Oncogene (2003) 22, 9107–9120 & 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc ORIGINAL PAPERS Induction of apoptosis by antibiotic 0II proceeds through a caspase-mediated mitochondrial amplification loop in an entirely Bax-dependent manner

Aram Prokop1,4, Wolf Wrasidlo2,4, Holger Lode2, Ralf Herold1, Florian Lang3,5,Gu¨ nter Henze1, Bernd Do¨ rken3, Thomas Wieder3,5,6 and Peter T Daniel*,3,6

1Department of Pediatric Oncology/Hematology, University Medical Center Charite´, Humboldt University of Berlin, Berlin 13353, Germany; 2Department of General Pediatrics, University Medical Center Charite´, Humboldt University of Berlin, Berlin 13353, Germany; 3Department of Hematology, Oncology and Tumor Immunology, University Medical Center Charite´, Humboldt University of Berlin, Berlin 13125, Germany

Calicheamicin 0II is a member of the enediyne class of Keywords: calicheamicin; apoptosis; caspase-3; Bax; antitumor antibiotics that bind to DNA and induce Bcl-2; cytochrome c; mitochondria; BJAB; Jurkat; apoptosis. These compounds differ, however, from con- HCT116 ventional anticancer drugs as they bind in a sequence- specific manner noncovalently to DNA and cause sequence-selective oxidation of and bending of the DNA helix. Calicheamicin is clinically employed as Introduction immunoconjugate to antibodies directed against, for example, CD33 in the case of gemtuzumab ozogamicin. In contrast to necrosis, apoptosis is a morphologically Here, we show by the use of the unconjugated drug that and biochemically defined form of cell death that plays a calicheamicin-induced apoptosis is independent from role under different physiological conditions, such as death-receptor/FADD-mediated signals. Moreover, cali- cell turnover, the immune response or embryonic cheamicin triggers apoptosis in a p53-independent manner development. In addition, the apoptotic program may as shown by the use of p53 knockout cells. Cell death be triggered by various, unphysiological death stimuli, proceeds via activation of mitochondrial permeability for example, ionizing radiation (Belka et al., 2000, transition, cytochrome c release and activation of 2001), chemotherapeutic anticancer drugs (Engels et al., caspase-9 and -3. The overexpression of Bcl-xL or Bcl-2 2000; Wieder et al., 2001a) and by natural compounds strongly inhibited calicheamicin-induced apoptosis. (Wieder et al., 2001b). So far, three major pathways of Knockout of Bax abrogated cell death after calicheamicin apoptotic cell death have been defined: (i) the death treatment. Thus, the activation of mitochondria and receptor-mediated signalling cascade (Daniel et al., execution of cell death occur through a fully Bax- 2001), (ii) the mitochondrial pathway involving the dependent mechanism. Interestingly, caspase inhibition apoptosome (Daniel, 2000; Rudner et al., 2001), and (iii) by the pancaspase-inhibitor zVAD-fmk interfered with the endoplasmic reticulum-dependent stress response mitochondrial activation by calicheamicin. This places (Belka and Budach, 2002; Rudner et al., 2002). caspase activation upstream of the mitochondria and In previous studies, we demonstrated that inactivation indicates that calicheamicin-triggered apoptosis is en- of apoptotic pathways results in both apoptotic hanced through death receptor-independent activation of resistance in vitro and clinical resistance to anticancer the caspase cascade, that is, an amplification loop that is treatment (Mrozek et al., 2003; Gu¨ ner et al., 2003; required for full activation of the mitochondrial pathway. Sturm et al., 2001; Sturm et al., 2003). One of the main Oncogene (2003) 22, 9107–9120. doi:10.1038/sj.onc.1207196 regulatory steps of apoptotic cell death is controlled by the ratio of anti- and proapoptotic members of the Bcl-2 family of proteins function as a rheostat and determine the susceptibility to apoptosis (for reviews, see, Daniel, 2000; Daniel et al., 2003). For example, the over- expression of antiapoptotic Bcl-2 family members that *Correspondence: PT Daniel, Molecular Hematology and Oncology, University Medical Center Charite´ , Campus Berlin-Buch, Humboldt can tilt the rheostat towards survival, thereby confering University of Berlin, Berlin 13125, Germany drug resistance, at least in some cellular tumor model 4These authors contributed equally to the present study systems (Rudner et al., 2001; Belka and Budach, 2002; 5Current address: Department of Physiology I, Eberhard-Karls Radetzki et al., 2002; von Haefen et al., 2003). On the University Tubingen, Tu¨ bingen 72076, Germany 6These authors share senior authorship other hand, forced expression of proapoptotic Bax or Received 9 May 2003; revised 28 August 2003; accepted 10 September Bak or, for example, the BH3-only protein Nbk/Bik, is 2003 sufficient to increase the sensitivity of malignant cells to Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9108 apoptosis and to overcome drug resistance (Hemmati Results et al., 2002; Radetzki et al., 2002; von Haefen et al., 2002). Interestingly, these in vitro data are in line with Dose-dependent induction of apoptosis by calicheamicin signalling analyses performed in primary carcinoma, WII lymphoma and leukemia cells. Upon activation of the mitochondria, cell death proceeds via the formation of Depending on their concentration, many cytostatic the mitochondrial apoptosome and initiation of the substances cause necrotic, membrane damaging as et al caspase cascade. Consequently, loss of proapoptotic Bax well as apoptotic cell death (Wieder ., 1998; et al results in the failure of cancer cells to activate caspase-3 Friedrich ., 2001). We therefore determined the in vivo (Prokop et al., 2000). unspecific, cytotoxic effect of calicheamicin WII (further Calicheamicin is a naturally occurring hydrophobic designated as calicheamicin) in BJAB cells by the enediyne antibiotic that was isolated in Micromonospora determination of LDH release into the culture echinospora calichensis (Maiese et al., 1989). Calichea- medium. As shown in Figure 1a, calicheamicin did micin and other naturally occurring exhibit not significantly reduce viability of BJAB cells after significant activity against Gram-positive and -negative 24 h of incubation at concentrations p100 pM, thereby indicating that a necrosis-like mechanism, for bacteria. In addition, calicheamicin exerts a potent example, through direct membrane damage, does not antitumor activity and has been shown to induce apoptotic cell death in the picomolar range. Calichea- play a role for its death-inducing potency. After 48 h, however, 25% of the cells showed LDH release, which micin binds to the minor groove of the DNA helix, can be attributed to secondary necrosis that follows preferentially to the 30 ends of oligopurine tracts, and apoptosis in vitro. Calicheamicin potently induced DNA causes sequence-selective oxidation of and fragmentation in up to 60% of the cells. Apoptosis bending of the DNA helix by an induced-fit mechanism induction was concentration-dependent with a half- of DNA target recognition (Salzberg and Dedon, 2000). maximum concentration of 10 pM (Figure 1b), and In addition, calicheamicin and other naturally occurring DNA fragmentation was already observed after 24 h enediyne antibiotics, including , neocarci- (Figure 1c). The percentage of cells showing loss of nostatin, kedarcidin and dynemicin constitute a unique membrane integrity at 48 h (Figure 1a) corresponds well class of reactive compounds, which can undergo aromatization to produce biradicals. Interestingly, to the number of cells showing DNA fragmentation at 24 h (Figure 1c). Thus, DNA fragmentation precedes phosphodiester bond breakage of DNA was shown to membrane damage induced in calicheamicin-induced be dispensable for the apoptosis-inducing activity (Hiatt et al., 1994). Furthermore, prevention of apoptosis was cells. This indicates that calicheamicin induces apoptotic cell death in BJAB cells and this is followed by observed in analogs that were electronically stabilized to secondary necrosis. impair aromatic rearrangements and generation of biradicals (Hiatt et al., 1994). Thus, apart from enforcing a conformational change in the DNA Calicheamicin-induced apoptosis occurs independently of structure and the induction of DNA strand breaks, CD95/Fas signalling additional signalling events appear to be critically involved in the activation of apoptosis by calicheamicin To investigate the involvement of CD95/Fas or other and its analogs. death receptor systems in calicheamicin-induced cell In clinical anticancer therapy, the enediyne calichea- death, we employed BJAB cells overexpressing a micin is currently successfully employed as an immuno- dominant-negative FADD (FADD-dn) mutant. Both conjugate to antibodies serving as vehicles for the mock and FADD-dn transfectants express endogenous targeting of calicheamicin to tumor cells (Linenberger FADD (27 kDa), while the FADD-dn transfectants also et al., 2001; van der Velden et al., 2001). So far, the show a lower band representing the truncated FADD- mechanisms of apoptosis induction by these compounds dn (Wieder et al., 2001a). To provide a positive control remains enigmatic. Paradoxically, the induction of for the biological activity of the FADD-dn, we apoptotic cell death in PC12 cells by neocarcinostatin functionally characterized the FADD-dn-overexpressing was shown to be potentiated in cells overexpressing the BJAB cells. Vector-transfected BJAB cells were very antiapoptotic factor Bcl-2 (Cortazzo and Schor, 1996; sensitive towards challenge with recombinant human Schor et al., 1999). This phenomenon was suggested to soluble Fas ligand (CD95 ligand, CD178): apoptosis as occur in consequence of changes in the cellular redox determined by phosphatidylserine exposure on the outer state induced by Bcl-2 leading to enhanced activation of membrane (as determined by Annexin-V–FITC binding the prodrug neocarcinostatin. to propidium iodide (PI) -negative cells) reached While it is well accepted that enediyne antibiotics are approximately 45% and DNA fragmentation 65% very efficient inducers of apoptosis in the picomolar after Fas ligand exposure. In contrast, BJAB/FADD- range (Nicolaou et al., 1993, 1994b), the molecular dn cells were completely resistant against this death mechanisms of cell death induction and execution are stimulus (Figure 2a and c). In the presence of Fas far from being clear. The aim of the present study was, ligand only 10% of BJAB/FADD-dn cells died, thereby therefore, to unravel the proapoptotic signalling path- clearly demonstrating the efficiency of FADD-dn in ways induced by synthetic calicheamicin WII, an analog these cells. In contrast, neither phosphatidylserine of the naturally occurring calicheamicin. exposure (Figure 2b) nor DNA fragmentation

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9109 (Figure 2d) were significantly influenced by FADD-dn treatment of BJAB cells in the presence of an overexpression when the cells were treated with 1– antagonistic anti-CD95/Fas antibody did neither impair 100 pM calicheamicin. CD95/Fas-independent cell death calicheamicin-induced phosphatidylserine exposure was also observed in another experimental approach: (Figure 3b) nor DNA fragmentation (Figure 3d). The efficiency of the blocking anti-CD95/Fas antibody was demonstrated by blocking the activation of the CD95/ Fas pathway by an agonistic anti-CD95/Fas antibody employed as CD95/Fas receptor agonist. Phosphatidyl- serine exposure on the outer plasma membrane layer (Figure 3a) and DNA fragmentation (Figure 3c) in- duced upon treatment with agonistic anti-CD95/Fas were inhibited by approximately 73%. Thus, we conclude that calicheamicin-induced apoptosis in BJAB cells proceeds independently of CD95/Fas and FADD signalling.

Calicheamicin-induced apoptosis is mediated by the loss of mitochondrial membrane potential, cytochrome c release and caspase-9 and -3 processing To determine the involvement of the mitochondrial pathway, we measured mitochondrial activation after treatment of BJAB cells with calicheamicin. Exposure to calicheamicin resulted in a dramatic loss of the mitochondrial membrane potential as determined by the cationic dye JC-1 (Figure 4a). After 24 h of incubation, activation of the mitochondria was observed in 58% of the population, thereby clearly preceding DNA fragmentation that only reached 15% at this time point (Figure 4b). To gain further insight into calicheamicin-induced cell death, we investigated cytochrome c release and consecutive processing of caspase-9 and -3 by means of Western blot analysis. Treatment of BJAB cells led to the release of mitochondrial cytochrome c into the cytosol after incubation with 10–100 pM calicheamicin for 48 h (Figure 5a). Cytochrome c release was paralleled by procaspase-9 processing and appearance of the the 37 kDa cleavage product of procaspase-9 at 48 and 72 h. (Figure 5b). Weak caspase-3 processing to the active p17 subunit was detectable as early as 48 h and then reached a maximum at 72 h. Thus, cytochrome c release and procaspase-9 processing precede caspase-3 activation. To further investigate the functional role of caspases in the calicheamicin-mediated death signalling pathway, we employed the broad-spectrum caspase inhibitor

Figure 1 Cytotoxic versus proapoptotic effects of calicheamicin. (a) BJAB cells were treated with different concentrations of calicheamicin for 24 h (open circles) or 48 h (closed circles). Then, viability was determined by the LDH release assay. Values are given as % of control7s.d. (n ¼ 3). (b) Dose response: BJAB cells were treated with different concentrations of calicheamicin. After 72 h of incubation, DNA fragmentation was measured by flow cytometric analysis by assessing the cellular DNA content. Values are given as percentages of cells with hypodiploid DNA7s.d. (n ¼ 3). (c) Time course: BJAB cells were treated with 100 pM calicheamicin. After different times of incubation, DNA fragmen- tation was measured by flow cytometric analysis of cellular DNA content. Values are given as percentages of cells with hypodiploid DNA7s.d. (n ¼ 3)

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9110

Figure 2 Overexpression of dominant-negative FADD in BJAB cells does not impair calicheamicin-induced phosphatidylserine exposure and DNA fragmentation. Vector- or FADD-dn-transfected BJAB cells were either left untreated (control) or incubated with soluble recombinant human Fas ligand for 24 h. Then, phosphatidylserine exposure (a) and DNA fragmentation (c) were measured by flow cytometric analysis. Additionally, vector- or FADD-dn-transfected BJAB cells were treated with different concentrations of calicheamicin for 48 h. Then, phosphatidylserine exposure (b) and DNA fragmentation (d) were measured likewise. Values of phosphatidylserine exposure and DNA fragmentation are given as percentages of annexin-V-positive/ PI-negative cells7s.d. (n ¼ 3) and as percentages of cells with hypodiploid DNA7s.d. (n ¼ 3), respectively

zVAD-fmk (von Haefen et al., 2002). Interestingly, Role of antiapoptotic Bcl-2 and Bcl-xL and proapoptotic mitochondrial permeability transition after treatment of Bax on calicheamicin-induced mitochondrial permeability BJAB cells with calicheamicin was inhibited by zVAD- transition and DNA fragmentation fmk in a concentration-dependent manner (Figure 6a). Likewise, calicheamicin-induced DNA fragmentation Mitochondrial apoptosis is controlled by anti- and was inhibited by zVAD-fmk (Figure 6b). Inhibition of proapoptotic members of the Bcl-2 protein family (for both apoptotic hallmarks reached about 50%. We review, see, Martinou and Green, 2001). Indeed, the therefore conclude that caspase activation after cali- overexpression of Bcl-2 in Jurkat cells completely cheamicin treatment plays a functional role in death blocked calicheamicin-induced loss of mitochondrial signalling upstream of the mitochondria as well as in the membrane potential (Figure 7b) and DNA fragmenta- execution of cell death downstream of the mitochondria. tion (Figure 7d). In these experiments, we employed A similar mechanism was observed previously during epirubicin as a positive control since we had shown paclitaxel (taxol)-induced apoptosis that relies on a previously that epirubicin treatment led to hierarchical caspase-mediated amplification loop for full activation activation of the mitochondria followed by the activa- of the mitochondrial apoptosis pathway (von Haefen tion of caspase-3 and DNA fragmentation (Wieder et al., et al., 2003). 2001a). As expected, epirubicin-induced the loss of

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9111

Figure 4 Calicheamicin-induced mitochondrial permeability tran- sition precedes DNA fragmentation. BJAB cells were incubated with 100 pM calicheamicin. After different times of incubation, mitochondrial permeability transition (a) and DNA fragmentation (b) were measured in parallel by flow cytometric analysis on the single cell level. Values of mitochondrial permeability transition and DNA fragmentation are given as percentages of cells with low DC 7s.d. (n ¼ 3) and as percentages of cells with hypodiploid Figure 3 Antagonistic anti-Fas/CD95 antibody does not impair m DNA7s.d. (n ¼ 3), respectively. Mitochondrial permeability tran- calicheamicin-induced phosphatidylserine exposure and DNA sition and DNA fragmentation of the medium controls (time 0 h) fragmentation. BJAB cells were either left untreated (control) or did not change over time incubated with an agonistic anti-CD95/Fas antibody in the presence or absence of 5 mg/ml blocking, antagonistic anti-Fas/ CD95 antibody as indicated in the figure. After 24 h of incubation, mitochondrial membrane potential as well as DNA phosphatidylserine exposure (a) and DNA fragmentation (c) were measured by flow cytometry. Additionally, BJAB cells were treated fragmentation were strongly inhibited in Bcl-2-over- with 10 pM calicheamicin in the presence or absence of 5 mg/ml expressing Jurkat cells (Figure 7a and c). antagonistic anti-Fas/CD95 antibody as indicated in the figure. These data were confirmed in BJAB cells overexpres- After 48 h of incubation, phosphatidylserine exposure (b) and sing Bcl-xL (Figure 8). Although the efficiency of Bcl-xL DNA fragmentation (d) were measured. Values of phosphatidyl- was slightly lower in BJAB B cells as compared with serine exposure and DNA fragmentation are given as percentages of annexin-V-positive/PI-negative cells7s.d. (n ¼ 3) and as percen- Bcl-2 overexpression in the Jurkat T cells, calicheamicin- tages of cells with hypodiploid DNA7s.d. (n ¼ 3), respectively induced mitochondrial activation and DNA fragmenta- tion were inhibited by 40–65% by Bcl-xL (Figure 8b

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9112

Figure 5 Calicheamicin induces cytochrome c release and caspase- 3 activation. BJAB cells were incubated with different concentra- tions of calicheamicin or with 1 mg/ml epirubicin as positive control. After 48 or 72 h of incubation, cytosolic extracts were prepared and 30 mg of cytosolic protein were separated by sodium dodecylsulfate polyacrylamide gel electrophoresis and subjected to Western blot analysis. Immunoblots developed with anticyto- chrome c antibody (a) or anticaspase-3 or -9 antibody (b) are shown. Arrows indicate the positions of cytochrome c, procaspase- 3 and -9, the 17 kDa active subunit of caspase-3 (p17) and the 37 kDa cleavage product of procaspase-9. The experiment was repeated twice and similar results were obtained

Figure 6 Calicheamicin-induced mitochondrial permeability tran- sition and DNA fragmentation are inhibited by the broad- and d). Epirubicin was again used as a positive control spectrum caspase inhibitor zVAD-fmk. (a) BJAB cells were to check the cellular system functionally. In Bcl-x - incubated in the presence of 100 pM calicheamicin. Some cultures L were preincubated for 2 h with 10, 20 or 40 mM zVAD-fmk or 0.2% overexpressing BJAB cells, epirubicin-induced loss of DMSO (0 mM zVAD-fmk) as control. After 48 h of incubation, the mitochondrial membrane potential and DNA fragmen- percentages of cells showing low DCm were determined by flow tion were reduced by 67 and 78%, respectively (Figure cytometric analysis. Inhibition of the calicheamicin-induced effects 8a and c). is given in % of control7s.d. (n ¼ 3). In these experiments, the percentage of cells with low DCm after calicheamicin treatment was In recent studies, evidence was provided that Bax is a 54%. (b) BJAB cells were incubated in the presence of 100 pM key player in different forms of apoptosis initiated via calicheamicin. Some cultures were preincubated for 2 h with 10, 20 the mitochondrial pathway (Martinou and Green, 2001; or 40 mM zVAD-fmk or 0.2% DMSO (0 mM zVAD-fmk) as control. Roucou and Martinou, 2001; Smaili et al., 2001; After 48 h of incubation, the percentages of cells showing hypoploid DNA were determined by flow cytometric analysis. Bosanquet et al., 2002; Hemmati et al., 2002; von Inhibition of the calicheamicin-induced effects is given in % of Haefen et al., 2002). To this end, we investigated the control7s.d. (n ¼ 3). In these experiments, the percentage of cells influence of Bax knockout on calicheamicin-mediated with hypodiploid DNA after calicheamicin treatment was 36% cell death in human colorectal cancer HCT116 cells (Figure 9). Bax-proficient HCT116 wild-type cells were susceptible to treatment with calicheamicin. At picomo- system (Figure 9a and c). Notably, HCT116 cells express lar concentrations of calicheamicin, mitochondrial significant levels of Bak (Gillissen et al., 2003). As Bak activation and DNA fragmentation were observed in has been shown in a variety of reports to be a key 40 and 28% of the cell population, respectively. In mediator of mitochondrial apoptosis, our present data contrast, HCT116 Bax knockout cells cells showed only indicate that calicheamicin operates in a Bak-indepen- marginal mitochondrial activation of 13% (Figure 9b) dent but entirely Bax-dependent manner. and DNA fragmentation of 9% (Figure 9d). Bax Given the central role of Bax and to gain a more dependency was also observed in the case of epirubi- complete picture of how calicheamicin induces the cin-induced apoptosis. This demonstrates both the Bax activation of the mitochondrial pathway, we studied dependency of anthracycline-induced apoptosis and the the role of p53, a transcriptional activator of the Bax efficiency of the Bax knockout in our experimental gene. HCT116 p53 k.o. cells (Bunz et al., 1998) were

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9113

Figure 7 Overexpression of Bcl-2 blocks calicheamicin-induced mitochondrial permeability transition and DNA fragmentation. Vector- or Bcl-2-transfected Jurkat cells were either left untreated (control), or incubated with 1 mg/ml epirubicin (Epi) for 48 h. Then, mitochondrial permeability transition (a) and DNA fragmentation (c) were measured by flow cytometric analysis. Additionally, vector- or Bcl-2-transfected Jurkat cells were treated with different concentrations of calicheamicin for 48 h. Then, mitochondrial permeability transition (b) and DNA fragmentation (d) were measured by flow cytometry. Values of mitochondrial permeability transition and DNA fragmentation are given as percentages of cells with low DCm7s.d. (n ¼ 3) and as percentages of cells with hypodiploid DNA7s.d. (n ¼ 3), respectively compared with HCT116 control cells that carry two induced by the majority of conventional anticancer wild-type p53 alleles (HCT116 p53 wt). The induction of drugs. apoptosis was determined by flow cytometric analysis of genomic DNA fragmentation on the single cell level. Apoptosis occurred to an identical extent in the p53- Discussion deficient HCT116 p53 k.o. cells (49%) as compared with the HCT116 p53 wt cells (48%) (Figure 10a). In Calicheamicin is a naturally occurring enediyne anti- contrast, the anthracyclin drug epirubicin-induced biotic with high antileukemic potency and represents the apoptotic cell death in a clearly p53-dependent manner clinically most promising candidate of this class of (Figure 10b). The HCT116 p53 wt cells exhibited an antitumor agents (Thorson et al., 2000). The tremen- apoptosis rate of 59%, whereas only 27% of the p53 k.o. dous potency of calicheamicin or its synthetic WII analog cells underwent apoptotic DNA fragmentation upon against leukemic cells is, however, accompanied by exposure to epirubicin. This corroborates that mechan- cytotoxic side effects. Therefore, calicheamicin is em- isms of calicheamicin-induced cell death differ from that ployed as an immunoconjugate to focus its cell killing

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9114

Figure 8 Overexpression of Bcl-xL in BJAB cells inhibits calicheamicin-induced mitochondrial permeability transition and DNA fragmentation. Vector- or Bcl-xL-transfected BJAB cells were either left untreated (control), or incubated with 1 mg/ml epirubicin (Epi) for 48 h. Then, mitochondrial permeability transition (a) and DNA fragmentation (c) were measured by flow cytometric analysis by the use of the cationic dye JC-1. Additionally, vector- or Bcl-xL-transfected BJAB cells were treated with different concentrations of calicheamicin for 48 h. Then, mitochondrial permeability transition (b) and DNA fragmentation (d) were determined. Values of mitochondrial permeability transition and DNA fragmentation are given as percentages of cells with low DCm7s.d. (n ¼ 3) and as percentages of cells with hypodiploid DNA (n ¼ 3), respectively

activity to cancer cells. Conjugates to CD33 that is isms leading to apoptosis in cancer cells remained expressed on more than 90% of myeloid leukemic blasts enigmatic. This is surprising in view of the impressive are currently evaluated in therapy of acute myeloid apoptosis-inducing potency of this class of antitumor leukemias (reviewed in Nabhan and Tallman, 2002). In agents. In the present study, we therefore employed the this vein treatment of a patient with an anti-CD33 synthetic and highly effective calicheamicin WII to calicheamicin immunoconjugate (gemtuzumab ozoga- investigate apoptosis signalling by this drug. Here, we micin, CMA-676, Mylotarg) led to molecular remission show that calicheamicin triggers the mitochondrial of Philadelphia/bcr-abl-positive acute myeloid leukemia pathway of apoptosis. A model for calicheamicin- (de Vetten et al., 2000). Meanwhile, gemtuzumab induced apoptosis induction is provided in Figure 11. ozogamicin was successfully applied in phase I and In contrast to a previous finding suggesting a phase II trials in patients with acute myeloid leukemias paradoxically enhanced apoptosis induction in cells (Sievers et al., 1999; van der Velden et al., 2001). overexpressing Bcl-2, both Bcl-2 and Bcl-xL strongly Although calicheamicin has already found its way to inhibited calicheamicin-induced cell death in the present the clinic as immunoconjugate, the molecular mechan- study. Moreover, calicheamicin induced apoptosis in an

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9115

Figure 9 Bax knockout in HCT116 cells abrogates calicheamicin-induced mitochondrial permeability transition and DNA fragmentation. HCT116 wild-type or HCT116 Bax knockout cells were either left untreated (control), or incubated with 1 mg/ml epirubicin (Epi) for 48 h. Then, mitochondrial permeability transition (a) and DNA fragmentation (c) were measured by flow cytometric analysis. Additionally, HCT116 wild type or HCT116 Bax À/À cells were incubated in the presence of different concentrations of calicheamicin for 48 h. Then, mitochondrial permeability transition (b) and DNA fragmentation (d) were measured by flow cytometry. Values of mitochondrial permeability transition and DNA fragmentation are given as percentages of cells with low DCm7s.d. (n ¼ 3) and as percentages of cells with hypodiploid DNA7s.d. (n ¼ 3), respectively entirely Bax-dependent manner: the knockout of proa- DNA fragmentation without affecting membrane integ- poptotic Bax in HCT116 cells almost completely rity, thereby ruling out unspecific toxic effects, that can abrogated apoptosis induction by calicheamicin. Inter- be observed at high concentrations of other hydro- estingly, the activation of the mitochondria occurred in phobic substances, for example, ceramides (Wieder et al., a caspase-dependent manner. The pancaspase inhibitor 1997) or alkylphosphocholines (Wieder et al., 1998). zVAD-fmk not only interfered with cell death execution Apoptotic hallmarks, such as phosphatidylserine ex- and inhibited apoptotic DNA fragmentation, but also posure and DNA fragmentation, were already observed inhibited mitochondrial activation by calicheamicin. at very low concentrations of calicheamicin ranging Nevertheless, calicheamicin induced apoptosis in a from 10 to 100 pM. Neither overexpression of a death-receptor-independent manner. Thus, death recep- dominant-negative mutant of the adaptor protein tor-induced apoptosis that is known to trigger caspase FADD nor blocking of CD95/Fas by an antagonistic activation upstream of the mitochondria (Belka et al., antibody significantly inhibited calicheamicin-induced 2000; Daniel, 2000; Daniel et al., 2001), could be ruled apoptosis. Thus, apoptosis signalling after calicheamicin out as cause for the mitochondrial activation in the treatment proceeds in a CD95/Fas-independent manner present case. This places caspase activation upstream of through the mitochondrial pathway. the mitochondria and may indicate a unique property of Mitochondrial apoptosis is the most prominent endiyene compounds. Calicheamicin induces apoptotic intrinsic apoptosis signalling pathway (for reviews, see,

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9116

Figure 10 p53 knockout in HCT116 cells does not affect calicheamicin-induced DNA fragmentation. HCT116 wild-type or HCT116 p53 knockout cells were incubated in the presence of different concentrations of calicheamicin for 48 h (a). Then, genomic DNA fragmentation was measured by flow cytometry. The number of cells undergoing DNA fragmentation is given as Figure 11 Signalling model of calicheamicin-induced apoptosis. percentages of cells with hypodiploid DNA7s.d. (n 3), respec- Calicheamicin binds to the minor groove of the DNA helix, ¼ 0 tively. Alternatively, cells were incubated with 1 mg/ml epirubicin preferentially to the 3 ends of oligopurine tracts and causes (Epi) for 48 h (b) sequence-selective oxidation of deoxyribose and bending of the DNA helix by an induced-fit mechanism of DNA target recogni- tion. In addition, calicheamicin forms biradicals that induce DNA strand breaks and additional, so far undefined damage signals. These events trigger a nuclear stress signal that ultimately initiates Kroemer and Reed, 2000; Martinou and Green, 2001). the mitochondrial apoptosis signalling cascade through death It is triggered by various cytotoxic drugs such as taxol or receptor- and FADD-independent mechanisms. The activation of epirubicin (Wieder et al., 2001a; von Haefen et al., mitochondria occurs through an entirely Bax-dependent mechan- ism that can be inhibited by either Bcl-2 or Bcl-xL. Mitochondrial 2003), natural compounds such as piceatannol (Wieder activation for apoptosis results in release of cytochrome c into the et al., 2001b), truncated ceramides (von Haefen et al., cytosol, processing of procaspase-9 and subsequent activation of 2002) or ionizing radiation (Belka et al., 2000). In a first procaspase-3. Notably, inhibition of caspase activities by the series of experiments, we demonstrated that challenge pancaspase-inhibitor zVAD-fmk partially interferes with mito- chondrial membrane potential breakdown. This suggests that with calicheamicin led to a dramatic decrease of the caspase processing might be initiated upstream of the mitochon- mitochondrial membrane potential, which was observed dria. Alternatively, caspases might mediate an autoamplification in approximately 80% of BJAB cells. Mitochondrial loop in calicheamicin-induced apoptosis that is required for the full permeability transition preceded DNA fragmentation, activation of the mitochondrial pathway as described earlier for thereby indicating that mitochondrial activation plays paclitaxel-induced cell death. Casp: caspase an important role in calicheamicin signalling. This

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9117 assumption was substantiated by measurements show- the notion of calicheamicin differing from conventional ing that calicheamicin-induced mitochondrial activation anticancer drugs. was accompanied by cytochrome c release into the Caspases are the main executioners of apoptotic cell cytosol and cleavage of caspase-9 that preceded proces- death (Cohen, 1997). Furthermore, evidence has been sing of procaspase-3 to the active 17 kDa subunit. Both provided that caspases are also involved in intracellular cytochrome c release and procaspase-9 processing are signalling pathways (for review, see, Salvesen and Dixit, generally accepted as essential parts of mitochondrial 1997). For example, a caspase-3-driven mitochondrial signalling and execution (Li et al., 1997; Liu et al., 1996; feedback loop has been proposed during cytotoxic drug- Raisova et al., 2000; von Haefen et al., 2002). This is and UV radiation-induced apoptosis (Slee et al., 2000). well in line with our finding that calicheamicin-induced Moreover, a caspase-3- and -8-driven feedback ampli- cell death can be inhibited by Bcl-2 and Bcl-xL, as these fication loop was recently shown to be required for full proteins are key regulators of mitochondrial cytochrome release of cytochrome c during paclitaxel-induced c release. Specifically, employing two independent apoptosis (von Haefen et al., 2003). Interestingly, the cellular systems we did not observe the previously broad-spectrum caspase inhibitor zVAD-fmk almost suggested paradoxical enhancement of calicheamicin- completely inhibited mitochondrial activation and DNA induced apoptosis in Bcl-2-overexpressing PC12 cells fragmentation upon calicheamicin WII treatment. Thus, (Cortazzo and Schor, 1996). This indicates that cali- calicheamicin-induced mitochondrial permeability tran- cheamicin-induced apoptosis critically depends on the sition proceeds in a caspase-dependent manner. The activation of the mitochondrial pathway of apoptosis. necessity of caspases for mitochondrial activation This model is further supported by the fact that loss of distinguishes calicheamicin-mediated from, for example, Bax impaired calicheamicin-induced apoptosis. Bax ceramide-induced cell death. In the case of ceramide, it directly mediates the release of cytochrome c and has has been recently demonstrated that signalling events been implicated in the formation of cytochrome c upstream of mitochondria are caspase independent, releasing channels in the outer mitochondrial membrane whereas the execution of cell death downstream of or the opening of VDAC pores (Tsujimoto and Shimizu, mitochondria proceeds in a caspase-dependent manner 2000; von Ahsen et al., 2000; van Loo et al., 2002). The (Utzand Anderson, 2000; Daniel et al., 2001; Ameisen, effect of Bax loss or Bcl-2/Bcl-xL overexpression was 2002; Distelhorst, 2002; Koonin and Aravind, 2002). even more pronounced when apoptotic DNA fragmen- This caspase dependency of mitochondrial apoptosis tation was investigated as compared with DCm break- is of special interest, since caspases are generally down. This indicates that loss of DCm can occur in the considered to act as initiators and executioners that absence of apoptosis execution. are activated either downstream of mitochondrial In a recent review, Martinou and Green have put apoptosis (Engels et al., 2000; Wieder et al., 2001a) or forward two models for breaking the mitochondrial link mitochondria to the the death receptor pathway barrier: (i) after a death signal BH3-only proteins (Belka et al., 2000, 2001). Even while caspases act (where BH3 stands for Bcl-2 homology domain 3) upstream of the mitochondria in the latter case, we activate members of the Bax subfamily leading to their could clearly exclude death receptor initiated signalling insertion in the outer mitochondrial membrane, where events in calicheamicin-induced apoptosis. Thus, cas- they form large channels and (ii) death stimuli activate a pases appear to be initiators of calicheamicin-induced pathway involving classical permeability transition pore mitochondrial activation and execution of apoptosis in a openers such as Ca2 þ . The latter pathway was suggested Bax-dependent manner. to operate independently of members of the Bax The mechanisms of caspase activation by calicheami- subfamily of Bcl-2 homologs (Martinou and Green, cin remain, however, to be elucidated. Caspase activa- 2001). In this context, we found that HCT116 Bax tion taking place upstream of the mitochondria has knockout cells were insensitive towards calicheamicin, recently been described as potential link between nuclear whereas HCT116 wild-type cells died by apoptosis after stress, for example by the topoisomerase II inhibitor challenge with the drug. This is worth noting since these etoposide, and the mitochondrial apoptosis pathway cells express significant amounts of Bak (Gillissen et al., (Lassus et al., 2002; Robertson et al., 2002). This may 2003). Thus, Bak is not sufficient to mediate sensitivity indicate that the key steps of caspase activation are for caliceamicin in this model. This indicates that initiated upstream of the mitochondria and are then calicheamicin triggers cell death preferentially via Bax utilized to amplify the mitochondrial cell death pathway and not via Bak, that is, in an entirely Bax-dependent in a Bcl-2-dependent manner (Marsden et al., 2002). manner. Nevertheless, zVAD-fmk is a rather weak inhibitor of Interestingly, cell death induction by calicheamicin caspase-2 and this indirectly suggests that caspase-2 is was independent from p53. Both p53 wild-type and p53 not the key effector in this setting. knockout HCT116 cells underwent apoptosis to an Given the interaction of calicheamicin and other identical extent when exposed to calicheamicin. This was enediyne compounds with the DNA, we favor the surprising in view of the established role of p53 as direct concept of a nuclear stress signal that initiates apoptosis. transcriptional activator of Bax and indirect transcrip- In this line, there is recent evidence that caspase tional activator of the Bak gene. Apart from pointing activation could potentially be initiated out of the cell out the potential usefulness of this compound in the cycle by direct interaction with cell cycle regulating treatment of p53 mutated cancers, this finding supports proteins: caspase-3 was shown to interact with p21

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9118 physically (Suzuki et al., 1999) or cyclin D3 (Mendel- Control vector- and Bcl-2-transfected Jurkat cells (Belka et al., sohn et al., 2002). Another possibility how caspase 2000) stably overexpressing the antiapoptotic protein Bcl-2 activation may take place upstream of the mitochondria were a kind gift of K Schulze-Osthoff (University of Mu¨ nster, could be a mitochondrial amplification loop (Slee et al., Germany). Cells were grown in RPMI 1640 medium supple- 2000; von Haefen et al., 2003). There, caspase-3 and -8 mented with 10% fetal calf serum, 0.56 g/l L-glutamine, 100 000 U/l penicillin and 0.1 g/l streptomycin. Media and activation is initiated downstream of the mitochondria culture reagents were from Life Technologies GmbH (Karls- but then serves as secondary amplifier to achieve ruhe, Germany). BJAB and Jurkat cells were subcultured every optimal mitochondrial activation, possibly via a Bid- 3–4 days by dilution of the cells to a concentration of 1 Â 105 dependent mechanism. In the case of calicheamicin, the cells/ml. Human colorectal cancer HCT116 wild-type cells, exact signaling requirements and caspases involved in HCT116 p53 knockout (Bunz et al., 1998) and HCT116 Bax this loop are under current investigation. knockout cells (Zhang et al., 2000) were a kind gift of Bert In summary, the data presented in this study provide Vogelstein (Johns Hopkins University School of Medicine, experimental evidence how calicheamicin mediates Baltimore, MD, USA). Confluent HCT116 cells were sub- apoptosis in a p53-independent manner through an cultured every 4–5 days after detaching the cells with 0.1% entirely Bax-dependent activation of the mitochondrial trypsin, 0.02% ethylenediamine tetraacetic acid (EDTA) in phosphate-buffered saline (PBS). HCT116 cell clones were pathway. These insights into cell death pathways after cultured under the same conditions as described for the BJAB calicheamicin treatment not only help to understand the and Jurkat cells above. impressive potency of this anticancer compound, but also serve as basis for the currently undertaken efforts to Measurement of cell death elucidate the upstream events in enediyne-induced apoptosis initiation. Cytotoxicity of calicheamicin was measured by the release of lactate dehydrogenase (LDH) as described previously (Frie- drich et al., 2001). After incubation with different concentra- tions of calicheamicin for 24 and 48 h, LDH activity released Materials and methods by BJAB cells was measured in the cell culture supernatants using the Cytotoxicity Detection Kit from Boehringer-Man- Materials nheim (Mannheim, Germany). The supernatants were centri- fuged at 300 g for 5 min. Cell-free supernatants (20 ml) were Polyclonal rabbit anti-human caspase-3 (developed against diluted with 80 ml PBS and 100 ml reaction mixture containing human recombinant protein) and monoclonal mouse anti- 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chlor- human cytochrome c (clone 7H8. 2C12) antibodies were from ide (INT), sodium lactate, NAD þ and diaphorase were added. Pharmingen (Hamburg, Germany) and were used at 1 : 2000 Then, time-dependent formation of the reaction product was and 1 : 1000, respectively. Polyclonal goat anti-human caspase- quantified photometrically at 490 nm. The maximum amount 9 was from R&D Systems GmbH (Wiesbaden-Nordenstadt, of LDH activity released by the cells was determined by lysis of Germany) and was used at 1 : 1000. Secondary anti-rabbit, the cells using 0.1% Triton X-100 in culture medium and was anti-goat and anti-mouse horseradish peroxidase (HRP)- set as 100% cell death. conjugated antibodies were from Promega (Mannheim, Germany) or Southern Biotechnology Associates (Birming- ham, AL, USA) and were used at 1 : 5000. Monoclonal, Measurement of CD95/Fas-mediated cell death antagonistic anti-CD95/Fas antibody (clone SM1/23) was For determination of CD95/Fas-mediated cell death, 1 Â 105 from Bender Med Systems (Vienna, Austria) and was used at BJAB cells/ml were treated with either recombinant human a concentration of 5 mg/ml. RNase A was from Roth soluble Fas ligand (rhs Super-Fas ligand; Alexis, Gru¨ nberg, (Karlsruhe, Germany). The broad-spectrum caspase inhibitor Germany) at 0.1 mg/ml or with an agonistic, that is, cross- zVAD-fmk (in which z stands for benzyloxycarbonyl and fmk linking, anti-CD95/Fas antibody (clone anti-APO-1 IgG3) at for fluoromethyl ketone) was from Calbiochem-Novabiochem 1 mg/ml for 24 h (Dhein et al., 1992). Then, cell death was GmbH (Bad Soden, Germany) and was dissolved in dimethyl assessed by measurement of phosphatidylserine exposure or by sulfoxide (DMSO) to give a 20 mM stock solution. According measurement of DNA fragmentation as described below. In to the manufacturer, this inhibitor was synthesized as a methyl some experiments, a blocking anti-CD95/Fas antibody (clone ester to enhance cell permeability. DMSO (0.2%) (vehicle) was SM1/23) was added at 5 mg/ml. added to controls. This treatment did not induce significant apoptotic DNA fragmentation (data not shown). Epirubicin Measurement of phosphatidylserine exposure was purchased from Pharmacia Upjohn (Erlangen, Germany). Calicheamicin WII was prepared by total synthesis in 49 steps Cell death was determined by staining cells with annexin-V– as described previously (Nicolaou et al., 1994a) and was FITC and counterstaining with PI. During apoptosis, the dissolved in DMSO to give a 0.1 mM stock solution. phospholipid phosphatidylserine is exposed to the outer leaflet Calicheamicin was then further diluted into growth medium of the plasma membrane (Fadok et al., 2001; Schlegel and to achieve picomolar concentrations. Williamson, 2001). Annexin-V–FITC then binds to phospha- tidylserine leading to an increase of the fluorescence. On the other hand, PI is excluded from cells with intact membranes. Cell culture PI positivity is therefore a sign of cell necrosis, whereas cells Control vector- and FADD-dn-transfected BJAB cells stably which are annexin-V–FITC positive but PI negative are expressing a dominant-negative FADD mutant lacking the N- generally defined as apoptotic (Vermes et al., 1995). For the terminal death effector domain have been described earlier Annexin-V/PI assay, 1 Â 105 cells were washed twice with ice- (Wieder et al., 2001a). Bcl-xL-transfected BJAB cells were a cold PBS and then resuspended in binding buffer (10 mM N-(2- kind gift of S Fulda (University of Ulm, Germany) and have hydroxyethyl)piperazin-N0-3(propansulfonicacid) (HEPES)/ been characterized in a previous study (Fulda et al., 2001). NaOH (pH 7.4), 140 mM NaCl, 2.5 mM CaCl2) at a concentra-

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9119 tion of 1 Â 106 cells/ml. Next, 5 ml of Annexin-V–FITC (BD 2002). After incubation of BJAB cells in medium supplemen- Pharmingen, Heidelberg, Germany) and 10 mlof50mg/ml PI ted with different concentrations of calicheamicin, cells were (Sigma-Aldrich, Taufkirchen, Germany) were added to the harvested in PBS, equilibrated in hypotonic buffer (20 mM cells. Analyses were performed on an FACScan (Becton HEPES (pH 7.4), 10 mM KCl, 2 mM MgCl2,1mM EDTA) and Dickinson, Heidelberg, Germany) using the CellQuest analysis centrifuged at 300 g for 5 min. The resulting cell pellets were software. then dissolved in hypotonic buffer containing phenylmethyl sulfonylfluoride (final concentration 0.1 mM) and incubated on Measurement of DNA fragmentation ice for 15 min. Cells were homogenized by passing the cells through a syringe (G 20) approximately 20 times. The DNA fragmentation was measured essentially as described membranes were pelleted by twofold centrifugation at (Daniel et al., 1999; Essmann et al., 2000). After treatment 10 000 g,41C for 10 min and the supernatant of the second with different concentrations of calicheamicin, cells were centrifugation was used as cytosolic extract. Protein concen- detached by trypsination and collected by centrifugation at tration was determined using the bicinchoninic acid assay from 300 g for 5 min. Cells were first washed with complete cell Pierce (Rockford, IL, USA). Then, Western blot analyses with culture medium to stop tryptic digestion and then with PBS at 30 mg, cytosolic protein were performed as described below and 41C. Cells were fixed in PBS/0.74% (v/v) formaldehyde on ice conducted with anticytochrome c antibodies or anticaspase- for 30 min, pelleted, incubated with ethanol/PBS (2 : 1, v/v) for 3/-9 antibodies. 15 min, pelleted and resuspended in PBS containing 40 mg/ml RNase A. RNA was digested for 30 min at 371C. Cells were pelleted again and finally resuspended in PBS containing Western blot analysis 50 mg/ml PI. Nuclear DNA fragmentation was quantified by flow cytometric determination of hypodiploid DNA. Data Cytosolic protein (30 mg) were loaded per lane and were were collected and analysed using an FACScan (Becton separated by SDS–PAGE. Then, blotting of proteins onto Dickinson; Heidelberg, Germany) equipped with the CELL- nitrocellulose membranes (Schleicher and Schuell, Dassel, Quest software. Data are given in % hypoploidy (subG1), Germany) was performed exactly as described (Wieder et al., which reflects the number of apoptotic cells. 2001a). After blotting, the membrane was blocked for 1 h in PBST (PBS, 0.05% Tween-20) containing 3% nonfat dry milk and incubated with primary antibody for 1 h. After Measurement of the mitochondrial permeability transition the membrane had been washed three times in PBST, After incubation with different concentrations of calicheamicin secondary antibody in PBST was applied for 1 h. Finally, the WII, mitochondrial permeability transition was determined by membrane was washed in PBST again and the ECL enhanced staining the cells with 5,50,6,60-tetrachloro-1,10,3,30-tetraethyl- chemiluminescence system from Amersham Buchler benzimidazolylcarbocyanin iodide (JC-1; Molecular Probes, (Braunschweig, Germany) was used to visualize the protein Leiden, The Netherlands) exactly as described elsewhere bands in question. (Wieder et al., 2001a). Mitochondrial permeability transition was then quantified by flow cytometric determination of cells with decreased fluorescence, that is, with mitochondria Acknowledgements displaying a lower membrane potential. Data were collected This work was supported by grants from the Deutsche Jose´ and analysed using an FACScan (Becton Dickinson) equipped Carreras Leuka¨ mie-Stiftung, the Verein zur Fo¨ rderung der with the CELLQuest software. Data are given in % cells with Tagesklinik and the Deutsche Krebshilfe. We thank B Vogelstein (Johns Hopkins University School of Medicine, low DCm, which reflects the number of cells undergoing mitochondrial apoptosis. Baltimore, MD, USA) for kindly providing HCT116 wild- type, p53 À/À and Bax À/À cells and the congeneic controls, S Fulda (University of Ulm, Germany) for kindly providing Determination of cytochrome c release and processing of BJAB cells overexpressing Bcl-x and K Schulze-Osthoff caspase-3 L (University of Du¨ sseldorf, Germany) for kindly providing Cytosolic extracts were prepared according to a method Jurkat cells overexpressing Bcl-2. The excellent technical described previously (Raisova et al., 2001; von Haefen et al., assistance of A Richter is gratefully acknowledged.

References

Ameisen JC. (2002). Cell Death Differ., 9, 367–393. Cortazzo M and Schor NF. (1996). Cancer Res., 56, Belka C and Budach W. (2002). Int. J. Radiat. Biol., 78, 1199–1203. 643–658. Daniel PT. (2000). Leukemia, 14, 2035–2044. Belka C, Rudner J, Wesselborg S, Stepczynska A, Marini P, Daniel PT, Schulze-Osthoff K, Belka C and Gu¨ ner D. (2003). Lepple-Wienhues A, Faltin H, Bamberg M, Budach W and Essays Biochem., 39, (in press). Schulze-Osthoff K. (2000). Oncogene, 19, 1181–1190. Daniel PT, Sturm I, Ritschel S, Friedrich K, Do¨ rken B, Belka C, Schmid B, Marini P, Durand E, Rudner J, Faltin H, Bendzko P and Hillebrand T. (1999). Anal. Biochem., 266, Bamberg M, Schulze-Osthoff K and Budach W. (2001). 110–115. Oncogene, 20, 2190–2196. Daniel PT, Sturm I, Wieder T and Schulze-Osthoff K. (2001). Bosanquet AG, Sturm I, Wieder T, Essmann F, Bosanquet Leukemia, 15, 1022–1032. MI, Head DJ, Do¨ rken B and Daniel PT. (2002). Leukemia, de Vetten MP, Jansen JH, van der Reijden BA, Berger MS, 16, 1035–1044. Zijlmans JM and Lowenberg B. (2000). Br. J. Haematol., BunzF, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown 111, 277–279. JP, Sedivy JM, Kinzler KW and Vogelstein B. (1998). Dhein J, Daniel PT, Trauth BC, Oehm A, Moller P and Science, 282, 1497–1501. Krammer PH. (1992). J. Immunol., 149, 3166–3173. Cohen GM. (1997). Biochem. J., 326, 1–16. Distelhorst CW. (2002). Cell Death Differ., 9, 6–19.

Oncogene Calicheamicin activates mitochondria via caspases and Bax A Prokop et al 9120 Engels IH, Stepczynska A, Stroh C, Lauber K, Berg C, Raisova M, Hossini A, Eberle J, Riebeling C, Wieder T, Sturm Schwenzer R, Wajant H, Janicke RU, Porter AG, Belka C, I, Daniel PT, Orfanos CE and Geilen CC. (2001). J. Invest. Gregor M, Schulze-Osthoff K and Wesselborg S. (2000). Dermatol., 177, 333–340. Oncogene, 19, 4563–4573. Robertson JD, Enoksson M, Suomela M, Zhivotovsky B and Essmann F, Wieder T, Otto A, Muller EC, Do¨ rken B and Orrenius S. (2002). J. Biol. Chem., 277, 29803–29809. Daniel PT. (2000). Biochem. J., 346, 777–783. Roucou X and Martinou JC. (2001). Cell Death Differ., 8, Fadok VA, Xue D and Henson P. (2001). Cell Death Differ., 8, 875–877. 582–587. Rudner J, Jendrossek V and Belka C. (2002). Apoptosis, 7, Friedrich K, Wieder T, Von Haefen C, Radetzki S, Janicke R, 441–447. Schulze-Osthoff K, Do¨ rken B and Daniel PT. (2001). Rudner J, Lepple-Wienhues A, Budach W, Berschauer J, Oncogene, 20, 2749–2760. Friedrich B, Wesselborg S, Schulze-Osthoff K and Belka C. Fulda S, Meyer E, Friesen C, Susin SA, Kroemer G and (2001). J. Cell Sci., 114, 4161–4172. Debatin KM. (2001). Oncogene, 20, 1063–1075. Salvesen GS and Dixit VM. (1997). Cell, 91, 443–446. Gillissen B, Essmann F, Graupner V, Sta¨ rck L, Radetzki S, Salzberg AA and Dedon PC. (2000). Biochemistry, 39, Do¨ rken B, Schulze-Osthoff K and Daniel PT. (2003). 7605–7612. EMBO J., 22, 3580–3590. Schlegel RA and Williamson P. (2001). Cell Death Differ., 8, Gu¨ ner D, Sturm I, Hemmati PG, Hermann S, Hauptmann S, 551–563. Wurm R, Budach V, Do¨ rken B, LorenzM and Daniel PT. Schor NF, Tyurina YY, Fabisiak JP, Tyurin VA, Lazo JS and (2003). Int. J. Cancer, 103, 445–454. Kagan VE. (1999). Brain Res., 831, 125–130. Hemmati PG, Gillissen B, von Haefen C, Wendt J, Sta¨ rck L, Sievers EL, Appelbaum FR, Spielberger RT, Forman SJ, Gu¨ ner D, Do¨ rken B and Daniel PT. (2002). Oncogene, 21, Flowers D, Smith FO, Shannon-Dorcy K, Berger MS and 3149–3161. Bernstein ID. (1999). Blood, 93, 3678–3684. Hiatt A, Merlock R, Mauch S and Wrasidlo W. (1994). Slee EA, Keogh SA and Martin SJ. (2000). Cell Death Differ., Bioorg. Med. Chem., 2, 315–322. 7, 556–565. Koonin EV and Aravind L. (2002). Cell Death Differ., 9, Smaili SS, Hsu YT, Sanders KM, Russell JT and Youle RJ. 394–404. (2001). Cell Death Differ., 8, 909–920. Kroemer G and Reed JC. (2000). Nat. Med., 6, 513–519. Sturm I, Bosanquet A, Hermann S, Gu¨ ner D, Do¨ rken B and Lassus P, Opitz-Araya X and Lazebnik Y. (2002). Science, Daniel P. (2003). Cell Death Differ, 10, 477–484. 297, 1352–1354. Sturm I, Petrowsky H, VolzR, LorenzM, RadetzkiS, Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Hillebrand T, Wolff G, Hauptmann S, Do¨ rken B and Daniel Alnemri ES and Wang X. (1997). Cell, 91, 479–489. PT. (2001). J. Clin. Oncol., 19, 2272–2281. Linenberger ML, Hong T, Flowers D, Sievers EL, Gooley TA, Suzuki A, Tsutomi Y, Miura M and Akahane K. (1999). Bennett JM, Berger MS, Leopold LH, Appelbaum FR and Oncogene, 18, 1239–1244. Bernstein ID. (2001). Blood, 98, 988–994. Thorson JS, Sievers EL, Ahlert J, Shepard E, Whitwam RE, Liu X, Kim CN, Yang J, Jemmerson R and Wang X. (1996). Onwueme KC and Ruppen M. (2000). Curr. Pharm. Des., 6, Cell, 86, 147–157. 1841–1879. Maiese WM, Lechevalier MP, Lechevalier HA, Korshalla J, Tsujimoto Y and Shimizu S. (2000). Cell Death Differ., 7, Kuck N, Fantini A, Wildey MJ, Thomas J and Greenstein 1174–1181. M. (1989). J. Antibiot., 42, 558–563. UtzPJ and Anderson P. (2000). Cell Death Differ., 7, 589–602. Marsden VS, O’Connor L, O’Reilly LA, Silke J, Metcalf D, van der Velden VH, te Marvelde JG, Hoogeveen PG, Ekert PG, Huang DC, Cecconi F, Kuida K, Tomaselli KJ, Bernstein ID, Houtsmuller AB, Berger MS and van Dongen Roy S, Nicholson DW, Vaux DL, Bouillet P, Adams JM and JJ. (2001). Blood, 97, 3197–3204. Strasser A. (2002). Nature, 419, 634–637. van Loo G, Saelens X, van Gurp M, MacFarlane M, Martinou JC and Green DR. (2001). Nat. Rev. Mol. Cell. Biol., Martin SJ and Vandenabeele P. (2002). Cell Death Differ., 2, 63–67. 9, 1031–1042. Mendelsohn AR, Hamer JD, Wang ZB and Brent R. (2002). Vermes I, Haanen C, Steffens-Nakken H and Reutelingsperger Proc. Natl. Acad. Sci. USA, 99, 6871–6876. C. (1995). J. Immunol. Methods, 184, 39–51. Mrozek A, Petrowsky H, Sturm I, Krauss J, Hermann S, von Ahsen O, Waterhouse NJ, Kuwana T, Newmeyer DD and Hauptmann S, LorenzM and Daniel P. (2003). Cell Death Green DR. (2000). Cell Death Differ., 7, 1192–1199. Differ, 10, 461–467. von Haefen C, Wieder T, Essmann F, Schulze-Osthoff K, Nabhan C and Tallman MS. (2002). Clin. Lymphoma, 2 Do¨ rken B and Daniel PT. (2003). Oncogene, 22, 2236–2247. (Suppl. 1), S19–S23. von Haefen C, Wieder T, Gillissen B, Sta¨ rck L, Graupner V, Nicolaou KC, Li T, Nakada M, Hummel W, Hiatt A and Do¨ rken B and Daniel PT. (2002). Oncogene, 21, 4009–4019. Wrasidlo W. (1994a). Angew. Chem., 33, 183–186. Wieder T, Essmann F, Prokop A, SchmelzK, Schulze-Osthoff Nicolaou KC, Pitsinos EN, Theodorakis EA, Saimoto H and K, Beyaert R, Do¨ rken B and Daniel PT. (2001a). Blood, 97, Wrasidlo W. (1994b). Chem. Biol., 1, 57–66. 1378–1387. Nicolaou KC, Stabila P, Esmaeli-Azad B, Wrasidlo W Wieder T, Geilen CC, Kolter T, Sadeghlar F, Sandhoff K, and Hiatt A. (1993). Proc. Natl. Acad. Sci. USA, 90, Brossmer R, Ihrig P, Perry D, Orfanos CE and Hannun YA. 3142–3146. (1997). FEBS Lett., 411, 260–264. Prokop A, Wieder T, Sturm I, Essmann F, Seeger K, Wuchter Wieder T, Orfanos CE and Geilen CC. (1998). J. Biol. Chem., C, Ludwig W-D, Henze G, Do¨ rken B and Daniel PT. (2000). 273, 11025–11031. Leukemia, 14, 1606–1613. Wieder T, Prokop A, Bagci B, Essmann F, Bernicke D, Radetzki S, Kohne CH, von Haefen C, Gillissen B, Sturm I, Schulze-Osthoff K, Do¨ rken B, SchmalzHG, Daniel PT and Do¨ rken B and Daniel PT. (2002). Oncogene, 21, 227–238. Henze G. (2001b). Leukemia, 15, 1735–1742. Raisova M, Bektas M, Wieder T, Daniel P, Eberle J, Orfanos Zhang L, Yu J, Park BH, Kinzler KW and Vogelstein B. CE and Geilen CC. (2000). FEBS Lett., 473, 27–32. (2000). Science, 290, 989–992.

Oncogene