Apoptotic Signal of Fas Is Not Mediated by Ceramide

12-O-tetradecanoylphorbol 13-acetate (TPA), A23187, N⁶, 2′-O-dibutyryladenosine 3′,5′-cyclic monophosphate (Bt₂cAMP), forskolin, and PDTC were purchased from Sigma Chemical Co (St Louis, MO). Anti-Fas antibody CH-1126 was purchased from Upstate Biotech Inc (Lake Placid, NY). C₂-ceramide and C₆-ceramide were obtained from Biomol (Plymouth Meeting, PA). ICE inhibitor z-VAD-FK was purchased from Kamiya (Thousand Oaks, CA). The two regioisomers with C3 or D3 symmetry of water-soluble carboxylic acid C₆₀derivatives (carboxylfullerenes) were synthesized as previously described.27 Both C₆₀ (C3) and C₆₀(D3) are effective scavengers of oxgen radicals, with complete elimination of hydroxyl radicals and superoxide radicals at concentrations of 5 to 50 μmol/L, and are potent inhibitors of neuronal apoptosis, which is associated with increased intracellular free radical production.27 H-89 and H-85 were purchased from Seikagaku (Tokyo, Japan). Human T lymphoblastomas CEM (ATCC CCL 119) and human T-cell leukemia Jurkat (ATCC TIB 152) were obtained from American Type Culture Collection (Rockville, MD). Recombinant human TNF-α was purchased from R&D (Minneapolis, MN).

CMV-RelA(p65) and κB-TATA-CAT²⁸ were kind gifts of Dr Warren C. Greene (University of California, San Francisco, CA). HA-JNK129 was obtained from Michael Karin (University of California, San Diego, CA). SEK-negative mutants SEK1(K→R) and SEK1(A→L)30 were obtained from Dr Leonard I. Zon (Harvard Medical School, Boston, MA). Green fluorescence protein expression vector pEGFP-N1 was obtained from Clontech (Palo Alto, CA).

T cells (1.6 × 10⁷) were washed once with STBS (25 mmol/L Tris-HCl, pH 7.4, 137 mmol/L NaCl, 5 mmol/L KCl, 0.6 mmol/L Na₂HPO₄, 0.7 mmol/L CaCl₂, 0.5 mmol/L MgCl₂) and incubated with a total of 10 μg DNA in 1.2 mL STBS containing 0.5 mg/mL diethylaminoethyl (DEAE)-dextran for 20 minutes at room temperature. T cells were then treated with 15% dimethylsulfoxide for 3 minutes and washed once with STBS.31,32 After 24 to 48 hours, transfected cells were then stimulated with anti-Fas or ceramide, and the cell death was quantitated.

The inhibition of JNK by the dominant negative form of SEK was performed with cotransfection of HA-JNK1. T cells were activated as indicated, washed twice with phosphate-buffered saline (PBS), and lysed in ice-cold lysis buffer (20 mmol/L Tris-HCl, pH 8.0, 1% Triton X-100, 10% glycerol, 137 mmol/L NaCl, 1.5 mmol/L MgCl₂, 1 mmol/L EDTA, 50 mmol/L NaF, 1 mmol/L Na₃VO₄, and 1 mmol/L phenylmethylsulfonyl fluoride). Detergent-insoluble material was removed by centrifugation at 14,000g for 10 minutes at 4°C. For each immuno-precipitation, 200 μg of cell lysate was mixed with anti-HA 12CA5 (Boehring Mannheim, Mannheim, Germany) and incubated at 4°C for 2 hours. Twenty microliters of protein A-sepharose (Pharmacia, Uppsala, Sweden) was then added and incubated for an additional 2 hours at 4°C. Immune complexes were washed three times with lysis buffer and once with kinase buffer (30 mmol/L Tris-HCl, pH 8.0, 20 mmol/L MgCl₂, 2 mmol/L MnCl₂). Immune complexes were then incubated in kinase buffer (30 μL) containing 2 mmol/L ATP, GST-c-Jun(1-79), and 5 μCi of (γ-³²P)ATP for 30 minutes at 30°C. Incubations were terminated by adding 15 μL of 3× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer and boiling for 3 minutes. The reaction products were resolved on 15% SDS-PAGE followed by autoradiography and quantitated by Phosphorimager (Molecular Dynamics, Sunnyvale, CA).33 

All cultures (except those treated with ceramide) were performed in RPMI with 10% fetal calf serum (both from GIBCO, Grand Island, NY), 10 mmol/L glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin, and 2 × 10⁻⁵ mol/L 2-ME.34 For apoptosis induced with ceramide, serum free-medium were used throughout the experiment. The extent of apoptosis was determined by propidium iodide (PI) staining or annexin V staining. Cells were treated with different inducers and/or inhibitors, washed with PBS, and fixed with ethanol. DNA content was determined by staining with 20 μg/mL PI and analyzed by FACScan (Becton Dickinson, Mountain View, CA). The fraction of cells with sub-G₁ DNA content was assessed with the CELLFIT program (Becton Dickinson). For annexin V staining, the treated cells were washed, resuspended in annexin V-fluorescein isothiocyanate (FITC) (1 μg/mL; Clontech), incubated at room temperature for 15 minutes in the dark, and analyzed on FACScan. For the ceramide/Fas sensitivity of cell transiently transfected with CMV-p65 or pEBG-SEK(A→L), survival was monitored with green fluorescence protein expression vector pEGFP-N1. Treated cells were examined using a Nikon Diaphot 2000 fluorescence microscope (Tokyo, Japan) 48 hours after transfection.

Ceramide is a potent inducer of cell death in T-lymphoma cells Jurkat and CEM (not shown for CEM). Cell death was assessed by DNA fragmentation as represented by sub-G₁ fraction (Fig 1B). DNA fragmentation induced by ceramide never exceeded 45%, despite extensive cell death. A similar observation was also earlier reported.35 We first identified a few reagents that could block ceramide-initiated cell death. cAMP was an effective antagonist of ceramide-induced cell death (Fig 1C). Forskolin (10 μmol/L) suppressed DNA fragmentation initiated by ceramide by at least 50%. A similar extent of inhibition was observed with dibutyryl cAMP (Bt₂cAMP) at 1 mmol/L (Fig 2A). There was a dose-dependent inhibition of cAMP on ceramide-triggered cell death. The inhibition of ceramide-induced cell death could be detected with Bt₂cAMP as low as 100 μmol/L. In contrast, apoptosis triggered by anti-Fas antibody CH-11 was resistant to cAMP. Treatment with forskolin up to 50 μmol/L or Bt₂cAMP up to 2.5 mmol/L did not reduce the apoptosis triggered by Fas (Fig 2B). cAMP agonists by itself did not induce any T-cell death at the highest concentrations used here.

The observation that Fas-initiated cell death cannot be prevented by cAMP is consistent with an earlier report,36 yet is in direct contrast to the prominent inhibitory effect of cAMP recently documented.37 Because DNA fragmentation may not well represent actual apoptosis,38,39 we also quantitated cell death by annexin V staining. Figure 3 shows that Fas-induced apoptosis was associated with an extensive phosphatidylserine translocation, which was clearly not inhibited by cAMP. cAMP is an antagonist for ceramide-induced cell death but not for Fas-initiated apoptosis.

We have futher observed that two malonic acid derivatives of C₆₀ (carboxyfullerens)27 were antagonist of ceramide-induced cell death (Figs 1D and 2A). The pure carbon sphere of C₆₀ (buckminsterfullerene) is known for its avid reactivity with free radicals, yet the usage is limited by its water insolublility. The two regioisoforms with C3 or D3 symmetry of water-soluble C₆₀ derivatives remain as potent free radical scavengers and are effective antagonists of apoptotic neronal death induced by serum deprivation, glutamate receptor agonists, and amyloid peptide.27 Ceramide-triggered apoptosis was suppressed by 50% in the presence of 10 μmol/L C₆₀ (D3) and was inhibited by 90% with 100 μmol/L C₆₀ (D3). We have repeatedly observed that C₆₀ (D3) was more effective than C₆₀ (C3) in the prevention of ceramide-induced cell death (Fig 2A). In contrast, C₆₀ (D3) and C₆₀ (C3) (at 100 μmol/L each) had no preventive effect on Fas-triggered cell death as analyzed by both sub-G₁ fraction and annexin V binding (Figs 2B and 3D). The same contrast was found with metal chelator PDTC. High concentration of PDTC antagonized ceramide-induced apoptosis (Figs 1E and 2A), but did not inhibit Fas-initiated cell death (Fig 2B). Of all the inhibitors analyzed in this study, PDTC was the only reagent that by itself increased spontaneous cell death at the concentration used (1% to 3% over control, data not shown). Because such increase was minute, the differential effect of PDTC on ceramide- and Fas-initiated apoptosis was still prominent. As a control, the potent ICE proteases inhibitor z-VAD-FK completely abolished cell death induced by either ceramide or Fas (Figs 1, 2, and 3), consistent with the notion that both ceramide- and Fas-induced cell death are known to involve caspases such as ICE.7,40 

Therefore, we have identified three types of reagents that antagonized ceramide-induced cell death but did not prevent Fas-induced cell death. Between 50% and 90% of ceramide-triggered cell death was suppressed by these inhibitors. If ceramide is the major apoptotic messenger downstream of Fas, a lesser but still prominent inhibition on Fas-induced cell death by these inhibitors should be detectable. The distinct effect on Fas- and ceramide-induced apoptosis suggest that ceramide is dissociated from Fas-induced death cascade.

SEK/JNK activation is shown to be essential for ceramide-initiated apoptosis.23 Despite the fact that JNK is activated after Fas engagement, its role in apoptosis is less certain. Lenczowski et al41 demonstrated that the activation of SEK/JNK is downstream of ICE protease and is not required for Fas-induced apoptosis. On the contrary, the prevention of JNK activation by SEK dominant negative mutant is shown to block Fas-induced cell death in L929 and 293 cells.42 We have thus studied the effect of the dominant negative mutant of SEK³⁰ on the same cell when treated with ceramide and anti-Fas. A transient transfection analysis was used to assess the effect on cell death in which SEK mutant was cotransfected with green fluorescence protein. Expression of the dominant negative mutants of SEK blocked the activation of the cotransfected HA-JNK1.32 Ceramide triggered cell death was suppressed by coexpression of SEK (A→L) in Jurkat cells (Fig 4), but the inhibition was less with SEK (K→R) mutant (data not shown). This is in contrast to the inability of SEK (A→L) to interfere with Fas-induced apoptosis (Fig 4). In both Jurkat and CEM cells (not shown for CEM), the inhibition of SEK activation prevented ceramide- but not Fas-induced apoptosis. JNK activation was not essential for Fas-mediated apoptosis at least in these two T-lymphoma cells.

In accordance with the recent finding that activation of NF-κB prevents the apoptotic signal of TNF-α, we also found that NF-κB activation antagonized ceramide-induced DNA fragmentation (Fig 5). Transient expression of NF-κB p65, which led to activation of κB-CAT (not shown), prevented ceramide-induced apoptosis. On the contrary, Fas-induced cell death was not affected by the activation of NF-κB in Jurkat cells. These results further showed that ceramide is not a mediator of Fas-induced apoptosis, for otherwise Fas would be equally sensitive to the inhibition by NF-κB.

In summary, we have presented evidence that ceramide is not essential for Fas-induced cell death. Ceramide-triggered cell death was sensitive to the inhibition of cAMP, yet its blockage did not prevent Fas-initiated apoptosis. Two other antagonists, C₆₀ and PDTC, also demonstrate such selectivity in suppressing ceramide-induced apoptosis. We have further shown that blocking SEK activation diminished ceramide-induced cell death in T lymphomas Jurkat and CEM, yet had no effect on Fas-triggered apoptosis on the same cells (Fig 4). A similar distinction was also found with NF-κB activation (Fig 5). Therefore, a large fraction of Fas-triggered death pathways must be ceramide-independent. Even though ceramide has been suggested as the mediator for Fas-initiated apoptosis, the linkage between ceramide and Fas-induced cell death is less than confirmative. Recent biochemical analysis have questioned the role of ceramide in Fas-induced apoptosis. In one study, the direct measurement of sphingosine-based ceramide failed to detect induction of ceramide up to 2 hours after Fas triggering.25 In another study, ceramide increase was found to be slower than Fas-induced cell death and can be inhibited by Z-VAD-FMK.24 Our results are fully in support of these observations and suggest that ceramide is not the mediator of Fas-initiated apoptosis.

We also examined the effect of cAMP, C₆₀, and PDTC on TNF-α–induced apoptosis, which is reported to be mediated by ceramide.14,16 The study was performed on L929 cells because TNF-α alone does not induce cell death in Jurkat cells. The effect of these inhibitors on TNF-α–induced cell death was not as unequivocal as on Fas-mediated apoptosis. PDTC (5 μmol/L) induced spontaneous cell death (70%) in L929 cells, and its inhibitory effect on TNF-α–induced cell death cannot assessed. TNF-α–triggered apoptosis was completely prevented by 50 μmol/L of C₆₀(C3) and C₆₀ (D3), yet was resistant to treatment with forskolin and Bt₂cAMP (Fig 6). In contrast to the NF-κB–independent Fas-mediated apoptosis (Fig 5), TNF-α–induced cell death is also known to be suppressed by the activation of NF-κB.43-45 Some of the discrepancy may be attributed to the difference in apoptotic signaling between Fas and TNF-α previously reported.36 For example, the different sensitivity to C₆₀ (Figs 2 and 6) is well correlated with the involvement of reactive oxgen radicals in TNF-α–induced, but not in Fas-induced, cell death.36 It may also be noted that the mixed effect of the inhibitors on TNF-α–mediated apoptosis is consistent with the observation that ceramide contributes to a fraction, but not all, of TNF-α–induced cell death.35 

The distinction between the Fas-induced apoptotic cascade and ceramide-induced death pathway shows that death processes induced by different stimulation can be clearly independent of each other. The combination of different stimulations may hence be more effective in the induction of apoptosis, especially in tumor cells for therapeutic purposes, notably, ceramide-mediated cell death initiated by TNF-α, daunorubicin, and γ-irradiation, which are reagents widely used in tumor treatment.20-22 The dissociation of Fas-induced death signals from ceramide may indicate an additional dimension in manipulation of apoptosis.

The authors thank Dr Leonard Zon for SEK(A→L) and SEK(K→R), Dr Warren C. Greene for κB-TATA-CAT, and Dr Michael Karin for HA-JNK1. We also thank Douglas Platt for editorial correction of the manuscript.