Department of Biochemistry and Molecular Biology, School of Biomedical Sciences, Monash University, Melbourne, Australia, 3800

Correspondence to: A Lawen, Department of Biochemistry and Molecular Biology, Monash University, PO Box 13D, Melbourne, Australia, 3800; E-mail: alfons.lawen@med.monash.edu.au

^aJD Ly and F Vaillant contributed equally to this publication.

^bCurrent address: Department of Veterinary Pathology, University of Glasgow Veterinary School, Bearsden Road, Glasgow G61 1QH, UK.

Permeability transition, and a subsequent drop in mitochondrial membrane potential (_m), have been suggested to be mechanisms by which cytochrome c is released from the mitochondria into the cytosol during apoptosis. Furthermore, a drop in _m has been suggested to be an obligate early step in the apoptotic pathway. Didemnin B, a branched cyclic peptolide described to have immunosuppressive, anti-tumour, and anti-viral properties, induces rapid apoptosis in a range of mammalian cell lines. Induction of apoptosis by didemnin B in cultured human pro-myeloid HL-60 cells is the fastest and most complete ever described with all cells being apoptotic after 3 h of treatment. By utilizing the system of didemnin B-induced apoptosis in HL-60 cells, and the potent inhibitors of mitochondrial permeability transition, cyclosporin A and bongkrekic acid, we show that permeability transition as determined by changes in _m and mitochondrial Ca²⁺ fluxing, is not a requirement for apoptosis or cytochrome c release. In this system, changes in mitochondrial membrane potential and cytochrome c release are shown to be dependent on caspase activation, and to occur concurrently with the release of caspase-9 from mitochondria, genomic DNA fragmentation and apoptotic body formation. Oncogene (2001) 20, 4085-4094.

didemnin B; apoptosis; mitochondria; cytochrome c; permeability transition; caspases

AIF, apoptosis inducing factor; CyA, cyclosporin A; BA, bongkrekic acid; JC-1, 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolcarbocyanine iodide; PT, permeability transition; PTP, permeability transition pore; CCCP, carbonyl cyanide m-chlorophenyl hydrazone; DAPI, 4,6'-diamidino-2-phenylindole; mtDNA, mitochondrial DNA; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-DL-Asp-fluoromethylketone; Ac-DEVD-cho, acetyl-Asp-Glu-Val-L-Asp-aldehyde; Ac-YVAD-cmk, acetyl-Tyr-Val-Ala-Asp-chloromethylketone

Introduction

The role of the mitochondrial membrane potential (_m) in apoptosis remains unclear, mostly due to the different model systems used. Cells devoid of mitochondrial (mt)DNA (⁰ cells) can still undergo apoptosis induced by staurosporine (Jacobson et al., 1993; Jiang et al., 1999), anti-Fas antibodies (Gamen et al., 1995), tumour necrosis factor- plus cycloheximide (Marchetti et al., 1996), or didemnin B (this paper) indicating that there is no general breakdown in apoptotic signalling in cells lacking a functional electron transport chain. Mitochondria have however been implicated in apoptosis through the release of several proteins during the onset of apoptosis; these include AIF (Susin et al., 1999), Apaf-1 (Zou et al., 1997), caspase-9 (Li et al., 1997; Krajewski et al., 1999) as well as cytochrome c (Liu et al., 1996; Kluck et al., 1997a; Yang et al., 1997). Cytochrome c release appears to occur upstream of caspase-3 activation, and to be necessary for caspase-9 induced activation of caspase-3 (Li et al., 1997; Kluck et al., 1997b; Slee et al., 1999).

There has been some debate as to the mechanism of cytochrome c release from mitochondria during apoptosis. Direct rupture of the outer mitochondria membrane via activation of the mitochondrial permeability transition pore (PTP; Scarlett and Murphy, 1997) has been suggested as a possible route of escape, however, cytochrome c release has been reported to occur prior to any mitochondrial membrane breakage (Zhuang et al., 1998). The PTP operates in three distinct states: a closed state with intact _m, a low-conductance state, where the pore is only permeable to molecules <300 Da and _m decreases reversibly, and finally the classic high-conductance state in which the pore is permeable to molecules <1500 Da. This latter state leads to the irreversible collapse of _m due to rupture of the outer mitochondrial membrane (Ichas and Mazat, 1998), and it is this state which had originally been suggested to be an early necessary step in apoptosis (Zamzami et al., 1995a, 1996). Inhibition of the PTP with bongkrekic acid (BA), a potent inhibitor of the adenine nucleotide translocator, can lead to inhibition of dexamethasone-induced apoptosis (Marchetti et al., 1996). Another inhibitor of the PTP, cyclosporin A (CyA) can inhibit PTP function (Fournier et al., 1987) through binding to mitochondrial cyclophilin D (Petronilli et al., 1993), and consequently, can inhibit apoptosis in several systems (Zamzami et al., 1995a, 1996). Furthermore, permeability transition (PT) has been demonstrated to be involved in the release of AIF from mitochondria during dexamethasone-induced apoptosis (Susin et al., 1999). A comprehensive review on the various hypotheses presently discussed for the mechanism of cytochrome c release can be found in Desagher and Martinou (2000).

Didemnin B is a cyclic branched peptolide described to have potent anti-neoplastic (Rinehart et al., 1983), anti-viral (Weed and Stringfellow, 1983), and immunosuppressive (Montgomery and Zukoski, 1985) properties. The biological action of didemnin B is obscure, however, didemnin B may interact with elongation factor 1 (Crews et al., 1994), and a palmitoyl thioesterase (Crews et al., 1996; Meng et al., 1998). We have previously shown that didemnin B induces extremely rapid apoptosis in unsynchronized cultures of the human pro-myeloid cell line HL-60, with all cells displaying morphological characteristics of apoptosis within about 3 h (Grubb et al., 1995). This extremely rapid and complete induction can be inhibited by pre-incubation of the cells with Zn²⁺ (Grubb et al., 1995), tyrosine kinases inhibitors (Johnson et al., 1996), z-VAD-fmk (Johnson et al., 1999) or by rapamycin (Johnson and Lawen, 1999). Didemnin B-induced apoptosis is not restricted to HL-60 cells, and has been described to induce apoptosis in human lymphoblastoid Daudi cells (Johnson et al., 1999). We have since used the drug to induce apoptosis in a wide range of mammalian cell lines including human lymphoblastoid Jurkat and MOLT-4, human melanoma MM96, SK-MEL-3 and SK-MEL-28, and human cervical cancer-derived HeLa cells.

Since the induction of apoptosis by didemnin B is a very fast event in HL-60 cells, we have used this system to study the minimal requirements for apoptosis to occur. In this paper we analyse the involvement of mitochondrial functions in the apoptotic process, and demonstrate that didemnin B-induced apoptosis does not require mitochondrial permeability transition for release of mitochondrial cytochrome c, an event which appears to be totally caspase-dependent in the system studied.

Results

Mitochondrial membrane potential and permeability transition

It has been suggested (Marchetti et al., 1996), that apoptosis requires the mitochondria to undergo high-conductance PT, a process which concomitantly proceeds with irreversible loss of mitochondrial membrane potential (_m). We sought to use our system of didemnin B-induced apoptosis in HL-60 cells to analyse the requirements for PT and _m dissipation during apoptosis, and, to compare cells depleted in mitochondrial DNA (and hence lacking an intact electron transport chain; ⁰) with their wild type ⁺ parental cells. We decided to use the mitochondrial dye JC-1, which is a more reliable fluorescent probe than either 3,3'dihexyloxacarbocyanine iodide [DiOC₆(3)] or rhodamine 123 for measuring _m in intact cells (Salvioli et al., 1997). JC-1 monomers, which emit at 538 nm when excited at 488 nm, are sequestered due to their charge into respiring mitochondria, where the monomers form J-aggregates which emit at 590 nm when excited at 488 nm. The _m for the ⁰ cells was 60.6±3% that of ⁺ cells. When incubated with 1 M didemnin B, ⁰ HL-60 cells underwent apoptosis at a similar rate to the ⁺ parental strain (Figure 1). Complete apoptosis (100% of cells displaying apoptotic morphology) was observed by 3.5 h. These data show that the lower _m of ⁰ cells does not predispose these cells to undergo more rapid apoptosis than the parental cells upon exposure to didemnin B.

Cells loaded with JC-1 and continuously treated with 1 M didemnin B show a decrease in _m between 1 and 2 h of treatment as determined by fluorescent spectrophotometric analysis (Figure 2). CCCP, an agent which rapidly dissipates the _m, was used at a concentration of 50 M, to determine the fluorescence of completely depolarized mitochondria and such gave a baseline for the experiments. To verify these observations, and the assay system, HL-60 cells were also treated with 1 M didemnin B prior to the addition of the DiOC₆(3), and then analysed by FACS analysis. The use of a completely different analysis technique revealed that decreases in _m are relatively early events (clearly visible after 1 h) in didemnin B-induced apoptosis (Figure 3), and not dependent on PT. The gradual loss of _m observed in didemnin B-treated HL-60 cells occurs before the onset of any significant morphological apoptosis. Pre-incubation of HL-60 cells with 100 M z-VAD-fmk prior to the addition of didemnin B, completely inhibits the morphological appearance of apoptosis (Johnson et al., 1999) and the decline in _m (data not shown).

The human lymphoblastoid Daudi cell line was used to determine if the effects seen in HL-60 cells were cell line specific. Since the kinetics of didemnin B-induced apoptosis in Daudi cells has been reported to be slower than in HL-60 cells (Johnson et al., 1999), _m of Daudi cells was determined over a 7-h period post didemnin B treatment. Daudi cells treated with didemnin B showed a gradual loss of _m, which began, after 2 h incubation in the presence of didemnin B (Figure 4). The loss of _m measured in Daudi cells also serves as a further control for the functionality of the assay used.

Loss of _m in apoptosis had been linked to the occurrence of high-conductance PT. PT can be specifically inhibited by CyA, which targets cyclophilin D. Preincubation of HL-60 cells with 1 M CyA prior to addition of didemnin B did not protect cells from undergoing apoptosis (Figure 5), nor did CyA have any effect on the membrane potential of didemnin B-treated HL-60 cells (Figure 3).

We also tested bongkrekic acid (BA), an inhibitor of the adenine nucleotide translocase which has been associated with PT (Henderson and Lardy, 1970; Klingenberg et al., 1970), and unlike CyA, is not known to have non-mitochondrial anti-apoptotic side effects. Fifty M BA or the vehicle [1.6 mM (NH₄)₂SO₄] had no detectable effects on HL-60 cells. Pre-incubation of HL-60 cells with BA for 30 min prior to the addition of didemnin B had no significant effect on the induction of apoptosis as judged by morphology and DNA fragmentation (Figure 5), or on the _m over the time course of the experiment (data not shown).

Electron microscopic examination of ⁺ HL-60 cells before and after treatment with 1 M didemnin B revealed the presence of intact mitochondria in control cells. Cells treated for 3 h with didemnin B showed mitochondria, which are not grossly swollen, and have apparently intact membranes. However, these mitochondria appear to have a disorganized internal membrane arrangement, indicating that there are mitochondrial changes occurring as a result of didemnin B treatment (Figure 6).

As 1 M CyA had no effect on didemnin B-induced apoptosis, it was important to determine that 1 M CyA could affect closure of the PTP, and protect against mitochondrial depolarization under the experimental conditions used. We therefore established an assay to assess PT in vivo. This assay measured mitochondrial Ca²⁺ using the fluorophore Rhod-2-AM as described in Material and methods. HL-60 mitochondria were analysed for their ability to cope with a transient rise in cytosolic Ca²⁺ induced with 10 M thapsigargin. Compared to the control HL-60 cells, incubation with 1 M didemnin B for up to 4 h did not significantly lower mitochondrial Rhod-2 fluorescence after thapsigargin treatment. Pre-treatment of cells for 30 min with 1 M CyA induced a transient peak in mitochondrial Ca²⁺ fluorescence at time zero (Table 1), suggesting that CyA is inhibiting PT at this concentration, by inhibiting the efflux of mitochondrial Ca²⁺. A similar effect was observed when cells were incubated with both CyA and didemnin B (Table 1).

Didemnin B-induced cytochrome c release

Mitochondria have also been deemed to play a central role in apoptosis by releasing several 'apoptotic factors' into the cytosol during apoptosis. AIF (Susin et al., 1999), Apaf-1 (Zou et al., 1997), caspase-9 (Li et al., 1997) and cytochrome c (Kluck et al., 1997a) have all been reported to be released during apoptosis, possibly via outer mitochondrial membrane breakage caused by opening of the PTP (Scarlett and Murphy, 1997). HL-60 cells incubated with 1 M didemnin B showed rapid and total translocation of cytochrome c from the mitochondria to the cytosol (Figure 7a,b). Pre-incubation with CyA or BA, which had no inhibitory effect on didemnin B-induced apoptosis, also had no inhibitory effect on didemnin B-induced cytochrome c release from the mitochondria (Figure 7c-e). Western blots for cytochrome c show that didemnin B induces the release of cytochrome c from the mitochondria into the cytosol over a time course, which corresponds with induction of both morphological apoptosis and DNA fragmentation. Cytochrome c is first observed in the cytosol after 2 h incubation with didemnin B, which is 2 h prior to any significant change in _m.

We have shown elsewhere (Johnson et al., 1999) that didemnin B-induced apoptosis is not sensitive to the inhibitor of caspase-1 Ac-YVAD-cmk, or the caspase-3 inhibitor, Ac-DEVD-cho. These two peptide inhibitors also had no effect on didemnin B-induced release of cytochrome c into the cytosol (Figure 8c,d). The general caspase inhibitor z-VAD-fmk on the other hand, completely inhibits apoptosis induced by 1 M didemnin B in HL-60 cells (Johnson et al., 1999), and the didemnin B-induced appearance of cytosolic cytochrome c in HL-60 cells (Figure 8b). Didemnin B-induced apoptosis in the human melanoma cell line MM96, which shows similar kinetics of induction of apoptosis to HL-60 cells (data not shown), also results in the early release of cytochrome c (Figure 9). Like HL-60 cells, pre-incubation of MM96 cells with 100 M z-VAD-fmk prior to the addition of 1 M didemnin B, protects MM96 cells from undergoing morphological apoptosis, and inhibits mitochondrial cytochrome c release (Figure 9), indicating that this phenomenon is not restricted to HL-60 cells. The inhibition of cytochrome c release by z-VAD-fmk was also not restricted to didemnin B-induced apoptosis. One hundred M z-VAD-fmk inhibits both the onset of apoptosis, and mitochondrial cytochrome c release induced by 70 M etoposide (Figure 10). These data suggest that the caspase-dependent release of cytochrome c is not restricted to didemnin B-induced apoptosis. The release of cytochrome c into the cytosol is known to lead to the activation of effector caspases, possibly via the formation of the apoptosome complex with pro-caspase-9 (Slee et al., 1999).

Discussion

Comparison of the time course for irreversible loss of _m with the appearance of morphological apoptosis, DNA fragmentation and cytochrome c release, suggests that _m change occurs concurrently with a number of apoptosis associated events in HL-60 cells upon didemnin B-treatment. This is in general agreement with a number of previous reports on _m changes during apoptosis (Kluck et al., 1997b; Bossy-Wetzel et al., 1998; Krohn et al., 1999). Work by Zamzami et al. (1995b) has suggested that cells which have a lowered _m could be induced to more readily undergo apoptosis than the same cell type with a high _m. The use of ⁰ cells provides some evidence that this is not the case for HL-60 cells undergoing didemnin B-induced apoptosis. It is interesting to note that during didemnin B-induced apoptosis, the _m decreases to a level comparable to ⁰ viable cells. Therefore, our results strongly imply that _m is not a determining factor in didemnin B-induced apoptosis.

There were no significant thapsigargin induced Rhod-2 fluorescence changes between control cells, and cells treated for up to 4 h with 1 M didemnin B, suggesting that PT must still be regulated in cells undergoing didemnin B-induced apoptosis. The ability to regulate PT even after the release of mitochondrial cytochrome c suggests that the inner mitochondrial membrane remains intact during apoptosis. These observations allow for the possibility of cytochrome c release through the formation of Bax containing pores in the outer mitochondrial membrane (Saito et al., 2000). Cytochrome c has also been suggested to be released through the formation of ANT/Bax complexes (Brenner et al., 2000). This pore would span both the inner and outer mitochondrial membranes, and lead to a rapid and complete loss of _m; however, our data do not support this idea, or the direct rupture of the mitochondrial membrane by uncontrolled PT (Scarlett and Murphy, 1997).

Caspases have been reported to play a role in the dissipation of _m (Zhai et al., 2000) during apoptosis. The actual mechanism whereby didemnin B-induces the decrease in _m remains obscure. We have earlier shown (Johnson et al., 1999) that didemnin B-induced apoptosis in HL-60 cells can be blocked by z-VAD-fmk. As well, the decrease in _m can be suppressed by zVAD-fmk, thus suggesting that caspases play a role in the degeneration of _m during didemnin B-induced apoptosis when induced to undergo apoptosis by didemnin B or etoposide. How caspases effect _m remains obscure; however cytochrome c release is caspase dependent in HL-60 and MM96 cells. One possibility may be that the physical translocation of cytochrome c out of the mitochondria leads to the disruption of proton pumping by the mitochondrial electron transport chain. Others have previously reported on the contribution of caspases to the release of cytochrome c (Chen et al., 2000; Rytomaa et al., 2000; Robertson et al., 2000); however, in our system, cytochrome c release can be completely blocked by z-VAD-fmk. These results are in contrast to data by Goldstein et al. (2000), where the authors could not inhibit cytochrome c release from HeLa cells by z-VAD-fmk upon actinomycin, staurosporine, or UV-exposure. On the other hand, these authors see, like us, complete inhibition of cytochrome c release upon exposure to tumour necrosis factor.

The results presented here are consistent with previous observations (Bossy-Wetzel et al., 1998; Krohn et al., 1999), in that loss of _m is not a pre-requisite for either mitochondrial cytochrome c release, or apoptosis. However, in these systems PT was not inhibited, and thus may still have played a role in the induction of apoptosis. By the use of the potent PT inhibitors CyA and BA, we show that neither PT nor, a functional PTP is required for didemnin B-induced apoptosis. These data are similar to data reported on Bid-induced cytochrome c release (Kim et al., 2000).

Apoptotic signalling via the mitochondria appears to occur via several distinct pathways. Apoptosis may be dependent on the induction of PT for the mitochondrial release of cytochrome c, and thus can be inhibited with inhibitors of the PTP, such as CyA (Scorrano et al., 1999). Or signalling may be independent of PT, and cytochrome c release is dependent on caspases (Kirsch et al., 1999), or Bid (Slee et al., 2000). The results presented here for didemnin B-induced apoptosis in HL-60 cells are consistent with apoptotic signalling via a PT independent, caspase dependent pathway, as the decrease in _m observed during didemnin B-induced apoptosis is not inhibitable by the PT-inhibitors CyA and BA, but can be completely inhibited by z-VAD-fmk. This suggests that PT is not a determining factor in _m dissipation during didemnin B-induced apoptosis.

The general caspase inhibitor z-VAD-fmk is a potent inhibitor of didemnin B-induced apoptosis (Johnson et al., 1999) and cytochrome c release in all the cell lines tested (HL-60, Daudi, MM96), indicating a role for caspases in the mitochondrial release of cytochrome c. The inhibition of etoposide-induced cytochrome c release indicates that the caspase-dependent release of cytochrome c is not restricted to didemnin B-induced apoptosis. The more specific caspase inhibitors Ac-DEVD-cho, Ac-YVAD-cmk and Ac-LEHD-cho have no inhibitory effect on didemnin B-induced apoptosis in HL-60 cells. This suggests that caspases-1, -3 and -9 are not important signalling intermediates for didemnin B-induced apoptosis in this cell line (Johnson et al., 1999). Neither Ac-DEVD-cho, Ac-YVAD-cmk or Ac-LEHD-cho showed any protective effect against cytochrome c release or had any effect on _m, which is consistent with previous observations (Pastorino et al., 1998). As caspase-3 appears to act downstream of cytochrome c release (Li et al., 1997; Kluck et al., 1997a,b), its inhibition by Ac-DEVD-cho should not effect cytochrome c release. Western blot analysis for pro-caspase-9 in subcellular fractions collected from HL-60 cells treated with the vehicle, ethanol for 3 h revealed that pro-caspase-9 localizes with cytochrome c and the mitochondrial porin protein, to a subcellular fraction enriched in mitochondria and membranes. However, HL-60 cells incubated for 3 h with 1 M didemnin B show the movement of caspase-9 from the membrane-enriched fraction into the cytosolic fraction (data not shown). Previous reports of the localization and release of pro-caspase-9 from the mitochondria (Krajewski et al., 1999) support our observations (data not shown), which suggest that pro-caspase-9 is present in the mitochondria and released into the cytosol upon didemnin B treatment. While didemnin B-induced apoptosis in HL-60 cells is dependent on caspase activation, this(ese) caspase(s) appear to be active upstream from mitochondria. Doxorubicin-induced apoptosis, cytochrome c release and _m changes in Jurkat cells also require the activation of caspases upstream from the mitochondria (Gamen et al., 2000). However, Ac-DEVD-cho and Ac-LEHD-cho both inhibit apoptosis, cytochrome c release and _m dissipation caused by doxorubicin, which is not evident in didemnin B-induced apoptosis. Recent observations that the cleavage of Bid by caspase-8 can lead to the mitochondrial release of cytochrome c (Zhai et al., 2000), suggest that the release of cytochrome c from the mitochondria may occur via the interaction of Bid with mitochondrial surface proteins such as Bax.

In conclusion, our data suggest that didemnin B-induced apoptosis is a caspase dependent process, which does not require permeability transition to induce _m dissipation, and the mitochondrial release of cytochrome c and caspase-9 but may involve Bid in its signalling pathway.

Materials and methods

Cell lines and culture

The human leukaemia cell lines HL-60 and Daudi, and the human melanoma cell line MM96 were maintained in RPMI 1640 medium (Gibco, Melbourne, Australia), and supplemented and cultured as described (Grubb et al., 1995; Johnson et al., 1996). Cells were passaged three times weekly. Didemnin B (kind gift of NIH, Bethesda, USA) and etoposide (Sigma-Aldrich, Castle Hill, Australia) were dissolved in ethanol, and were added to cultures at a final concentration of 1 M and 70 M respectively (2.5 to 3.5´10⁵ viable cells/ml as determined by Trypan blue exclusion). Where indicated, cells were also pre-incubated for 30 min with either 1 M cyclosporin A (CyA) dissolved in ethanol or 50 M bongkrekic acid (BA; Lijmbach et al., 1970; de Bruijn et al., 1973), dissolved in 1.6 mM NH₄OH. Peptide caspase inhibitors were used as described elsewhere (Johnson et al., 1999).

Cells devoid of mitochondrial DNA (⁰) were generated as described previously (Larm et al., 1994; Vaillant and Nagley, 1995). The presence of mtDNA was analysed by PCR using mtDNA specific primers (Vaillant and Nagley, 1995). The two primer pairs used were H15680-L14820 and H3728-L2826, which gives rise to an 861 bp and 903 bp fragment respectively. Measurement of cytochrome c oxidase activity was carried out as described (Vaillant and Nagley, 1995). HL-60 ⁰ clones were selected on the bases of having no detectable cytochrome c oxidase activity, or detectable mtDNA.

Western blots

For Western blots to detect cytochrome c, 3´10⁶ cells were suspended in 0.8 ml of homogenization buffer (20 mM HEPES (pH 7.5), 10 mM KCl, 1.5 mM MgCl₂, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 250 mM sucrose, 100 g/ml digitonin, 1 mM PMSF, 10 g/ml aprotinin, 10 g/ml leupeptin). Cells were incubated on ice for 3 min then dounced on ice at a rate of one stroke/second for two min. Samples were centrifuged at 25 000 g for 20 min at 4°C. Supernatant was collected, and the pellet resuspended in 200 l homogenization buffer. Protein concentration was determined as described (Bradford, 1976), and total protein was precipitated with 10% TCA. Precipitated protein was collected by centrifugation at 10 000 g for 15 min at 4°C. The pellets were resuspended in 0.4 M NaOH to give a final protein concentration of 1 mg/ml. Fifteen g of protein were loaded per lane, and separated in polyacrylamide gels by electrophoresis, transferred to nitrocellulose, and probed using mouse monoclonal anti-cytochrome c antibody (PharMingen, San Diego, CA, USA). Blots were developed using ECL (Amersham Life Science, Buckinghamshire, England) before exposure to X-ray films (Fuji Photo Film Co. Ltd., Tokyo, Japan).

Morphological determination of apoptosis

At the times indicated, 200 l samples of the cell culture were collected for determination of apoptosis as described previously (Grubb et al., 1995; Johnson et al., 1996). Nuclear morphology was determined using 0.1 g/ml 4',6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich, Castle Hill, Australia) as described (Russel et al., 1975). The slides were observed under a fluorescence microscope at an excitation wavelength of 280 nm. The percentage of apoptotic nuclei was determined by counting more than 400 cells from at least three separate determinations each. The percentages of apoptotic cells are presented as averages±standard deviation.

DNA fragmentation assays

Cells were grown and treated as above. A total of 2´10⁶ cells/sample was lysed and treated as described (Herrmann et al., 1994), before DNA fragments were separated by agarose gel electrophoresis in 1.8% gels.

Measurement of mitochondrial membrane potential

Thirty minutes prior to each time point 3 M 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolcarbocyanine iodide (JC-1; Molecular Probes, Eugene, OR, USA) dissolved in ethanol, was added to 1´10⁶ cells/ml. Cells were then pelleted, resuspended in phosphate buffered saline at room temperature, and the fluorescence was then determined using a Hitachi F-4000 fluorescence spectrophotometer. To determine the 'green-red' ratio of low energy mitochondria to high energy mitochondria, samples were excited at 488 nm, and the fluorescent units (FU) recorded at 535 nm and 590 nm. The ratio was determined by dividing the FU at 590 nm by the FU at 535 nm. All values are expressed as the average of six independent experiments±standard deviation. Carbonyl cyanide m-chlorophenyl hydrazone (CCCP; Sigma-Aldrich, Castle Hill, Australia) was used at a concentration of 50 M as a positive control for membrane depolarization.

Flow cytometry was performed as follows: 1´10⁶ cells were treated with either 1 M didemnin B, 1 M CyA, 1 M didemnin B+1 M CyA or absolute ethanol (vehicle) as described above. At each time point, cells were washed once in PBS, then resuspended in PBS+2% FCS with 50 nM 3,3'-dihexyloxacarbocyanide iodide [DiOC₆(3) PBS, Molecular Probes, Eugene, OR, USA] for 30 min at 37°C, followed by analysis on a flow cytometer (FACScan Becton Dickinson, USA). Cell viability was determined by the ability to exclude propidium iodide (5 g/ml; Sigma Aldrich, Castle Hill, Australia). Forward and side scattering, as well as two fluorescence channels for DiOC₆(3), and propidium iodide were measured for each sample. Each histogram represents 25 000 events. Control cells incubated for 10 min with 50 M CCCP were used to determine a basal level for depolarized mitochondria.

Measurement of mitochondrial damage using calcium fluorophores

Prior to use, Rhod-2 acetoxylmethyl ester (Rhod-2 AM) (Molecular Probes, Eugene, OR, USA) was converted to the colourless, non-fluorescent compound dihydrorhod-2 AM as per the manufacturer's instructions. Esterase cleavage of dihydroRhod-2 AM into dihydroRhod-2 allows it to be sequestered into the mitochondria as a result of a net positive charge. The dihydroRhod-2 is rapidly oxidized into the active calcium fluorophore Rhod-2 in the mitochondrial environment. The cells to be used were pelleted and resuspended in RPMI-1640 media (non-supplemented) at a concentration of 1´10⁷ cells/ml at room temperature. DihydroRhod-2 AM was added to a final concentration of 5 M and incubated with the cells for 25 min at room temperature in the dark. The cells were then washed once in PBS at 37°C, then resuspended in supplemented RPMI-1640 media at a concentration of 5´10⁵ cells/ml and allowed to recover for 2 h prior to the addition of didemnin B. At each time point 3´10⁶ cells were collected and pelleted by centrifugation. The cell sample was resuspended in 1 ml of PBS at room temperature and read on a fluorescent spectrophotometer with excitation at 540 nm and emission at 578 nm.

To measure mitochondrial damage, mitochondria in intact control cells or cells induced to undergo apoptosis with didemnin B were tested for their ability to transiently uptake Ca²⁺. The background level of Rhod-2 fluorescence was determined, and then cells were treated with 10 M thapsigargin (Sigma-Aldrich, Castle Hill, Australia) dissolved in DMSO. The resulting Ca²⁺ spike was measured as a function of time and expressed as an increase in Rhod-2 fluorescence. To determine if thapsigargin induced Rhod-2 fluorescence was due to mitochondrial calcium uptake, cells were pre-treated for either 1 min, 15 min or 2 h, with 1 M CyA prior to addition of thapsigargin.

Electron microscopic analysis of mitochondria

A total of 3´10⁶ HL-60 cells treated with didemnin B was harvested and the washed pellet was fixed with 1.5% glutaraldehyde in phosphate buffered saline (PBS) overnight. The fixed cells were washed three times in PBS, then embedded in 2% agar. The agar block was treated for 1 h at room temperature with osmic acid (1% in 0.1 M cacodylate buffer). After dehydration by successive passages in 30, 70, 90 and 100% (v/v) ethanol, the samples were treated overnight in propylene oxide/resin (1 : 1) and embedded in Epon/Araldite resin. Sections were cut and further stained with uranyl acetate and lead citrate prior to examination using a JEOL-100S transmission electron microscope.

Acknowledgements

We thank the National Cancer Institute, Drug Synthesis and Chemistry Branch, Bethesda, (USA) for the gift of didemnin B, Prof JA Duine from Delft University of Technology (Netherlands) for the gift of bongkrekic acid, Dr René Traber from Novartis Pharma Ltd (Basel, Switzerland) for the gift of cyclosporin A, Dr Donald Nicholson from Department of Biochemistry and Molecular Biology, Merck Frosst Centre for Therapeutic Research, Quebec, Canada for the gift of anti-caspase-3 antibody. We also thank Mr Barry Veitch for his invaluable assistance with the electron microscopy work.

Bradford MM. (1976). Anal. Biochem. 72, 248-254. Article MEDLINE

Brenner C, Cadiou H, Vieira HL, Zamzami N, Marzo I, Xie Z, Leber B, Andrews D, Duclohier H, Reed JC, Kroemer G. (2000). Oncogene 19, 329-336. MEDLINE

Bossy-Wetzel E, Newmeyer DD, Green DR. (1998). EMBO J. 17, 37-49. Article MEDLINE

Chen Q, Gong B, Almasan A. (2000). Cell. Death Differ. 7, 227-233. MEDLINE

Crews CM, Collins JL, Lane WS, Snapper ML, Schreiber SL. (1994). J. Biol. Chem. 269, 15411-15414. MEDLINE

Crews CM, Lane WS, Schreiber SL. (1996). Proc. Natl. Acad. Sci. USA 93, 4316-4319. MEDLINE

de Bruijn J, Frost DJ, Nugteren DH, Graudemer A, Lijmbach GWM, Cox HC, Berends W. (1973). Tetrahedron 29, 1541-1547.

Desagher S, Martinou JC. (2000). Trends Cell. Biol. 10, 369-377. Article MEDLINE

Fournier N, Ducet G, Crevat A. (1987). J. Bioenerg. Biomembr. 19, 297-303. MEDLINE

Gamen S, Anel A, Montoya J, Marzo I, Piñeiro A, Naval J. (1995). FEBS Lett. 376, 15-18. MEDLINE

Gamen S, Anel A, Perez-Galan P, Lasierra P, Johnson D, Pineiro A, Naval J. (2000). Exp. Cell. Res. 258, 223-235. Article MEDLINE

Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. (2000). Nat. Cell. Biol. 2, 156-162. Article MEDLINE

Grubb DR, Wolvetang EJ, Lawen A. (1995). Biochem. Biophys. Res. Commun. 215, 1130-1136. MEDLINE

Henderson PJ, Lardy HA. (1970). J. Biol. Chem. 245, 1319-1326. MEDLINE

Herrmann M, Lorenz H-M, Voll R, Grünke M, Woith W, Kalden JR. (1994). Nucl. Acid. Res. 22, 5506-5507.

Ichas F, Mazat JP. (1998). Biochim. Biophys. Acta 1366, 33-50. MEDLINE

Jacobson MD, Burne JF, King MP, Miyashita T, Reed JC, Raff MC. (1993). Nature 361, 365-369. MEDLINE

Jiang S, Cai J, Wallace DC, Jones DP. (1999). J. Biol. Chem. 274, 29905-29911. MEDLINE

Johnson KL, Vaillant F, Lawen A. (1996). FEBS Lett. 383, 1-5. MEDLINE

Johnson KL, Grubb DR, Lawen A. (1999). J. Cell. Biochem. 72, 269-278. MEDLINE

Johnson KL, Lawen A. (1999). Immunol. Cell. Biol. 77, 242-248. MEDLINE

Kim TH, Zhao Y, Barber MJ, Kuharsky DK, Yin XM. (2000). J. Biol. Chem. 275, 39474-39481. MEDLINE

Kirsch DG, Doseff A, Chau BN, Lim DS, de Souza-Pinto NC, Hansford R, Kastan MB, Lazebnik YA, Hardwick JM. (1999). J. Biol. Chem. 274, 21155-21161. Article MEDLINE

Klingenberg M, Grebe K, Heldt HW. (1970). Biochem. Biophys. Res. Commun. 39, 344-351. MEDLINE

Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD. (1997a). Science 275, 1132-1136. Article MEDLINE

Kluck RM, Martin SJ, Hoffman BM, Zhou JS, Green DR, Newmeyer DD. (1997b). EMBO J. 16, 4639-4649. Article MEDLINE

Krajewski S, Krajewski M, Ellerby LM, Welsh K, Xie Z, Deveraux QL, Salvesen GS, Bredesen DE, Rosenthal RE, Fiskum G, Reed JC. (1999). Proc. Natl. Acad. Sci. USA 96, 5752-5757. Article MEDLINE

Krohn AJ, Wahlbrink T, Prehn JH. (1999). J. Neurosci. 19, 7394-7404. MEDLINE

Larm JA, Vaillant F, Linnane AW, Lawen A. (1994). J. Biol. Chem. 269, 30097-30100. MEDLINE

Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. (1997). Cell 91, 479-489. MEDLINE

Lijmbach GWM, Cox HC, Berends W. (1970). Tetrahedron 26, 5993-5999.

Liu X, Kim CN, Yang J, Jemmerson R, Wang X. (1996). Cell 86, 147-157. MEDLINE

Marchetti P, Castedo M, Susin SA, Zamzami N, Hirsch T, Macho A, Haeffner A, Hirsch F, Geuskens M, Kroemer G. (1996). J. Exp. Med. 184, 1155-1160. MEDLINE

Meng L, Sin N, Crews CM. (1998). Biochemistry 37, 10488-10492. MEDLINE

Montgomery DW, Zukoski CF. (1985). Transplantation 40, 49-56. MEDLINE

Pastorino JG, Chen S-T, Tafani M, Snyder JW, Farber JL. (1998). J. Biol. Chem. 273, 7770-7775. Article MEDLINE

Petronilli V, Cola C, Massari S, Colonna R, Bernardi P. (1993). J. Biol. Chem. 268, 21939-21945. MEDLINE

Rinehart Jr KL, Gloer JB, Wilson GR, Hughes Jr RG, Li LH, Renis HE, McGoven JP. (1983). Fed. Proc. 42, 87-90. MEDLINE

Robertson JD, Gogvadze V, Zhivotovsky B, Orrenius S. (2000). J. Biol. Chem. 275, 32438-32443. MEDLINE

Russel WC, Newman C, Williamson DH. (1975). Nature 253, 4671-462.

Rytomaa M, Lehmann K, Downward J. (2000). Oncogene 19, 4461-4468. MEDLINE

Saito M, Korsmeyer SJ, Schlesinger PH. (2000). Nat. Cell. Biol. 2, 553-555. Article MEDLINE

Salvioli S, Ardizzoni A, Franceschi C, Cossarizza A. (1997). FEBS Lett. 411, 77-82. Article MEDLINE

Scarlett JL, Murphy MP. (1997). FEBS Lett. 418, 282-286. MEDLINE

Scorrano L, Petronilli V, Di Lisa F, Bernardi P. (1999). J. Biol. Chem. 274, 22581-22585. Article MEDLINE

Slee EA, Harte MT, Kluck RM, Wolf BB, Casiano CA, Newmeyer DD, Wang HG, Reed JC, Nicholson DW, Alnemri ES, Green DR, Martin SJ. (1999). J Cell Biol. 144, 281-292. MEDLINE

Slee EA, Keogh SA, Martin SJ. (2000). Cell Death Differ. 7, 556-565. MEDLINE

Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM, Mangion J, Jacotot E, Costantini P, Loeffler M, Larochette N, Goodlett DR, Aebersold R, Siderovski DP, Penninger JM, Kroemer G. (1999). Nature 397, 441-446. Article MEDLINE

Vaillant F, Nagley P. (1995). Hum. Mol. Genet. 4,, 903-914.

Weed SD, Stringfellow DA. (1983). Antiviral Res. 3, 269-274. MEDLINE

Yang J, Liu X, Bhalla K, Kim CN, Ibrado AM, Cai J, Peng T-I, Jones DP, Wang X. (1997). Science 275, 1129-1132. Article MEDLINE

Zamzami N, Marchetti P, Castedo M, Decaudin D, Macho A, Hirsch T, Susin SA, Petit PX, Mignotte B, Kroemer G. (1995a). J. Exp. Med. 182, 367-377. MEDLINE

Zamzami N, Marchetti P, Castedo M, Hirsch T, Susin SA, Masse B, Kroemer G. (1996). FEBS Lett. 384, 53-57. Article MEDLINE

Zamzami N, Marchetti P, Castedo M, Zanin C, Vayssiere JL, Petit PX, Kroemer G. (1995b). J. Exp. Med. 181, 1661-1672. MEDLINE

Zhai D, Huang X, Han X, Yang F. (2000). FEBS Lett. 472, 293-296. MEDLINE

Zhuang J, Dinsdale D, Cohen GM. (1998). Cell Death Differ. 5, 953-962. MEDLINE

Zou H, Henzel WJ, Liu X, Lutschg A, Wang X. (1997). Cell 90, 405-413. MEDLINE

Figure 1 Effect of didemnin B on ⁺ and ⁰ HL-60 cells. Cells were incubated in the presence of 1 M didemnin B. Samples were collected as described for both (a). ⁺ HL-60 cells () and (b). ⁰ HL-60 cells (). Results are expressed as the mean of four separate experiments±standard deviation. The large error bars represent minor variations between different experiments. Inset: Cellular DNA was subjected to PCR amplification as described in Materials and methods. mtDNA specific primer pairs H15680-L14820 (lanes 1 and 2) and H3728-L2826 (lane 3 and 4) give rise to a 861 bp and 903 bp fragment respectively. Cells were ⁺ (lanes 1 and 3) and ⁰ cells (lanes 2 and 4). Bacteriophage X174/HaeIII fragments were loaded as a size standard as indicated at the right-hand side in lane ' X'

Figure 2 Dissipation of _m during didemnin B-induced apoptosis in HL-60 cells. _m was determined as described in Materials and methods using the mitochondrial marker JC-1 in the presence of 1 M didemnin B (), ethanol () or 50 M CCCP (). Results have been plotted against time, where 0 h represents the time of addition of didemnin B. Control values have been set at 100%. The data represent the mean of nine experiments±standard deviation

Figure 3 Effect of the PT inhibitor CyA on the dissipation of _m during didemnin B-induced apoptosis in HL-60 cells. _m was determined as described in Materials and methods using the mitochondrial marker DiOC₆(3) in the presence of ethanol, 1 M didemnin B, 1 M CyA, 1 M didemnin B+1 M CyA. Fifty M CCCP was used to determine the fluorescence of fully depolarized mitochondria. Samples were analysed by flow cytometry, excitation at 488 nm, and the resulting emission was recorded at 530 nm

Figure 4 Dissipation of _m during didemnin B-induced apoptosis in Daudi cells. _m was determined as described in Materials and methods using the mitochondrial marker JC-1 in the presence of 1 M didemnin B () or ethanol () or 50 M CCCP (). Results have been plotted against time, where 0 h represents the time of addition of didemnin B. Control values have been set at 100%. The data represent the mean of nine experiments±standard deviation

Figure 5 Effect of mitochondrial permeability transition pore inhibitors on didemnin B-induced apoptosis. ⁺ HL-60 cells were incubated with 1 M didemnin B, or 1 M didemnin B following 30 min pre-treatment with 1 M CyA or 50 M BA. Genomic DNA was extracted and treated as described in Materials and methods. `M' /HindIII standard

Figure 6 Electron micrographs of ⁺ HL-60 mitochondria in whole cells treated with didemnin B. ⁺ HL-60 cells were incubated with 0.1% (v/v) ethanol (a) or 1 M didemnin B (b) for 3 h prior to collection for electron micrographs. Both images at same magnification, bar represents 1 m

Figure 7 Release of cytochrome c from cells treated with 1 M didemnin B. HL-60 cells were treated with 1 M didemnin B and samples were collected at 0, 2 and 4 h (a). Release of cytochrome c was determined by Western blot as described in Materials and methods. Five M Staurosporine was used as a positive control for cytochrome c release (b). To determine the effect of PT inhibitors on cytochrome c release in HL-60 cells, cells were pre-treated for 30 min with 0.1% (v/v) ethanol (c), 1 M CyA (d) or 50 M BA (e) prior to addition of 1 M didemnin B. Western blots were preformed as described in Materials and methods. Samples were collected at zero, 2 h and 4 h after addition of didemnin B. Control lanes (C1, C2 and C3) represent cells incubated with vehicle (ethanol or NH₄OH) alone

Figure 8 Effect of caspase inhibitors on didemnin B-induced cytochrome c release. HL-60 cells were pre-treated as described (Johnson et al., 1999) with 0.1% (v/v) DMSO (a), 100 M z-VAD-fmk (b), 300 M Ac-YVAD-cmk (c), or 300 M Ac-DEVD-cho (d) prior to addition of 1 M didemnin B. Western blots were preformed as described in Materials and methods. Control lanes represent cells incubated with vehicle (DMSO) alone

Figure 9 Effect of caspase inhibitor z-VAD-fmk on didemnin B-induced cytochrome c release in MM96 cells. MM96 cells were pre-treated as described (Johnson et al., 1999) with either 0.1% (v/v) DMSO or 100 M z-VAD-fmk prior to addition of 1 M didemnin B. Western blots were preformed as described in Materials and methods on gels containing either mitochondrial fraction (a), or cytosolic fraction (b)

Figure 10 Effect of caspase inhibitor z-VAD-fmk on etoposide-induced cytochrome c release in HL-60 cells. HL-60 cells were pre-treated as described (Johnson et al., 1999) with either 0.1% (v/v) DMSO or 100 M z-VAD-fmk prior to addition of 70 M etoposide. Western blots were preformed as described in Materials and methods on gels containing either mitochondrial fraction (a), or cytosolic fraction (b)

Table 1 Effect of didemnin B and CyA on thapsigargin-induced mitochondrial Ca²⁺ flux in HL-60 cells