e.g., by the CD95L/ CD95 receptor system). We analyzed DISC formation by co-immunoprecipitation studies with FADD and caspase-8. As a positive control, wild type Jurkat cells were exposed to CD95/Fas ligand at a concentration of 100 ng/ml for 4, 6 and 12 h. Co-immunoprecipitation analyses of CD95/Fas ligandtreated cells showed almost complete processing of caspase8 (p43/p41 cleavage product) and binding of processed caspase-8 to FADD after 4 h (Fig. 4a). A co-immunoprecipitation of processed caspase-8 with FADD was detected after 6 h but was absent at 12 h. In parallel, wild type Jurkat

123

(%)

100

80

m

Similarly, the loss of either FADD or caspase-8 inhibited the breakdown of the mitochondrial membrane potential (DWm) in Ino-C2-PAF-treated cells. Loss of DWm was evident in 72% of wild type Jurkat cells treated with 10 lM Ino-C2-PAF for 42 h but in only 24% of Jurkat FADDnegative and 10% of caspase-8-negative cells (Fig. 3a). To further analyze the mechanism of cell death induction by Ino-C2-PAF, we determined the extent of cytochrome c release and caspase processing by immunoblotting. As shown in Fig. 3b, Ino-C2-PAF induced cytochrome c release in Jurkat cells 24 h after exposure. This coincided with processing of procaspase-3 to the active p17 subunit and generation of the 37-kDa cleavage product of procaspase-9. Notably, cytochrome c release was associated with caspase-8 processing and Bid cleavage. To address whether these effects were related to death receptor signaling, we also performed these analyses in FADD- and caspase-8deficient Jurkat cells (Fig. 3b). FADD deficiency impaired Bid cleavage and the processing of caspase-8, -3, and -9, indicating that these events are secondary to the formation of the FADD/caspase-8 complex. Likewise, loss of caspase-8 blocks Bid processing and the cleavage of caspase-3 and -9. Notably, loss of FADD or caspase-8 impaired the release of cytochrome c (Fig. 3b). These findings establish that FADD and caspase-8 act upstream of the mitochondrial pathway in Ino-C2-PAF-induced cell death.

Jurkat wt Jurkat FADD -/Jurkat caspase-8 -/-

A

cells with low Δψ

Ino-C2-PAF-induced breakdown of mitochondrial membrane potential, caspase processing and Bid cleavage is dependent on FADD and caspase-8

Apoptosis (2011) 16:636–651

60

40

20

0 0

5

10

15

20

µ

B

wt

FADD-/-

C8-/kDa

FADD

28 55

casp-8

43/41

t-Bid

15

cyt c

14

casp-9

47 37

casp-3

19/17

ß-actin

42 0

5

10

0

5 10

Ino-C2-

0

5

10

µ

Fig. 3 Breakdown of the mitochondrial membrane potential DWm after treatment with Ino-C2-PAF is dependent on FADD and caspase8. a Wild type (Jurkat A3), FADD-deficient (FADD-/-) and caspase-8-deficient (caspase-8-/-) Jurkat cells were cultured in the presence or absence of Ino-C2-PAF for 42 h. Dissipation of DWm was determined by flow cytometric measurement of DWm after loading with the cationic dye JC-1. Cells exhibiting loss of DWm are defined by a decrease in JC-1 red fluorescence. Dose response curves. Mean ± standard deviation from triplicate experiments are shown. b Cytosolic extracts were separated by SDS-PAGE, and the proteolytic processing of BID, caspase-8, -3 and cytochrome c release was detected by immunoblotting after treatment with Ino-C2PAF for 48 h. Similar results were obtained in a second trial of the experiment. ß-actin served as a loading control

cells were treated with Ino-C2-PAF or Glc-PAF for 4–12 h. Caspase-8 processing was observed beginning after 6 h in cultures exposed to Ino-PAF and Glc-PAF and reached a maximum after 12 h. Weak co-immunoprecipitation of processed caspase-8 with FADD was also found. Co-immunoprecipitation analyses of CD95/Fas ligand

Apoptosis (2011) 16:636–651

A

641

medium

InoPAF 20 µM

P S

P

B

GlcPAF FasL 20 µM 100 ng/ml

S P

S P

S

kDa 55 43 41

4h

medium

InoPAF 20 µM

P S

P

24

12 h

24

D

43 41 20

FasLigand expression (x fold rel. to 18S rRNA)

80 60 40 20

15

ed iu 0n m + In FA g/m oC Sb l 2PA lo ck F + FA Sb lo ck

0

FA

SL

SL

m 5

Ju ed rk T c at e m lls ed iu m

µ

F

µ

Jurkat wt Jurkat Fas -/100

Jurkat

E wt

Fas-/-

Fas

kDa 43

ß-actin

42

hypodiploid cells (%)

at tiv

10

µ

10

0

ac

Jurkat wt

100

FA

C

S

6h

55

12 h

S P

kDa 24

hypodiploid cells (%)

43 41

S P

4h

55

6h

GlcPAF FasL 20 µM 100 ng/ml

80

60

40

20

µ

Fa

slig

an

d

10

m

0

ed

ng

iu

/m

m

l

0

treated cells showed that binding of FADD to caspase-8 was not detectable (Fig. 4b). Jurkat cells were treated with InoC2-PAF or Glc-PAF for 4–12 h and FADD binding to caspase-8 was observed after 6 h (Fig. 4b). Thus, both coimmunoprecipitation studies revealed a DISC formation that was both delayed and attenuated compared to cells exposed to the natural CD95/Fas ligand. To clarify whether this was related to a low rate of induction of the CD95/Fas

ligand by Ino-C2-PAF or Glc-PAF, quantitative RT-PCR was employed along with a functional approach. Activated peripheral blood T-cells that were restimulated by crosslinking of the CD3e-chain [21] served as a positive control for the induction of CD95/Fas ligand expression. As shown in Fig. 4c, exposure of Jurkat cells to Ino-C2-PAF or GlcPAF failed to induce CD95/Fas ligand mRNA expression at all doses (up to 20 lM Ino-C2-PAF and 40 lM Glc-PAF)

123

642 b Fig. 4 Ino-C2-PAF- and Glc-PAF-induced formation of the FADD/

caspase-8 complex is independent of CD95/Fas ligand expression, death ligand/receptor interaction and the presence of CD95/Fas receptor. a Jurkat cells were treated in the presence or absence of InoC2-PAF or Glc-PAF. The CD95/Fas ligand served as a positive control for FADD/caspase-8 death-inducing signaling complex (DISC) formation. After incubation for 4, 6, or 12 h, cells were solubilized in Triton X-100 buffer and lysates were immunoprecipitated with antiFADD antibody. Immune complexes were resolved by SDS-PAGE and the presence of caspase-8 was detected by immunoblotting, as indicated by arrows. Similar results were obtained in three separate trials of the experiment. S supernatant, P precipitate. b Jurkat cells were treated in the presence or absence of Ino-C2-PAF or Glc-PAF. The CD95/Fas ligand served as a positive control for FADD/caspase-8 death-inducing signaling complex formation. After incubation for 4, 6, or 12 h, cells were solubilized in Triton X-100 buffer and lysates were immunoprecipitated with anti-caspase-8 antibody. Immune complexes were resolved by SDS-PAGE and the presence of FADD was detected by immunoblotting, as indicated by arrows. Similar results were obtained in three separate trials of the experiment. S supernatant, P precipitate. c Wild type Jurkat cells (Jurkat A3) were cultured in the presence or absence of Ino-C2-PAF or Glc-PAF for 24 h. Detection of CD95/Fas ligand by quantitative real-time PCR. Activated T cells served as a positive control. d Blocking of CD95/Fas ligand-induced apoptosis. Cells were incubated in the presence or absence of recombinant CD95/ Fas ligand and in the absence (-Fas block) or presence (?Fas block) of a blocking anti-CD95 antibody. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation after 72 h. The percentages of cells displaying a sub-G1, hypodiploid DNA content are shown. Data is presented as the mean ± standard deviation from triplicate experiments. e Immunoblot for CD95/Fas of wt and Fas deficient Jurkat cells. Similar results were obtained in a second trial of the experiment. ß-actin served as a loading control. f Fas-deficient Jurkat cells (Jurkat Fas-/-) or the vector control (Jurkat mock) were incubated in the presence of Ino-C2-PAF for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. The mean values for triplicate cultures are shown

(Fig. 4c). To exclude the participation of CD95/Fas ligand involvement in Ino-C2-PAF-induced apoptosis, we inhibited a putative CD95/Fas receptor/ligand interaction by blocking with an anti-CD95 antibody. To verify the efficacy of the block, we performed a control experiment in which we treated Jurkat cells with CD95/Fas ligand in the presence or absence of the blocking anti-CD95 antibody. CD95/Fas ligand-induced apoptosis was reduced from 46 to 13% in Jurkat cells when the blocking anti-CD95 antibody was added. In contrast, apoptosis was actually enhanced by the blocking antibody in the presence of Ino-C2-PAF (Fig. 4d). We then analyzed DNA fragmentation in Fas-deficient Jurkat cells (Fig. 4e) in comparison to the vector control in order to study the role of the CD95/Fas receptor in Ino-C2PAF induced apoptosis. After a 72 h incubation with 20 lM Ino-C2-PAF, similar induction of apoptotic cell death was observed in Jurkat Fas-deficient cells compared to the wild type (Fig. 4f). Together, these analyses suggest that Glc-PAF and Ino-C2-PAF induce low rates of DISC formation through a CD95-ligand and receptor-independent mechanism.

123

Apoptosis (2011) 16:636–651

Expression of Bcl-2, targeted at the mitochondria or the ER, inhibited Ino-C2-PAF-induced apoptotic cell death Overexpression of Bcl-2 inhibits apoptosis after exposure to a number of drugs [16, 18]. To clarify the mechanism of Ino-C2-PAF- and Glc-PAF-induced apoptosis, Jurkat cells stably expressing different versions of Bcl-2 were used. The induction of apoptosis was determined by measurement of genomic DNA fragmentation and was compared to the apoptosis in mock-transfected cells 72 h after drug treatment. The generation of hypodiploid, sub-G1 cells was completely inhibited in Bcl-2-wt-, Bcl-2-Mito- and Bcl-2ER-overexpressing Jurkat cells relative to mock-transfected cells following exposure to Ino-C2-PAF. Ino-C2-PAF induced apoptosis in 72% of cells at 20 lM after 72 h, whereas only 4% of the Bcl-2-wt, 4% of Bcl-2-Mitooverexpressing and 16% of the Bcl-2-ER-overexpressing Jurkat cells were apoptotic (Fig. 5a). Bcl-2 interferes with the release of proapoptotic proteins from mitochondria in response to proapoptotic stimuli. Therefore, we analyzed dissipation of the mitochondrial membrane potential (loss of DWm) in Ino-C2-PAF-treated Bcl-2 overexpressing Jurkat cells after 42 h. The mitochondrial permeability transition was determined by flow cytometry based on JC-1 red fluorescence. Figure 5b shows that Bcl-2 completely blocked the mitochondrial permeability transition in cells treated with Ino-C2-PAF for 42 h. Breakdown of DWm was evident in 63% of the mock-transfected Jurkat cells upon cultivation with 20 lM of Ino-C2-PAF, compared with only 6% of the Bcl-2-wt, 8% of the Bcl-2-Mito and 12% of the Bcl-2-ER-transfected cells (Fig. 5b). Overexpression of Bcl-2, localized at the mitochondrion or the ER impaired the processing of caspase-8 and Bid, and prevented the release of cytochrome c from the mitochondria as well as the processing of caspase-3 (Fig. 5c), indicating that caspase activation both upstream and downstream of the mitochondria is necessary for effective Ino-C2-PAFinduced apoptotic cell death. Glc-PAF-induced FADD- and caspase-8-dependent cell death that is inhibited by Bcl-2 To determine whether this model of FADD/caspase-8triggered activation of a mitochondrial apoptosis pathway is a general feature of glycosidated PAF analogs, we analyzed the requirements for Glc-PAF-induced apoptosis in the Jurkat model. As shown in Fig. 6a, loss of FADD reduced apoptotic DNA fragmentation from 81 to 30% after 72 h in cells treated with 20 lM of Glc-PAF. Under the same conditions, loss of caspase-8 reduced apoptosis to 19%. Apoptotic DNA fragmentation was observed in 48% of mock-transfected Jurkat cells after treatment with

Apoptosis (2011) 16:636–651

643

B

Jurkat mock Jurkat Bcl-2 Jurkat Bcl-2 Mito Jurkat Bcl-2 ER

A

100

cells with low Δψ m (%)

100

hypodiploid cells (%)

Jurkat mock Jurkat Bcl-2 wt Jurkat Bcl-2 Mito Jurkat Bcl-2 ER

80

60

40

20

80

60

40

20

0

0 0

10

5

15

0

20

Ino-C2-PAF [µM]

C

mock

Bcl2 wt

5

10

15

20

Ino-C2-PAF [µM]

Bcl2 Mito

Bcl2 ER

kDa 26

Bcl-2

55 casp-8

43/41

t-BID

15

cyt-c

14

casp-3

19 17

ß-actin

42 0

5 10

0

5 10

0

5 10

0

5 10

Ino-C2-PAF [µM] Fig. 5 Bcl-2 overexpression inhibits apoptosis, breakdown of mitochondrial membrane potential and caspase activation upon treatment with Ino-C2-PAF. a Jurkat cells overexpressing Bcl-2 ectopically (Bcl-2 wt), at the mitochondria (Bcl-2-Mito), at the ER (Bcl-ER) and mock-transfected Jurkat cells (Bcl-2-mock) were cultured in the presence or absence of Ino-C2-PAF for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. The mean values from three different cultures are shown. b Jurkat cells overexpressing Bcl-2 ectopically (Bcl-2 wt), at the mitochondria (Bcl-2-Mito), at the ER (Bcl-ER) and mock-transfected Jurkat cells (Bcl-2-mock) were cultured in the presence or absence of Ino-C2-PAF for 42 h. Dissipation of DWm was determined by flow

cytometric measurement of DWm following loading with the cationic dye JC-1. Cells with a loss of DWm are defined by a decrease in JC-1 red fluorescence. c Immunoblot of Bcl-2, t-Bid, cytochrome c and caspase-8 and -3 processing. Jurkat cells overexpressing Bcl-2 ectopically (Bcl-2 wt), at the mitochondria (Bcl-2-Mito), at the ER (Bcl-ER) and mock-transfected Jurkat cells (Bcl-2-mock) were cultured in the presence or absence of Ino-C2-PAF for 48 h with or without Ino-C2-PAF. Cytosolic extracts were separated and subjected to immunoblotting. Molecular weights are indicated at right. Similar results were obtained from two independent experiments. ß-actin served as a loading control

20 lM of Glc-PAF, and in 81% of cells after treatment with 40 lM Glc-PAF. In comparison, 40 lM Glc-PAF induced apoptotic DNA fragmentation in only 8% of Bcl2-wt, 14% of Bcl-2-ER and 22% of Bcl-2-Mito cells (Fig. 6b).

Inhibition of Ino-C2-PAF- and Glc-PAF-induced apoptosis by Bcl-xL in Jurkat cells To investigate whether Bcl-xL would also prevent Ino-C2-PAF- and Glc-PAF-induced apoptotic cell death,

123

644

Apoptosis (2011) 16:636–651

B Jurkat wt Jurkat FADD-/Jurkat caspase-8 -/-

Jurkat mock Jurkat Bcl-2 wt Jurkat Bcl-2 Mito Jurkat Bcl-2 ER

100

100

80

80

hypodipoid cells (%)

hypodiploid cells (%)

A

60

40

60

40

20

20

0

0 0

10

20

30

40

0

Glc-PAF [µM]

10

20

30

40

Glc-PAF [µM]

Fig. 6 Glc-PAF-induced apoptotic cell death is dependent on FADD and caspase-8 and is inhibited by Bcl-2. a Wild type (Jurkat A3), FADD-deficient (FADD-/-) and caspase-8-deficient (C8-/-) Jurkat cells were cultured in the presence or absence of Glc-PAF for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages of cells displaying a subG1, hypodiploid DNA content are shown. Data presented are the mean ± standard deviation from three independent experiments.

b Jurkat cells overexpressing Bcl-2 ectopically (Bcl-2 wt), at the mitochondria (Bcl-2-Mito), at the ER (Bcl-2-ER) and mock-transfected Jurkat cells (Bcl-2-mock) were cultured in the presence or absence of Glc-PAF for 42 h. Dissipation of DWm was determined by flow cytometric measurement of DWm following loading with the cationic dye JC-1. Cells with DWm loss are defined by a decrease in JC-1 red fluorescence. Mean ± standard deviations from three independent experiments are shown

Bcl-xL-overexpressing Jurkat and mock cells (Fig. 7a) were incubated with Ino-C2-PAF and Glc-PAF, respectively. Subsequently, DNA fragmentation was determined after 72 h. Compared to Bcl-2-overexpressing Jurkat cells, Jurkat cells overexpressing Bcl-xL showed higher levels of Ino-C2-PAF-induced apoptosis (Fig. 7b). Similar results were observed after treatment with Glc-PAF (Fig. 7c).

were refractory to Ino-C2-PAF-induced apoptosis. Apoptotic DNA fragmentation was observed in 27% of mock-transfected BJAB cells, but only in 13% of Bcl-xLoverexpressing BJAB cells, after treatment with 20 lM Ino-C2-PAF for 72 h (Fig. 8c).

Discussion Ino-C2-PAF-induced cell death in BJAB cells is dependent on FADD and caspase-8 and is inhibited by Bcl-xL To investigate whether these effects were cell-type specific, we employed a second, independent lymphoid cell line (the Burkitt-like lymphoma line, BJAB). To corroborate the dependency of Ino-C2-PAF-induced apoptosis on proteins in the death receptor pathway, we analyzed BJAB cells that stably expressed a dominant negative FADD mutant (FADD-DN) or cowpox cytokine response modifier A (crmA). Apoptotic DNA fragmentation was observed in 27% of mock-transfected BJAB cells after treatment with 20 lM Ino-C2-PAF for 72 h, compared to 8% of FADDDN cells and 11% of crmA-expressing cells (Fig. 8a). As in Jurkat cells, Bcl-xL-overexpressing BJAB cells (Fig. 8b)

123

In preclinical and clinical trials it has been demonstrated that alkyl-phospholipids possess impressive antiproliferative activity [8, 9, 22, 23]. In our experiments we could confirm that He-PC, Glc-PC, Glc-PAF and Ino-C2-PAF displayed good antiproliferative properties. As previously shown cytotoxic effects of Glc-PC and Glc-PAF was observed at least at 30 lM [24] and has also been described for He-PC in HaCaT cells [25]. In our experimental system the antiproliferative effects are already observed at concentrations where cytotoxic effects do not occur. We therefore assume that Glc-PC and He-PC exert their antiproliferative effects via non-apoptotic routes like autophagy [26]. In this study we could show that a novel subfamily of this group, the glycosidated phospholipids, exerts its activity by inducing apoptosis in two different

Apoptosis (2011) 16:636–651

645

A mock

kDa

Bcl-xL

Bcl-x L

30

ß-actin

42

80

hypodiploid cells (%)

100

hypodiploid cells (%)

100

80

Jurkat mock Jurkat Bcl-xL

C

Jurkat mock Jurkat Bcl-xL

B

Jurkat

60

40

20

60

0 0

5

10

15

20

Ino-C2-PAF [µM]

40

20

0 0

10

20

30

40

Glc-PAF [µM] Fig. 7 Ino-C2-PAF- and Glc-PAF-induced cell death in Jurkat cells can be inhibited by Bcl-xL. Overexpression of Bcl-xL in Jurkat cells suppressed Ino-C2-PAF and Glc-PAF-mediated DNA fragmentation. a Immunoblot for Bcl-xL of mock- and Jurkat cells overexpressing Bcl-xL. Similar results were obtained in a second trial of the experiment. ß-actin served as a loading control. b Bcl-xL-transfected Jurkat and mock-transfected cells were cultured for 72 h in the presence or absence of Ino-C2-PAF. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages

of cells displaying a sub-G1, hypodiploid DNA content are shown. Data presented are the mean ± standard deviation from three independent experiments. c Bcl-xL-transfected Jurkat cells and mock-transfected cells were cultured for 72 h in the presence or absence of Glc-PAF. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages of cells displaying a sub-G1, hypodiploid DNA content are shown. Data presented are the mean ± standard deviation from three independent experiments

leukemia cell line systems. Notably, both Ino-C2-PAF and Glc-PAF trigger apoptosis by induction of proteins of the death receptor pathway that requires secondary activation of the mitochondrial, intrinsic apoptosis-signaling machinery. This mechanism is in strong contrast to conventional anticancer agents that initiate a death receptorindependent mode of apoptosis signaling [16, 18]. The presented data demonstrate that the activation of Ino-C2-PAF- and Glc-PAF-induced cell death requires FADD and caspase-8. In fact, in FADD-/- or caspase8-/- Jurkat cells incubated with glycosidated phospholipids the breakdown of the mitochondrial membrane potential followed by the release of cytochrome c out of the mitochondria, the activation of caspases and the apoptotic DNA fragmentation was inhibited. Nonetheless, FADD- or caspase-8 deficient cells could not entirely rescue Ino-C2PAF-induced apoptosis indicating that other proteins might be activated in this process. Therefore, we propose that

redundant proteins like caspase-10, caspase-2 or unknown proteins are involved and could provide to eliminate cells by apoptosis in the absence of caspase-8. Ino-C2-PAFinduced cell death is also supported by the fact that Jurkat cells overexpressing the cell-intrinsic caspase-8 inhibitor c-FLIP or treated with the caspase-8 inhibitor z-IETD-fmk are protected against apoptosis. As already demonstrated for edelfosine, Ino-C2-PAFinduced apoptosis is independent of CD95/Fas death-ligand expression and CD95/Fas death-ligand/receptor interaction [27, 28]. However, Ino-C2-PAF promotes cell death in the absence of CD95/Fas receptor and it is clearly in contrast to the mechanism of action of edelfosine, which induces apoptosis through intracellular activation of CD95/Fas receptor [28]. Consequently, the apoptotic activity of InoC2-PAF is characterized by the formation of an atypical DISC. This was confirmed by co-immunoprecipitation analysis, where caspase-8 is associated with FADD

123

646

Apoptosis (2011) 16:636–651

A

BJAB mock BJAB FADD-DN BJAB crmA

hypodiploid cells (%)

50

B

BJAB mock

Bcl-xL

40

ß-actin

Bcl-xL

kDa 30 42

30 20 10 0 0

5

10

15

20

µ BJAB mock BJAB Bcl-xL

C

hypodiploid cells (%)

50 40 30 20 10 0 0

5

10

15

20

µ Fig. 8 Ino-C2-PAF-induced cell death in BJAB cells is dependent on FADD and caspase-8 and can be inhibited by Bcl-xL. Inhibition of caspase-8, overexpression of dominant negative FADD (FADD-DN) and overexpression of Bcl-xL in BJAB cells suppressed Ino-C2-PAFmediated DNA fragmentation. a BJAB mock, BJAB crmA and BJAB FADD-DN cells were cultured for 72 h in the presence or absence of Ino-C2-PAF. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages represent cells with hypodiploid DNA. Data presented are the mean ± standard deviation from three independent experiments. b Immunoblot for Bcl-xL of mock- and BJAB cells overexpressing Bcl-xL. Similar results were obtained in a second trial of the experiment. ß-actin served as a loading control. c Bcl-xL-transfected BJAB cells and mock-transfected cells were cultured in the presence or absence of Ino-C2-PAF for 72 h. Apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation. Percentages of cells displaying a subG1, hypodiploid DNA content are shown. Data presented are the mean ± standard deviation from three independent experiments

excluding death domain containing CD95/Fas. In addition, we show that Jurkat cells deficient in Fas expression were not protected against Ino-C2-PAF-induced cell death. The atypical DISC has recently been proposed for Yersinia outer protein P (YopP)-induced cell death in dendritic cells, which is independent of death domain containing receptors, but mediated by caspase-8 at the level of DISC [29]. Remarkably, synthetic phospholipids analogs like edelfosine, perifosine and erucylphosphocholine (ErPC)

123

trigger cell death through alternative mechanisms. Overall, it appears that lipid rafts are critical membrane portals for the cellular entry and anticancer activity of alkylphospholipids [30]. Edelfosine and perifosine induce the formation of lipid raft clusters, which recruit DISC components and mediate apoptosis in a Fas/CD95 death receptordependent way [28, 31, 32]. Furthermore, in contrast to our data, ErPC-induced cell death is independent from FADD and caspase-8 since caspase-8- and FADD-negative Jurkat cells, as well as BJAB cells expressing FADD-DN, were sensitive to the apoptotic action of ErPC [33]. Moreover, overexpression of Bcl-2 inhibits apoptosis after exposure to Ino-C2-PAF and Glc-PAF. Expression of Bcl-2 at the ER or the mitochondria inhibited Ino-C2-PAF-induced loss of the mitochondrial membrane potential, cytochrome c release, caspase-8- and -3 processing, Bid cleavage and DNA fragmentation. This results points to a relevant effect of the mitochondria in apoptosis signaling after Ino-C2-PAF treatment, as it has been shown for other synthetic phospholipids [5, 23, 28, 32]. In our study, the expression of Bcl2 at the ER inhibited Ino-C2-PAF-induced caspase-3 and -8 activation as well as Bid cleavage and DNA fragmentation with almost similar potency as Bcl-2-wt and Bcl-2 localized at the mitochondria. This results points to a relevant effect of the ER in apoptosis signaling after Ino-C2-PAF treatment. Bcl-2 localized at the ER completely blocked InoC2-PAF- and Glc-PAF-induced loss of mitochondrial membrane potential, which indicates a molecular interaction between the ER and the mitochondria upstream of the change in mitochondrial membrane potential in InoC2-PAF-induced apoptosis. Recently, it has been shown that ER-targeted Bcl-2 interferes with ErPC-induced mitochondrial damage, suggesting that the ER is involved in apoptosis signaling upstream of the mitochondria and that there is a crosstalk between both organelles [33]. Nieto-Miguel and coworkers have shown that edelfosine induces ER stress in solid tumor cells and that Bcl-xL-overexpression as well as Bcl-2 localization to the mitochondria can inhibit edelfosineinduced cell death [34]. Chandra et al. demonstrated that active caspase-8 becomes associated with the membranes in apoptosis caused by multiple stimuli. In MDA-MB-231 breast cancer cells treated with etoposide (VP16), active caspase-8 is detected only in the membrane fraction of both mitochondria and ER, as revealed by fractionation studies [35]. After ER stress, such as treatment with tunicamycin, the ER-bound protein RTN3 recruited endogenous FADD to the ER membrane and subsequently initiated the caspase-8 cascade [36]. Upton and collegues found that ER stress induces proteolytic activation of the BH3-only protein Bid as a critical apoptotic switch. Further, they identified caspase-2 as the premitochondrial protease that cleaves Bid in response to ER stress [37].

Apoptosis (2011) 16:636–651

This could explain the CD95/Fas death receptor-independent activation of FADD and caspase-8 at the membrane of the ER. In addition to its generally accepted role in membrane-bound DISC formation, FADD may also play a role in the transduction of apoptotic signals in cellular compartments other than the plasma membrane [36]. How Bcl-2 blocks Ino-C2-PAF-induced caspase-8 activation and Bid cleavage is not clear. Similar results were described in TRAIL-induced apoptosis. The authors proposed that soluble forms of the Bcl-2 protein suppressed the activation of caspase-8 or that Bcl-2 blocks the recruitment of caspase-8 to the receptor [38]. Here we suppose that the induction of effective apoptosis accompanied by a detectable level of processed caspase-8 and Bid require amplification loops via caspases downstream of the mitochondrion. The extent of cell death induction is dependent on the cell type. Our data showed that, in contrast to Jurkat cells, only part of the growth inhibition induced by Ino-C2-PAF in BJAB cells was caused by induction of apoptosis. A reason could be the expression of the CD40 ligand on BJAB cells, which induces an autocrine antiapoptotic signal when cells are exposed to cytotoxic drugs [39]. The cell type-dependent activation of apoptosis by alkyl-phospholipids has also been shown for leukemic versus solid tumor cells [40]. Taken together, we have demonstrated that Ino-C2-PAF and Glc-PAF, although distinguished by the formation of an atypical DISC, induced FADD- and caspase-8-dependent apoptosis in Jurkat cells. These findings should open up attractive opportunities for therapeutic applications of phospholipid analogs like our glycosidated phospholipids, especially due to their membrane targeting properties, in combination with conventional anti cancer therapeutics, can potentiate apoptotic cell death preferentially in tumor cells [41–43]. Further, we demonstrate that Bcl-2 potently inhibits Ino-C2-PAF- and Glc-PAF-induced apoptosis (Fig. 9). However, it is still unclear whether the ER is activated and cooperates specifically with the mitochondrial apoptotic pathway during Ino-C2-PAF- and Glc-PAFmediated apoptosis, as previously shown for edelfosine [34]. The details of the molecular mechanism underlying this process remain to be elucidated.

647

Ino-C2-PAF plasma membrane

FADD c-Flip

ER

pro-caspase 8 active caspase-8 z-IETD-fmk

Bcl-2

Bid mitochondrion amplification loops

ΔΨm cytochrome c active caspase-9

active caspase-3

apoptotic cell death

Fig. 9 A model for CD95-ligand- and receptor-independent activation of FADD and caspase-8 by Ino-C2-PAF and Glc-PAF. Treatment of cells with Ino-C2-PAF and Glc-PAF leads to the activation of FADD (Fas-associated via death domain) and caspase-8 which triggers apoptosis via Bid (BH3-interacting-domain death agonist) cleavage and the mitochondrial dependence cell death pathway. This process is inhibited by c-FLIP and z-VAD-fmk. This pathway leads to disruption of the outer mitochondrial membrane, resulting in the release of cytochrome c. This in turn promotes activation of caspase9, followed by the activation of effector caspase-3. The breakdown of mitochondrial membrane potential and consequent DNA fragmentation is strongly inhibited by the overexpression of Bcl-2 localized both at the mitochondrion or the ER. The induction of effective apoptosis by Ino-C2-PAF and Glc-PAF requires several amplification loops via caspases (dashed arrows). Some unknown and maybe redundant proteins have to be involved, because lack of FADD cannot completely protect cells against apoptotic cell death

in turn leads to activation of the mitochondrial cell death pathway via BID. The Ino-C2-PAF-induced apoptotic cell death in Jurkat cells can be completely inhibited by the overexpression of Bcl-2 located at the ER and the mitochondria. Bcl-xL also prevents apoptosis in the presence of Ino-C2-PAF, but it is less effective than Bcl-2.

Conclusion Materials and methods Synthetic glycosidated phospholipids induce a CD95/Fas ligand- and receptor-independent atypical death inducing signaling pathway. The present study provides experimental evidence that Ino-C2-PAF- and Glc-PAF-induced apoptosis proceeds via activation of members of the deathinducing signaling complex, FADD and Caspase-8, which

Cell culture Wild type Jurkat A3 cells were made neomycin resistant and treated with three cycles of exposure to the frameshifting mutagen ICR-191 to isolate clones harboring

123

648

recessive mutations that were resistant to killing by Fas antibody. ICR-191 treated clones were serially diluted in 96-well plates in the presence of Fas Antibody for 3–5 weeks. They are a clone with a mutation in the cysteine protease caspase-8/FLICE and clone with a mutation in the adaptor FADD. The caspase-8-/- cell line is functionally defective for caspase-8 and the Jurkat FADD-/- cell line is functionally defective for FADD [44, 45]. The cell lines are completely resistant to Fas-induced apoptosis and generated by J. Blenis, Harvard Medical School, Boston, USA. Stably transfected Jurkat cells overexpressing c-Flip (Jurkat c-Flip), mock-transfected cells (pcDNA3-mock) were kindly provided by C. Belka, University of Munich, Germany, as were Jurkat cells deficient for CD95/Fas receptor (Jurkat Fas-/-), as well as mock-transfected and stably Bcl-xL (prcCMV-Bcl-xL) transfected Jurkat cells. Jurkat cells stably ectopic expressing of Bcl-2 (Bcl-2-wt), localized at the mitochondrion (Bcl-2-Mito), localized at the ER (Bcl-2 ER) and the respective vector control (Jurkat-mock) were kindly provide by C. Belka and described previously [46]. Mocktransfected BJAB cells (pcDNA3-mock) or BJAB cells stably expressing the crmA (pcDNA3-crmA) or a dominant negative FADD mutant lacking the N-terminal death effector domain (pcDNA3-FADD-DN) were provided by K. Schulze-Osthoff, University of Du¨sseldorf, Germany and described before [16, 18]. Bcl-xL overexpressing (pcDNA3-Bcl-xL) BJAB cells and corresponding mocktransfected cells (pcDNA3-mock) were obtained from S. Fulda, University of Ulm, Germany. All cells were grown in RPMI 1640 medium supplemented with 2 mM glutamine, 10% fetal calf serum (FCS), 106 U/l penicillin and 0.1 g/l streptomycin at 37°C with 5% CO2 in a fully humidified atmosphere. Media and culture reagents were purchased from Invitrogen (Karlsruhe, Germany). Antibodies and reagents Polyclonal rabbit anti-human actin antiserum was purchased from Sigma-Aldrich (Taufkirchen, Germany) and used at a 1:100 dilution. Polyclonal goat anti-human caspase-9 and anti-human caspase-3 were obtained from R&D Systems (Wiesbaden-Nordenstadt, Germany) and used at 1:1,000 or 1:2,000 dilutions, respectively. Monoclonal murine anti-human cytochrome c antibody (Clone 7H8.2C12, IgG2b) was obtained from BD-Bioscience (Heidelberg, Germany) and used at a dilution of 1:500. The CD95/Fas-receptor-blocking monoclonal antibody (antiAPO-1/Fas; clone SM1/23, IgG2b) was purchased from Bender MedSystems (Vienna, Austria). Monoclonal murine anti-human caspase-8 hybridoma supernatant was provided by K. Schulze-Osthoff and was used at 1:20.

123

Apoptosis (2011) 16:636–651

Rabbit human anti-FADD antibody was purchased from Upstate Biotechnology, Inc. (Charlottesville, USA) and was used at a concentration of 4 lg/ml in immunoprecipitation assays. Polyclonal rabbit anti-human Bid antibody was purchased from BD Bioscience and used at a dilution of 1:500. All antibodies used for immunoblotting were diluted in Western blocking reagent from Roche (Mannheim, Germany). Secondary anti-rabbit, anti-goat and antimouse horseradish peroxidase (HRP)-conjugated antibodies were obtained from Southern Biotechnology Associates (Birmingham, USA) and used at a concentration of 1:50,000. Caspase-8 inhibitor (z-IETD-fmk) was obtained from Calbiochem (Darmstadt, Germany) and used at a concentration of 20 lM. Recombinant CD95/Fas ligand was purchased from Axxora (Gru¨nberg, Germany). RNase A was obtained from Roth (Karlsruhe, Germany). Propidium iodide (PI) was purchased from Sigma-Aldrich. InoC2-PAF, Glc-PAF, Glc-PC and He-PC were dissolved in distilled water for delivery. Measurement of apoptotic cell death by flow cytometry Apoptotic DNA fragmentation was determined at the single-cell level by measuring the DNA content of individual cells [47]. After 72 h of drug treatment, Jurkat cells were collected by centrifugation at 3009g for 5 min and washed with PBS at 4°C. Cells were fixed in PBS/0.75% (v/v) formaldehyde on ice for 30 min, pelleted, incubated with ethanol/PBS (2:1, v/v) for 15 min, pelleted and resuspended in PBS containing 40 lg/ml RNase A. RNA was digested for 30 min at 37°C. Cells were pelleted again and finally resuspended in PBS containing 50 lg/ml PI. DNA fragmentation was quantified by flow cytometric determination of hypodiploid DNA content (sub G1 population). Data were collected and analyzed using a FACScan flow cytometer (Becton-Dickinson; Heidelberg, Germany) equipped with CellQuestTM software. Cell debris was excluded by gating the cells using the FSC threshold. To distinguish between clumped and mitotic cells, a doublet discrimination module was employed, thereby excluding doublets. Data are given in % hypodiploid cells (subG1), which reflects the proportion of apoptotic cells. Quantification of necrotic cell death by flow cytometry The determination of vital and death cells can be detected by evaluation of cellular viability. Living cells have intact cell membranes and an active cell metabolism that exclude PI, while cells with damaged cell membranes or impaired metabolism allow PI to enter the cell. Cells were treated with different concentrations of Ino-C2-PAF or Glc-PAF for 24 h, or left untreated. Cells were harvested by incubation with 4 mM of EDTA for 10 min at 37°C and

Apoptosis (2011) 16:636–651

subsequent by centrifugation at 3009g for 5 min. Cells were then washed with PBS and resuspended in 200 ll PBS containing 50 lg/ml PI. Data were collected and analyzed using a FACScan flow cytometer equipped with CellQuestTM software (Becton-Dickinson). Data are given in PI positive % of cells which reflects the number of necrotic cells. Measurement of cell proliferation The MTT [3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide] assay is based on the ability of a mitochondrial dehydrogenase enzyme produced in viable cells to cleave the tetrazolium rings of the pale yellow MTT and form a dark blue formazan crystals, which are largely impermeable to cell membranes and thus accumulate in healthy cells. Solubilization of the cells by the addition of a detergent results in the liberation of the crystals, which are then solubilized. The number of surviving and proliferating cells is directly proportional to the concentration of formazan. The cytotoxic and antiproliferative effect of Ino-C2-PAF, Glc-PAF, He-PC and Glc-PC on Jurkat cells was measured by this assay. Cells were dispensed in 96-well flat bottom microtiter plates at a density of 1 9 104 cells per well and treated with the indicated concentrations of Ino-C2-PAF and Glc-PAF for 72 h. Cells were then incubated with MTT (0.5 mg/ml) for 4 h at 37°C. The formazan crystals in the cells were solubilized with a solution containing 0.04 M HCl in absolute isopropanol. The level of MTT formazan was determined by measuring absorbance at 490 nm with a Infinite 200 reader (TECAN, Salzburg, Austria). Measurement of the mitochondrial permeability transition The mitochondrial permeability transition was determined by staining the cells with the cationic dye 5,50 ,6,60 -tetrachloro-1,10 ,3,30 -tetraethyl-benzimidazolylcarbocyanin iodide (JC-1; Molecular Probes, Leiden, Netherlands) as previously described [20]. The cells (1 9 105) were resuspended in 500 ll phenol-red-free RPMI 1640 and JC1 was added to a final concentration of 2.5 lg/ml. The cells were grown for 30 min at 37°C with moderate shaking; control cells were incubated in the absence of JC-1 dye. The cells were then harvested by centrifugation at 3009g for 5 min, washed with PBS and resuspended in 200 ll PBS at 4°C. The mitochondrial permeability transition was then quantified by flow cytometric determination of cells with decreased red fluorescence as measured in the FL-2 channel (550–650 nm), i.e., with mitochondria displaying a low membrane potential (DWm). Data were collected and analyzed using a FACScan flow cytometer

649

equipped with CellQuestTM software. Data are given in % of cells with low DWm, which reflects the proportion of cells undergoing mitochondrial apoptosis. Immunoblotting Cells were washed twice with PBS and lysed in buffer containing 10 mM Tris–HCl pH 7.5, 300 mM NaCl, 1% Triton X-100, 2 mM MgCl2, 5 lM EDTA, and the Complete Mini protease inhibitor cocktail (Roche). Protein concentration was determined using the bicinchoninic acid assay (BCA) from Pierce (Rockford, USA) and 20 lg of protein was separated on an SDS-PAGE gel [17]. Subsequently, blotting of proteins onto nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany) was performed as previously described [47]. Quantitative real-time PCR for CD95/Fas ligand RNA was isolated using the RNeasy kit (Qiagen, Hilden, Germany) and 250 ng of total RNA was reverse transcribed as described previously [48]. PCR and detection were performed in triplicate in a 25 ll reaction mix on an ABI Prism 7000 system (Applied Biosystems, Foster City, USA) with 18S rRNA used as an internal reference. The expression of CD95/FasL mRNA is expressed as a ratio to expression of 18S rRNA. Analysis of DISC formation by immunoprecipitation Cells (1 9 107) were mock treated, treated with compound or treated with 100 ng/ml CD95/Fas ligand as a positive control for DISC formation. After incubation for 4, 6, and 12 h, cells were washed twice with ice-cold PBS. Cell pellets were incubated in 500 ll lysis buffer (30 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1% Triton X-100, 10% [v/v] glycerol, and a protease inhibitor cocktail (Complete Mini, Roche)) for 30 min. For immunoprecipitation of FADD, 30 ll protein A-Sepharose beads and 4 lg anti-FADD antibody were incubated for 1 h, followed by addition of 500 ll of the cell lysate. Immunoprecipitation was performed overnight at 4°C. Protein A-Sepharose beads were then washed five times with ice-cold lysis buffer and boiled for 5 min in SDS-loading buffer. For co-immunoprecipitation of caspase-8 supernatants were separated by SDS-PAGE and followed by immunoblotting of caspase-8. For immunoprecipitation of caspase-8, 30 ll protein G-Sepharose beads and 4 lg anti-caspase-8 antibody were incubated for 1 h, followed by addition of 500 ll of the cell lysate. Immunoprecipitation was performed overnight at 4°C. Protein G-Sepharose beads were then washed five times with ice-cold lysis buffer and boiled for 5 min in SDS-loading buffer. For co-immunoprecipitation of

123

650

FADD supernatants were separated by SDS-PAGE and followed by an immunoblot of FADD.

References 1. Guner D, Belka C, Daniel PT (2003) Disruption of cell death signaling in cancer: impact on disease prognosis and response to therapy. Curr Med Chem Anticancer Agents 3:319–326 2. Hajra KM, Liu JR (2004) Apoptosome dysfunction in human cancer. Apoptosis 9:691–704 3. Danker K, Reutter W, Semini G (2010) Glycosidated phospholipids: uncoupling of signalling pathways at the plasma membrane. Br J Pharmacol 160:36–47 4. de Mendoza AE, Campanero MA, de la Iglesia-Vicente J, Gajate C, Mollinedo F, Blanco-Prieto MJ (2009) Antitumor alkyl ether lipid edelfosine: tissue distribution and pharmacokinetic behavior in healthy and tumor-bearing immunosuppressed mice. Clin Cancer Res 15:858–864 5. Gajate C, Gonzalez-Camacho F, Mollinedo F (2009) Lipid raft connection between extrinsic and intrinsic apoptotic pathways. Biochem Biophys Res Commun 380:780–784 6. Gajate C, Mollinedo F (2002) Biological activities, mechanisms of action and biomedical prospect of the antitumor ether phospholipid ET-18-OCH(3) (edelfosine), a proapoptotic agent in tumor cells. Curr Drug Metab 3:491–525 7. Gajate C, Santos-Beneit AM, Macho A et al (2000) Involvement of mitochondria and caspase-3 in ET-18-OCH(3)-induced apoptosis of human leukemic cells. Int J Cancer 86:208–218 8. Mollinedo F, de la Iglesia-Vicente J, Gajate C et al (2010) Lipid raft-targeted therapy in multiple myeloma. Oncogene 29: 3748–3757 9. Mollinedo F, Gajate C, Morales AI et al (2009) Novel antiinflammatory action of edelfosine lacking toxicity with protective effect in experimental colitis. J Pharmacol Exp Ther 329:439–449 10. Noseda A, White JG, Godwin PL, Jerome WG, Modest EJ (1989) Membrane damage in leukemic cells induced by ether and ester lipids: an electron microscopic study. Exp Mol Pathol 50:69–83 11. Mickeleit M, Wieder T, Buchner K, Geilen C, Mulzer J, Reutter W (1995) Glc-PC, a new type of glucosidic phospholipid. Angew Chem Int Ed 34:2667–2669 12. Mickeleit M, Wieder T, Arnold M, Geilen C, Mulzer J, Reutter W (1998) A glucose-containing ether lipid (Glc-PAF) as an antiproliferative analog of the platelet activating factor. Angew Chem Int Ed 37:351–353 13. Fischer A, Muller D, Zimmermann-Kordmann M et al (2006) The ether lipid inositol-C2-PAF is a potent inhibitor of cell proliferation in HaCaT cells. Chembiochem 7:441–449 14. Belka C, Rudner J, Wesselborg S et al (2000) Differential role of caspase-8 and BID activation during radiation- and CD95induced apoptosis. Oncogene 19:1181–1190 15. Eischen CM, Kottke TJ, Martins LM et al (1997) Comparison of apoptosis in wild-type and Fas-resistant cells: chemotherapyinduced apoptosis is not dependent on Fas/Fas ligand interactions. Blood 90:935–943 16. Klopfer A, Hasenjager A, Belka C, Schulze-Osthoff K, Dorken B, Daniel PT (2004) Adenine deoxynucleotides fludarabine and cladribine induce apoptosis in a CD95/Fas receptor, FADD and caspase-8-independent manner by activation of the mitochondrial cell death pathway. Oncogene 23:9408–9418 17. Newton K, Strasser A (2000) Ionizing radiation and chemotherapeutic drugs induce apoptosis in lymphocytes in the absence of

123

Apoptosis (2011) 16:636–651

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Fas or FADD/MORT1 signaling. Implications for cancer therapy. J Exp Med 191:195–200 von Haefen C, Wieder T, Essmann F, Schulze-Osthoff K, Dorken B, Daniel PT (2003) Paclitaxel-induced apoptosis in BJAB cells proceeds via a death receptor-independent, caspases-3/-8-driven mitochondrial amplification loop. Oncogene 22:2236–2247 Wesselborg S, Engels IH, Rossmann E, Los M, Schulze-Osthoff K (1999) Anticancer drugs induce caspase-8/FLICE activation and apoptosis in the absence of CD95 receptor/ligand interaction. Blood 93:3053–3063 Wieder T, Essmann F, Prokop A et al (2001) Activation of caspase-8 in drug-induced apoptosis of B-lymphoid cells is independent of CD95/Fas receptor-ligand interaction and occurs downstream of caspase-3. Blood 97:1378–1387 Starck L, Scholz C, Dorken B, Daniel PT (2005) Costimulation by CD137/4-1BB inhibits T cell apoptosis and induces Bcl-xL and c-FLIP(short) via phosphatidylinositol 3-kinase and AKT/ protein kinase B. Eur J Immunol 35:1257–1266 Mollinedo F, de la Iglesia-Vicente J, Gajate C et al (2010) In vitro and In vivo selective antitumor activity of Edelfosine against mantle cell lymphoma and chronic lymphocytic leukemia involving lipid rafts. Clin Cancer Res 16:2046–2054 Mollinedo F, Fernandez-Luna JL, Gajate C et al (1997) Selective induction of apoptosis in cancer cells by the ether lipid ET-18OCH3 (Edelfosine): molecular structure requirements, cellular uptake, and protection by Bcl-2 and Bcl-X(L). Cancer Res 57:1320–1328 Wiese A, Wieder T, Mickeleit M et al (2000) Structure-dependent effects of glucose-containing analogs of platelet activating factor (PAF) on membrane integrity. Biol Chem 381:135–144 Wieder T, Orfanos CE, Geilen CC (1998) Induction of ceramidemediated apoptosis by the anticancer phospholipid analog, hexadecylphosphocholine. J Biol Chem 273:11025–11031 Tsujimoto Y, Shimizu S (2005) Another way to die: autophagic programmed cell death. Cell Death Differ 12(Suppl 2): 1528–1534 Cabaner C, Gajate C, Macho A, Munoz E, Modolell M, Mollinedo F (1999) Induction of apoptosis in human mitogen-activated peripheral blood T-lymphocytes by the ether phospholipid ET-18-OCH3: involvement of the Fas receptor/ligand system. Br J Pharmacol 127:813–825 Gajate C, Fonteriz RI, Cabaner C et al (2000) Intracellular triggering of Fas, independently of FasL, as a new mechanism of antitumor ether lipid-induced apoptosis. Int J Cancer 85:674–682 Grobner S, Adkins I, Schulz S et al (2007) Catalytically active Yersinia outer protein P induces cleavage of RIP and caspase-8 at the level of the DISC independently of death receptors in dendritic cells. Apoptosis 12:1813–1825 van der Luit AH, Vink SR, Klarenbeek JB et al (2007) A new class of anticancer alkylphospholipids uses lipid rafts as membrane gateways to induce apoptosis in lymphoma cells. Mol Cancer Ther 6:2337–2345 Gajate C, Del Canto-Janez E, Acuna AU et al (2004) Intracellular triggering of Fas aggregation and recruitment of apoptotic molecules into Fas-enriched rafts in selective tumor cell apoptosis. J Exp Med 200:353–365 Gajate C, Mollinedo F (2007) Edelfosine and perifosine induce selective apoptosis in multiple myeloma by recruitment of death receptors and downstream signaling molecules into lipid rafts. Blood 109:711–719 Jendrossek V, Muller I, Eibl H, Belka C (2003) Intracellular mediators of erucylphosphocholine-induced apoptosis. Oncogene 22:2621–2631 Nieto-Miguel T, Fonteriz RI, Vay L, Gajate C, Lopez-Hernandez S, Mollinedo F (2007) Endoplasmic reticulum stress in the

Apoptosis (2011) 16:636–651

35.

36.

37.

38.

39.

40.

41.

proapoptotic action of edelfosine in solid tumor cells. Cancer Res 67:10368–10378 Chandra D, Choy G, Deng X, Bhatia B, Daniel P, Tang DG (2004) Association of active caspase 8 with the mitochondrial membrane during apoptosis: potential roles in cleaving BAP31 and caspase 3 and mediating mitochondrion-endoplasmic reticulum cross talk in etoposide-induced cell death. Mol Cell Biol 24:6592–6607 Xiang R, Liu Y, Zhu L, Dong W, Qi Y (2006) Adaptor FADD is recruited by RTN3/HAP in ER-bound signaling complexes. Apoptosis 11:1923–1932 Upton JP, Austgen K, Nishino M et al (2008) Caspase-2 cleavage of BID is a critical apoptotic signal downstream of endoplasmic reticulum stress. Mol Cell Biol 28:3943–3951 Lamothe B, Aggarwal BB (2002) Ectopic expression of Bcl-2 and Bcl-xL inhibits apoptosis induced by TNF-related apoptosisinducing ligand (TRAIL) through suppression of caspases-8, 7, and 3 and BID cleavage in human acute myelogenous leukemia cell line HL-60. J Interferon Cytokine Res 22:269–279 Voorzanger-Rousselot N, Blay JY (2004) Coexpression of CD40 and CD40L on B lymphoma and carcinoma cells: an autocrine anti-apoptotic role. Leuk Lymphoma 45:1239–1245 Nieto-Miguel T, Gajate C, Mollinedo F (2006) Differential targets and subcellular localization of antitumor alkyl-lysophospholipid in leukemic versus solid tumor cells. J Biol Chem 281: 14833–14840 Marini P, Denzinger S, Schiller D et al (2006) Combined treatment of colorectal tumours with agonistic TRAIL receptor

651

42.

43.

44.

45.

46.

47.

48.

antibodies HGS-ETR1 and HGS-ETR2 and radiotherapy: enhanced effects in vitro and dose-dependent growth delay in vivo. Oncogene Marini P, Schmid A, Jendrossek V et al (2005) Irradiation specifically sensitises solid tumour cell lines to TRAIL mediated apoptosis. BMC Cancer 5:5 von Haefen C, Gillissen B, Hemmati PG et al (2004) Multidomain Bcl-2 homolog Bax but not Bak mediates synergistic induction of apoptosis by TRAIL and 5-FU through the mitochondrial apoptosis pathway. Oncogene 23:8320–8332 Juo P, Kuo CJ, Yuan J, Blenis J (1998) Essential requirement for caspase-8/FLICE in the initiation of the Fas-induced apoptotic cascade. Curr Biol 8:1001–1008 Juo P, Woo MS, Kuo CJ et al (1999) FADD is required for multiple signaling events downstream of the receptor Fas. Cell Growth Differ 10:797–804 Rudner J, Lepple-Wienhues A, Budach W et al (2001) Wild-type, mitochondrial and ER-restricted Bcl-2 inhibit DNA damageinduced apoptosis but do not affect death receptor-induced apoptosis. J Cell Sci 114:4161–4172 von Haefen C, Wieder T, Gillissen B et al (2002) Ceramide induces mitochondrial activation and apoptosis via a Baxdependent pathway in human carcinoma cells. Oncogene 21: 4009–4019 Sturm I, Stephan C, Gillissen B et al (2006) Loss of the tissuespecific proapoptotic BH3-only protein Nbk/Bik is a unifying feature of renal cell carcinoma. Cell Death Differ 13:619–627

123