Introduction

Two distinct apoptotic pathways are known to operate in cells: death receptor-initiated and mitochondrial-mediated (Hengartner, 2000; Kroemer and Reed, 2000). Ligation of cell surface death receptors including Fas (APO-1, CD95), TNFR1, DR3 (TRAMP), DR4 (TRAIL-R1), or DR5 (TRAIL-R2) by their specific ligands triggers a death receptor apoptotic process. FADD is first recruited directly or indirectly to the death receptor, followed by the association of procaspase-8 (FLICE or MACH), which is then cleaved and activated at the receptor complex (for a review, see Nagata, 1997; Krueger et al., 2001; Thome and Tschopp, 2001). In contrast, cytotoxic drugs, UV, and -irradiation are known to trigger another apoptotic pathway initiated in mitochondria (Kroemer and Reed, 2000). The mitochondrial released cytochrome c, together with dATP and apoptotic protease-activating factor-1 (Apaf1), form the complexes that activate procaspase-9 (for a review, see Wang, 2001). Following activation of caspase-8 and caspase-9, death receptor and mitochondrial pathways converge to activate downstream effector caspases.

Reactive oxygen species (ROS) have been implicated in the induction or enhancement of apoptosis. ROS is produced upon stress stimulation such as UV, -irradiation, or cytotoxic drugs. There is a very close link between ROS production and stress-induced apoptosis (Simizu et al., 1998). In contrast, the role of ROS in Fas-induced cell death remains uncertain (Hug et al., 1994; Schulze-Osthoff et al., 1994; Gulbins et al., 1996; Um et al., 1996; Jayanthi et al., 1999; Alleva et al., 2001; Aronis et al., 2003). Mitochondria are a major source of ROS (Droge, 2002). Mitochondria are also the site where many ROS-metabolizing enzymes are situated, including glutathione peroxidase and manganese superoxide dismutase (MnSOD). Inhibition of ROS by overexpression of MnSOD or glutathione peroxidase, or by antioxidants, prevents cell death triggered by TNF-, cytotoxic drugs, UV, and -irradiation (Manna et al., 1998; Simizu et al., 1998; Nomura et al., 1999).

UV, cycloheximide, c-Jun N-terminal kinase, and cytotoxic drugs are also reported to trigger FADD-dependent, but FasL-independent, apoptosis in cancer cells (Kamitani et al., 1997; Rehemtulla et al., 1997; Aragane et al., 1998; Bennett et al., 1998; Micheau et al., 1999; Tang et al., 1999; Chen and Lai, 2001; Han et al., 2001). The mechanism of how FADD–capase-8 cascade is activated under those stress stimuli is not completely clear. An interestingly possibility is the clustering of Fas receptor and its interaction with FADD induced by these stress stimuli, as suggested by several studies (Rehemtulla et al., 1997; Tang et al., 1999; Micheau et al., 1999). In the present study, we demonstrated that -irradiation, similar to cisplatin, induced apoptosis in Jurkat cells by the FADD-dependent but Fas ligand (FasL)-independent pathway. We provided additional evidences that DNA-damaging reagents trigger cell death by promoting Fas receptor aggregation. In contrast to the classical death receptor pathway, -irradiation- and cisplatin-triggered FADD-mediated death was ROS-dependent. We have further illustrated that inhibition of ROS blocks clustering of the Fas receptor. Our results suggest a novel coupling mechanism with the Fas receptor pathway activated by ROS in leukemia cells with high death receptor expression.

Results

DNA damage-induced apoptosis was caspase-8- and FADD-dependent in Jurkat cells

DNA-damaging reagents are known to trigger cell death mainly through mitochondrial pathway. We used Z-IETD, c-FLIP_L, and DN-FADD, to assess whether FADD–caspase-8 is involved in the apoptosis of Jurkat cells induced by -irradiation and cisplatin. Z-IETD is a caspase-8-specific inhibitor, the expression of c-FLIP_L prevents the activation of caspase-8 at death receptor complex (Krueger et al., 2001; Thome and Tschopp, 2001), while DN-FADD expression competes with wild-type FADD (Boldin et al., 1995; Chinnaiyan et al., 1995). -Irradiation-mediated apoptosis, as well as anti-Fas antibody CH11-triggered death, was abrogated in the presence of Z-IETD (25 M) in Jurkat cells (Figure 1a and b). Apoptosis triggered by cisplatin was also sensitive to inhibition by Z-IETD (Figure 1c). The requirement of caspase-8 in DNA damage-induced apoptosis was further examined in Jurkat cells transducted with c-FLIP_L. Apoptosis induced by -irradiation and cisplatin was inhibited in Jurkat cells expressing c-FLIP_L but not in the YFP control (Figure 1d). Transient transfection with DN-FADD also effectively inhibited subsequent cisplatin- and -irradiation induced apoptosis in Jurkat cells (Figure 1e, not shown for -irradiation). The inhibition of cisplatin/-irradiation-induced apoptosis by DN-FADD, c-FLIP_L, and Z-IETD suggests that the FADD-mediated death receptor apoptotic pathway plays a major role in DNA damage-triggered cell death.

Figure 1.

-irradiation- and cisplatin-induced cell death was inhibited by caspase-8 inhibitor IETD, c-FLIP_L, and DN-FADD in Jurkat cells. (a) and (b) Jurkat cells were -irradiated (IR) or treated with CH11 (50 ng/ml) in the absence or presence of Z-IETD (25 M) for 19 h. Cells were then stained with PI and were analysed by FACScan (Becton Dickinson). Cells with sub-G₁ DNA content were assessed using the CELLFIT program (Becton Dickinson). (c) Jurkat cells were treated with cisplatin (CDDP) at different concentrations or CH11, with or without Z-IETD, and cell death quantitated 22 h later. (d) Jurkat cells transduced with YFP or c-FLIP_L were -irradiated or stimulated with cisplatin, and the extent of apoptosis quantitated 20 h later. (e) Jurkat cells were transfected with pEGFP (2 g) with or without pcDNA3-DN-FADD by electroporation. After 36 h, cells were treated with cisplatin for another 8 h. GFP-positive cells were gated and the fraction of cells stained with PE–Annexin V was determined

Full figure and legend (132K)

To avoid unknown cross-inhibition of other caspases by Z-IETD, the specificity of caspase-8 in cisplatin-mediated cell death was further examined using the JB6 cell line, the mutant Jurkat cell line deficient in caspase-8 (Kawahara et al., 1998). In contrast to wild-type Jurkat, -irradiation- and cisplatin- triggered apoptosis was nearly abrogated in JB6 cells (Figure 2a). Therefore, caspase-8 is pivotal in the Jurkat cell death induced by DNA damage.

Figure 2.

Requirement of caspase-8 and Fas in -irradiation- and cisplatin-triggered apoptosis. (a) Jurkat cells and caspase-8-deficient JB6 cells (Kawahara et al., 1998) were treated with -irradiation, cisplatin, or CH11 (25 ng/ml) at the indicated doses. Apoptotic cells were determined 19 h later using PI staining and FACS analysis. (b) L1210 and L1210-Fas+ cells were treated with the indicated amounts of cisplatin or FasL (5 ng/ml plus 5 g/ml of anti-His antibody, R&D), and cell deaths were quantitated 24 h later

Full figure and legend (43K)

We also employed mouse leukemia cells L1210 and L1210 cell transfected with Fas (L1210-Fas+) to examine the role of Fas in cisplatin-induced death (Figure 2b). L1210 cells were relatively resistant to cisplatin-triggered apoptosis. The expression of Fas on L1210 enhanced cisplatin-mediated cytotoxicity (Figure 2b) as well as -irradiation-triggered death (data not shown). Together, these results suggest that a majority of cisplatin- and -irradiation-induced death in Fas-expressing cells are mediated by the Fas–FADD apoptotic pathway.

DNA damage-induced cell death was independent of Fas–FasL interaction

We examined the total Fas expression after DNA damage. The expression of Fas protein and mRNA was already high in untreated Jurkat cells, and was not significantly increased by -irradiation at the doses of used in the present study (Figure 3a and b). DNA damage may increase surface Fas expression by promoting Fas transport from the Golgi (Bennett et al., 1998). We used flow cytometry to determine surface Fas expression after treatment of Jurkat cells with cisplatin and -irradiation. Surface Fas expression was constitutively high in Jurkat cells, but could be further increased after stimulation with TPA/A23187 (Figure 3c). In contrast, treatment of Jurkat cells with cisplatin (600 M) or -irradiation (30 Gy) did not significantly increase surface Fas levels. The involvement of Fas in apoptosis induced by cisplatin and -irradiation was apparently not due to increased export of Fas to the cell surface.

Figure 3.

Fas and FasL expression after DNA damage. Jurkat cells were treated with -irradiation, different concentrations of cisplatin, or TPA/A23187. (a) Fas, Bid, and tBid protein levels after -irradiation. Total cell lysates were isolated at the indicated time points after irradiation (30 Gy), and the contents of Fas, Bid, and tBid were assessed by immunoblots. (b) Fas mRNA expression after -irradiation. Total RNA was isolated at the indicated times after ionization irradiation and Fas mRNA was determined by RT–PCR. (c) Surface expression of Fas was determined by FACS using PE-conjugated anti-Fas antibody DX2 (eBioscience). unstained background, dotted curve; untreated Jurkat cells, shaded curve; Jurkat cells treated with cisplatin (600 M) for 4.5 h, bold curve; cells stimulated with T/A for 18 h, light curve. (d) Surface expression of FasL was quantitated using PE-labeled anti-FasL. untreated Jurkat cells, shaded curve; Jurkat cells treated with cisplatin (300 M) for 4.5 h, bold curve; Jurkat cells treated with cisplatin (1 mM) for 4.5 h, light curve

Full figure and legend (125K)

We next explored whether the activation of the Fas death pathway in -irradiation- and cisplatin-triggered apoptosis was mediated by a FasL–Fas interaction by assessing the induction of FasL expression after DNA damage. High concentration of cisplatin (1 mM) promoted surface FasL expression in Jurkat cells. However, we failed to detect any increase of surface FasL at lower concentrations of cisplatin (300 M) and doses of -irradiation that triggered extensive cell death (Figure 3d, not shown for -irradiation).

We also used a soluble Fas–Fc fusion protein and the antagonistic anti-Fas antibody ZB4 to block the binding of FasL to Fas in order to examine the possible role of FasL in DNA damage-induced death. Preincubation of Jurkat cells with ZB4 (250 ng/ml) effectively suppressed CH11-induced apoptosis (Figure 4a), indicating a disruption in the Fas–FasL interaction. In contrast, ZB4 was ineffective in preventing cisplatin- and -irradiation-induced death. Similarly, Fas–Fc (200 ng/ml) blocked FasL-induced cell death, but had little effect on cisplatin- and -irradiation-induced apoptosis in Jurkat cells (Figure 4b). The poor induction of FasL by cisplatin and -irradiation and the failure of Fas–Fc and ZB4 to protect Jurkat cells from apoptosis support the notion that cisplatin- and -irradiation-triggered apoptosis is not mediated through a FasL–Fas interaction.

Figure 4.

DNA damage-induced cell death was not mediated by Fas–FasL interaction. Jurkat cells were stimulated with CH11 antibody (50 ng/ml), sFasL (25 ng/ml), indicated doses of cisplatin, or -irradiation in the absence or presence of ZB4 (250 ng/ml) (a) or Fas–Fc (200 ng/ml) (b). Cell death was determined 22 h later by quantitation of hypodiploid cells

Full figure and legend (51K)

DNA damage-induced Fas aggregation on Jurkat cells

Previous reports had indicated that cytotoxic drugs induce cell death by triggering Fas clustering in a FasL-independent manner (Aragane et al., 1998; Bennett et al., 1998; Micheau et al., 1999; Tang et al., 1999). We examined the surface Fas distribution before and after -irradiation and cisplatin treatment by confocal laser scanning microscope. Fas were mostly evenly distributed on the surface of the resting Jurkat cells (Figure 5a). Treatment of Jurkat cells with anti-Fas antibody CH11 led to increased aggregation of Fas on cell surface (Figure 5b). The clustered Fas were colocalized with FADD. For Jurkat cells treated with cisplatin, there was a similar increased clustering of Fas, colocalized with FADD, on the cell surface (Figure 5c). -irradiation-triggered Fas aggregation was indistinguishable from that induced by cisplatin and CH11.

Figure 5.

DNA damage-induced aggregation of Fas. Jurkat cells were treated with CH11 (50 ng/ml) or cisplatin (300 M) for 2 h. The cells were attached to slides and stained by anti-Fas antibody (C-20, Santa Cruz) and anti-FADD antibody (clone 1, PharMingen). The cells were then analysed by a Zeiss confocal laser scanning microscope LSM 510 with a 63 objective lens. (a), Untreated Jurkat cells; (b), Jurkat cells treated with CH11; (c), Jurkat cells treated with cisplatin. Bar indicates 10 m

Full figure and legend (97K)

A number of structural and signaling proteins are cleaved by caspases during the apoptotic processes (Kothakota et al., 1997; Widmann et al., 1998; Stegh et al., 2000; Lavastre et al., 2002), which could lead to reorganization of membrane proteins. Fas aggregation as a consequence of caspase activation was now investigated. Fas aggregation was examined in caspase-8-deficient Jurkat cell lines JB6, because of the pivotal role of caspase-8 in cisplatin-triggered apoptosis (Figure 1). Figure 6 illustrates that cisplatin triggered clustering of Fas receptors both in parent Jurkat cells and in JB6 cells. A similar Fas clustering was found in Jurkat and JB6 cells after -irradiation (data not shown). Fas aggregation was therefore not dependent on the activation of caspase-8.

Figure 6.

DNA damage-triggered Fas clustering does not require caspase-8. (a) Jurkat cells and (b) capase-8-deficient JB6 cells were treated with cisplatin (300 M) or CH11 (50 ng/ml) for 4 h, and Fas expression examined using confocal microscopy as described in Figure 5. Bar indicates 10 m

Full figure and legend (52K)

ROS are required for cisplatin-induced apoptosis

The Fas–FADD pathway and ROS production are two apparently nonoverlapping pathways. Ionization irradiation and cisplatin treatment of Jurkat cells led to the generation of hydrogen peroxide and superoxide anion, as measured by the oxidation of fluorescence dyes DCFDA and DHE, respectively (Figure 7a, not shown for cisplatin). We next investigated whether apoptosis induced by cisplatin/-irradiation, mediated by FADD-caspase-8, was also dependent on ROS. We used three different free radicals scavengers to assess the role of ROS in cisplatin-induced apoptosis. C₆₀(D3) is an effective scavenger of hydroxyl and superoxide radicals (Dugan et al., 1997; Hsu et al., 1998), MnTBAP is a superoxide dismutase mimic, and N-acetylcysteine (NAC) detoxifies free radicals to prevent DNA damage (Malins et al., 2002). C₆₀(D3) effectively suppressed apoptosis induced by -irradiation and cisplatin (Figure 7b). MnTBAP also inhibited -irradiation- and cisplatin-triggered apoptosis (Figure 7c). Treatment with NAC effectively prevented cisplatin-induced death of Jurkat cells (Figure 7d). Cell death activated by -irradiation was also attenuated by NAC (data not shown). In contrast, NAC, MnTBAP, and C₆₀(D3) had no effect on anti-Fas antibody CH11-triggered apoptosis (Figure 7b–d). Despite that FADD-caspase-8 is involved in both DNA damage- and CH11-induced death (Figure 1), ROS is required for DNA damage-, but not CH11-, triggered apoptosis of Jurkat cells.

Figure 7.

DNA damage induces ROS production and removal of ROS prevents apoptosis. (a) DNA damage-stimulated ROS generation. Jurkat cells were -irradiated and the production of H₂O₂ and O₂^- was measured by the oxidation of DCFDA and DHE, respectively, at the indicated time points after irradiation. (b–d) ROS scavengers inhibit -irradiation- and cisplatin-induced apoptosis in Jurkat cells. Jurkat cells were pretreated with or without (b) C₆₀(D3) (100 M), (c) MnTBAP (300 M), or (d) NAC (5 mM) for 1.5 h. Cells were then -irradiated, or treated with cisplatin or CH11 (100 ng/ml) and the extent of apoptosis determined 22 h later. (e) Jurkat cells transduced with YFP or MnSOD-myc were -irradiated or stimulated with cisplatin, and the extent of apoptosis quantitated 20 h later. Expression of MnSOD was detected by anti-myc antibody

Full figure and legend (126K)

Mitochondria are the major ROS production sites. MnSOD, the mitochondrial form of superoxide dismutase, was overexpressed in Jurkat cells by retroviral transduction (Figure 7e). MnSOD expression conferred resistance in Jurkat cells to apoptosis induced by -irradiation and cisplatin (Figure 7e), suggesting that mitochondrial ROS production is involved in DNA damage-triggered apoptosis. Fas-mediated apoptotic pathway is also amplified in mitochondria by the cleavage of Bid into tBid in Jurkat cells (Li et al., 1998; Luo et al., 1998). We monitored the processing of Bid following -irradiation. The cleavage of Bid and appearance of tBid were evident 6 h after ionization irradiation (Figure 3a). In contrast, the clustering of Fas was visible on the cell surface of Jurkat 2 h after -irradiation (Figure 5). Therefore, generation of ROS in mitochondria, but not the processing of Bid, is required for DNA damage-induced Fas aggregation.

ROS scavenger prevented the clustering of Fas receptor

ROS scavengers inhibited DNA damage-induced death but not Fas-initiated apoptosis (Figure 7), therefore, DNA damage-induced apoptotic process that involves ROS must be prior to the initiation of Fas apoptotic signals. We tested whether exogenous ROS triggered FAS clustering in Jurkat cells. Treatment with tert-butyl hydrogen peroxide (TBHP) for 1.5 h led to clear Fas aggregation on the surface of Jurkat cells (Figure 8a) and subsequent apoptosis (data not shown). In contrast, TBHP induced Fas clustering, but not cell death, in caspase-8-deficient JB6 cells (data not shown). We also examined the effect of antioxidants on -irradiation and cisplatin-triggered Fas aggregation. Treatment of C₆₀(D3), MnTBAP, or NAC on Jurkat cells effectively abrogated Fas aggregation induced by -irradiation (Figure 8b, data not shown for MnTBAP and NAC). Similarly, cisplatin-triggered Fas clustering was prevented by C₆₀(D3) (Figure 8b). Our results reveal that ROS participate in -irradiation and in cisplatin-induced Fas clustering, which constitutes a major difference between DNA damage- and Fas-initiated apoptosis.

Figure 8.

ROS is critical for Fas clustering. (a) Exogenous ROS induced Fas aggregation. Jurkat cells were treated with TBHP (200 M) and the expression of Fas was analysed using confocal microscopy 1.5 h later. Bar indicates 10 m. (b, c) ROS scavengers prevented Fas aggregation induced by -irradiation and cisplatin. Jurkat cells were pretreated with or without C₆₀(D3) (100 M) for 1.5 h, and were then subjected to -irradiation (b) or cisplatin treatment (c)

Full figure and legend (171K)

Discussion

Death receptor and mitochondria initiate two different apoptotic processes. In different reports, cytotoxic reagents either trigger cell death mediated by mitochondria, or induce apoptosis in Fas-dependent manner. The exact mechanisms on how two distinct apoptotic cascades are simultaneously activated by cytotoxic drugs are not understood. For the Fas-mediated apoptotic pathway, cytotoxic drugs are reported to promote FasL expression and FasL–Fas interaction (Hug et al., 1997, Kasibhatla et al., 1998). In the present study, we used -irradiation and cisplatin to trigger DNA damage. We illustrated that high doses of cisplatin induced FasL surface expression, but -irradiation and lower concentrations of cisplatin, which still triggered death (Figure 1), were unable to increase FasL on Jurkat cells (Figure 3). The capacity of cisplatin and -irradiation to promote FasL expression, therefore, is dependent on the type of cancer cells and on the dosage of the drugs. The observation that cisplatin- and -irradiation-triggered Jurkat cell death was not prevented by Fas–Fc and Fas-antagonist antibody ZB4 (Figure 4), further exclude a role of FasL–Fas engagement in cisplatin- and -irradiation-induced apoptosis. Our results are consistent with reports that apoptosis triggered by genotoxic drugs cannot be prevented by blocking the FasL–Fas interaction (Eischen et al., 1997; Micheau et al., 1999; Landowski et al., 1999; Ferreira et al., 2000), suggesting that FasL–Fas engagement is not necessarily involved in apoptosis induction operated in cancer cells treated with cytotoxic drugs.

Despite of being FasL-independent, -irradiation- and cisplatin-induced apoptosis in Jurkat and L1210-Fas+ cells was mostly mediated by the FADD-caspase-8 apoptotic cascade (Figure 1). In mutant Jurkat cells lacking caspase-8, JB6, the ability of -irradiation and cisplatin to trigger apoptosis was largely diminished (Figure 2a). In addition, the cytotoxicity of cisplatin and -irradiation was largely compromised in the absence of Fas (Figure 2b). Fas expression increased the sensitivity of L1210 cells to cisplatin, suggesting that lower extents of DNA damage, which are unable to trigger the mitochondrial death pathway, activate Fas–FADD pathway. FADD-dependent, but FasL-independent, cell death has been reported in apoptosis induced by UV, cycloheximide, c-Jun N-terminal kinase, and cytotoxic drugs (Kamitani et al., 1997; Rehemtulla et al., 1997; Aragane et al., 1998; Bennett et al., 1998; Micheau et al., 1999; Tang et al., 1999; Chen and Lai, 2001; Han et al., 2001). Through the use of same cell line with selective mutation, we further confirmed the critical role of caspase-8 and Fas in DNA damage-induced apoptosis. Previous studies with UV, cisplatin, etoposide, and vinblastine illustrate a cell death linked to clustering of Fas receptor and its recruitment of FADD (Rehemtulla et al., 1997; Tang et al., 1999; Micheau et al., 1999). We demonstrated that direct DNA damage by -irradiation also promoted Fas aggregation. In addition, we provided further support that Fas receptor aggregation could be a common mechanism linking genotoxic cell damage to apoptotic death receptor pathway. Our results suggest that, to effectively trigger Fas clustering, a likely parameter that discriminates Fas death pathway against mitochondrial death pathway in cancer cells is the expression level of Fas. Presumably, in cancer cells with high cell surface Fas expression, the use of the Fas–FADD pathway may become the default apoptotic pathway responsive to DNA damage, due to the susceptibility to Fas aggregation. Further studies are required to define the correlation between Fas expression and Fas clustering triggered by DNA damage.

In the present study, inhibition of ROS production had no effect on Fas-initiated cell death (Figure 7). ROS are produced mainly in mitochondria, yet their involvement in Fas-mediated apoptosis remains unsettled. Previous studies have illustrated conflicting roles of ROS in Fas-mediated cell death, ranging from a complete dissociation from Fas-induced apoptosis (Hug et al., 1994; Schulze-Osthoff et al., 1994; Alleva et al., 2001), to a key mediator of Fas-initiated cell death (Gulbins et al., 1996; Um et al., 1996; Jayanthi et al., 1999), or to an inhibitor of Fas-triggered apoptosis (Aronis et al., 2003). The discrepancy between these studies may reflect the complicated nature of ROS and the dependence of Fas-mediated death processes on the intracellular environment. It is likely that ROS, produced in larger quantities in macrophages as a stress response to Fas engagement, play a more dominant role in monocytes than in T lymphocytes (Um et al., 1996). Conceivably, it will be difficult to separate Fas-induced and DNA-damaging reagents-triggered death processes for ROS involvement in monocytes and neuroglioma cells (Um et al., 1996; Jayanthi et al., 1999). Our observation that NAC, MnTBAP, and C₆₀(D3) did not interfere with CH11-triggered apoptosis (Figure 7) may seem contradict with the report that decreased cellular ROS production by oligomycin or FCCP enhances Fas-induced apoptosis (Aronis et al., 2003). We speculate that a reason for such discrepancy is the concentration-dependent physiological effect for ROS. It is likely that a minimum amount of endogenous ROS is essential for cell viability, and a complete disruption of mitochondrial redox status by oligomycin and FCCP leads to increased cell death. In contrast, the antioxidants used in the present study, effective in reducing stress stimuli-triggered ROS burst (Alleva et al., 2001), are likely not as effective as oligomycin and FCCP in the abrogation of basal ROS production. Together, the observations that CH11-induced death is not affected by conventional antioxidants (Hug et al., 1994; Schulze-Osthoff et al., 1994; Alleva et al., 2001) and is augmented by oligomycin and FCCP (Aronis et al., 2003), suggest that intracellular basal ROS production, but not Fas-triggered ROS increase, is participated in the apoptotic processes.

The failure of NAC, MnTBAP, and C₆₀(D3) to interfere with CH11-triggered apoptosis (Figure 7), in contrast to the effective prevention of DNA damage-induced apoptosis by the same ROS scavengers, enables us to place the ROS-sensitive stage upstream of a Fas-mediated apoptotic pathway. ROS have been implicated in FasL expression (Bauer et al., 1998), but FasL was proved to be dispensable for cisplatin-induced death in the present study (Figure 4). We further identified that ROS participate at the stage of cisplatin-triggered Fas receptor aggregation (Figure 8). Together, the process from DNA damage to apoptosis is summarized by a proposed schematic diagram (Figure 9). In such a model, DNA damage triggers generation of ROS, leading to increased Fas clustering in cancer cell with high surface Fas expression, ending with FADD recruitment and caspase-8 activation. The mechanism by which ROS influences Fas clustering is unknown at present and requires investigation. Presumably, ROS activate signals that eventually result in cytoskeleton reorganization and membrane protein clustering (Wang et al., 2001; van Wetering et al., 2002). We are in the progress of determining the exact cellular processes involved in cisplatin-mediated apoptosis, from ROS generation to Fas aggregation.

Figure 9.

Hypothetical model on how DNA damage leads to Fas-dependent apoptosis on Jurkat cells. DNA damage, mediated by -irradiation and cisplatin, triggers ROS production, which was inhibited by C₆₀, MnTBAP, and NAC. In cancer cells with high level of Fas expression, ROS may trigger clustering of Fas in a mechanism yet to be characterized. Fas aggregation recruits FADD and activates caspase-8, leading to irreversible death. The dominance of Fas-initiated apoptotic pathway in DNA damage-induced apoptosis is supported by the sensitivity to inhibition by DN-FADD, c-FLIP, and Z-IETD

Full figure and legend (62K)

Even though death receptors and mitochondria represent two distinct apoptotic initiators, the two death pathways are known to communicate with each other in the different stages of apoptotic process. Death receptor apoptotic signals are amplified by their coupling to mitochondrial pathway through Bid protein cleavage (Li et al., 1998; Luo et al., 1998). The apoptotic processes initiated from another death receptor, TRAIL-receptor, involves the mitochondrial release of Smac/DIABLO (Deng et al., 2002) and neutralization of the caspase inhibitor IAP, representing a different cross-talk between death receptor and mitochondria. Results from the present study reveal a new coupling mechanism from mitochondria, the major site for ROS production, to death receptors, in cells with high Fas expression. Therefore, DNA-damaging reagents trigger production of ROS, which in turn induce the aggregation of Fas receptor and activation of FADD-caspase 8 cascade in Fas-expressing cells. Increased sensitivity to DNA-damaging reagents as apoptosis was triggered by Fas-dependent pathway (Figure 2), in contrast to mitochondrial cascade, further supports the critical role of Fas in the removal of leukemia cell.

Materials and methods

Reagents

Jurkat cell line (H6. 2 clone) was a gift of Dr Daniel Olive (INSERM U119, Marseille, France). JB6 cells (Kawahara et al., 1998), the caspase-8-deficient Jurkat cells, and the parent Jurkat cells, were gifts of Dr Shigekazu Nagata (Osaka University Medical School, Osaka, Japan). Anti-Fas (C-20) antibody was purchased from Santa Cruz Biotech (Santa Cruz, CA, USA). PE-conjugated anti-Fas antibody (DX2) and anti-FasL antibody (NOK-1) were purchased from eBioscience (San Diego, CA, USA). sFasL and anti-Fas antibodies, CH11 and ZB4, were purchased from Upstate Biotechnology (Lake Placid, NY, USA). Caspase-8 inhibitors Z-IETD-FMK and MnTBAP (Mn(III)tetrakis(4-benzoic acid)porphyrin chloride) were obtained from Calbiochem (La Jolla, CA, USA). FITC-conjugated anti-Fas antibody, human Fas-Fc, His-tagged FasL, and anti-His antibody were purchased from R&D (Minneapolis, MN, USA). The regioisomer with D3 symmetry of water-soluble carboxylic acid C₆₀ derivatives (carboxylfullerenes) was synthesized as previously described (Dugan et al., 1997). Dichlorodihydrofluorescein diacetate (DCFCA) and dihydroethidium (DHE) were obtained from Molecular Probes (Eugene, OR, USA). pcDNA3-DN-FADD was gift of Dr Vishva Dixit (Geneteh, South San Francisco, USA). MnSOD cDNA was isolated from murine EL4 cells by RT–PCR using 5' primer AGG GTA ccg gtc gtg taa acc tca and 3' primer AGT CTA GAG CAC CCC AGT CAT AGTG, and was subcloned into pcDNA4 for myc tagging.

Retroviral infection and transient transfection

cDNA of human c-FLIP_L has been previously described (Yeh et al., 1998). pGCIRES-YFP, similar to the reported pGCIRES-GFP (Costa et al., 2000), was a gift from Dr Gina Costa (Stanford University, Stanford, CA, USA) and was obtained through Dr Nan-Shih Liao (Academia Sinica, Taipei, Taiwan). Myc-tagged c-FLIPL and Myc-tagged MnSOD were subcloned into pGCIRES-YFP. Retroviruses were produced by transfecting Phoenix-Eco cells (from Dr Garry P Nolan, Stanford University) with 10 g of pGC-IRES-YFP, pGC-c-FLIPL-IRES-YFP, or pGC-MnSOD-IRES-YFP plasmids. Phoenix cell supernatants containing retrovirus were collected 48 h after transfection. Virus stocks with titers greater than 1 10⁶ were used for spin infection on Eco-Jurkat cells (also gift of Dr Garry Nolan). At 24 h after infection, YFP-expressing Jurkat cells were isolated by sorting on FACStar Plus (Becton Dickinson, Mountain View, CA, USA).

For transient transfection, Jurkat T cells (1 10⁷) were washed, resuspended in 0.6 ml of RPMI medium containing 1% glucose, 10% fetal calf serum, and 18 g plasmid DNA. The electroporation was performed in a Bio-Rad Gene Pulser II (Hercules, CA, USA) at 260 mV and 975 F. The cells were left on ice for 15 min, washed twice with PBS, and incubated for 36 h before treatment with cisplatin.

Cell death measurement

Apoptosis was determined by propidium iodide (PI) staining. At the indicated times after treatment, cells were washed and resuspended in hypotonic fluorochrome solution (20 g/ml PI, 0.1% sodium citrate, 0.1% Triton X-100) (Nicoletti et al., 1991). Cells were placed in the dark overnight at 4°C, and the DNA content was analysed by FACScan (Becton Dickinson, Mountain View, CA, USA). Fractions of cell population with sub-G₁ DNA content were determined using the CELLFIT program (Becton Dickinson). In order to study apoptosis in cells transiently transfected with DN-FADD, EGFP (Clonetech) was cotransfected. After cells were harvested, the green cells (GFP-positive) were gated on FACScan, and the fractions of cell stained with PE-conjugated Annexin V quantitated.

Confocal microscopy

After stimulation, Jurkat cells were adhered to slides precoated with poly-L-lysine (Sigma) and fixed with 3.7%. paraformaldehyde for 30 min, followed by methanol permeabilization for 5 min. The cells were stained with primary antibodies including FITC-conjugated anti-Fas (DX2, R&D), anti-Fas (C-20, Santa Cruz), or anti-FADD antibody (clone 1, PharMingen). For unlabeled primary antibodies, secondary antibody of either FITC-conjugated anti-mouse Ig (Sigma) or rhodamine red X-conjugated goat anti-rabbit Ig (Jackson, West Grove, PA, USA) was used. The Fas expression was analysed by Zeiss confocal laser scanning microscope LSM 510 with a 63 objective lens. The use of different anti-Fas antibodies was not observed to affect Fas aggregation.

Determination of ROS production

Jurkat cells were preloaded with either DHE (2 M) or DCFDA (2 M) for 30 min. Jurkat cells were then equally divided, untreated, or -irradiated immediately or after 60 and 90 min. The reactions were terminated at 120 min, and the increase of DHE or DCFDA fluorescence determined on FACScan to give ROS contents 0, 30, 60, and 120 min after DNA damage.

References

1. Alleva R, Tomasetti M, Andera L, Gellert N, Borghi B, Weber C, Michael P, Murphy MP and Neuzil J. (2001). FEBS Lett., 503, 46–50. | Article | PubMed | ISI | ChemPort |
2. Aragane Y, Kulms D, Metze D, Wilkes G, Poppelmann B, Luger TA and Schwarz T. (1998). J. Cell Biol., 140, 171–182. | Article | PubMed | ISI | ChemPort |
3. Aronis A, Andr's Melendez J, Golan O, Shilo S, Dicter N and Tirosh O. (2003). Cell Death Diff., 10, 335–344.
4. Bauer MKA, Vogt M, Los M, Siegel J, Wesselborg S and Schulze-Osthoff K. (1998). J. Biol. Chem., 273, 8048–8055. | Article | PubMed | ISI | ChemPort |
5. Bennett M, McDonald K, Chan SW, Luzio JP, Simari R and Weissberg P. (1998). Science, 282, 290–293. | Article | PubMed | ISI | ChemPort |
6. Boldin MP, Varfolomeev EE, Pancer Z, Mett IL, Camonis JH and Wallach D. (1995). J. Biol. Chem., 270, 7795–7798. | Article | PubMed | ISI | ChemPort |
7. Chen Y and Lai MZ. (2001). J. Biol. Chem., 276, 8350–8357. | PubMed |
8. Chinnaiyan AM, O'Roueke K, Tewari M and Dixit VM. (1995). Cell, 81, 505–512. | Article | PubMed | ISI | ChemPort |
9. Costa GL, Benson JM, Seroogy CM, Achacoso P, Fathman CG and Nolan GP. (2000). J. Immunol., 164, 3581–3590. | PubMed | ISI | ChemPort |
10. Deng Y, Lin Y and Wu X. (2002). Gene Dev., 16, 33–45. | Article | PubMed | ISI | ChemPort |
11. Droge W. (2002). Physiol. Rev., 82, 47–95. | PubMed | ISI | ChemPort |
12. Dugan LL, Tureysky DM, Du C, Lobner D, Wheeler M, Almli CR, Shen CKF, Luh TY, Choi DW and Lin TS. (1997). Proc. Natl. Acad. Sci. USA, 94, 9434–9439. | Article | PubMed | ChemPort |
13. Eischen CM, Kottke TJ, Martins LM, Basi GS, Tung JS, Earnshaw WC, Leibson PJ and Kaufmann SH. (1997). Blood, 90, 935–943. | PubMed | ISI | ChemPort |
14. Ferreira CG, Tolis C, Span SW, Peters GJ, van Lopik T, Kummer AJ, Pinedo HM and Giaccone G. (2000). Clin. Cancer Res., 6, 203–212. | PubMed |
15. Gulbins E, Brenner B, Schlottmann K, Welsch J, Heinle H, Koppenhoefer U, Linderkamp O, Coggeshall KM and Lang F. (1996). Immunology, 89, 205–212. | Article | PubMed | ISI | ChemPort |
16. Han Z, Pantazis P, Wyche JH, Kouttab N, Kidd VJ and Hendrickson EA. (2001). J. Biol. Chem., 276, 38748–38754. | Article | PubMed | ISI | ChemPort |
17. Hengartner MO. (2000). Nature, 407, 770–776. | Article | PubMed | ISI | ChemPort |
18. Hsu SC, Wu CC, Luh TY, Chou CK, Han SH and Lai MZ. (1998). Blood, 91, 2658–2663. | PubMed | ISI | ChemPort |
19. Hug H, Enari M and Nagata S. (1994). FEBS Lett., 351, 311–313. | Article | PubMed | ChemPort |
20. Hug H, Strand S, Grambihler A, Galle J, Hack V, Stremmel W, Krammer PH and Galle PR. (1997). J. Biol. Chem., 272, 28191–28193. | Article | PubMed | ISI | ChemPort |
21. Jayanthi S, Ordonez S, McCoy MT and Cadet JL. (1999). Brain Res. Mol. Brain Res., 72, 158–165. | Article | PubMed | ChemPort |
22. Kamitani T, Nguyen HP and Yeh ETH. (1997). J. Biol. Chem., 272, 22307–22314. | Article | PubMed | ChemPort |
23. Kasibhatla S, Brunner T, Genestier L, Echeverri F, Mahboubi A and Green DR. (1998). Mol. Cell, 1, 543–551. | Article | PubMed | ISI | ChemPort |
24. Kawahara A, Ohsawa Y, Matsumura H, Uchiyama Y and Nagata S. (1998). J. Cell Biol., 143, 1353–1360. | Article | PubMed | ISI | ChemPort |
25. Kothakota S, Azuma T, Reinhard C, Klippel A, Tang J, Chu K, McGarry TJ, Kirschner MW, Koths K, Kwiatkowski DJ and Williams LT. (1997). Science, 278, 294–298. | Article | PubMed | ISI | ChemPort |
26. Kroemer G and Reed JC. (2000). Nat. Med., 6, 513–519. | Article | PubMed | ISI | ChemPort |
27. Krueger A, Baumann S, Krammer PH and Kirchhoff S. (2001). Mol. Cell. Biol., 21, 8247–8254. | Article | PubMed | ISI | ChemPort |
28. Landowski TH, Shain KH, Oshiro MM, Buyuksal I, Painter JS and Dalton WS. (1999). Blood, 94, 265–274. | PubMed | ISI | ChemPort |
29. Lavastre V, Pelletier M, Saller R, Hostanska K and Girard D. (2002). J. Immunol., 168, 1419–1427. | PubMed | ISI | ChemPort |
30. Li H, Zhu H, Xu CJ and Yuan J. (1998). Cell, 94, 491–501. | Article | PubMed | ISI | ChemPort |
31. Luo X, Budihardjo I, Zou H, Slaughter C and Wang X. (1998). Cell, 94, 481–490. | Article | PubMed | ISI | ChemPort |
32. Malins DC, Hellstrom KE, Anderson KM, Johnson PM and Vinson MA. (2002). Proc. Natl. Acad. Sci. USA, 99, 5937–5941. | PubMed |
33. Manna SK, Zhang HJ, Yan T, Oberley LW and Aggarwal BB. (1998). J. Biol. Chem., 272, 13245–13254.
34. Micheau O, Solary E, Hammann A and Dimanche-Boitrel MT. (1999). J. Biol. Chem., 274, 7987–7992. | Article | PubMed | ISI | ChemPort |
35. Nagata S. (1997). Cell, 88, 355–365. | Article | PubMed | ISI | ChemPort |
36. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F and Riccardi C. (1991). J. Immunol. Methods, 139, 271–279. | Article | PubMed | ISI | ChemPort |
37. Nomura K, Imai H, Koumurai T, Arai M and Nakagawa Y. (1999). J. Biol. Chem., 274, 29294–29302. | Article | PubMed | ISI | ChemPort |
38. Rehemtulla A, Hamilton CA, Chinnaiyan AM and Dixit VM. (1997). J. Biol. Chem., 272, 25783–25786. | Article | PubMed | ISI | ChemPort |
39. Schulze-Osthoff K, Krammer PH and Droge W. (1994). EMBO J., 13, 4587–4596. | PubMed | ChemPort |
40. Simizu S, Takada M, Umezawa K and Imoto M. (1998). J. Biol. Chem., 273, 26900–26907. | Article | PubMed | ChemPort |
41. Stegh AH, Herrmann H, Lampel S, Weisenberger D, Andra K, Seper M, Wiche G, Krammer PH and Peter ME. (2000). Mol. Cell. Biol., 15, 5665–5679. | Article |
42. Tang D, Lahti JM, Grenet J and Kidd VJ. (1999). J. Biol. Chem., 274, 7245–7252. | Article | PubMed | ISI | ChemPort |
43. Thome M and Tschopp J. (2001). Nat. Rev. Immunol., 1, 50–58. | Article | PubMed | ChemPort |
44. Um HD, Orenstein JM and Wahl SM. (1996). J. Immunol., 156, 3469–3477. | PubMed | ISI | ChemPort |
45. van Wetering S, van Buul JD, Quik S, Mul FP, Anthony EC, ten Klooster JP, Collard JG and Hordijk PL. (2002). J. Cell Sci., 115, 1837–1846. | PubMed | ISI | ChemPort |
46. Wang J, Boja ES, Tan W, Tekle E, Fales HM, English S, Mieyal JJ and Chock PB. (2001). J. Biol. Chem., 276, 47763–47766. | PubMed | ISI | ChemPort |
47. Wang X. (2001). Gene Dev., 15, 2922–2933. | PubMed | ChemPort |
48. Widmann C, Gibson S and Johnson GL. (1998). J. Biol. Chem., 273, 7141–7147. | Article | PubMed | ISI | ChemPort |
49. Yeh JH, Hsu SC, Han SH and Lai MZ. (1998). J. Exp. Med., 188, 1795–1802. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Dr Shigekazu Nagata for JB6 cells, Dr. Daniel Olive for Jurkat cells, Dr Vishva Dixit for DN-FADD, Dr Gina Costa and Dr Nan-Shih Liao for pGC-IRES-YFP vector, Dr Garry P Nolan for Phoenix-Eco and Jurkat-Eco cells, and Dr Ken Deen for editing the manuscript.

This project was supported by grant NHRI-EX92-9217BI from National Health Research Institute, a grant from Academia Sinica, Taiwan, ROC.